Fix bug with media files, don't encode thumbnail when not necessary

Update USAGE.md

.github/ISSUE_TEMPLATE/issue-template.md

@@ -0,0 +1,18 @@
+---
+name: Issue template
+about: General
+title: ''
+labels: ''
+assignees: ''
+
+---
+
+sist2 version:
+
+Platform (Linux or Docker):
+
+Elasticsearch version:
+
+Command with arguments:  `ex: "scan ~/Documents -o ./i2 --threads 3 -q 1.0`
+
+If the issue is related to the `scan` module, please attach the files necessary to reproduce the error or email them to me[at]simon987.net.

.gitignore

@@ -11,7 +11,9 @@ Makefile
 LOG
 sist2*
 index.sist2/
-bundle.css
+bundle*.css
 bundle.js
 *.a
 vgcore.*
+build/
+third-party/

.gitmodules

@@ -1,18 +1,6 @@
-[submodule "argparse"]
-	path = argparse
+[submodule "third-party/libscan"]
+	path = third-party/libscan
+	url = https://github.com/simon987/libscan
+[submodule "third-party/argparse"]
+	path = third-party/argparse
 	url = https://github.com/cofyc/argparse
-[submodule "cJSON"]
-	path = cJSON
-	url = https://github.com/DaveGamble/cJSON
-[submodule "lib/mupdf"]
-	path = lib/mupdf
-	url = git://git.ghostscript.com/mupdf.git
-[submodule "lib/onion"]
-	path = lib/onion
-	url = https://github.com/davidmoreno/onion
-[submodule "lib/ffmpeg"]
-	path = lib/ffmpeg
-	url = https://git.ffmpeg.org/ffmpeg.git
-[submodule "lmdb"]
-	path = lmdb
-	url = https://github.com/LMDB/lmdb

.teamcity/settings.kts

@@ -0,0 +1,69 @@
+import jetbrains.buildServer.configs.kotlin.v2019_2.*
+import jetbrains.buildServer.configs.kotlin.v2019_2.buildSteps.ExecBuildStep
+import jetbrains.buildServer.configs.kotlin.v2019_2.buildSteps.exec
+import jetbrains.buildServer.configs.kotlin.v2019_2.triggers.vcs
+import jetbrains.buildServer.configs.kotlin.v2019_2.vcs.GitVcsRoot
+
+/*
+The settings script is an entry point for defining a TeamCity
+project hierarchy. The script should contain a single call to the
+project() function with a Project instance or an init function as
+an argument.
+
+VcsRoots, BuildTypes, Templates, and subprojects can be
+registered inside the project using the vcsRoot(), buildType(),
+template(), and subProject() methods respectively.
+
+To debug settings scripts in command-line, run the
+
+    mvnDebug org.jetbrains.teamcity:teamcity-configs-maven-plugin:generate
+
+command and attach your debugger to the port 8000.
+
+To debug in IntelliJ Idea, open the 'Maven Projects' tool window (View
+-> Tool Windows -> Maven Projects), find the generate task node
+(Plugins -> teamcity-configs -> teamcity-configs:generate), the
+'Debug' option is available in the context menu for the task.
+*/
+
+version = "2019.2"
+
+project {
+
+    vcsRoot(HttpsGithubComSimon987sist2refsHeadsMaster)
+
+    buildType(Build)
+}
+
+object Build : BuildType({
+    name = "Build"
+
+    artifactRules = """
+        sist2
+        sist2_scan
+    """.trimIndent()
+
+    vcs {
+        root(HttpsGithubComSimon987sist2refsHeadsMaster)
+    }
+
+    steps {
+        exec {
+            name = "Build"
+            path = "./ci/build.sh"
+            dockerImage = "simon987/general_ci"
+            dockerImagePlatform = ExecBuildStep.ImagePlatform.Linux
+            dockerPull = true
+        }
+    }
+
+    triggers {
+        vcs {
+        }
+    }
+})
+
+object HttpsGithubComSimon987sist2refsHeadsMaster : GitVcsRoot({
+    name = "https://github.com/simon987/sist2#refs/heads/master"
+    url = "https://github.com/simon987/sist2"
+})

CMakeLists.txt

@@ -2,126 +2,123 @@ cmake_minimum_required(VERSION 3.7)
 set(CMAKE_C_STANDARD 11)
 
 project(sist2 C)
-list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_LIST_DIR}/CMakeModules")
+
+option(SIST_DEBUG "Build a debug executable" on)
+
+add_subdirectory(third-party/libscan)
+set(ARGPARSE_SHARED off)
+add_subdirectory(third-party/argparse)
 
 add_executable(
         sist2
         src/main.c
         src/sist.h
         src/io/walk.h src/io/walk.c
-        src/parsing/media.h src/parsing/media.c
-        src/parsing/pdf.h src/parsing/pdf.c
         src/io/store.h src/io/store.c
         src/tpool.h src/tpool.c
         src/parsing/parse.h src/parsing/parse.c
         src/io/serialize.h src/io/serialize.c
         src/parsing/mime.h src/parsing/mime.c src/parsing/mime_generated.c
-        src/parsing/text.h src/parsing/text.c
         src/index/web.c src/index/web.h
         src/web/serve.c src/web/serve.h
         src/index/elastic.c src/index/elastic.h
         src/util.c src/util.h
-        src/ctx.h src/types.h src/parsing/font.c src/parsing/font.h
+        src/ctx.h src/types.h
+        src/log.c src/log.h
 
         # argparse
-        argparse/argparse.h argparse/argparse.c
+        third-party/argparse/argparse.h third-party/argparse/argparse.c
 
-        # cJSON
-        cJSON/cJSON.h cJSON/cJSON.c
+        src/cli.c src/cli.h
+        src/stats.c src/stats.h src/ctx.c)
 
-        # LMDB
-        lmdb/libraries/liblmdb/lmdb.h lmdb/libraries/liblmdb/mdb.c
-        lmdb/libraries/liblmdb/midl.h lmdb/libraries/liblmdb/midl.c
-        src/cli.c src/cli.h)
+target_link_directories(sist2 PRIVATE BEFORE ${_VCPKG_INSTALLED_DIR}/${VCPKG_TARGET_TRIPLET}/lib/)
+set(CMAKE_FIND_LIBRARY_SUFFIXES .a .lib)
 
-find_package(PkgConfig REQUIRED)
-set(ENV{PKG_CONFIG_PATH} "$ENV{PKG_CONFIG_PATH}:/usr/local/lib/pkgconfig/")
+find_package(lmdb CONFIG REQUIRED)
+find_package(cJSON CONFIG REQUIRED)
+find_package(unofficial-glib CONFIG REQUIRED)
+find_package(unofficial-mongoose CONFIG REQUIRED)
+find_library(UUID_LIB NAMES uuid)
+find_package(CURL CONFIG REQUIRED)
 
-find_package(LibMagic REQUIRED)
-find_package(FFmpeg REQUIRED)
-find_package(OpenSSL REQUIRED)
-find_package(Freetype REQUIRED)
+#find_package(OpenSSL REQUIRED)
 
-pkg_check_modules(GLIB REQUIRED glib-2.0)
-pkg_check_modules(GOBJECT REQUIRED gobject-2.0)
-pkg_check_modules(UUID REQUIRED uuid)
 
-include_directories(${LIBMAGIC_INCLUDE_DIRS})
-link_directories(${LIBMAGIC_LIBRARY_DIRS})
-add_definitions(${LIBMAGIC_CFLAGS_OTHER})
-
-link_directories(${UUID_LIBRARY_DIRS})
-include_directories(${UUID_INCLUDE_DIRS})
-add_definitions(${UUID_CFLAGS_OTHER})
-
-include_directories(${GLIB_INCLUDE_DIRS})
-link_directories(${GLIB_LIBRARY_DIRS})
-add_definitions(${GLIB_CFLAGS_OTHER})
-
-include_directories(${GOBJECT_INCLUDE_DIRS})
-link_directories(${GOBJECT_LIBRARY_DIRS})
-add_definitions(${GOBJECT_CFLAGS_OTHER})
-
-link_directories(${FFMPEG_LIBRARY_DIRS})
-include_directories(${FFMPEG_INCLUDE_DIRS})
-
-include_directories(${OPENSSL_INCLUDE_DIR})
-link_directories(${OPENSSL_CRYPTO_LIBRARY})
-
-list(REMOVE_ITEM GLIB_LIBRARIES pcre)
-list(REMOVE_ITEM GOBJECT_LIBRARIES pcre)
-list(REMOVE_ITEM UUID_LIBRARIES pcre)
-
-include_directories(${FREETYPE_INCLUDE_DIRS})
-add_definitions(${FREETYPE_CFLAGS_OTHER})
-
-include_directories(
-        ${PROJECT_SOURCE_DIR}/
-        ${PROJECT_SOURCE_DIR}/lmdb/libraries/liblmdb/
-        ${PROJECT_SOURCE_DIR}/lib/onion/src/
-        ${PROJECT_SOURCE_DIR}/lib/mupdf/include/
+target_include_directories(
+        sist2 PUBLIC
+        ${CMAKE_SOURCE_DIR}/third-party/onion/src/
+        ${CMAKE_SOURCE_DIR}/third-party/utf8.h/
+        ${CMAKE_SOURCE_DIR}/third-party/libscan/
+        ${CMAKE_SOURCE_DIR}/
 )
 
-target_compile_options(sist2
+target_compile_options(
+        sist2
         PRIVATE
-        -O3
-#        -march=native
-        -fno-stack-protector
-        -fomit-frame-pointer
-        )
+        -fPIC
+)
 
-TARGET_LINK_LIBRARIES(
+if (SIST_DEBUG)
+    target_compile_options(
+            sist2
+            PRIVATE
+            -g
+            -fstack-protector
+            -fno-omit-frame-pointer
+            -fsanitize=address
+            -O2
+    )
+    target_link_options(
+            sist2
+            PRIVATE
+            -fsanitize=address
+#            -static
+    )
+    set_target_properties(
+            sist2
+            PROPERTIES
+            OUTPUT_NAME sist2_debug
+    )
+else ()
+#    set(VCPKG_BUILD_TYPE release)
+    target_compile_options(
+            sist2
+            PRIVATE
+            -Ofast
+            -fno-stack-protector
+            -fomit-frame-pointer
+    )
+endif ()
+
+add_dependencies(
+        sist2
+        scan
+        argparse
+)
+
+target_link_libraries(
         sist2
 
-        ${GLIB_LIBRARIES}
-        ${GOBJECT_LIBRARIES}
-        ${UUID_LIBRARIES}
-
-        # ffmpeg
-#        ${PROJECT_SOURCE_DIR}/lib/libavcodec.a
-#        ${PROJECT_SOURCE_DIR}/lib/libavformat.a
-#        ${PROJECT_SOURCE_DIR}/lib/libavutil.a
-#        ${PROJECT_SOURCE_DIR}/lib/libswscale.a
-#        ${PROJECT_SOURCE_DIR}/lib/libswresample.a
-        ${FFMPEG_LIBRARIES}
-        swscale
-
-        # mupdf
-        ${PROJECT_SOURCE_DIR}/lib/libmupdf.a
-        ${PROJECT_SOURCE_DIR}/lib/libmupdf-third.a
-
-        # onion
-        ${PROJECT_SOURCE_DIR}/lib/libonion_static.a
+        z
+        lmdb
+        cjson
+        argparse
+        unofficial::glib::glib
+        unofficial::mongoose::mongoose
+#        OpenSSL::SSL OpenSSL::Crypto
+        CURL::libcurl
 
+        ${UUID_LIB}
         pthread
-        curl
-        m
-        bz2
         magic
+
+        scan
 )
 
 add_custom_target(
         before_sist2
         COMMAND ${CMAKE_CURRENT_SOURCE_DIR}/scripts/before_build.sh
 )
+
 add_dependencies(sist2 before_sist2)

Docker/Dockerfile

@@ -0,0 +1,22 @@
+FROM ubuntu:19.10
+MAINTAINER simon987 <me@simon987.net>
+
+RUN apt update
+RUN apt install -y libglib2.0-0 libcurl4 libmagic1 libharfbuzz-bin libopenjp2-7 libarchive13 liblzma5 libzstd1 liblz4-1 \
+ curl libtiff5 libpng16-16 libpcre3
+
+RUN mkdir -p /usr/share/tessdata && \
+    cd /usr/share/tessdata/ && \
+    curl -o /usr/share/tessdata/hin.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/hin.traineddata &&\
+    curl -o /usr/share/tessdata/jpn.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/jpn.traineddata &&\
+    curl -o /usr/share/tessdata/eng.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/eng.traineddata &&\
+    curl -o /usr/share/tessdata/fra.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/fra.traineddata &&\
+    curl -o /usr/share/tessdata/rus.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/rus.traineddata &&\
+    curl -o /usr/share/tessdata/spa.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/spa.traineddata && ls -lh
+
+ADD sist2 /root/sist2
+
+ENV LANG C.UTF-8
+ENV LC_ALL C.UTF-8
+
+ENTRYPOINT ["/root/sist2"]

Docker/build.sh

@@ -0,0 +1,14 @@
+rm ./sist2 sist2_debug
+cp ../sist2.gz .
+gzip -d sist2.gz
+strip sist2
+
+version=$(./sist2 --version)
+
+echo "Version ${version}"
+docker build . -t simon987/sist2:${version} -t simon987/sist2:latest
+
+docker push simon987/sist2:${version}
+docker push simon987/sist2:latest
+
+docker run --rm simon987/sist2 -v

README.md

@@ -1,85 +1,135 @@
 ![GitHub](https://img.shields.io/github/license/simon987/sist2.svg)
 [![CodeFactor](https://www.codefactor.io/repository/github/simon987/sist2/badge?s=05daa325188aac4eae32c786f3d9cf4e0593f822)](https://www.codefactor.io/repository/github/simon987/sist2)
+[![Development snapshots](https://ci.simon987.net/app/rest/builds/buildType(Sist2_Build)/statusIcon)](https://files.simon987.net/artifacts/Sist2/Build/)
 
 # sist2
 
 sist2 (Simple incremental search tool)
 
+*Warning: sist2 is in early development*
+
+![sist2.png](docs/sist2.png)
+
 ## Features
 
+* Fast, low memory usage, multi-threaded
+* Mobile-friendly Web interface
+* Portable (all its features are packaged in a single executable)
+* Extracts text and metadata from common file types \*
+* Generates thumbnails \*
+* Incremental scanning
+* Manual tagging from the UI and automatic tagging based on file attributes via [user scripts](docs/scripting.md)
+* Recursive scan inside archive files \*\*
+* OCR support with tesseract \*\*\*
+* Stats page & disk utilisation visualization
+
+
+\* See [format support](#format-support)    
+\*\* See [Archive files](#archive-files)    
+\*\*\* See [OCR](#ocr)    
+
+![stats](docs/stats.png)
+
+
+## Getting Started
+
+1. Have an Elasticsearch (>= 6.X.X) instance running
+    1. Download [from official website](https://www.elastic.co/downloads/elasticsearch)
+    1. *(or)* Run using docker:
+        ```bash
+       docker run -d --name es1 --net sist2_net -p 9200:9200 \
+            -e "discovery.type=single-node" elasticsearch:7.5.2
+        ```
+    1. *(or)* Run using docker-compose:
+        ```yaml
+          elasticsearch:
+            image: docker.elastic.co/elasticsearch/elasticsearch:7.5.2
+            environment:
+              - discovery.type=single-node
+              - "ES_JAVA_OPTS=-Xms1G -Xmx2G"
+        ```
+1. Download sist2 executable
+    1. Download the [latest sist2 release](https://github.com/simon987/sist2/releases) *
+    1. *(or)* Download a [development snapshot](https://files.simon987.net/artifacts/Sist2/Build/) *(Not recommended!)*
+    1. *(or)* `docker pull simon987/sist2:latest`
+
+1. See [Usage guide](docs/USAGE.md)
+   
+
+\* *Windows users*: **sist2** runs under [WSL](https://en.wikipedia.org/wiki/Windows_Subsystem_for_Linux)    
 
 
 ## Example usage
 
-See help page `sist2 --help` for more details.
+See [Usage guide](docs/USAGE.md) for more details
 
-**Scan a directory**
-```bash
-sist2 scan ~/Documents -o ./orig_idx/
-sist2 scan --threads 4 --content-size 16384 /mnt/Pictures
-sist2 scan -i ./orig_idx/ -o ./updated_idx/ ~/Documents
-```
+1. Scan a directory: `sist2 scan ~/Documents -o ./docs_idx`
+1. Push index to Elasticsearch: `sist2 index ./docs_idx`
+1. Start web interface: `sist2 web ./docs_idx`
 
-**Push index to Elasticsearch or file**
-```bash
-sist2 index --force-reset ./my_idx
-sist2 index --print ./my_idx > raw_documents.ndjson
-```
-
-**Start web interface**
-```bash
-sist2 web --bind 0.0.0.0 --port 4321 ./my_idx1 ./my_idx2 ./my_idx3
-```
 
 ## Format support
 
 File type | Library | Content | Thumbnail | Metadata
 :---|:---|:---|:---|:---
-pdf,xps,cbz,cbr,fb2,epub | MuPDF | yes | yes, `png` | *planned* |
-`audio/*` | libav | - | yes, `jpeg` | ID3 tags |
-`video/*` | libav | - | yes, `jpeg` | *planned* |
-`image/*` | libav | - | yes, `jpeg` | *planned* |
+pdf,xps,fb2,epub | MuPDF | text+ocr | yes | title |
+cbz,cbr | *(none)* | - | yes | - |
+`audio/*` | ffmpeg | - | yes | ID3 tags |
+`video/*` | ffmpeg | - | yes | title, comment, artist |
+`image/*` | ffmpeg | - | yes | [Common EXIF tags](https://github.com/simon987/sist2/blob/efdde2734eca9b14a54f84568863b7ffd59bdba3/src/parsing/media.c#L190) |
+raw, rw2, dng, cr2, crw, dcr, k25, kdc, mrw, pef, xf3, arw, sr2, srf, erf  | LibRaw | - | yes | Common EXIF tags |
 ttf,ttc,cff,woff,fnt,otf | Freetype2 | - | yes, `bmp` | Name & style |
 `text/plain` | *(none)* | yes | no | - |
-docx, xlsx, pptx |  | *planned* | no | *planned* |
+html, xml | *(none)* | yes | no | - |
+tar, zip, rar, 7z, ar ...  | Libarchive | yes\* | - | no |
+docx, xlsx, pptx | *(none)* | yes | if embedded | creator, modified_by, title |
+mobi, azw, azw3 | libmobi | yes | no | author, title |
 
+\* *See [Archive files](#archive-files)*
+ 
+### Archive files
+**sist2** will scan files stored into archive files (zip, tar, 7z...) as if
+they were directly in the file system. Recursive (archives inside archives)
+scan is also supported.
 
+**Limitations**:
+* Support for parsing media files with formats that require *seek* (e.g. `.gif`, `.mp4` w/ fragmented metadata etc.) 
+  is limitted (see `--mem-buffer` option)
+* Archive files are scanned sequentially, by a single thread. On systems where
+**sist2** is not I/O bound, scans might be faster when larger archives are split
+ into smaller parts.
+ 
+ 
+### OCR
 
+You can enable OCR support for pdf,xps,fb2,epub file types with the
+`--ocr <lang>` option. Download the language data files with your
+package manager (`apt install tesseract-ocr-eng`) or directly [from Github](https://github.com/tesseract-ocr/tesseract/wiki/Data-Files).
 
-## Installation
+The `simon987/sist2` image comes with common languages 
+(hin, jpn, eng, fra, rus, spa) pre-installed.
 
-1. Download runtime dependencies
+Examples
+```bash
+sist2 scan --ocr jpn ~/Books/Manga/
+sist2 scan --ocr eng ~/Books/Textbooks/
+```
 
-    `apt install curl bzip2`
-    
-1. Download binary
-    
-    Get [the latest release](https://github.com/simon987/sist2/releases) from GitHub
-
-1. (Optional) Add to search path
-    
-   `mv sist2 /usr/bin/`
 
 ## Build from source
 
+You can compile **sist2** by yourself if you don't want to use the pre-compiled
+binaries (GCC 7+ required).
+
 1. Install compile-time dependencies
 
-    *(Debian)*
-    ```bash
-    apt install git cmake pkg-config libglib2.0-dev\
-        libssl-dev uuid-dev libavformat-dev libswscale-dev \
-        python3 libmagic-dev libfreetype6-dev libcurl-dev \
-        libbz2-dev yasm
-    ```
-    *(Archlinux)*
-    ```bash
-    pacman -S git ffmpeg pkg-config cmake openssl curl \
-        bzip2 yasm libutil-linux
-    ```
+   ```bash
+   vcpkg install lmdb cjson glib libarchive[core,bzip2,libxml2,lz4,lzma,lzo] pthread tesseract libxml2 ffmpeg zstd gtest mongoose libuuid libmagic libraw
+   ```
+
 2. Build
     ```bash
-    git clone --recurse-submodules https://github.com/simon987/sist2
-    ./scripts/get_static_libs.sh
-    cmake .
+    git clone --recursive https://github.com/simon987/sist2/
+    cmake -DCMAKE_TOOLCHAIN_FILE=<VCPKG_ROOT>/scripts/buildsystems/vcpkg.cmake .
     make
-    ```
+    ```

ci/build.sh

@@ -0,0 +1,17 @@
+#!/usr/bin/env bash
+
+VCPKG_ROOT="/vcpkg"
+
+rm *.gz
+
+rm -rf CMakeFiles CMakeCache.txt
+cmake -DSIST_DEBUG=off -DVCPKG_BUILD_TYPE=release -DCMAKE_TOOLCHAIN_FILE="${VCPKG_ROOT}/scripts/buildsystems/vcpkg.cmake" .
+make -j 12
+strip sist2
+gzip -9 sist2
+
+rm -rf CMakeFiles CMakeCache.txt
+cmake -DSIST_DEBUG=on -DVCPKG_BUILD_TYPE=debug -DCMAKE_TOOLCHAIN_FILE="${VCPKG_ROOT}/scripts/buildsystems/vcpkg.cmake" .
+make -j 12
+cp /usr/lib/x86_64-linux-gnu/libasan.so.2.0.0 libasan.so.2
+tar -czf sist2_debug.tar.gz sist2_debug libasan.so.2

docs/USAGE.md

@@ -0,0 +1,338 @@
+# Usage
+
+*More examples (specifically with docker/compose) are in progress*
+
+* [scan](#scan)
+    * [options](#scan-options)
+    * [examples](#scan-examples)
+    * [index format](#index-format)
+* [index](#index)
+    * [options](#index-options)
+    * [examples](#index-examples)
+* [web](#web)
+    * [options](#web-options)
+    * [examples](#web-examples)
+    * [rewrite_url](#rewrite_url)
+    * [link to specific indices](#link-to-specific-indices)
+* [exec-script](#exec-script)
+* [tagging](#tagging)
+
+```
+Usage: sist2 scan [OPTION]... PATH
+   or: sist2 index [OPTION]... INDEX
+   or: sist2 web [OPTION]... INDEX...
+   or: sist2 exec-script [OPTION]... INDEX
+Lightning-fast file system indexer and search tool.
+
+    -h, --help                    show this help message and exit
+    -v, --version                 Show version and exit
+    --verbose                     Turn on logging
+    --very-verbose                Turn on debug messages
+
+Scan options
+    -t, --threads=<int>           Number of threads. DEFAULT=1
+    -q, --quality=<flt>           Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best. DEFAULT=5
+    --size=<int>                  Thumbnail size, in pixels. Use negative value to disable. DEFAULT=500
+    --content-size=<int>          Number of bytes to be extracted from text documents. Use negative value to disable. DEFAULT=32768
+    --incremental=<str>           Reuse an existing index and only scan modified files.
+    -o, --output=<str>            Output directory. DEFAULT=index.sist2/
+    --rewrite-url=<str>           Serve files from this url instead of from disk.
+    --name=<str>                  Index display name. DEFAULT: (name of the directory)
+    --depth=<int>                 Scan up to DEPTH subdirectories deep. Use 0 to only scan files in PATH. DEFAULT: -1
+    --archive=<str>               Archive file mode (skip|list|shallow|recurse). skip: Don't parse, list: only get file names as text, shallow: Don't parse archives inside archives. DEFAULT: recurse
+    --ocr=<str>                   Tesseract language (use tesseract --list-langs to see which are installed on your machine)
+    -e, --exclude=<str>           Files that match this regex will not be scanned
+    --fast                        Only index file names & mime type
+    --treemap-threshold=<str>     Relative size threshold for treemap (see USAGE.md). DEFAULT: 0.0005
+    --mem-buffer=<int>            Maximum memory buffer size per thread in MB for files inside archives (see USAGE.md). DEFAULT: 2000
+
+Index options
+    -t, --threads=<int>           Number of threads. DEFAULT=1
+    --es-url=<str>                Elasticsearch url with port. DEFAULT=http://localhost:9200
+    -p, --print                   Just print JSON documents to stdout.
+    --script-file=<str>           Path to user script.
+    --batch-size=<int>            Index batch size. DEFAULT: 100
+    -f, --force-reset             Reset Elasticsearch mappings and settings. (You must use this option the first time you use the index command)
+
+Web options
+    --es-url=<str>                Elasticsearch url. DEFAULT=http://localhost:9200
+    --bind=<str>                  Listen on this address. DEFAULT=localhost:4090
+    --auth=<str>                  Basic auth in user:password format
+    --tag-auth=<str>              Basic auth in user:password format for tagging
+
+Exec-script options
+    --script-file=<str>           Path to user script.
+Made by simon987 <me@simon987.net>. Released under GPL-3.0
+```
+
+## Scan
+
+### Scan options
+
+* `-t, --threads` 
+      Number of threads for file parsing. **Do not set a number higher than `$(nproc)` or `$(Get-WmiObject Win32_ComputerSystem).NumberOfLogicalProcessors` in Windows!**
+* `-q, --quality` 
+    Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best. *Does not affect PDF thumbnails quality*
+* `--size` 
+    Thumbnail size in pixels.
+* `--content-size` 
+    Number of bytes of text to be extracted from the content of files (plain text and PDFs).
+    Repeated whitespace and special characters do not count toward this limit.
+* `--incremental`
+    Specify an existing index. Information about files in this index that were not modified (based on *mtime* attribute)
+    will be copied to the new index and will not be parsed again.
+* `-o, --output` Output directory. 
+* `--rewrite-url` Set the `rewrite_url` option for the web module (See [rewrite_url](#rewrite_url)) 
+* `--name` Set the `name` option for the web module
+* `--depth` Maximum scan dept. Set to 0 only scan files directly in the root directory, set to -1 for infinite depth
+* `--archive` Archive file mode.
+    * skip: Don't parse
+    * list: Only get file names as text
+    * shallow: Don't parse archives inside archives.
+    * recurse: Scan archives recursively (default)
+* `--ocr` See [OCR](../README.md#OCR)
+* `-e, --exclude` Regex pattern to exclude files. A file is excluded if the pattern matches any 
+    part of the full absolute path.
+    
+    Examples: 
+    * `-e ".*\.ttf"`: Ignore ttf files
+    * `-e ".*\.(ttf|rar)"`: Ignore ttf and rar files
+    * `-e "^/mnt/backups/"`: Ignore all files in the `/mnt/backups/` directory
+    * `-e "^/mnt/Data[12]/"`: Ignore all files in the `/mnt/Data1/` and `/mnt/Data2/` directory
+    * `-e "(^/usr/)|(^/var/)|(^/media/DRIVE-A/tmp/)|(^/media/DRIVE-B/Trash/)"` Exclude the
+     `/usr`, `/var`, `/media/DRIVE-A/tmp`, `/media/DRIVE-B/Trash` directories
+* `--fast` Only index file names and mime type
+* `--treemap-threshold` Directories smaller than (`treemap-threshold` * `<total size of the index>`)
+    will not be considered for the disk utilisation visualization; their size will be added to
+    the parent directory. If the parent directory is still smaller than the threshold, it will also be "merged upwards"
+    and so on.
+    
+    In effect, smaller `treemap-threshold` values will yield a more detailed 
+    (but also a more cluttered and harder to read) visualization. 
+    
+* `--mem-buffer` Maximum memory buffer size in MB (per thread) for files inside archives. Media files 
+    larger than this number will be read sequentially and no *seek* operations will be supported.
+
+    To check if a media file can be parsed without *seek*, execute `cat file.mp4 | ffprobe -`
+
+### Scan examples
+
+Simple scan
+```bash
+sist2 scan ~/Documents
+
+sist2 scan \
+    --threads 4 --content-size 16000000 --quality 1.0 --archive shallow \
+    --name "My Documents" --rewrite-url "http://nas.domain.local/My Documents/" \
+    ~/Documents -o ./documents.idx/
+```
+
+Incremental scan
+```
+sist2 scan --incremental ./orig_idx/ -o ./updated_idx/ ~/Documents
+```
+
+### Index format
+
+A typical `binary` type index structure looks like this:
+```
+documents.idx/
+├── descriptor.json
+├── _index_139965416830720
+├── _index_139965425223424
+├── _index_139965433616128
+├── _index_139965442008832
+├── _index_139965442008832
+├── treemap.csv
+├── agg_mime.csv
+├── agg_date.csv
+├── add_size.csv
+├── thumbs
+|   ├── data.mdb
+|   └── lock.mdb
+└── tags
+    ├── data.mdb
+    └── lock.mdb
+```
+
+The `_index_*` files contain the raw binary index data and are not meant to be
+read by other applications. The format is generally compatible across different 
+sist2 versions.
+
+The `thumbs/` folder is a [LMDB](https://en.wikipedia.org/wiki/Lightning_Memory-Mapped_Database)
+database containing the thumbnails.
+
+The `descriptor.json` file contains general information about the index. The 
+following fields are safe to modify manually: `root`, `name`, [rewrite_url](#rewrite_url) and `timestamp`.
+
+The `.csv` are pre-computed aggregations necessary for the stats page.
+
+
+*Advanced usage*
+
+Instead of using the `scan` module, you can also import an index generated
+by a third party application. The 'external' index must have the following format:
+
+```
+my_index/
+├── descriptor.json
+├── _index_0
+└── thumbs
+    ├── data.mdb
+    └── lock.mdb
+```
+
+*descriptor.json*:
+```json
+{
+    "uuid": "<valid UUID4>",
+    "version": "_external_v1",
+    "root": "(optional)",
+    "name": "<name>",
+    "rewrite_url": "(optional)",
+    "type": "json",
+    "timestamp": 1578971024
+}
+```
+
+*_index_0*: NDJSON format (One json object per line)
+
+```json
+{
+  "_id": "unique uuid for the file",
+  "index": "index uuid4 (same one as descriptor.json!)",
+  "mime": "application/x-cbz",
+  "size": 14341204,
+  "mtime": 1578882996,
+  "extension": "cbz",
+  "name": "my_book",
+  "path": "path/to/books",
+  "content": "text contents of the book",
+  "title": "Title of the book",
+  "tag": ["genre.fiction", "author.someguy", "etc..."],
+  "_keyword": [
+    {"k": "ISBN", "v": "ABCD34789231"}
+  ],
+  "_text": [
+    {"k": "other", "v": "This will be indexed as text"}
+  ]
+}
+```
+
+You can find the full list of supported fields [here](../src/io/serialize.c#L90)
+
+The `_keyword.*` items will be indexed and searchable as **keyword** fields (only full matches allowed).
+The `_text.*` items will be indexed and searchable as **text** fields (fuzzy searching allowed)
+
+
+*thumbs/*:
+
+LMDB key-value store. Keys are **binary** 128-bit UUID4s (`_id` field)
+and values are raw image bytes.
+
+Importing an external `binary` type index is technically possible but
+it is currently unsupported and has no guaranties of back/forward compatibility.
+
+
+## Index
+### Index options
+ * `--es-url` 
+ Elasticsearch url and port. If you are using docker, make sure that both containers are on the
+ same network.
+ * `-p, --print` 
+    Print index in JSON format to stdout.
+ * `--script-file` 
+    Path to user script. See [Scripting](scripting.md).
+ * `--batch-size=<int>` 
+    Index batch size. Indexing is generally faster with larger batches, but payloads that
+    are too large will fail and additional overhead for retrying with smaller sizes may slow
+    down the process.
+ * `-f, --force-reset` 
+    Reset Elasticsearch mappings and settings.
+    **(You must use this option the first time you use the index command)**.
+    
+### Index examples
+
+**Push to elasticsearch**
+```bash
+sist2 index --force-reset --batch-size 1000 --es-url http://localhost:9200 ./my_index/
+sist2 index ./my_index/
+```
+
+**Save index in JSON format**
+```bash
+sist2 index --print ./my_index/ > my_index.ndjson
+```
+
+**Inspect contents of an index**
+```bash
+sist2 index --print ./my_index/ | jq | less
+```
+
+## Web
+
+### Web options
+ * `--es-url=<str>` Elasticsearch url.
+ * `--bind=<str>` Listen on this address.
+ * `--auth=<str>` Basic auth in user:password format
+ * `--tag-auth=<str>` Basic auth in user:password format. Works the same way as the 
+    `--auth` argument, but authentication is only applied the `/tag/` endpoint.
+ 
+### Web examples
+
+**Single index**
+```bash
+sist2 web --auth admin:hunter2 --bind 0.0.0.0:8888 my_index
+```
+
+**Multiple indices**
+```bash
+# Indices will be displayed in this order in the web interface
+sist2 web index1 index2 index3 index4
+```
+
+### rewrite_url
+
+When the `rewrite_url` field is not empty, the web module ignores the `root`
+field and will return a HTTP redirect to `<rewrite_url><path>/<name><extension>`
+instead of serving the file from disk. 
+Both the `root` and `rewrite_url` fields are safe to manually modify from the 
+`descriptor.json` file.
+
+### Link to specific indices
+
+To link to specific indices, you can add a list of comma-separated index name to 
+the URL: `?i=<name>,<name>`. By default, indices with `"(nsfw)"` in their name are
+not displayed.
+
+## exec-script
+
+The `exec-script` command is used to execute a user script for an index that has already been imported to Elasticsearch with the `index` command. Note that the documents will not be reset to their default state before each execution as the `index` command does: if you make undesired changes to the documents by accident, you will need to run `index` again to revert to the original state.
+
+
+# Tagging
+
+### Manual tagging
+
+You can modify tags of individual documents directly from the 
+ `web` interface. Note that you can setup authentication for this feature
+ with the `--tag-auth` option (See [web options](#web-options))
+
+![manual_tag](manual_tag.png)
+
+Tags that are manually added are saved both in the 
+ index folder (in `/tags/`) and in Elasticsearch*. When re-`index`ing, 
+ they are read from the index and automatically applied.
+ 
+You can safely copy the `/tags/` database to another index.
+
+See [Automatic tagging](#automatic-tagging) for information about tag 
+ hierarchies and tag colors.
+
+\* *It can take a few seconds to take effect in new search queries, and the page needs 
+    to be reloaded for the tag tab to update*
+
+
+### Automatic tagging
+
+See [scripting](docs/scripting.md) documentation.

docs/scripting.md

@@ -0,0 +1,142 @@
+## User scripts
+
+*This document is under construction, more in-depth guide coming soon*
+
+During the `index` step, you can use the `--script-file <script>` option to
+modify documents or add user tags. This option is mainly used to
+implement automatic tagging based on file attributes.
+
+The scripting language used 
+([Painless Scripting Language](https://www.elastic.co/guide/en/elasticsearch/painless/7.4/index.html)) 
+is very similar to Java, but you should be able to create user scripts
+without programming experience at all if you're somewhat familiar with
+regex.
+
+This is the base structure of the documents we're working with:
+```json
+{
+  "_id": "e171405c-fdb5-4feb-bb32-82637bc32084",
+  "_index": "sist2",
+  "_type": "_doc",
+  "_source": {
+    "index": "206b3050-e821-421a-891d-12fcf6c2db0d",
+    "mime": "application/json",
+    "size": 1799,
+    "mtime": 1545443685,
+    "extension": "md",
+    "name": "README",
+    "path": "sist2/scripting",
+    "content": "..."
+  }
+}
+```
+
+**Example script**
+
+This script checks if the `genre` attribute exists, if it does
+it adds the `genre.<genre>` tag. 
+```Java
+ArrayList tags = ctx._source.tag = new ArrayList();
+
+if (ctx._source?.genre != null) {
+    tags.add("genre." + ctx._source.genre.toLowerCase());
+}
+```
+
+You can use `.` to create a hierarchical tag tree:
+
+![scripting/genre_example](genre_example.png)
+
+
+To use regular expressions, you need to add this line in `/etc/elasticsearch/elasticsearch.yml`
+```yaml
+script.painless.regex.enabled: true
+```
+Or, if you're using docker add `-e "script.painless.regex.enabled=true"`
+
+**Tag color**
+
+You can specify the color for an individual tag by appending an 
+hexadecimal color code (`#RRGGBBAA`) to the tag name.
+
+### Examples
+
+If `(20XX)` is in the file name, add the `year.<year>` tag:
+```Java
+ArrayList tags = ctx._source.tag = new ArrayList();
+
+Matcher m = /[\(\.+](20[0-9]{2})[\)\.+]/.matcher(ctx._source.name);
+if (m.find()) {
+    tags.add("year." + m.group(1));
+}
+```
+
+Use default *Calibre* folder structure to infer author.
+```Java
+ArrayList tags = ctx._source.tag = new ArrayList();
+
+// We expect the book path to look like this:
+//  /path/to/Calibre Library/Author/Title/Title - Author.pdf
+
+if (ctx._source.name.contains("-") && ctx._source.extension == "pdf") {
+    String[] names = ctx._source.name.splitOnToken('-');
+    tags.add("author." + names[1].strip());
+}
+```
+
+If the file matches a specific pattern `AAAA-000 fName1 lName1, <fName2 lName2>...`, add the `actress.<actress>` and 
+`studio.<studio>` tag:
+```Java
+ArrayList tags = ctx._source.tag = new ArrayList();
+
+Matcher m = /([A-Z]{4})-[0-9]{3} (.*)/.matcher(ctx._source.name);
+if (m.find()) {
+    tags.add("studio." + m.group(1));
+
+    // Take the matched group (.*), and add a tag for
+    //  each name, separated by comma
+    for (String name : m.group(2).splitOnToken(',')) {
+        tags.add("actress." + name);
+    }
+}
+```
+
+Set the name of the last folder (`/path/to/<studio>/file.mp4`) to `studio.<studio>` tag
+```Java
+ArrayList tags = ctx._source.tag = new ArrayList();
+
+if (ctx._source.path != "") {
+    String[] names = ctx._source.path.splitOnToken('/');
+    tags.add("studio." + names[names.length-1]);
+}
+```
+
+Parse `EXIF:F Number` tag
+```Java
+if (ctx._source?.exif_fnumber != null) {
+    String[] values = ctx._source.exif_fnumber.splitOnToken(' ');
+    String aperture = String.valueOf(Float.parseFloat(values[0]) / Float.parseFloat(values[1]));
+    if (aperture == "NaN") {
+        aperture = "0,0";
+    }
+    tags.add("Aperture.f/" + aperture.replace(".", ","));
+}
+```
+
+Display year and months from `EXIF:DateTime` tag
+```Java
+if (ctx._source?.exif_datetime != null) {
+    SimpleDateFormat parser = new SimpleDateFormat("yyyy:MM:dd HH:mm:ss");
+    Date date = parser.parse(ctx._source.exif_datetime);
+
+    SimpleDateFormat yp = new SimpleDateFormat("yyyy");
+    SimpleDateFormat mp = new SimpleDateFormat("MMMMMMMMM");
+
+    String year = yp.format(date);
+    String month = mp.format(date);
+
+    tags.add("Month." + month);
+    tags.add("Year." + year);
+}
+
+```

schema/mappings.json

@@ -1,31 +1,58 @@
 {
   "properties": {
+    "_tie": {
+      "type": "keyword",
+      "doc_values": true
+    },
+    "_depth": {
+      "type": "integer"
+    },
     "path": {
       "type": "text",
       "analyzer": "path_analyzer",
-      "copy_to": "suggest-path"
+      "copy_to": "suggest-path",
+      "fielddata": true,
+      "fields": {
+        "nGram": {
+          "type": "text",
+          "analyzer": "my_nGram"
+        },
+        "text": {
+          "type": "text",
+          "analyzer": "content_analyzer"
+        }
+      }
     },
     "suggest-path": {
       "type": "completion",
-      "analyzer": "keyword"
+      "analyzer": "case_insensitive_kw_analyzer"
     },
     "mime": {
       "type": "keyword"
     },
+    "thumbnail": {
+      "type": "keyword",
+      "index": false
+    },
     "videoc": {
-      "type": "keyword"
+      "type": "keyword",
+      "index": false
     },
     "audioc": {
-      "type": "keyword"
+      "type": "keyword",
+      "index": false
     },
     "duration": {
-      "type": "float"
+      "type": "float",
+      "index": false
     },
     "width": {
-      "type": "integer"
+      "type": "integer",
+      "index": false
     },
     "height": {
-      "type": "integer"
+      "type": "integer",
+      "index": false
     },
     "mtime": {
       "type": "integer"
@@ -70,6 +97,23 @@
       "analyzer": "my_nGram",
       "type": "text"
     },
+    "_keyword.*":  {
+      "type": "keyword"
+    },
+    "_text.*":  {
+      "analyzer": "content_analyzer",
+      "type": "text",
+      "fields": {
+        "nGram": {
+          "type": "text",
+          "analyzer": "my_nGram"
+        }
+      }
+    },
+    "_url": {
+      "type": "keyword",
+      "index": false
+    },
     "content": {
       "analyzer": "content_analyzer",
       "type": "text",
@@ -80,6 +124,44 @@
           "analyzer": "my_nGram"
         }
       }
+    },
+    "tag": {
+      "type": "keyword",
+      "copy_to": "suggest-tag"
+    },
+    "suggest-tag": {
+      "type": "completion",
+      "analyzer": "case_insensitive_kw_analyzer"
+    },
+    "exif_make": {
+      "type": "text"
+    },
+    "exif_model": {
+      "type": "text"
+    },
+    "exif:software": {
+      "type": "text"
+    },
+    "exif_exposure_time": {
+      "type": "keyword"
+    },
+    "exif_fnumber": {
+      "type": "keyword"
+    },
+    "exif_iso_speed_ratings": {
+      "type": "keyword"
+    },
+    "exif_focal_length": {
+      "type": "keyword"
+    },
+    "exif_user_comment": {
+      "type": "text"
+    },
+    "author": {
+      "type": "text"
+    },
+    "modified_by": {
+      "type": "text"
     }
   }
 }

schema/pipeline.json

@@ -0,0 +1,10 @@
+{
+  "description": "Copy _id to _tie, save path depth",
+  "processors": [
+    {
+      "script": {
+        "source": "ctx._tie = ctx._id; ctx._depth = ctx.path.length() == 0 ? 0 : 1 + ctx.path.length() - ctx.path.replace(\"/\", \"\").length();"
+      }
+    }
+  ]
+}

schema/settings.json

@@ -1,7 +1,8 @@
 {
   "index": {
-    "refresh_interval": "-1",
-    "codec": "best_compression"
+    "refresh_interval": "30s",
+    "codec": "best_compression",
+    "number_of_replicas": 0
   },
   "analysis": {
     "tokenizer": {
@@ -21,16 +22,24 @@
           "lowercase"
         ]
       },
+      "case_insensitive_kw_analyzer": {
+        "tokenizer": "keyword",
+        "filter": [
+          "lowercase"
+        ]
+      },
       "my_nGram": {
         "tokenizer": "my_nGram_tokenizer",
         "filter": [
-          "lowercase"
+          "lowercase",
+          "asciifolding"
         ]
       },
       "content_analyzer": {
         "tokenizer": "standard",
         "filter": [
-          "lowercase"
+          "lowercase",
+          "asciifolding"
         ]
       }
     }

scripts/before_build.sh

@@ -1,14 +1,16 @@
-#!/bin/bash
+#!/usr/bin/env bash
 
 rm -rf index.sist2/
 
-rm web/js/bundle.js 2> /dev/null
-cat `ls -v web/js/*.min.js` > web/js/bundle.js
-cat web/js/{util,dom,search}.js >> web/js/bundle.js
+rm src/static/js/bundle.js 2> /dev/null
+cat `ls src/static/js/*.min.js` > src/static/js/bundle.js
+cat src/static/js/{util,dom}.js >> src/static/js/bundle.js
 
-rm web/css/bundle.css 2> /dev/null
-cat web/css/*.min.css > web/css/bundle.css
-cat web/css/main.css >> web/css/bundle.css
+rm src/static/css/bundle*.css 2> /dev/null
+cat src/static/css/*.min.css > src/static/css/bundle.css
+cat src/static/css/light.css >> src/static/css/bundle.css
+cat src/static/css/*.min.css > src/static/css/bundle_dark.css
+cat src/static/css/dark.css >> src/static/css/bundle_dark.css
 
 python3 scripts/mime.py > src/parsing/mime_generated.c
 python3 scripts/serve_static.py > src/web/static_generated.c

scripts/get_static_libs.sh

@@ -1,39 +0,0 @@
-#!/bin/bash
-cd lib
-
-cd mupdf
-HAVE_X11=no HAVE_GLUT=no make -j 4
-cd ..
-
-mv mupdf/build/release/libmupdf.a .
-mv mupdf/build/release/libmupdf-third.a .
-
-# ffmpeg
-cd ffmpeg
-./configure --disable-shared --enable-static --disable-ffmpeg --disable-ffplay \
- --disable-ffprobe --disable-doc\
- --disable-manpages --disable-postproc --disable-avfilter \
- --disable-alsa --disable-lzma --disable-xlib --disable-debug\
- --disable-vdpau --disable-vaapi --disable-sdl2 --disable-network
-make -j 4
-cd ..
-
-mv ffmpeg/libavcodec/libavcodec.a .
-mv ffmpeg/libavformat/libavformat.a .
-mv ffmpeg/libavutil/libavutil.a .
-mv ffmpeg/libswresample/libswresample.a .
-mv ffmpeg/libswscale/libswscale.a .
-
-# onion
-cd onion
-mkdir build 2> /dev/null
-cd build
-cmake -DONION_USE_SSL=false -DONION_USE_PAM=false -DONION_USE_PNG=false -DONION_USE_JPEG=false \
--DONION_USE_JPEG=false -DONION_USE_XML2=false -DONION_USE_SYSTEMD=false -DONION_USE_SQLITE3=false \
--DONION_USE_REDIS=false -DONION_USE_GC=false -DONION_USE_TESTS=false -DONION_EXAMPLES=false \
--DONION_USE_BINDINGS_CPP=false ..
-make -j 4
-cd ../..
-
-mv onion/build/src/onion/libonion_static.a .
-cd ..

scripts/index_static.py

@@ -1,6 +1,9 @@
+import json
+
 files = [
     "schema/mappings.json",
     "schema/settings.json",
+    "schema/pipeline.json",
 ]
 
 
@@ -9,6 +12,7 @@ def clean(filepath):
 
 
 for file in files:
-    with open(file, "rb") as f:
-        data = f.read()
+    with open(file, "r") as f:
+        data = json.dumps(json.load(f), separators=(",", ":")).encode()
+    data += b'\0'
     print("char %s[%d] = {%s};" % (clean(file), len(data), ",".join(str(int(b)) for b in data)))

scripts/mime.csv

@@ -2,14 +2,18 @@ application/arj, arj
 application/base64, mme
 application/binhex, hqx
 application/book, boo|book
+application/CDFV2-corrupt,
 application/CDFV2, sdv
 application/clariscad, ccad
 application/commonground, dp
+application/csv,
+application/dicom, dcm
 application/drafting, drw
+application/epub+zip, epub
 application/freeloader, frl
 application/futuresplash, spl
 application/groupwise, vew
-application/gzip, gz
+application/gzip, gz|tgz
 application/hta, hta
 application/i-deas, unv
 application/iges, iges|igs
@@ -17,7 +21,6 @@ application/inf, inf
 application/java-archive, jar
 application/java, class
 application/javascript,
-application/x-archive, a
 application/json, json
 application/marc, mrc
 application/mbedlet, mbd
@@ -27,7 +30,9 @@ application/msword, doc|dot|w6w|wiz|word
 application/netmc, mcp
 application/octet-stream, bin|dump|gpg
 application/oda, oda
+application/ogg, ogv
 application/pdf, pdf
+application/pgp-keys,
 application/pgp-signature, pgp
 application/pkcs7-signature, p7s
 application/pkix-cert, cer|crt
@@ -43,6 +48,10 @@ application/vda, vda
 application/vnd.fdf, fdf
 application/vnd.font-fontforge-sfd, sfd
 application/vnd.hp-hpgl, hgl|hpg|hpgl
+application/vnd.iccprofile, icm
+application/vnd.iccprofile, icm
+application/vnd.lotus-1-2-3,
+application/vnd.ms-cab-compressed, cab
 application/vnd.ms-excel, xlb|xlc|xll|xlm|xls|xlw
 application/vnd.ms-fontobject, eot
 application/vnd.ms-opentype, otf
@@ -54,45 +63,75 @@ application/vnd.ms-project, mpp
 application/vnd.oasis.opendocument.base, odb
 application/vnd.oasis.opendocument.formula, odf
 application/vnd.oasis.opendocument.graphics, odg
+application/vnd.oasis.opendocument.presentation, odp
+application/vnd.oasis.opendocument.spreadsheet, ods
 application/vnd.oasis.opendocument.text, odt
+application/vnd.openxmlformats-officedocument.presentationml.presentation, pptx
+application/vnd.openxmlformats-officedocument.spreadsheetml.sheet, xlsx
+application/vnd.openxmlformats-officedocument.wordprocessingml.document, docx
+application/vnd.symbian.install,
+application/vnd.tcpdump.pcap, pcap
 application/vnd.wap.wmlc, wmlc
 application/vnd.wap.wmlscriptc, wmlsc
 application/vnd.xara, web
 application/vocaltec-media-desc, vmd
 application/vocaltec-media-file, vmf
+application/warc, warc
+application/winhelp, hlp
 application/wordperfect6.0, w60
 application/wordperfect6.1, w61
 application/wordperfect, wp|wp5|wp6|wpd
 application/x-123, wk1
+application/x-7z-compressed, 7z
 application/x-aim, aim
+application/x-apple-diskimage,
+application/x-arc,
+application/x-archive, a
+application/x-atari-7800-rom, a78
 application/x-authorware-bin, aab
 application/x-authorware-map, aam
 application/x-authorware-seg, aas
+application/x-avira-qua,
 application/x-bcpio, bcpio
 application/x-bittorrent, torrent
 application/x-bsh, bsh
 application/x-bytecode.python, pyc
 application/x-bzip2, boz|bz2
 application/x-bzip, bz
+application/x-cbr, cbr
+application/x-cbz, cbz
 application/x-cdlink, vcd
 application/x-chat, cha|chat
+application/x-chrome-extension,
 application/x-cocoa, cco
 application/x-conference, nsc
+application/x-coredump,
 application/x-cpio, cpio
 application/x-dbf, dbf
 application/x-dbt,
+application/x-debian-package, deb
 application/x-deepv, deepv
-application/x-director, dcr|dir|dxr
+application/x-director, dir|dxr
+application/x-dmp, dmp
+application/x-dosdriver,
 application/x-dosexec, dll
 application/x-dvi, dvi
 application/x-elc, elc
+application/x-empty,
 application/x-envoy, env|evy
 application/x-esrehber, es
 application/x-excel, xla|xld|xlk|xlt|xlv
 application/x-executable, exe
+application/x-font-gdos,
+application/x-font-pf2, pf2
+application/x-font-pfm, pfm
 application/x-font-sfn,
-application/x-font-ttf, ttf
+application/x-font-ttf, ttf|ttc
+application/x-fptapplication/x-dbt,
 application/x-freelance, pre
+application/x-gamecube-rom,
+application/x-gdbm,
+application/x-gettext-translation,
 application/x-git,
 application/x-gsp, gsp
 application/x-gss, gss
@@ -102,46 +141,68 @@ application/x-hdf, hdf
 application/x-helpfile, help
 application/x-httpd-imap, imap
 application/x-ima, ima
+application/x-innosetup,
 application/x-internett-signup, ins
 application/x-inventor, iv
 application/x-ip2, ip
 application/x-java-applet,
 application/x-java-commerce, jcm
 application/x-java-image,
+application/x-java-jmod, jmod
 application/x-java-keystore,
+application/x-kdelnk,
 application/x-koan, skd|skm|skp|skt
 application/x-latex, latex|ltx
 application/x-livescreen, ivy
 application/x-lotus, wq1
+application/x-lz4+json, jsonlz4
+application/x-lz4, lz4
+application/x-lz4, lz4
+application/x-lzh-compressed,
 application/x-lzh, lzh
+application/x-lzip, lz
+application/x-lzma, lzma
+application/x-lzop, lzo
 application/x-lzx, lzx
 application/x-mach-binary, jnilib|dylib
 application/x-mach-executable,
 application/x-magic-cap-package-1.0, mc$
 application/x-mathcad, mcd
+application/x-maxis-dbpf,
 application/x-meme, mm
 application/x-midi, midi
 application/x-mif, mif
 application/x-mix-transfer, nix
 application/xml, opf
+application/x-mobipocket-ebook, mobi
+application/vnd.amazon.mobi8-ebook, azw|azw3
+application/x-msaccess, accdb
+application/x-ms-compress-szdd, fon
 application/x-ms-pdb, pdb
+application/x-ms-reader, lit
+application/x-n64-rom, z64
 application/x-navi-animation, ani
 application/x-navidoc, nvd
 application/x-navimap, map
 application/x-navistyle, stl
+application/x-nes-rom, nes
 application/x-netcdf, cdf|nc
 application/x-newton-compatible-pkg, pkg
+application/x-nintendo-ds-rom,
 application/x-object, o
 application/x-omcdatamaker, omcd
 application/x-omc, omc
 application/x-omcregerator, omcr
 application/x-pagemaker, pm4|pm5
 application/x-pcl, pcl
+application/x-pgp-keyring,
 application/x-pixclscript, plx
 application/x-pkcs7-certreqresp, p7r
 application/x-pkcs7-signature, p7a
 application/x-project, mpc|mpt|mpv|mpx
 application/x-qpro, wb1
+application/x-rar, rar
+application/x-rpm, rpm
 application/x-sdp, sdp
 application/x-sea, sea
 application/x-seelogo, sl
@@ -149,12 +210,17 @@ application/x-setupscript,
 application/x-sharedlib, so
 application/x-shar, shar
 application/x-shockwave-flash, swf
+application/x-snappy-framed,
 application/x-sprite, spr|sprite
 application/x-sqlite3,
+application/x-stargallery-thm,
+application/x-stuffit, sit
 application/x-sv4cpio, sv4cpio
 application/x-sv4crc, sv4crc
 application/x-tar, tar
 application/x-tbook, sbk|tbk
+application/x-terminfo,
+application/x-terminfo2,
 application/x-texinfo, texi|texinfo
 application/x-tex-tfm, tfm
 application/x-ustar, ustar
@@ -163,16 +229,22 @@ application/x-vnd.audioexplosion.mzz, mzz
 application/x-vnd.ls-xpix, xpix
 application/x-vrml, vrml
 application/x-wais-source, src|wsrc
+application/x-wine-extension-ini,
 application/x-wintalk, wtk
 application/x-world, svr
 application/x-wri, wri
 application/x-x509-ca-cert, der
 application/x-xz, xz
+application/x-zip,
+application/x-zstd, zst
 application/zip, zip
+application/zlib, z
+!audio/basic, au
 audio/it, it
 audio/make, funk|my|pfunk
 audio/midi, kar
 audio/mid, rmi
+audio/mp4, m4b
 audio/mpeg, m2a|mpa
 audio/ogg, ogg
 audio/s3m, s3m
@@ -180,7 +252,10 @@ audio/tsp-audio, tsi
 audio/tsplayer, tsp
 audio/vnd.qcelp, qcp
 audio/voxware, vox
+audio/x-aiff, aiff|aif
+audio/x-flac, flac
 audio/x-gsm, gsd|gsm
+audio/x-hx-aac-adts,
 audio/x-jam, jam
 audio/x-liveaudio, lam
 audio/x-m4a, m4a
@@ -194,17 +269,24 @@ audio/x-nspaudio, lma
 audio/x-pn-realaudio, ram|rm|rmm|rmp
 audio/x-psid, sid
 audio/x-realaudio, ra
+audio/x-s3m,
 audio/x-twinvq-plugin, vqe|vql
 audio/x-twinvq, vqf
 audio/x-voc, voc
 audio/x-wav, wav
+!audio/x-xbox360-executable, xex
+!audio/x-xbox-executable, xbe
 font/otf,
 font/sfnt,
+font/woff2, woff2
+font/woff, woff
+image/bmp,
 image/cmu-raster, rast
 image/fif, fif
 image/florian, flo|turbot
 image/g3fax, g3
 image/gif, gif
+image/heic, heic
 image/ief, ief|iefs
 image/jpeg, jfif|jfif-tbnl|jpe|jpeg|jpg
 image/jutvision, jut
@@ -213,6 +295,9 @@ image/pict, pic|pict
 image/png, png|x-png
 !image/svg, svg
 !image/svg+xml,
+image/tiff,
+!image/vnd.adobe.photoshop, psd
+!image/vnd.djvu, djvu
 image/vnd.fpx, fpx
 image/vnd.microsoft.icon,
 image/vnd.rn-realflash, rf
@@ -220,9 +305,15 @@ image/vnd.rn-realpix, rp
 image/vnd.wap.wbmp, wbmp
 image/vnd.xiff, xif
 image/webp, webp
+image/wmf,
+image/x-3ds, 3ds
+image/x-award-bioslogo,
 image/x-cmu-raster, ras
+image/x-cur, tga
 image/x-dwg, dwg|dxf|svf
 image/x-eps,
+image/x-exr, exr
+image/x-gem,
 image/x-icns,
 !image/x-icon, ico
 image/x-jg, art
@@ -236,34 +327,33 @@ image/x-portable-graymap, pgm
 image/x-portable-pixmap, ppm
 image/x-quicktime, qif|qti|qtif
 image/x-rgb, rgb
+image/x-tga,
 image/x-tiff, tif|tiff
-image/tiff,
+image/x-win-bitmap,
 !image/x-xcf, xcf
 !image/x-xpixmap, xpm
+image/x-xwindowdump, xwd
+message/news,
 message/rfc822, mht|mhtml|mime
 model/vnd.dwf, dwf
+model/vnd.gdl, gdl
+model/vnd.gs.gdl, gdsl
 model/vrml, wrz
 model/x-pov, pov
 text/asp, asp
 text/css, css
-text/x-sass, sass
-text/x-scss, scss
 text/html, acgi|htm|html|htmls|htx|shtml
 text/javascript, js
 text/mcf, mcf
 text/pascal, pas
-text/plain, com|cmd|conf|def|g|idc|list|lst|mar|sdml|text|txt|md|groovy|license|properties|desktop|ini|rst|cmake|ipynb|readme|less|lo|go|yml|d|cs|hpp|srt
+text/PGP,
+text/plain, com|cmd|conf|def|g|idc|list|lst|mar|sdml|text|txt|md|groovy|license|properties|desktop|ini|rst|cmake|ipynb|readme|less|lo|go|yml|d|cs|hpp|srt|nfo|sfv|m3u|csv|eml
 text/richtext, rt|rtf|rtx
+text/rtf,
 text/scriplet, wsc
-text/x-awk, awk
-!video/x-jng, jng
-video/x-mng, mng
-image/x-cur, tga
-image/x-xwindowdump, xwd
-!image/vnd.adobe.photoshop, psd
 text/tab-separated-values, tsv
 text/troff, man|me|ms|roff|t|tr
-text/uri-list, uni|unis|uri|uris
+text/uri-list, uji|unis|uri|uris
 text/vnd.abc, abc
 text/vnd.fmi.flexstor, flx
 text/vnd.wap.wmlscript, wmls
@@ -272,6 +362,7 @@ text/webviewhtml, htt
 text/x-Algol68,
 text/x-asm, asm|s
 text/x-audiosoft-intra, aip
+text/x-awk, awk
 text/x-bcpl,
 text/x-c, c|cc|h
 text/x-c++, cpp|cxx|c++
@@ -286,23 +377,31 @@ text/x-makefile, am|mak
 text/xml, xml|pom|iml|plist
 text/x-m, m
 text/x-msdos-batch, bat
+text/x-ms-regedit, reg
+text/x-objective-c,
 text/x-pascal, p
 text/x-perl, pl
 text/x-php, php
+text/x-po, po
 text/x-python, py
 text/x-ruby, rb
+text/x-sass, sass
+text/x-scss, scss
 text/x-server-parsed-html, ssi
 text/x-setext, etx
 text/x-sgml, sgm|sgml
 text/x-shellscript, sh
 text/x-speech, talk
+text/x-tcl,
 text/x-tex, tex
 text/x-uil, uil
 text/x-uuencode, uue
 text/x-vcalendar, vcs
+text/x-vcard, vcf
 video/animaflex, afl
 video/avi, avi
 video/avs-video, avs
+video/MP2T,
 video/mp4, mp4
 video/mpeg, m1v|m2v|mpe|mpeg|mpg
 video/quicktime, moov|mov|qt
@@ -317,43 +416,35 @@ video/x-atomic3d-feature, fmf
 video/x-dl, dl
 video/x-dv, dif|dv
 video/x-fli, fli
+video/x-flv, flv
 video/x-isvideo, isu
+!video/x-jng, jng
+video/x-m4v, m4v
+video/x-matroska, mkv
+video/x-mng, mng
 video/x-motion-jpeg, mjpg
-video/x-ms-asf, asf|asx
+video/x-ms-asf, asf|asx|wmv
+video/x-msvideo, divx
 video/x-qtc, qtc
 video/x-sgi-movie, movie|mv
-application/x-7z-compressed, 7z
-application/vnd.openxmlformats-officedocument.wordprocessingml.document, docx
-text/x-po, po
-application/x-rpm, rpm
-application/x-debian-package, deb
-application/vnd.iccprofile, icm
-application/dicom, dcm
-image/x-exr, exr
-application/vnd.iccprofile, icm
-video/x-matroska, mkv
-application/x-empty,
-model/vnd.gdl, gdl
-model/vnd.gs.gdl, gdsl
-font/woff, woff
-font/woff2, woff2
-application/epub+zip, epub
-application/x-mobipocket-ebook, mobi
-audio/x-flac, flac
-application/x-rar, rar
-video/x-msvideo, divx
-video/x-flv, flv
-application/x-kdelnk,
-text/x-tcl,
-application/ogg, ogv
-application/vnd.openxmlformats-officedocument.spreadsheetml.sheet, xlsx
-application/vnd.ms-cab-compressed, cab
-audio/mp4, m4b
-!image/vnd.djvu, djvu
-application/x-ms-reader, lit
-application/CDFV2-corrupt,
-text/x-vcard, vcf
-application/x-innosetup,
-application/winhelp, hlp
-image/x-tga,
-application/x-wine-extension-ini,
+x-epoc/x-sisx-app,
+application/x-zstd-dictionary,
+application/vnd.ms-outlook, msg
+image/x-olympus-orf, orf
+image/x-nikon-nef, nef
+image/x-fuji-raf, raf
+image/x-panasonic-raw, rw2|raw
+image/x-adobe-dng, dng
+image/x-canon-cr2, cr2
+image/x-canon-crw, crw
+image/x-dcraw,
+image/x-kodak-dcr, dcr
+image/x-kodak-k25, k25
+image/x-kodak-kdc, kdc
+image/x-minolta-mrw, mrw
+image/x-pentax-pef, pef
+image/x-sigma-x3f, xf3
+image/x-sony-arw, arw
+image/x-sony-sr2, sr2
+image/x-sony-srf, srf
+image/x-epson-erf, erf

scripts/mime.py

@@ -12,18 +12,19 @@ major_mime = {
     "audio": 7,
     "image": 8,
     "text": 9,
-    "application": 10
+    "application": 10,
+    "x-epoc": 11,
 }
 
 pdf = (
     "application/pdf",
-    "application/x-cbr",
-    "application/x-cbz",
+    "application/epub+zip",
     "application/vnd.ms-xpsdocument",
 )
 
 font = (
     "application/vnd.ms-opentype",
+    "application/x-ms-compress-szdd"
     "application/x-font-sfn",
     "application/x-font-ttf",
     "font/otf",
@@ -32,6 +33,68 @@ font = (
     "font/woff2"
 )
 
+# Archive "formats"
+archive = (
+    "application/x-tar",
+    "application/zip",
+    "application/x-rar",
+    "application/x-arc",
+    "application/x-warc",
+    "application/x-7z-compressed",
+)
+
+# Archive "filters"
+arc_filter = (
+    "application/gzip",
+    "application/x-bzip2",
+    "application/x-xz",
+    "application/x-zstd",
+    "application/x-lzma",
+    "application/x-lz4",
+    "application/x-lzip",
+    "application/x-lzop",
+)
+
+doc = (
+    "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
+    "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+    "application/vnd.openxmlformats-officedocument.presentationml.presentation"
+)
+
+mobi = (
+    "application/x-mobipocket-ebook",
+    "application/vnd.amazon.mobi8-ebook"
+)
+
+markup = (
+    "text/xml",
+    "text/html",
+    "text/x-sgml"
+)
+
+raw = (
+    "image/x-olympus-orf",
+    "image/x-nikon-nef",
+    "image/x-fuji-raf",
+    "image/x-panasonic-raw",
+    "image/x-adobe-dng",
+    "image/x-canon-cr2",
+    "image/x-canon-crw",
+    "image/x-dcraw",
+    "image/x-kodak-dcr",
+    "image/x-kodak-k25",
+    "image/x-kodak-kdc",
+    "image/x-minolta-mrw",
+    "image/x-pentax-pef",
+    "image/x-sigma-x3f",
+    "image/x-sony-arw",
+    "image/x-sony-sr2",
+    "image/x-sony-srf",
+    "image/x-minolta-mrw",
+    "image/x-pentax-pef",
+    "image/x-epson-erf",
+)
+
 cnt = 1
 
 
@@ -46,6 +109,18 @@ def mime_id(mime):
         mime_id += " | 0x40000000"
     elif mime in font:
         mime_id += " | 0x20000000"
+    elif mime in archive:
+        mime_id += " | 0x10000000"
+    elif mime in arc_filter:
+        mime_id += " | 0x08000000"
+    elif mime in doc:
+        mime_id += " | 0x04000000"
+    elif mime in mobi:
+        mime_id += " | 0x02000000"
+    elif mime in markup:
+        mime_id += " | 0x01000000"
+    elif mime in raw:
+        mime_id += " | 0x00800000"
     elif mime == "application/x-empty":
         return "1"
     return mime_id
@@ -55,7 +130,7 @@ def clean(t):
     return t.replace("/", "_").replace(".", "_").replace("+", "_").replace("-", "_")
 
 
-with open("mime.csv") as f:
+with open("scripts/mime.csv") as f:
     for l in f:
         mime, ext_list = l.split(",")
         if l.startswith("!"):
@@ -67,11 +142,11 @@ with open("mime.csv") as f:
     print("// **Generated by mime.py**")
     print("#ifndef MIME_GENERATED_C")
     print("#define MIME_GENERATED_C")
-    print("#include <glib-2.0/glib.h>\n")
+    print("#include <glib.h>\n")
     print("#include <stdlib.h>\n")
     # Enum
     print("enum mime {")
-    for mime, ext in mimes.items():
+    for mime, ext in sorted(mimes.items()):
         print("    " + clean(mime) + "=" + mime_id(mime) + ",")
     print("};")
 

scripts/serve_static.py

@@ -1,9 +1,12 @@
 files = [
-    "web/css/bundle.css",
-    "web/js/bundle.js",
-    "web/img/bg-bars.png",
-    "web/img/sprite-skin-flat.png",
-    "web/search.html",
+    "src/static/css/bundle.css",
+    "src/static/css/bundle_dark.css",
+    "src/static/js/bundle.js",
+    "src/static/js/search.js",
+    "src/static/img/sprite-skin-flat.png",
+    "src/static/img/sprite-skin-flat-dark.png",
+    "src/static/search.html",
+    "src/static/stats.html",
 ]
 
 

src/cli.c

@@ -1,30 +1,70 @@
 #include "cli.h"
+#include "ctx.h"
+#include <tesseract/capi.h>
 
 #define DEFAULT_OUTPUT "index.sist2/"
-#define DEFAULT_CONTENT_SIZE 4096
-#define DEFAULT_QUALITY 15
-#define DEFAULT_SIZE 200
+#define DEFAULT_CONTENT_SIZE 32768
+#define DEFAULT_QUALITY 5
+#define DEFAULT_SIZE 500
 #define DEFAULT_REWRITE_URL ""
 
 #define DEFAULT_ES_URL "http://localhost:9200"
+#define DEFAULT_BATCH_SIZE 100
 
-#define DEFAULT_BIND_ADDR "localhost"
-#define DEFAULT_PORT "4090"
+#define DEFAULT_LISTEN_ADDRESS "localhost:4090"
+#define DEFAULT_TREEMAP_THRESHOLD 0.0005
+
+#define DEFAULT_MAX_MEM_BUFFER 2000
+
+const char *TESS_DATAPATHS[] = {
+        "/usr/share/tessdata/",
+        "/usr/share/tesseract-ocr/tessdata/",
+        "./",
+        NULL
+};
 
 
 scan_args_t *scan_args_create() {
     scan_args_t *args = calloc(sizeof(scan_args_t), 1);
+
+    args->depth = -1;
+
     return args;
 }
 
-index_args_t *index_args_create() {
-    index_args_t *args = calloc(sizeof(index_args_t), 1);
+exec_args_t *exec_args_create() {
+    exec_args_t *args = calloc(sizeof(exec_args_t), 1);
     return args;
 }
 
-web_args_t *web_args_create() {
-    web_args_t *args = calloc(sizeof(web_args_t), 1);
-    return args;
+void scan_args_destroy(scan_args_t *args) {
+    if (args->name != NULL) {
+        free(args->name);
+    }
+    if (args->incremental != NULL) {
+        free(args->incremental);
+    }
+    if (args->path != NULL) {
+        free(args->path);
+    }
+    if (args->output != NULL) {
+        free(args->output);
+    }
+    free(args);
+}
+
+void index_args_destroy(index_args_t *args) {
+    //todo
+    free(args);
+}
+
+void web_args_destroy(web_args_t *args) {
+    //todo
+    free(args);
+}
+
+void exec_args_destroy(exec_args_t *args) {
+    free(args);
 }
 
 int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
@@ -35,17 +75,17 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
 
     char *abs_path = abspath(argv[1]);
     if (abs_path == NULL) {
-        fprintf(stderr, "File not found: %s", argv[1]);
+        fprintf(stderr, "File not found: %s\n", argv[1]);
         return 1;
     } else {
         args->path = abs_path;
     }
 
     if (args->incremental != NULL) {
-        abs_path = abspath(args->incremental);
+        args->incremental = abspath(args->incremental);
         if (abs_path == NULL) {
-            fprintf(stderr, "File not found: %s", args->incremental);
-            return 1;
+            sist_log("main.c", SIST_WARNING, "Could not open original index! Disabled incremental scan feature.");
+            args->incremental = NULL;
         }
     }
 
@@ -58,16 +98,13 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
 
     if (args->size == 0) {
         args->size = DEFAULT_SIZE;
-    } else if (args->size <= 0) {
-        fprintf(stderr, "Invalid size: %d\n", args->size);
+    } else if (args->size > 0 && args->size < 32) {
+        printf("Invalid size: %d\n", args->content_size);
         return 1;
     }
 
     if (args->content_size == 0) {
         args->content_size = DEFAULT_CONTENT_SIZE;
-    } else if (args->content_size <= 0) {
-        fprintf(stderr, "Invalid content-size: %d\n", args->content_size);
-        return 1;
     }
 
     if (args->threads == 0) {
@@ -90,6 +127,12 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
         return 1;
     }
 
+    if (args->depth <= 0) {
+        args->depth = G_MAXINT32;
+    } else {
+        args->depth += 1;
+    }
+
     if (args->name == NULL) {
         args->name = g_path_get_basename(args->output);
     }
@@ -97,32 +140,174 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
     if (args->rewrite_url == NULL) {
         args->rewrite_url = DEFAULT_REWRITE_URL;
     }
+
+    if (args->archive == NULL || strcmp(args->archive, "recurse") == 0) {
+        args->archive_mode = ARC_MODE_RECURSE;
+    } else if (strcmp(args->archive, "list") == 0) {
+        args->archive_mode = ARC_MODE_LIST;
+    } else if (strcmp(args->archive, "shallow") == 0) {
+        args->archive_mode = ARC_MODE_SHALLOW;
+    } else if (strcmp(args->archive, "skip") == 0) {
+        args->archive_mode = ARC_MODE_SKIP;
+    } else {
+        fprintf(stderr, "Archive mode must be one of (skip, list, shallow, recurse), got '%s'", args->archive);
+        return 1;
+    }
+
+    if (args->tesseract_lang != NULL) {
+        TessBaseAPI *api = TessBaseAPICreate();
+
+        char filename[128];
+        sprintf(filename, "%s.traineddata", args->tesseract_lang);
+        const char *path = find_file_in_paths(TESS_DATAPATHS, filename);
+        if (path == NULL) {
+            LOG_FATAL("cli.c", "Could not find tesseract language file!");
+        }
+
+        ret = TessBaseAPIInit3(api, path, args->tesseract_lang);
+        if (ret != 0) {
+            fprintf(stderr, "Could not initialize tesseract with lang '%s'\n", args->tesseract_lang);
+            return 1;
+        }
+        TessBaseAPIEnd(api);
+        TessBaseAPIDelete(api);
+
+        args->tesseract_path = path;
+    }
+
+    if (args->exclude_regex != NULL) {
+        const char *error;
+        int error_offset;
+
+        pcre *re = pcre_compile(args->exclude_regex, 0, &error, &error_offset, 0);
+        if (error != NULL) {
+            LOG_FATALF("cli.c", "pcre_compile returned error: %s (offset:%d)", error, error_offset)
+        }
+
+        pcre_extra *re_extra = pcre_study(re, 0, &error);
+        if (error != NULL) {
+            LOG_FATALF("cli.c", "pcre_study returned error: %s", error)
+        }
+
+        ScanCtx.exclude = re;
+        ScanCtx.exclude_extra = re_extra;
+    } else {
+        ScanCtx.exclude = NULL;
+    }
+
+    if (args->treemap_threshold_str == 0) {
+        args->treemap_threshold = DEFAULT_TREEMAP_THRESHOLD;
+    } else {
+        args->treemap_threshold = atof(args->treemap_threshold_str);
+    }
+
+    if (args->max_memory_buffer == 0) {
+        args->max_memory_buffer = DEFAULT_MAX_MEM_BUFFER;
+    }
+
+    LOG_DEBUGF("cli.c", "arg quality=%f", args->quality)
+    LOG_DEBUGF("cli.c", "arg size=%d", args->size)
+    LOG_DEBUGF("cli.c", "arg content_size=%d", args->content_size)
+    LOG_DEBUGF("cli.c", "arg threads=%d", args->threads)
+    LOG_DEBUGF("cli.c", "arg incremental=%s", args->incremental)
+    LOG_DEBUGF("cli.c", "arg output=%s", args->output)
+    LOG_DEBUGF("cli.c", "arg rewrite_url=%s", args->rewrite_url)
+    LOG_DEBUGF("cli.c", "arg name=%s", args->name)
+    LOG_DEBUGF("cli.c", "arg depth=%d", args->depth)
+    LOG_DEBUGF("cli.c", "arg path=%s", args->path)
+    LOG_DEBUGF("cli.c", "arg archive=%s", args->archive)
+    LOG_DEBUGF("cli.c", "arg tesseract_lang=%s", args->tesseract_lang)
+    LOG_DEBUGF("cli.c", "arg tesseract_path=%s", args->tesseract_path)
+    LOG_DEBUGF("cli.c", "arg exclude=%s", args->exclude_regex)
+    LOG_DEBUGF("cli.c", "arg fast=%d", args->fast)
+    LOG_DEBUGF("cli.c", "arg treemap_threshold=%f", args->treemap_threshold)
+    LOG_DEBUGF("cli.c", "arg max_memory_buffer=%d", args->max_memory_buffer)
+
+    return 0;
+}
+
+int load_script(const char *script_path, char **dst) {
+    struct stat info;
+    int res = stat(script_path, &info);
+
+    if (res == -1) {
+        fprintf(stderr, "Error opening script file '%s': %s\n", script_path, strerror(errno));
+        return 1;
+    }
+
+    int fd = open(script_path, O_RDONLY);
+    if (fd == -1) {
+        fprintf(stderr, "Error opening script file '%s': %s\n", script_path, strerror(errno));
+        return 1;
+    }
+
+    *dst = malloc(info.st_size + 1);
+    res = read(fd, *dst, info.st_size);
+    if (res < 0) {
+        fprintf(stderr, "Error reading script file '%s': %s\n", script_path, strerror(errno));
+        return 1;
+    }
+
+    *(*dst + info.st_size) = '\0';
+    close(fd);
+
     return 0;
 }
 
 int index_args_validate(index_args_t *args, int argc, const char **argv) {
 
+    LogCtx.verbose = 1;
+
     if (argc < 2) {
         fprintf(stderr, "Required positional argument: PATH.\n");
         return 1;
     }
 
+    if (args->threads == 0) {
+        args->threads = 1;
+    } else if (args->threads < 0) {
+        fprintf(stderr, "Invalid threads: %d\n", args->threads);
+        return 1;
+    }
+
     char *index_path = abspath(argv[1]);
     if (index_path == NULL) {
-        fprintf(stderr, "File not found: %s", argv[1]);
+        fprintf(stderr, "File not found: %s\n", argv[1]);
         return 1;
     } else {
         args->index_path = argv[1];
+        free(index_path);
     }
 
     if (args->es_url == NULL) {
         args->es_url = DEFAULT_ES_URL;
     }
+
+    if (args->script_path != NULL) {
+        if (load_script(args->script_path, &args->script) != 0) {
+            return 1;
+        }
+    }
+
+    if (args->batch_size == 0) {
+        args->batch_size = DEFAULT_BATCH_SIZE;
+    }
+
+    LOG_DEBUGF("cli.c", "arg es_url=%s", args->es_url)
+    LOG_DEBUGF("cli.c", "arg index_path=%s", args->index_path)
+    LOG_DEBUGF("cli.c", "arg script_path=%s", args->script_path)
+    LOG_DEBUGF("cli.c", "arg script=%s", args->script)
+    LOG_DEBUGF("cli.c", "arg print=%d", args->print)
+    LOG_DEBUGF("cli.c", "arg batch_size=%d", args->batch_size)
+    LOG_DEBUGF("cli.c", "arg force_reset=%d", args->force_reset)
+
     return 0;
 }
 
 int web_args_validate(web_args_t *args, int argc, const char **argv) {
 
+    LogCtx.verbose = 1;
+
     if (argc < 2) {
         fprintf(stderr, "Required positional argument: PATH.\n");
         return 1;
@@ -132,12 +317,53 @@ int web_args_validate(web_args_t *args, int argc, const char **argv) {
         args->es_url = DEFAULT_ES_URL;
     }
 
-    if (args->bind == NULL) {
-        args->bind = DEFAULT_BIND_ADDR;
+    if (args->listen_address == NULL) {
+        args->listen_address = DEFAULT_LISTEN_ADDRESS;
     }
 
-    if (args->port == NULL) {
-        args->port = DEFAULT_PORT;
+    if (args->credentials != NULL) {
+        char *ptr = strstr(args->credentials, ":");
+        if (ptr == NULL) {
+            fprintf(stderr, "Invalid --auth format, see usage\n");
+            return 1;
+        }
+
+        strncpy(args->auth_user, args->credentials, (ptr - args->credentials));
+        strcpy(args->auth_pass, ptr + 1);
+
+        if (strlen(args->auth_user) == 0) {
+            fprintf(stderr, "--auth username must be at least one character long");
+            return 1;
+        }
+
+        args->auth_enabled = TRUE;
+    } else {
+        args->auth_enabled = FALSE;
+    }
+
+    if (args->tag_credentials != NULL && args->credentials != NULL) {
+        fprintf(stderr, "--auth and --tag-auth are mutually exclusive");
+        return 1;
+    }
+
+    if (args->tag_credentials != NULL) {
+        char *ptr = strstr(args->tag_credentials, ":");
+        if (ptr == NULL) {
+            fprintf(stderr, "Invalid --tag-auth format, see usage\n");
+            return 1;
+        }
+
+        strncpy(args->auth_user, args->tag_credentials, (ptr - args->tag_credentials));
+        strcpy(args->auth_pass, ptr + 1);
+
+        if (strlen(args->auth_user) == 0) {
+            fprintf(stderr, "--tag-auth username must be at least one character long");
+            return 1;
+        }
+
+        args->tag_auth_enabled = TRUE;
+    } else {
+        args->tag_auth_enabled = FALSE;
     }
 
     args->index_count = argc - 1;
@@ -146,10 +372,64 @@ int web_args_validate(web_args_t *args, int argc, const char **argv) {
     for (int i = 0; i < args->index_count; i++) {
         char *abs_path = abspath(args->indices[i]);
         if (abs_path == NULL) {
-            fprintf(stderr, "File not found: %s", abs_path);
+            fprintf(stderr, "File not found: %s\n", args->indices[i]);
             return 1;
         }
     }
+
+    LOG_DEBUGF("cli.c", "arg es_url=%s", args->es_url)
+    LOG_DEBUGF("cli.c", "arg listen=%s", args->listen_address)
+    LOG_DEBUGF("cli.c", "arg credentials=%s", args->credentials)
+    LOG_DEBUGF("cli.c", "arg tag_credentials=%s", args->tag_credentials)
+    LOG_DEBUGF("cli.c", "arg auth_user=%s", args->auth_user)
+    LOG_DEBUGF("cli.c", "arg auth_pass=%s", args->auth_pass)
+    LOG_DEBUGF("cli.c", "arg index_count=%d", args->index_count)
+    for (int i = 0; i < args->index_count; i++) {
+        LOG_DEBUGF("cli.c", "arg indices[%d]=%s", i, args->indices[i])
+    }
+
     return 0;
 }
 
+index_args_t *index_args_create() {
+    index_args_t *args = calloc(sizeof(index_args_t), 1);
+    return args;
+}
+
+web_args_t *web_args_create() {
+    web_args_t *args = calloc(sizeof(web_args_t), 1);
+    return args;
+}
+
+int exec_args_validate(exec_args_t *args, int argc, const char **argv) {
+
+    if (argc < 2) {
+        fprintf(stderr, "Required positional argument: PATH.\n");
+        return 1;
+    }
+
+    char *index_path = abspath(argv[1]);
+    if (index_path == NULL) {
+        fprintf(stderr, "File not found: %s\n", argv[1]);
+        return 1;
+    } else {
+        args->index_path = argv[1];
+        free(index_path);
+    }
+
+    if (args->es_url == NULL) {
+        args->es_url = DEFAULT_ES_URL;
+    }
+
+    if (args->script_path == NULL) {
+        LOG_FATAL("cli.c", "--script-file argument is required");
+    }
+
+    if (load_script(args->script_path, &args->script) != 0) {
+        return 1;
+    }
+
+    LOG_DEBUGF("cli.c", "arg script_path=%s", args->script_path)
+    LOG_DEBUGF("cli.c", "arg script=%s", args->script)
+    return 0;
+}

src/cli.h

@@ -3,6 +3,8 @@
 
 #include "sist.h"
 
+#include "libscan/arc/arc.h"
+
 typedef struct scan_args {
     float quality;
     int size;
@@ -12,30 +14,72 @@ typedef struct scan_args {
     char *output;
     char *rewrite_url;
     char *name;
+    int depth;
     char *path;
+    char *archive;
+    archive_mode_t archive_mode;
+    char *tesseract_lang;
+    const char *tesseract_path;
+    char *exclude_regex;
+    int fast;
+    const char* treemap_threshold_str;
+    double treemap_threshold;
+    int max_memory_buffer;
 } scan_args_t;
 
+scan_args_t *scan_args_create();
+
+void scan_args_destroy(scan_args_t *args);
+
+int scan_args_validate(scan_args_t *args, int argc, const char **argv);
+
 typedef struct index_args {
     char *es_url;
     const char *index_path;
+    const char *script_path;
+    char *script;
     int print;
+    int batch_size;
     int force_reset;
+    int threads;
 } index_args_t;
 
 typedef struct web_args {
     char *es_url;
-    char *bind;
-    char *port;
+    char *listen_address;
+    char *credentials;
+    char *tag_credentials;
+    char auth_user[256];
+    char auth_pass[256];
+    int auth_enabled;
+    int tag_auth_enabled;
     int index_count;
     const char **indices;
 } web_args_t;
 
-scan_args_t *scan_args_create();
+typedef struct exec_args {
+    char *es_url;
+    const char *index_path;
+    const char *script_path;
+    char *script;
+} exec_args_t;
+
 index_args_t *index_args_create();
+
+void index_args_destroy(index_args_t *args);
+
 web_args_t *web_args_create();
 
-int scan_args_validate(scan_args_t *args, int argc, const char **argv);
+void web_args_destroy(web_args_t *args);
+
 int index_args_validate(index_args_t *args, int argc, const char **argv);
+
 int web_args_validate(web_args_t *args, int argc, const char **argv);
 
+exec_args_t *exec_args_create();
+
+void exec_args_destroy(exec_args_t *args);
+
+int exec_args_validate(exec_args_t *args, int argc, const char **argv);
+
 #endif

src/ctx.c

@@ -0,0 +1,6 @@
+#include "ctx.h"
+
+ScanCtx_t ScanCtx;
+WebCtx_t WebCtx;
+IndexCtx_t IndexCtx;
+LogCtx_t LogCtx;

src/ctx.h

@@ -2,8 +2,23 @@
 #define SIST2_CTX_H
 
 #include "sist.h"
+#include "tpool.h"
+#include "libscan/scan.h"
+#include "libscan/arc/arc.h"
+#include "libscan/comic/comic.h"
+#include "libscan/ebook/ebook.h"
+#include "libscan/font/font.h"
+#include "libscan/media/media.h"
+#include "libscan/ooxml/ooxml.h"
+#include "libscan/text/text.h"
+#include "libscan/mobi/scan_mobi.h"
+#include "libscan/raw/raw.h"
+#include "src/io/store.h"
 
-struct {
+#include <glib.h>
+#include <pcre.h>
+
+typedef struct {
     struct index_t index;
 
     GHashTable *mime_table;
@@ -11,10 +26,8 @@ struct {
 
     tpool_t *pool;
 
-    int tn_size;
     int threads;
-    int content_size;
-    float tn_qscale;
+    int depth;
 
     size_t stat_tn_size;
     size_t stat_index_size;
@@ -22,20 +35,49 @@ struct {
     GHashTable *original_table;
     GHashTable *copy_table;
 
-    pthread_mutex_t mupdf_mu;
-} ScanCtx;
+    pcre *exclude;
+    pcre_extra *exclude_extra;
+    int fast;
 
+    scan_arc_ctx_t arc_ctx;
+    scan_comic_ctx_t comic_ctx;
+    scan_ebook_ctx_t ebook_ctx;
+    scan_font_ctx_t font_ctx;
+    scan_media_ctx_t media_ctx;
+    scan_ooxml_ctx_t ooxml_ctx;
+    scan_text_ctx_t text_ctx;
+    scan_mobi_ctx_t mobi_ctx;
+    scan_raw_ctx_t raw_ctx;
+} ScanCtx_t;
 
-struct {
+typedef struct {
+    int verbose;
+    int very_verbose;
+    int no_color;
+} LogCtx_t;
+
+typedef struct {
     char *es_url;
-} IndexCtx;
+    int batch_size;
+    tpool_t *pool;
+    store_t *tag_store;
+    GHashTable *tags;
+} IndexCtx_t;
 
-struct {
+typedef struct {
     char *es_url;
     int index_count;
-    struct index_t indices[16];
-} WebCtx;
+    char *auth_user;
+    char *auth_pass;
+    int auth_enabled;
+    int tag_auth_enabled;
+    struct index_t indices[64];
+} WebCtx_t;
 
+extern ScanCtx_t ScanCtx;
+extern WebCtx_t WebCtx;
+extern IndexCtx_t IndexCtx;
+extern LogCtx_t LogCtx;
 
 
 #endif

src/index/elastic.c

@@ -1,16 +1,10 @@
 #include "elastic.h"
 #include "src/ctx.h"
 
-#include <stdlib.h>
 #include "web.h"
-#include <stdio.h>
-#include <string.h>
-#include <cJSON/cJSON.h>
-#include <src/ctx.h>
 
 #include "static_generated.c"
 
-#define BULK_INDEX_SIZE 100
 
 typedef struct es_indexer {
     int queued;
@@ -20,7 +14,18 @@ typedef struct es_indexer {
 } es_indexer_t;
 
 
-static es_indexer_t *Indexer;
+static __thread es_indexer_t *Indexer;
+
+void delete_queue(int max);
+void elastic_flush();
+
+void elastic_cleanup() {
+    elastic_flush();
+    if (Indexer != NULL) {
+        free(Indexer->es_url);
+        free(Indexer);
+    }
+}
 
 void print_json(cJSON *document, const char uuid_str[UUID_STR_LEN]) {
 
@@ -29,17 +34,22 @@ void print_json(cJSON *document, const char uuid_str[UUID_STR_LEN]) {
     cJSON_AddStringToObject(line, "_id", uuid_str);
     cJSON_AddStringToObject(line, "_index", "sist2");
     cJSON_AddStringToObject(line, "_type", "_doc");
-    cJSON_AddItemToObject(line, "_source", document);
+    cJSON_AddItemReferenceToObject(line, "_source", document);
 
     char *json = cJSON_PrintUnformatted(line);
 
     printf("%s\n", json);
 
-    cJSON_free(line);
+    cJSON_free(json);
+    cJSON_Delete(line);
+}
+
+void index_json_func(void *arg) {
+    es_bulk_line_t *line = arg;
+    elastic_index_line(line);
 }
 
 void index_json(cJSON *document, const char uuid_str[UUID_STR_LEN]) {
-
     char *json = cJSON_PrintUnformatted(document);
 
     size_t json_len = strlen(json);
@@ -51,7 +61,188 @@ void index_json(cJSON *document, const char uuid_str[UUID_STR_LEN]) {
     bulk_line->next = NULL;
 
     cJSON_free(json);
-    elastic_index_line(bulk_line);
+    tpool_add_work(IndexCtx.pool, index_json_func, bulk_line);
+}
+
+void execute_update_script(const char *script, const char index_id[UUID_STR_LEN]) {
+
+    if (Indexer == NULL) {
+        Indexer = create_indexer(IndexCtx.es_url);
+    }
+
+    cJSON *body = cJSON_CreateObject();
+    cJSON *script_obj = cJSON_AddObjectToObject(body, "script");
+    cJSON_AddStringToObject(script_obj, "lang", "painless");
+    cJSON_AddStringToObject(script_obj, "source", script);
+
+    cJSON *query = cJSON_AddObjectToObject(body, "query");
+    cJSON *term_obj = cJSON_AddObjectToObject(query, "term");
+    cJSON_AddStringToObject(term_obj, "index", index_id);
+
+    char *str = cJSON_Print(body);
+
+    char bulk_url[4096];
+    snprintf(bulk_url, 4096, "%s/sist2/_update_by_query?wait_for_completion=false", Indexer->es_url);
+    response_t *r = web_post(bulk_url, str);
+    LOG_INFOF("elastic.c", "Executed user script <%d>", r->status_code);
+    cJSON *resp = cJSON_Parse(r->body);
+
+    cJSON_free(str);
+    cJSON_Delete(body);
+    free_response(r);
+
+    cJSON *error = cJSON_GetObjectItem(resp, "error");
+    if (error != NULL) {
+        char *error_str = cJSON_Print(error);
+
+        LOG_ERRORF("elastic.c", "User script error: \n%s", error_str);
+        cJSON_free(error_str);
+    }
+
+    cJSON_Delete(resp);
+}
+
+#define ACTION_STR_LEN 91
+
+void *create_bulk_buffer(int max, int *count, size_t *buf_len) {
+    es_bulk_line_t *line = Indexer->line_head;
+    *count = 0;
+
+    size_t buf_size = 0;
+    size_t buf_cur = 0;
+    char *buf = malloc(8192);
+    size_t buf_capacity = 8192;
+
+    while (line != NULL && *count < max) {
+        char action_str[256];
+        snprintf(action_str, 256,
+                 "{\"index\":{\"_id\":\"%s\", \"_type\":\"_doc\", \"_index\":\"sist2\"}}\n", line->uuid_str);
+
+        size_t line_len = strlen(line->line);
+
+        while (buf_size + line_len + ACTION_STR_LEN > buf_capacity) {
+            buf_capacity *= 2;
+            buf = realloc(buf, buf_capacity);
+        }
+
+        buf_size += line_len + ACTION_STR_LEN;
+
+        memcpy(buf + buf_cur, action_str, ACTION_STR_LEN);
+        buf_cur += ACTION_STR_LEN;
+        memcpy(buf + buf_cur, line->line, line_len);
+        buf_cur += line_len;
+
+        line = line->next;
+        (*count)++;
+    }
+
+    if (buf_size + 1 > buf_capacity) {
+        buf = realloc(buf, buf_capacity + 1);
+    }
+
+    *(buf + buf_cur) = '\0';
+
+    *buf_len = buf_cur;
+    return buf;
+}
+
+void print_errors(response_t *r) {
+    char *tmp = malloc(r->size + 1);
+    memcpy(tmp, r->body, r->size);
+    *(tmp + r->size) = '\0';
+
+    cJSON *ret_json = cJSON_Parse(tmp);
+    if (cJSON_GetObjectItem(ret_json, "errors")->valueint != 0) {
+        cJSON *err;
+        cJSON_ArrayForEach(err, cJSON_GetObjectItem(ret_json, "items")) {
+            if (cJSON_GetObjectItem(cJSON_GetObjectItem(err, "index"), "status")->valueint != 201) {
+                char *str = cJSON_Print(err);
+                LOG_ERRORF("elastic.c", "%s\n", str);
+                cJSON_free(str);
+            }
+        }
+    }
+    cJSON_Delete(ret_json);
+    free(tmp);
+}
+
+void _elastic_flush(int max) {
+
+    if (max == 0) {
+        LOG_WARNING("elastic.c", "calling _elastic_flush with 0 in queue")
+        return;
+    }
+
+    size_t buf_len;
+    int count;
+    void *buf = create_bulk_buffer(max, &count, &buf_len);
+
+    char bulk_url[4096];
+    snprintf(bulk_url, 4096, "%s/sist2/_bulk?pipeline=tie", Indexer->es_url);
+    response_t *r = web_post(bulk_url, buf);
+
+    if (r->status_code == 0) {
+        LOG_FATALF("elastic.c", "Could not connect to %s, make sure that elasticsearch is running!\n", IndexCtx.es_url)
+    }
+
+    if (r->status_code == 413) {
+
+        if (max <= 1) {
+            LOG_ERRORF("elastic.c", "Single document too large, giving up: {%s}", Indexer->line_head->uuid_str)
+            free_response(r);
+            free(buf);
+            delete_queue(1);
+            if (Indexer->queued != 0) {
+                elastic_flush();
+            }
+            return;
+        }
+
+        LOG_WARNINGF("elastic.c", "Payload too large, retrying (%d documents)", count);
+
+        free_response(r);
+        free(buf);
+        _elastic_flush(max / 2);
+        return;
+
+    } else if (r->status_code == 429) {
+
+        free_response(r);
+        free(buf);
+        LOG_WARNING("elastic.c", "Got 429 status, will retry after delay")
+        usleep(1000000 * 20);
+        _elastic_flush(max);
+        return;
+
+    } else if (r->status_code != 200) {
+        print_errors(r);
+        delete_queue(Indexer->queued);
+
+    } else {
+
+        print_errors(r);
+        LOG_INFOF("elastic.c", "Indexed %d documents (%zukB) <%d>", count, buf_len / 1024, r->status_code);
+        delete_queue(max);
+
+        if (Indexer->queued != 0) {
+            elastic_flush();
+        }
+    }
+
+    free_response(r);
+    free(buf);
+}
+
+void delete_queue(int max) {
+    for (int i = 0; i < max; i++) {
+        es_bulk_line_t *tmp = Indexer->line_head;
+        Indexer->line_head = tmp->next;
+        if (Indexer->line_head == NULL) {
+            Indexer->line_tail = NULL;
+        }
+        free(tmp);
+        Indexer->queued -= 1;
+    }
 }
 
 void elastic_flush() {
@@ -60,45 +251,7 @@ void elastic_flush() {
         Indexer = create_indexer(IndexCtx.es_url);
     }
 
-    es_bulk_line_t *line = Indexer->line_head;
-
-    int count = 0;
-
-    size_t buf_size = 0;
-    size_t buf_cur = 0;
-    char *buf = malloc(1);
-
-    while (line != NULL) {
-        char action_str[512];
-        snprintf(action_str, 512, "{\"index\":{\"_id\": \"%s\"}}\n", line->uuid_str);
-        size_t action_str_len = strlen(action_str);
-
-        size_t line_len = strlen(line->line);
-        buf = realloc(buf, buf_size + line_len + action_str_len);
-        buf_size += line_len + action_str_len;
-
-        memcpy(buf + buf_cur, action_str, action_str_len);
-        buf_cur += action_str_len;
-        memcpy(buf + buf_cur, line->line, line_len);
-        buf_cur += line_len;
-
-        es_bulk_line_t *tmp = line;
-        line = line->next;
-        free(tmp);
-        count++;
-    }
-    buf = realloc(buf, buf_size + 1);
-    *(buf+buf_cur) = '\0';
-
-    Indexer->line_head = NULL;
-    Indexer->line_tail = NULL;
-    Indexer->queued = 0;
-
-    char bulk_url[4096];
-    snprintf(bulk_url, 4096, "%s/sist2/_bulk", Indexer->es_url);
-    response_t *r = web_post(bulk_url, buf, "Content-Type: application/x-ndjson");
-    printf("Indexed %3d documents (%zukB) <%d>\n", count, buf_cur / 1024, r->status_code);
-    free_response(r);
+    _elastic_flush(Indexer->queued);
 }
 
 void elastic_index_line(es_bulk_line_t *line) {
@@ -117,15 +270,14 @@ void elastic_index_line(es_bulk_line_t *line) {
 
     Indexer->queued += 1;
 
-    if (Indexer->queued >= BULK_INDEX_SIZE) {
+    if (Indexer->queued >= IndexCtx.batch_size) {
         elastic_flush();
     }
 }
 
 es_indexer_t *create_indexer(const char *url) {
 
-    size_t url_len = strlen(url);
-    char *es_url = malloc(url_len);
+    char *es_url = malloc(strlen(url) + 1);
     strcpy(es_url, url);
 
     es_indexer_t *indexer = malloc(sizeof(es_indexer_t));
@@ -138,24 +290,34 @@ es_indexer_t *create_indexer(const char *url) {
     return indexer;
 }
 
-void destroy_indexer() {
+void finish_indexer(char *script, char *index_id) {
 
     char url[4096];
 
     snprintf(url, sizeof(url), "%s/sist2/_refresh", IndexCtx.es_url);
-    response_t *r = web_post(url, "", NULL);
-    printf("Refresh index <%d>\n", r->status_code);
+    response_t *r = web_post(url, "");
+    LOG_INFOF("elastic.c", "Refresh index <%d>", r->status_code);
     free_response(r);
 
+    if (script != NULL) {
+        execute_update_script(script, index_id);
+        free(script);
+
+        snprintf(url, sizeof(url), "%s/sist2/_refresh", IndexCtx.es_url);
+        r = web_post(url, "");
+        LOG_INFOF("elastic.c", "Refresh index <%d>", r->status_code);
+        free_response(r);
+    }
+
     snprintf(url, sizeof(url), "%s/sist2/_forcemerge", IndexCtx.es_url);
-    r = web_post(url, "", NULL);
-    printf("Merge index <%d>\n", r->status_code);
+    r = web_post(url, "");
+    LOG_INFOF("elastic.c", "Merge index <%d>", r->status_code);
     free_response(r);
 
-    if (Indexer != NULL) {
-        free(Indexer->es_url);
-        free(Indexer);
-    }
+    snprintf(url, sizeof(url), "%s/sist2/_settings", IndexCtx.es_url);
+    r = web_put(url, "{\"index\":{\"refresh_interval\":\"1s\"}}");
+    LOG_INFOF("elastic.c", "Set refresh interval <%d>", r->status_code);
+    free_response(r);
 }
 
 void elastic_init(int force_reset) {
@@ -163,46 +325,81 @@ void elastic_init(int force_reset) {
     // Check if index exists
     char url[4096];
     snprintf(url, 4096, "%s/sist2", IndexCtx.es_url);
-    response_t *r = web_get(url);
+    response_t *r = web_get(url, 30);
     int index_exists = r->status_code == 200;
     free_response(r);
 
     if (!index_exists || force_reset) {
         r = web_delete(url);
-        printf("Delete index <%d>\n", r->status_code);
+        LOG_INFOF("elastic.c", "Delete index <%d>", r->status_code);
         free_response(r);
 
         snprintf(url, 4096, "%s/sist2", IndexCtx.es_url);
-        r = web_put(url, "", NULL);
-        printf("Create index <%d>\n", r->status_code);
+        r = web_put(url, "");
+        LOG_INFOF("elastic.c", "Create index <%d>", r->status_code);
         free_response(r);
 
         snprintf(url, 4096, "%s/sist2/_close", IndexCtx.es_url);
-        r = web_post(url, "", NULL);
-        printf("Close index <%d>\n", r->status_code);
+        r = web_post(url, "");
+        LOG_INFOF("elastic.c", "Close index <%d>", r->status_code);
+        free_response(r);
+
+        snprintf(url, 4096, "%s/_ingest/pipeline/tie", IndexCtx.es_url);
+        r = web_put(url, pipeline_json);
+        LOG_INFOF("elastic.c", "Create pipeline <%d>", r->status_code);
         free_response(r);
 
         snprintf(url, 4096, "%s/sist2/_settings", IndexCtx.es_url);
-        r = web_put(url, settings_json, "Content-Type: application/json");
-        printf("Update settings <%d>\n", r->status_code);
+        r = web_put(url, settings_json);
+        LOG_INFOF("elastic.c", "Update settings <%d>", r->status_code);
         free_response(r);
 
-        snprintf(url, 4096, "%s/sist2/_mappings", IndexCtx.es_url);
-        r = web_put(url, mappings_json, "Content-Type: application/json");
-        printf("Update mappings <%d>\n", r->status_code);
+        snprintf(url, 4096, "%s/sist2/_mappings/_doc?include_type_name=true", IndexCtx.es_url);
+        r = web_put(url, mappings_json);
+        LOG_INFOF("elastic.c", "Update mappings <%d>", r->status_code);
         free_response(r);
 
         snprintf(url, 4096, "%s/sist2/_open", IndexCtx.es_url);
-        r = web_post(url, "", NULL);
-        printf("Open index <%d>\n", r->status_code);
+        r = web_post(url, "");
+        LOG_INFOF("elastic.c", "Open index <%d>", r->status_code);
         free_response(r);
     }
 }
 
 cJSON *elastic_get_document(const char *uuid_str) {
     char url[4096];
-    snprintf(url, 4096, "%s/sist2/_source/%s", WebCtx.es_url, uuid_str);
+    snprintf(url, 4096, "%s/sist2/_doc/%s", WebCtx.es_url, uuid_str);
 
-    response_t *r = web_get(url);
-    return cJSON_Parse(r->body);
+    response_t *r = web_get(url, 3);
+    cJSON *json = NULL;
+    if (r->status_code == 200) {
+        json = cJSON_Parse(r->body);
+    }
+    free_response(r);
+    return json;
+}
+
+char *elastic_get_status() {
+    char url[4096];
+    snprintf(url, 4096,
+             "%s/_cluster/state/metadata/sist2?filter_path=metadata.indices.*.state", WebCtx.es_url);
+
+    response_t *r = web_get(url, 30);
+    cJSON *json = NULL;
+    char *status = malloc(128 * sizeof(char));
+    status[0] = '\0';
+
+    if (r->status_code == 200) {
+        json = cJSON_Parse(r->body);
+        const cJSON *metadata = cJSON_GetObjectItem(json, "metadata");
+        if (metadata != NULL) {
+            const cJSON *indices = cJSON_GetObjectItem(metadata, "indices");
+            const cJSON *sist2 = cJSON_GetObjectItem(indices, "sist2");
+            const cJSON *state = cJSON_GetObjectItem(sist2, "state");
+            strcpy(status, state->valuestring);
+        }
+    }
+    free_response(r);
+    cJSON_Delete(json);
+    return status;
 }

src/index/elastic.h

@@ -16,18 +16,21 @@ typedef struct es_indexer es_indexer_t;
 
 void elastic_index_line(es_bulk_line_t *line);
 
-void elastic_flush();
-
 void print_json(cJSON *document, const char uuid_str[UUID_STR_LEN]);
 
 void index_json(cJSON *document, const char uuid_str[UUID_STR_LEN]);
 
 es_indexer_t *create_indexer(const char* es_url);
 
-void destroy_indexer();
+void elastic_cleanup();
+void finish_indexer(char *script, char *index_id);
 
 void elastic_init(int force_reset);
 
 cJSON *elastic_get_document(const char *uuid_str);
 
+char *elastic_get_status();
+
+void execute_update_script(const char *script, const char index_id[UUID_STR_LEN]);
+
 #endif

src/index/web.c

@@ -1,4 +1,10 @@
 #include "web.h"
+#include "src/sist.h"
+
+#include <mongoose.h>
+#include <pthread.h>
+#include <curl/curl.h>
+
 
 size_t write_cb(char *ptr, size_t size, size_t nmemb, void *user_data) {
 
@@ -9,11 +15,104 @@ size_t write_cb(char *ptr, size_t size, size_t nmemb, void *user_data) {
 }
 
 void free_response(response_t *resp) {
-    free(resp->body);
+    if (resp->body != NULL) {
+        free(resp->body);
+    }
     free(resp);
 }
 
-response_t *web_get(const char *url) {
+#define SIST2_HEADERS "User-Agent: sist2\r\nContent-Type: application/json\r\n"
+
+
+void http_req_ev(struct mg_connection *nc, int ev, void *ptr) {
+
+    http_ev_data_t *ev_data = (http_ev_data_t *) nc->user_data;
+
+    switch (ev) {
+        case MG_EV_CONNECT: {
+            int connect_status = *(int *) ptr;
+            if (connect_status != 0) {
+                ev_data->done = TRUE;
+                ev_data->resp->status_code = 0;
+            }
+            break;
+        }
+        case MG_EV_HTTP_REPLY: {
+            struct http_message *hm = (struct http_message *) ptr;
+
+            //TODO: Check errors?
+
+            ev_data->resp->size = hm->body.len;
+            ev_data->resp->status_code = hm->resp_code;
+            ev_data->resp->body = malloc(hm->body.len + 1);
+            memcpy(ev_data->resp->body, hm->body.p, hm->body.len);
+            *(ev_data->resp->body + hm->body.len) = '\0';
+
+            ev_data->done = TRUE;
+            break;
+        }
+        case MG_EV_CLOSE: {
+            ev_data->done = TRUE;
+            break;
+        }
+        default:
+            break;
+    }
+}
+
+subreq_ctx_t *http_req(const char *url, const char *extra_headers, const char *post_data, const char *method) {
+
+    struct mg_str scheme;
+    struct mg_str user_info;
+    struct mg_str host;
+    unsigned int port;
+    struct mg_str path;
+    struct mg_str query;
+    struct mg_str fragment;
+
+    if (post_data == NULL) post_data = "";
+    if (extra_headers == NULL) extra_headers = "";
+    if (path.len == 0) path = mg_mk_str("/");
+    if (host.len == 0) host = mg_mk_str("");
+
+    // [scheme://[user_info@]]host[:port][/path][?query][#fragment]
+    mg_parse_uri(mg_mk_str(url), &scheme, &user_info, &host, &port, &path, &query, &fragment);
+
+    if (query.len > 0) path.len += query.len + 1;
+
+    subreq_ctx_t *ctx = malloc(sizeof(subreq_ctx_t));
+    mg_mgr_init(&ctx->mgr, NULL);
+
+    char address[8192];
+    snprintf(address, sizeof(address), "tcp://%.*s:%u", (int) host.len, host.p, port);
+    struct mg_connection *nc = mg_connect(&ctx->mgr, address, http_req_ev);
+    nc->user_data = &ctx->ev_data;
+    mg_set_protocol_http_websocket(nc);
+
+    ctx->ev_data.resp = calloc(1, sizeof(response_t));
+    ctx->ev_data.done = FALSE;
+
+    mg_printf(
+            nc, "%s %.*s HTTP/1.1\r\n"
+                  "Host: %.*s\r\n"
+                  "Content-Length: %zu\r\n"
+                  "%s\r\n"
+                  "%s",
+            method, (int) path.len, path.p,
+            (int) (path.p - host.p), host.p,
+            strlen(post_data),
+            extra_headers,
+            post_data
+    );
+
+    return ctx;
+}
+
+subreq_ctx_t *web_post_async(const char *url, const char *data) {
+    return http_req(url, SIST2_HEADERS, data, "POST");
+}
+
+response_t *web_get(const char *url, int timeout) {
     response_t *resp = malloc(sizeof(response_t));
 
     CURL *curl;
@@ -24,18 +123,23 @@ response_t *web_get(const char *url) {
     curl_easy_setopt(curl, CURLOPT_WRITEDATA, (void *) (&buffer));
     curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, write_cb);
     curl_easy_setopt(curl, CURLOPT_USERAGENT, "sist2");
+    curl_easy_setopt(curl, CURLOPT_TIMEOUT, timeout);
+
+    struct curl_slist *headers = curl_slist_append(headers, "Content-Type: application/json");
+    curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
 
     curl_easy_perform(curl);
     curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &resp->status_code);
 
     curl_easy_cleanup(curl);
+    curl_slist_free_all(headers);
 
     resp->body = buffer.buf;
     resp->size = buffer.cur;
     return resp;
 }
 
-response_t *web_post(const char *url, const char *data, const char *header) {
+response_t *web_post(const char *url, const char *data) {
 
     response_t *resp = malloc(sizeof(response_t));
 
@@ -49,18 +153,16 @@ response_t *web_post(const char *url, const char *data, const char *header) {
     curl_easy_setopt(curl, CURLOPT_POST, 1);
     curl_easy_setopt(curl, CURLOPT_USERAGENT, "sist2");
 
-    if (header != NULL) {
-        struct curl_slist *headers = NULL;
-        headers = curl_slist_append(headers, header);
-        curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
-    }
+    struct curl_slist *headers = curl_slist_append(headers, "Content-Type: application/json");
+    curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
 
     curl_easy_setopt(curl, CURLOPT_POSTFIELDS, data);
 
-    int r1 = curl_easy_perform(curl);
+    curl_easy_perform(curl);
     curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &resp->status_code);
 
     curl_easy_cleanup(curl);
+    curl_slist_free_all(headers);
 
     resp->body = buffer.buf;
     resp->size = buffer.cur;
@@ -69,7 +171,7 @@ response_t *web_post(const char *url, const char *data, const char *header) {
 }
 
 
-response_t *web_put(const char *url, const char *data, const char *header) {
+response_t *web_put(const char *url, const char *data) {
 
     response_t *resp = malloc(sizeof(response_t));
 
@@ -85,11 +187,8 @@ response_t *web_put(const char *url, const char *data, const char *header) {
     curl_easy_setopt(curl, CURLOPT_DNS_USE_GLOBAL_CACHE, 0);
     curl_easy_setopt(curl, CURLOPT_IPRESOLVE, CURLOPT_DNS_LOCAL_IP4 );
 
-    if (header != NULL) {
-        struct curl_slist *headers = NULL;
-        headers = curl_slist_append(headers, header);
-        curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
-    }
+    struct curl_slist *headers = curl_slist_append(headers, "Content-Type: application/json");
+    curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
 
     curl_easy_setopt(curl, CURLOPT_POSTFIELDS, data);
 
@@ -97,6 +196,7 @@ response_t *web_put(const char *url, const char *data, const char *header) {
     curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &resp->status_code);
 
     curl_easy_cleanup(curl);
+    curl_slist_free_all(headers);
 
     resp->body = buffer.buf;
     resp->size = buffer.cur;
@@ -118,13 +218,16 @@ response_t *web_delete(const char *url) {
     curl_easy_setopt(curl, CURLOPT_USERAGENT, "sist2");
 
     curl_easy_setopt(curl, CURLOPT_POSTFIELDS, "");
+    struct curl_slist *headers = curl_slist_append(headers, "Content-Type: application/json");
+    curl_easy_setopt(curl, CURLOPT_HTTPHEADER, headers);
 
     curl_easy_perform(curl);
     curl_easy_getinfo(curl, CURLINFO_RESPONSE_CODE, &resp->status_code);
 
     curl_easy_cleanup(curl);
+    curl_slist_free_all(headers);
 
     resp->body = buffer.buf;
     resp->size = buffer.cur;
     return resp;
-}
+}

src/index/web.h

@@ -2,6 +2,7 @@
 #define SIST2_WEB_H
 
 #include "src/sist.h"
+#include <mongoose.h>
 
 typedef struct response {
     char *body;
@@ -9,9 +10,20 @@ typedef struct response {
     int status_code;
 } response_t;
 
-response_t *web_get(const char *url);
-response_t *web_post(const char * url, const char * data, const char* header);
-response_t *web_put(const char *url, const char *data, const char *header);
+typedef struct {
+    response_t *resp;
+    int done;
+} http_ev_data_t;
+
+typedef struct {
+    http_ev_data_t ev_data;
+    struct mg_mgr mgr;
+} subreq_ctx_t;
+
+response_t *web_get(const char *url, int timeout);
+response_t *web_post(const char * url, const char * data);
+subreq_ctx_t *web_post_async(const char *url, const char *data);
+response_t *web_put(const char *url, const char *data);
 response_t *web_delete(const char *url);
 
 void free_response(response_t *resp);

src/io/serialize.c

@@ -1,7 +1,9 @@
 #include "src/ctx.h"
 #include "serialize.h"
+#include "src/parsing/parse.h"
+#include "src/parsing/mime.h"
 
-static __thread int IndexFd = -1;
+static __thread int index_fd = -1;
 
 typedef struct {
     unsigned char uuid[16];
@@ -34,15 +36,19 @@ void write_index_descriptor(char *path, index_descriptor_t *desc) {
     cJSON_AddStringToObject(json, "version", desc->version);
     cJSON_AddStringToObject(json, "root", desc->root);
     cJSON_AddStringToObject(json, "name", desc->name);
+    cJSON_AddStringToObject(json, "type", desc->type);
     cJSON_AddStringToObject(json, "rewrite_url", desc->rewrite_url);
     cJSON_AddNumberToObject(json, "timestamp", (double) desc->timestamp);
 
     int fd = open(path, O_CREAT | O_WRONLY, S_IRUSR | S_IWUSR);
-    if (fd == -1) {
-        perror(path);
+    if (fd < 0) {
+        LOG_FATALF("serialize.c", "Could not open index descriptor: %s", strerror(errno));
     }
     char *str = cJSON_Print(json);
-    write(fd, str, strlen(str));
+    int ret = write(fd, str, strlen(str));
+    if (ret == -1) {
+        LOG_FATALF("serialize.c", "Could not write index descriptor: %s", strerror(errno));
+    }
     free(str);
     close(fd);
 
@@ -54,8 +60,16 @@ index_descriptor_t read_index_descriptor(char *path) {
     struct stat info;
     stat(path, &info);
     int fd = open(path, O_RDONLY);
+
+    if (fd == -1) {
+        LOG_FATALF("serialize.c", "Invalid/corrupt index (Could not find descriptor): %s: %s\n", path, strerror(errno))
+    }
+
     char *buf = malloc(info.st_size + 1);
-    read(fd, buf, info.st_size);
+    int ret = read(fd, buf, info.st_size);
+    if (ret == -1) {
+        LOG_FATALF("serialize.c", "Could not read index descriptor: %s", strerror(errno));
+    }
     *(buf + info.st_size) = '\0';
     close(fd);
 
@@ -66,9 +80,14 @@ index_descriptor_t read_index_descriptor(char *path) {
     strcpy(descriptor.root, cJSON_GetObjectItem(json, "root")->valuestring);
     strcpy(descriptor.name, cJSON_GetObjectItem(json, "name")->valuestring);
     strcpy(descriptor.rewrite_url, cJSON_GetObjectItem(json, "rewrite_url")->valuestring);
-    descriptor.root_len = (short)strlen(descriptor.root);
+    descriptor.root_len = (short) strlen(descriptor.root);
     strcpy(descriptor.version, cJSON_GetObjectItem(json, "version")->valuestring);
     strcpy(descriptor.uuid, cJSON_GetObjectItem(json, "uuid")->valuestring);
+    if (cJSON_GetObjectItem(json, "type") == NULL) {
+        strcpy(descriptor.type, INDEX_TYPE_BIN);
+    } else {
+        strcpy(descriptor.type, cJSON_GetObjectItem(json, "type")->valuestring);
+    }
 
     cJSON_Delete(json);
     free(buf);
@@ -105,6 +124,32 @@ char *get_meta_key_text(enum metakey meta_key) {
             return "title";
         case MetaFontName:
             return "font_name";
+        case MetaParent:
+            return "parent";
+        case MetaExifMake:
+            return "exif_make";
+        case MetaExifSoftware:
+            return "exif_software";
+        case MetaExifExposureTime:
+            return "exif_exposure_time";
+        case MetaExifFNumber:
+            return "exif_fnumber";
+        case MetaExifFocalLength:
+            return "exif_focal_length";
+        case MetaExifUserComment:
+            return "exif_user_comment";
+        case MetaExifIsoSpeedRatings:
+            return "exif_iso_speed_ratings";
+        case MetaExifModel:
+            return "exif_model";
+        case MetaExifDateTime:
+            return "exif_datetime";
+        case MetaAuthor:
+            return "author";
+        case MetaModifiedBy:
+            return "modified_by";
+        case MetaThumbnail:
+            return "thumbnail";
         default:
             return NULL;
     }
@@ -113,13 +158,13 @@ char *get_meta_key_text(enum metakey meta_key) {
 
 void write_document(document_t *doc) {
 
-    if (IndexFd == -1) {
+    if (index_fd == -1) {
         char dstfile[PATH_MAX];
-        pid_t tid = syscall(SYS_gettid);
-        snprintf(dstfile, PATH_MAX, "%s_index_%d", ScanCtx.index.path, tid);
-        IndexFd = open(dstfile, O_CREAT | O_WRONLY | O_APPEND, S_IRUSR | S_IWUSR);
+        pthread_t self = pthread_self();
+        snprintf(dstfile, PATH_MAX, "%s_index_%lu", ScanCtx.index.path, self);
+        index_fd = open(dstfile, O_CREAT | O_WRONLY | O_APPEND, S_IRUSR | S_IWUSR);
 
-        if (IndexFd == -1) {
+        if (index_fd == -1) {
             perror("open");
         }
     }
@@ -127,8 +172,8 @@ void write_document(document_t *doc) {
     dyn_buffer_t buf = dyn_buffer_create();
 
     // Ignore root directory in the file path
-    doc->ext = doc->ext - ScanCtx.index.desc.root_len;
-    doc->base = doc->base - ScanCtx.index.desc.root_len;
+    doc->ext = (short) (doc->ext - ScanCtx.index.desc.root_len);
+    doc->base = (short) (doc->base - ScanCtx.index.desc.root_len);
     doc->filepath += ScanCtx.index.desc.root_len;
 
     dyn_buffer_write(&buf, doc, sizeof(line_t));
@@ -139,11 +184,11 @@ void write_document(document_t *doc) {
         dyn_buffer_write_char(&buf, meta->key);
 
         if (IS_META_INT(meta->key)) {
-            dyn_buffer_write_int(&buf, meta->intval);
+            dyn_buffer_write_int(&buf, meta->int_val);
         } else if (IS_META_LONG(meta->key)) {
-            dyn_buffer_write_long(&buf, meta->longval);
+            dyn_buffer_write_long(&buf, meta->long_val);
         } else {
-            dyn_buffer_write_str(&buf, meta->strval);
+            dyn_buffer_write_str(&buf, meta->str_val);
         }
 
         meta_line_t *tmp = meta;
@@ -152,24 +197,29 @@ void write_document(document_t *doc) {
     }
     dyn_buffer_write_char(&buf, '\n');
 
-    write(IndexFd, buf.buf, buf.cur);
+    int res = write(index_fd, buf.buf, buf.cur);
+    if (res == -1) {
+        LOG_FATALF("serialize.c", "Could not write document: %s", strerror(errno))
+    }
     ScanCtx.stat_index_size += buf.cur;
     dyn_buffer_destroy(&buf);
 }
 
-void serializer_cleanup() {
-    close(IndexFd);
+void thread_cleanup() {
+    close(index_fd);
+    cleanup_parse();
+    cleanup_font();
 }
 
-void read_index(const char *path, const char index_id[UUID_STR_LEN], index_func func) {
 
+void read_index_bin(const char *path, const char *index_id, index_func func) {
     line_t line;
     dyn_buffer_t buf = dyn_buffer_create();
 
     FILE *file = fopen(path, "rb");
     while (1) {
         buf.cur = 0;
-        fread((void *) &line, 1, sizeof(line_t), file);
+        size_t _ = fread((void *) &line, 1, sizeof(line_t), file);
         if (feof(file)) {
             break;
         }
@@ -180,82 +230,204 @@ void read_index(const char *path, const char index_id[UUID_STR_LEN], index_func
         char uuid_str[UUID_STR_LEN];
         uuid_unparse(line.uuid, uuid_str);
 
-        cJSON_AddStringToObject(document, "mime", mime_get_mime_text(line.mime));
-        cJSON_AddNumberToObject(document, "size", (double)line.size);
+        const char *mime_text = mime_get_mime_text(line.mime);
+        if (mime_text == NULL) {
+            cJSON_AddNullToObject(document, "mime");
+        } else {
+            cJSON_AddStringToObject(document, "mime", mime_get_mime_text(line.mime));
+        }
+        cJSON_AddNumberToObject(document, "size", (double) line.size);
         cJSON_AddNumberToObject(document, "mtime", line.mtime);
 
-        int c;
+        int c = 0;
         while ((c = getc(file)) != 0) {
             dyn_buffer_write_char(&buf, (char) c);
         }
         dyn_buffer_write_char(&buf, '\0');
 
+        if (IndexCtx.tags != NULL) {
+            const char *tags_string = g_hash_table_lookup(IndexCtx.tags, buf.buf);
+            if (tags_string != NULL) {
+                cJSON *tags_arr = cJSON_Parse(tags_string);
+                cJSON_AddItemToObject(document, "tag", tags_arr);
+            }
+        }
+
         cJSON_AddStringToObject(document, "extension", buf.buf + line.ext);
         if (*(buf.buf + line.ext - 1) == '.') {
             *(buf.buf + line.ext - 1) = '\0';
         } else {
             *(buf.buf + line.ext) = '\0';
         }
-        cJSON_AddStringToObject(document, "name", buf.buf + line.base);
-        *(buf.buf + line.base - 1) = '\0';
-        cJSON_AddStringToObject(document, "path", buf.buf);
+
+        char tmp[PATH_MAX * 3];
+
+        str_escape(tmp, buf.buf + line.base);
+        cJSON_AddStringToObject(document, "name", tmp);
+
+        if (line.base > 0) {
+            *(buf.buf + line.base - 1) = '\0';
+
+            str_escape(tmp, buf.buf);
+            cJSON_AddStringToObject(document, "path", tmp);
+        } else {
+            cJSON_AddStringToObject(document, "path", "");
+        }
 
         enum metakey key = getc(file);
+        size_t ret = 0;
         while (key != '\n') {
             switch (key) {
                 case MetaWidth:
-                case MetaHeight:
-                case MetaMediaDuration:
-                case MetaMediaBitrate: {
+                case MetaHeight: {
                     int value;
-                    fread(&value, sizeof(int), 1, file);
+                    ret = fread(&value, sizeof(int), 1, file);
                     cJSON_AddNumberToObject(document, get_meta_key_text(key), value);
                     break;
                 }
-                case MetaMediaAudioCodec:
-                case MetaMediaVideoCodec: {
-                    int value;
-                    fread(&value, sizeof(int), 1, file);
-                    const AVCodecDescriptor *desc = avcodec_descriptor_get(value);
-                    if (desc != NULL) {
-                        cJSON_AddStringToObject(document, get_meta_key_text(key), desc->name);
-                    }
+                case MetaMediaDuration:
+                case MetaMediaBitrate: {
+                    long value;
+                    ret = fread(&value, sizeof(long), 1, file);
+                    cJSON_AddNumberToObject(document, get_meta_key_text(key), (double) value);
                     break;
                 }
-
+                case MetaMediaAudioCodec:
+                case MetaMediaVideoCodec:
                 case MetaContent:
                 case MetaArtist:
                 case MetaAlbum:
                 case MetaAlbumArtist:
                 case MetaGenre:
                 case MetaFontName:
+                case MetaParent:
+                case MetaExifMake:
+                case MetaExifSoftware:
+                case MetaExifExposureTime:
+                case MetaExifFNumber:
+                case MetaExifFocalLength:
+                case MetaExifUserComment:
+                case MetaExifIsoSpeedRatings:
+                case MetaExifDateTime:
+                case MetaExifModel:
+                case MetaAuthor:
+                case MetaModifiedBy:
+                case MetaThumbnail:
                 case MetaTitle: {
                     buf.cur = 0;
                     while ((c = getc(file)) != 0) {
-                        dyn_buffer_write_char(&buf, (char) c);
+                        if (SHOULD_KEEP_CHAR(c) || c == ' ') {
+                            dyn_buffer_write_char(&buf, (char) c);
+                        }
                     }
                     dyn_buffer_write_char(&buf, '\0');
                     cJSON_AddStringToObject(document, get_meta_key_text(key), buf.buf);
                     break;
                 }
+                default:
+                LOG_FATALF("serialize.c", "Invalid meta key (corrupt index): %x", key)
             }
 
             key = getc(file);
         }
 
         func(document, uuid_str);
-        cJSON_free(document);
+        cJSON_Delete(document);
+    }
+    dyn_buffer_destroy(&buf);
+    fclose(file);
+}
+
+const char *json_type_copy_fields[] = {
+        "mime", "name", "path", "extension", "index", "size", "mtime", "parent",
+
+        // Meta
+        "title", "content", "width", "height", "duration", "audioc", "videoc",
+        "bitrate", "artist", "album", "album_artist", "genre", "title", "font_name",
+
+        // Special
+        "tag", "_url"
+};
+
+const char *json_type_array_fields[] = {
+        "_keyword", "_text"
+};
+
+void read_index_json(const char *path, UNUSED(const char *index_id), index_func func) {
+
+    FILE *file = fopen(path, "r");
+    while (1) {
+        char *line = NULL;
+        size_t len;
+        size_t read = getline(&line, &len, file);
+        if (read < 0) {
+            if (line) {
+                free(line);
+            }
+            break;
+        }
+
+        cJSON *input = cJSON_Parse(line);
+        if (input == NULL) {
+            LOG_FATALF("serialize.c", "Could not parse JSON line: \n%s", line)
+        }
+        if (line) {
+            free(line);
+        }
+
+        cJSON *document = cJSON_CreateObject();
+        const char *uuid_str = cJSON_GetObjectItem(input, "_id")->valuestring;
+
+        for (int i = 0; i < (sizeof(json_type_copy_fields) / sizeof(json_type_copy_fields[0])); i++) {
+            cJSON *value = cJSON_GetObjectItem(input, json_type_copy_fields[i]);
+            if (value != NULL) {
+                cJSON_AddItemReferenceToObject(document, json_type_copy_fields[i], value);
+            }
+        }
+
+        for (int i = 0; i < (sizeof(json_type_array_fields) / sizeof(json_type_array_fields[0])); i++) {
+            cJSON *arr = cJSON_GetObjectItem(input, json_type_array_fields[i]);
+            if (arr != NULL) {
+                cJSON *obj;
+                cJSON_ArrayForEach(obj, arr) {
+                    char key[1024];
+                    cJSON *k = cJSON_GetObjectItem(obj, "k");
+                    cJSON *v = cJSON_GetObjectItem(obj, "v");
+                    if (k == NULL || v == NULL || !cJSON_IsString(k) || !cJSON_IsString(v)) {
+                        char *str = cJSON_Print(obj);
+                        LOG_FATALF("serialize.c", "Invalid %s member: must contain .k and .v string fields: \n%s",
+                                   json_type_array_fields[i], str)
+                    }
+                    snprintf(key, sizeof(key), "%s.%s", json_type_array_fields[i], k->valuestring);
+                    cJSON_AddStringToObject(document, key, v->valuestring);
+                }
+            }
+        }
+
+        func(document, uuid_str);
+        cJSON_Delete(document);
+        cJSON_Delete(input);
+
     }
     fclose(file);
 }
 
+void read_index(const char *path, const char index_id[UUID_STR_LEN], const char *type, index_func func) {
+
+    if (strcmp(type, INDEX_TYPE_BIN) == 0) {
+        read_index_bin(path, index_id, func);
+    } else if (strcmp(type, INDEX_TYPE_JSON) == 0) {
+        read_index_json(path, index_id, func);
+    }
+}
+
 void incremental_read(GHashTable *table, const char *filepath) {
     FILE *file = fopen(filepath, "rb");
     line_t line;
 
     while (1) {
-        fread((void *) &line, 1, sizeof(line_t), file);
-        if (feof(file)) {
+        size_t ret = fread((void *) &line, 1, sizeof(line_t), file);
+        if (ret != 1 || feof(file)) {
             break;
         }
 
@@ -278,8 +450,8 @@ void incremental_copy(store_t *store, store_t *dst_store, const char *filepath,
     line_t line;
 
     while (1) {
-        fread((void *) &line, 1, sizeof(line_t), file);
-        if (feof(file)) {
+        size_t ret = fread((void *) &line, 1, sizeof(line_t), file);
+        if (ret != 1 || feof(file)) {
             break;
         }
 
@@ -289,6 +461,7 @@ void incremental_copy(store_t *store, store_t *dst_store, const char *filepath,
             size_t buf_len;
             char *buf = store_read(store, (char *) line.uuid, 16, &buf_len);
             store_write(dst_store, (char *) line.uuid, 16, buf, buf_len);
+            free(buf);
 
             char c;
             while ((c = (char) getc(file))) {
@@ -306,11 +479,11 @@ void incremental_copy(store_t *store, store_t *dst_store, const char *filepath,
 
                 if (IS_META_INT(key)) {
                     int val;
-                    fread(&val, sizeof(val), 1, file);
+                    ret = fread(&val, sizeof(val), 1, file);
                     fwrite(&val, sizeof(val), 1, dst_file);
                 } else if (IS_META_LONG(key)) {
                     long val;
-                    fread(&val, sizeof(val), 1, file);
+                    ret = fread(&val, sizeof(val), 1, file);
                     fwrite(&val, sizeof(val), 1, dst_file);
                 } else {
                     while ((c = (char) getc(file))) {
@@ -318,6 +491,10 @@ void incremental_copy(store_t *store, store_t *dst_store, const char *filepath,
                     }
                     fwrite("\0", sizeof(c), 1, dst_file);
                 }
+
+                if (ret != 1) {
+                    break;
+                }
             }
         } else {
             skip_meta(file);

src/io/serialize.h

@@ -2,7 +2,10 @@
 #define SIST2_SERIALIZE_H
 
 #include "src/sist.h"
+#include "store.h"
+
 #include <sys/syscall.h>
+#include <glib.h>
 
 typedef void(*index_func)(cJSON *, const char[UUID_STR_LEN]);
 
@@ -11,14 +14,14 @@ void incremental_copy(store_t *store, store_t *dst_store, const char *filepath,
 
 void write_document(document_t *doc);
 
-void read_index(const char *path, const char[UUID_STR_LEN], index_func);
+void read_index(const char *path, const char[UUID_STR_LEN], const char *type, index_func);
 
 void incremental_read(GHashTable *table, const char *filepath);
 
 /**
  * Must be called after write_document
  */
-void serializer_cleanup();
+void thread_cleanup();
 
 void write_index_descriptor(char *path, index_descriptor_t *desc);
 

src/io/store.c

@@ -1,25 +1,25 @@
 #include "store.h"
 #include "src/ctx.h"
 
-store_t *store_create(char *path) {
+store_t *store_create(char *path, size_t chunk_size) {
 
     store_t *store = malloc(sizeof(struct store_t));
+    store->chunk_size = chunk_size;
     pthread_rwlock_init(&store->lock, NULL);
 
     mdb_env_create(&store->env);
 
     int open_ret = mdb_env_open(store->env,
-                 path,
-                 MDB_WRITEMAP | MDB_MAPASYNC,
-                 S_IRUSR | S_IWUSR
+                                path,
+                                MDB_WRITEMAP | MDB_MAPASYNC,
+                                S_IRUSR | S_IWUSR
     );
 
     if (open_ret != 0) {
-        fprintf(stderr, "Error while opening store: %s", mdb_strerror(open_ret));
-        exit(1);
+        LOG_FATALF("store.c", "Error while opening store: %s (%s)\n", mdb_strerror(open_ret), path)
     }
 
-    store->size = (size_t) 1024 * 1024 * 5;
+    store->size = (size_t) store->chunk_size;
     ScanCtx.stat_tn_size = 0;
     mdb_env_set_mapsize(store->env, store->size);
 
@@ -42,6 +42,12 @@ void store_destroy(store_t *store) {
 
 void store_write(store_t *store, char *key, size_t key_len, char *buf, size_t buf_len) {
 
+    if (LogCtx.very_verbose) {
+        char uuid_str[UUID_STR_LEN];
+        uuid_unparse((unsigned char *) key, uuid_str);
+        LOG_DEBUGF("store.c", "Store write {%s} %lu bytes", uuid_str, buf_len)
+    }
+
     MDB_val mdb_key;
     mdb_key.mv_data = key;
     mdb_key.mv_size = key_len;
@@ -64,17 +70,19 @@ void store_write(store_t *store, char *key, size_t key_len, char *buf, size_t bu
         // Cannot resize when there is a opened transaction.
         //  Resize take effect on the next commit.
         pthread_rwlock_wrlock(&store->lock);
-        store->size += 1024 * 1024 * 5;
+        store->size += store->chunk_size;
         mdb_env_set_mapsize(store->env, store->size);
         mdb_txn_begin(store->env, NULL, 0, &txn);
         put_ret = mdb_put(txn, store->dbi, &mdb_key, &mdb_value, 0);
+
+        LOG_INFOF("store.c", "Updated mdb mapsize to %lu bytes", store->size)
     }
 
     mdb_txn_commit(txn);
     pthread_rwlock_unlock(&store->lock);
 
     if (put_ret != 0) {
-        printf("%s\n", mdb_strerror(put_ret));
+        LOG_ERROR("store.c", mdb_strerror(put_ret))
     }
 }
 
@@ -103,3 +111,40 @@ char *store_read(store_t *store, char *key, size_t key_len, size_t *ret_vallen)
     return buf;
 }
 
+GHashTable *store_read_all(store_t *store) {
+
+    int count = 0;
+
+    GHashTable *table = g_hash_table_new_full(g_str_hash, g_str_equal, free, free);
+
+    MDB_txn *txn = NULL;
+    mdb_txn_begin(store->env, NULL, MDB_RDONLY, &txn);
+
+    MDB_cursor *cur = NULL;
+    mdb_cursor_open(txn, store->dbi, &cur);
+
+    MDB_val key;
+    MDB_val value;
+
+    while (mdb_cursor_get(cur, &key, &value, MDB_NEXT) == 0) {
+        char *key_str = malloc(key.mv_size);
+        memcpy(key_str, key.mv_data, key.mv_size);
+        char *val_str = malloc(value.mv_size);
+        memcpy(val_str, value.mv_data, value.mv_size);
+
+        g_hash_table_insert(table, key_str, val_str);
+        count += 1;
+    }
+
+    LOG_DEBUGF("store.c", "Read tags for %d documents", count);
+
+    mdb_cursor_close(cur);
+    mdb_txn_abort(txn);
+    return table;
+}
+
+
+void store_copy(store_t *store, const char *destination) {
+    mkdir(destination, S_IWUSR | S_IRUSR | S_IXUSR);
+    mdb_env_copy(store->env, destination);
+}

src/io/store.h

@@ -4,16 +4,20 @@
 #include <pthread.h>
 #include <lmdb.h>
 
+#include <glib.h>
+
+#define STORE_SIZE_TN 1024 * 1024 * 5
+#define STORE_SIZE_TAG 1024 * 16
+
 typedef struct store_t {
     MDB_dbi dbi;
     MDB_env *env;
     size_t size;
+    size_t chunk_size;
     pthread_rwlock_t lock;
 } store_t;
 
-#include "src/sist.h"
-
-store_t *store_create(char *path);
+store_t *store_create(char *path, size_t chunk_size);
 
 void store_destroy(store_t *store);
 
@@ -21,4 +25,8 @@ void store_write(store_t *store, char *key, size_t key_len, char *buf, size_t bu
 
 char *store_read(store_t *store, char *key, size_t key_len, size_t *ret_vallen);
 
+GHashTable *store_read_all(store_t *store);
+
+void store_copy(store_t *store, const char *destination);
+
 #endif

src/io/walk.c

@@ -1,28 +1,50 @@
 #include "walk.h"
 #include "src/ctx.h"
+#include "src/parsing/parse.h"
 
-parse_job_t *create_parse_job(const char *filepath, const struct stat *info, int base) {
+#include <ftw.h>
+
+__always_inline
+parse_job_t *create_fs_parse_job(const char *filepath, const struct stat *info, int base) {
     int len = (int) strlen(filepath);
-
     parse_job_t *job = malloc(sizeof(parse_job_t) + len);
 
-    memcpy(&(job->filepath), filepath, len + 1);
+    strcpy(job->filepath, filepath);
     job->base = base;
     char *p = strrchr(filepath + base, '.');
     if (p != NULL) {
-        job->ext = (int)(p - filepath + 1);
+        job->ext = (int) (p - filepath + 1);
     } else {
         job->ext = len;
     }
 
-    memcpy(&(job->info), info, sizeof(struct stat));
+    job->vfile.info = *info;
+
+    memset(job->parent, 0, 16);
+
+    job->vfile.filepath = job->filepath;
+    job->vfile.read = fs_read;
+    job->vfile.reset = fs_reset;
+    job->vfile.close = fs_close;
+    job->vfile.fd = -1;
+    job->vfile.is_fs_file = TRUE;
 
     return job;
 }
 
+int sub_strings[30];
+#define EXCLUDED(str) (pcre_exec(ScanCtx.exclude, ScanCtx.exclude_extra, filepath, strlen(filepath), 0, 0, sub_strings, sizeof(sub_strings)) >= 0)
+
 int handle_entry(const char *filepath, const struct stat *info, int typeflag, struct FTW *ftw) {
-    if (typeflag == FTW_F && S_ISREG(info->st_mode)) {
-        parse_job_t *job = create_parse_job(filepath, info, ftw->base);
+
+    if (typeflag == FTW_F && S_ISREG(info->st_mode) && ftw->level <= ScanCtx.depth) {
+
+        if (ScanCtx.exclude != NULL && EXCLUDED(filepath)) {
+            LOG_DEBUGF("walk.c", "Excluded: %s", filepath)
+            return 0;
+        }
+
+        parse_job_t *job = create_fs_parse_job(filepath, info, ftw->base);
         tpool_add_work(ScanCtx.pool, parse, job);
     }
 

src/io/walk.h

@@ -3,8 +3,6 @@
 
 #define _XOPEN_SOURCE 500
 
-#include "src/sist.h"
-
 int walk_directory_tree(const char *);
 
 #endif

src/log.c

@@ -0,0 +1,112 @@
+#include "log.h"
+
+#include <pthread.h>
+#include <stdarg.h>
+
+const char *log_colors[] = {
+        "\033[34m", "\033[01;34m", "\033[0m",
+        "\033[01;33m", "\033[31m", "\033[01;31m"
+};
+
+const char *log_levels[] = {
+        "DEBUG", "INFO", "WARNING", "ERROR", "FATAL"
+};
+
+void vsist_logf(const char *filepath, int level, char *format, va_list ap) {
+
+    static int is_tty = -1;
+    if (is_tty == -1) {
+        is_tty = isatty(STDERR_FILENO);
+    }
+
+    char log_str[LOG_MAX_LENGTH];
+
+    unsigned long long pid = (unsigned long long) pthread_self();
+
+    char datetime[32];
+    time_t t;
+    struct tm result;
+    t = time(NULL);
+    localtime_r(&t, &result);
+    strftime(datetime, sizeof(datetime), "%Y-%m-%d %H:%M:%S", &result);
+
+    int log_len;
+    if (is_tty) {
+        log_len = snprintf(
+                log_str, sizeof(log_str),
+                "\033[%dm[%04llX]%s [%s] [%s %s] ",
+                31 + ((unsigned int) (pid)) % 7, pid, log_colors[level],
+                datetime, log_levels[level], filepath
+        );
+    } else {
+        log_len = snprintf(
+                log_str, sizeof(log_str),
+                "[%04llX] [%s] [%s %s] ",
+                pid, datetime, log_levels[level], filepath
+        );
+    }
+
+    size_t maxsize = sizeof(log_str) - log_len;
+    log_len += vsnprintf(log_str + log_len, maxsize, format, ap);
+
+    if (is_tty) {
+        log_len += sprintf(log_str + log_len, "\033[0m\n");
+    } else {
+        *(log_str + log_len) = '\n';
+        log_len += 1;
+    }
+
+    int ret = write(STDERR_FILENO, log_str, log_len);
+    if (ret == -1) {
+        LOG_FATALF("serialize.c", "Could not write index descriptor: %s", strerror(errno))
+    }
+}
+
+void sist_logf(const char *filepath, int level, char *format, ...) {
+    va_list ap;
+    va_start(ap, format);
+    vsist_logf(filepath, level, format, ap);
+    va_end(ap);
+}
+
+void sist_log(const char *filepath, int level, char *str) {
+
+    static int is_tty = -1;
+    if (is_tty == -1) {
+        is_tty = isatty(STDERR_FILENO);
+    }
+
+    char log_str[LOG_MAX_LENGTH];
+
+    unsigned long long pid = (unsigned long long) pthread_self();
+
+    char datetime[32];
+    time_t t;
+    struct tm result;
+    t = time(NULL);
+    localtime_r(&t, &result);
+    strftime(datetime, sizeof(datetime), "%Y-%m-%d %H:%M:%S", &result);
+
+    int log_len;
+    if (is_tty) {
+        log_len = snprintf(
+                log_str, sizeof(log_str),
+                "\033[%dm[%04llX]%s [%s] [%s %s] %s \033[0m\n",
+                31 + ((unsigned int) (pid)) % 7, pid, log_colors[level],
+                datetime, log_levels[level], filepath,
+                str
+        );
+    } else {
+        log_len = snprintf(
+                log_str, sizeof(log_str),
+                "[%04llX] [%s] [%s %s] %s \n",
+                pid, datetime, log_levels[level], filepath,
+                str
+        );
+    }
+
+    int ret = write(STDERR_FILENO, log_str, log_len);
+    if (ret == -1) {
+        LOG_FATALF("serialize.c", "Could not write index descriptor: %s", strerror(errno));
+    }
+}

src/log.h

@@ -0,0 +1,47 @@
+#ifndef SIST2_LOG_H
+#define SIST2_LOG_H
+
+
+#define LOG_MAX_LENGTH 8192
+
+#define SIST_DEBUG 0
+#define SIST_INFO 1
+#define SIST_WARNING 2
+#define SIST_ERROR 3
+#define SIST_FATAL 4
+
+#define LOG_DEBUGF(filepath, fmt, ...) \
+    if (LogCtx.very_verbose) {sist_logf(filepath, SIST_DEBUG, fmt, __VA_ARGS__);}
+#define LOG_DEBUG(filepath, str) \
+    if (LogCtx.very_verbose) {sist_log(filepath, SIST_DEBUG, str);}
+
+#define LOG_INFOF(filepath, fmt, ...) \
+    if (LogCtx.verbose) {sist_logf(filepath, SIST_INFO, fmt, __VA_ARGS__);}
+#define LOG_INFO(filepath, str) \
+    if (LogCtx.verbose) {sist_log(filepath, SIST_INFO, str);}
+
+#define LOG_WARNINGF(filepath, fmt, ...) \
+    if (LogCtx.verbose) {sist_logf(filepath, SIST_WARNING, fmt, __VA_ARGS__);}
+#define LOG_WARNING(filepath, str) \
+    if (LogCtx.verbose) {sist_log(filepath, SIST_WARNING, str);}
+
+#define LOG_ERRORF(filepath, fmt, ...) \
+    if (LogCtx.verbose) {sist_logf(filepath, SIST_ERROR, fmt, __VA_ARGS__);}
+#define LOG_ERROR(filepath, str) \
+    if (LogCtx.verbose) {sist_log(filepath, SIST_ERROR, str);}
+
+#define LOG_FATALF(filepath, fmt, ...) \
+    sist_logf(filepath, SIST_FATAL, fmt, __VA_ARGS__);\
+    exit(-1);
+#define LOG_FATAL(filepath, str) \
+    sist_log(filepath, SIST_FATAL, str);\
+    exit(-1);
+
+#include "sist.h"
+
+void sist_logf(const char *filepath, int level, char *format, ...);
+void vsist_logf(const char *filepath, int level, char *format, va_list ap);
+
+void sist_log(const char *filepath, int level, char *str);
+
+#endif

src/main.c

@@ -1,66 +1,186 @@
 #include "sist.h"
 #include "ctx.h"
 
+#include <third-party/argparse/argparse.h>
+#include <locale.h>
+
+#include "cli.h"
+#include "io/serialize.h"
+#include "io/store.h"
+#include "tpool.h"
+#include "io/walk.h"
+#include "index/elastic.h"
+#include "web/serve.h"
+#include "parsing/mime.h"
+#include "parsing/parse.h"
+
+#include "stats.h"
+
 #define DESCRIPTION "Lightning-fast file system indexer and search tool."
+
 #define EPILOG "Made by simon987 <me@simon987.net>. Released under GPL-3.0"
 
 
-static const char *const Version = "1.0.0";
+static const char *const Version = "2.7.4";
 static const char *const usage[] = {
         "sist2 scan [OPTION]... PATH",
         "sist2 index [OPTION]... INDEX",
         "sist2 web [OPTION]... INDEX...",
+        "sist2 exec-script [OPTION]... INDEX",
         NULL,
 };
 
-void global_init() {
-    curl_global_init(CURL_GLOBAL_NOTHING);
-    av_log_set_level(AV_LOG_QUIET);
-}
-
 void init_dir(const char *dirpath) {
     char path[PATH_MAX];
     snprintf(path, PATH_MAX, "%sdescriptor.json", dirpath);
 
     uuid_t uuid;
-    uuid_generate_time_safe(uuid);
+    uuid_generate(uuid);
     uuid_unparse(uuid, ScanCtx.index.desc.uuid);
     time(&ScanCtx.index.desc.timestamp);
     strcpy(ScanCtx.index.desc.version, Version);
+    strcpy(ScanCtx.index.desc.type, INDEX_TYPE_BIN);
 
     write_index_descriptor(path, &ScanCtx.index.desc);
 }
 
 void scan_print_header() {
-    printf("sist2 V%s\n", Version);
-    printf("---------------------\n");
-    printf("threads\t\t%d\n", ScanCtx.threads);
-    printf("tn_qscale\t%.1f/31.0\n", ScanCtx.tn_qscale);
-    printf("tn_size\t\t%dpx\n", ScanCtx.tn_size);
-    printf("output\t\t%s\n", ScanCtx.index.path);
+    LOG_INFOF("main.c", "sist2 v%s", Version)
 }
 
-void sist2_scan(scan_args_t *args) {
+void _store(char *key, size_t key_len, char *buf, size_t buf_len) {
+    store_write(ScanCtx.index.store, key, key_len, buf, buf_len);
+}
+
+void _log(const char *filepath, int level, char *str) {
+    if (level == LEVEL_FATAL) {
+        sist_log(filepath, level, str);
+        exit(-1);
+    }
+
+    if (LogCtx.verbose) {
+        if (level == LEVEL_DEBUG) {
+            if (LogCtx.very_verbose) {
+                sist_log(filepath, level, str);
+            }
+        } else {
+            sist_log(filepath, level, str);
+        }
+    }
+}
+
+void _logf(const char *filepath, int level, char *format, ...) {
+
+    va_list args;
+
+    va_start(args, format);
+    if (level == LEVEL_FATAL) {
+        vsist_logf(filepath, level, format, args);
+        exit(-1);
+    }
+
+    if (LogCtx.verbose) {
+        if (level == LEVEL_DEBUG) {
+            if (LogCtx.very_verbose) {
+                vsist_logf(filepath, level, format, args);
+            }
+        } else {
+            vsist_logf(filepath, level, format, args);
+        }
+    }
+    va_end(args);
+}
+
+void initialize_scan_context(scan_args_t *args) {
+
+    // Arc
+    ScanCtx.arc_ctx.mode = args->archive_mode;
+    ScanCtx.arc_ctx.log = _log;
+    ScanCtx.arc_ctx.logf = _logf;
+    ScanCtx.arc_ctx.parse = (parse_callback_t) parse;
+
+    // Comic
+    ScanCtx.comic_ctx.log = _log;
+    ScanCtx.comic_ctx.logf = _logf;
+    ScanCtx.comic_ctx.store = _store;
+    ScanCtx.comic_ctx.tn_size = args->size;
+    ScanCtx.comic_ctx.tn_qscale = args->quality;
+    ScanCtx.comic_ctx.cbr_mime = mime_get_mime_by_string(ScanCtx.mime_table, "application/x-cbr");
+    ScanCtx.comic_ctx.cbz_mime = mime_get_mime_by_string(ScanCtx.mime_table, "application/x-cbz");
+
+    // Ebook
+    pthread_mutex_init(&ScanCtx.ebook_ctx.mupdf_mutex, NULL);
+    ScanCtx.ebook_ctx.content_size = args->content_size;
+    ScanCtx.ebook_ctx.tn_size = args->size;
+    ScanCtx.ebook_ctx.tesseract_lang = args->tesseract_lang;
+    ScanCtx.ebook_ctx.tesseract_path = args->tesseract_path;
+    ScanCtx.ebook_ctx.log = _log;
+    ScanCtx.ebook_ctx.logf = _logf;
+    ScanCtx.ebook_ctx.store = _store;
+
+    // Font
+    ScanCtx.font_ctx.enable_tn = args->size > 0;
+    ScanCtx.font_ctx.log = _log;
+    ScanCtx.font_ctx.logf = _logf;
+    ScanCtx.font_ctx.store = _store;
+
+    // Media
+    ScanCtx.media_ctx.tn_qscale = args->quality;
+    ScanCtx.media_ctx.tn_size = args->size;
+    ScanCtx.media_ctx.log = _log;
+    ScanCtx.media_ctx.logf = _logf;
+    ScanCtx.media_ctx.store = _store;
+    ScanCtx.media_ctx.max_media_buffer = (long) args->max_memory_buffer * 1024 * 1024;
+    init_media();
+
+    // OOXML
+    ScanCtx.ooxml_ctx.content_size = args->content_size;
+    ScanCtx.ooxml_ctx.log = _log;
+    ScanCtx.ooxml_ctx.logf = _logf;
+    ScanCtx.ooxml_ctx.store = _store;
+
+    // MOBI
+    ScanCtx.mobi_ctx.content_size = args->content_size;
+    ScanCtx.mobi_ctx.log = _log;
+    ScanCtx.mobi_ctx.logf = _logf;
+
+    // TEXT
+    ScanCtx.text_ctx.content_size = args->content_size;
+    ScanCtx.text_ctx.log = _log;
+    ScanCtx.text_ctx.logf = _logf;
 
-    ScanCtx.tn_qscale = args->quality;
-    ScanCtx.tn_size = args->size;
-    ScanCtx.content_size = args->content_size;
-    ScanCtx.pool = tpool_create(args->threads, serializer_cleanup);
     ScanCtx.threads = args->threads;
+    ScanCtx.depth = args->depth;
+
     strncpy(ScanCtx.index.path, args->output, sizeof(ScanCtx.index.path));
     strncpy(ScanCtx.index.desc.name, args->name, sizeof(ScanCtx.index.desc.name));
-    strcpy(ScanCtx.index.desc.root, args->path);
+    strncpy(ScanCtx.index.desc.root, args->path, sizeof(ScanCtx.index.desc.root));
+    strncpy(ScanCtx.index.desc.rewrite_url, args->rewrite_url, sizeof(ScanCtx.index.desc.rewrite_url));
     ScanCtx.index.desc.root_len = (short) strlen(ScanCtx.index.desc.root);
+    ScanCtx.fast = args->fast;
 
-    init_dir(ScanCtx.index.path);
+    // Raw
+    ScanCtx.raw_ctx.tn_qscale = args->quality;
+    ScanCtx.raw_ctx.tn_size = args->size;
+    ScanCtx.raw_ctx.log = _log;
+    ScanCtx.raw_ctx.logf = _logf;
+    ScanCtx.raw_ctx.store = _store;
+}
+
+
+void sist2_scan(scan_args_t *args) {
 
     ScanCtx.mime_table = mime_get_mime_table();
     ScanCtx.ext_table = mime_get_ext_table();
 
+    initialize_scan_context(args);
+
+    init_dir(ScanCtx.index.path);
+
     char store_path[PATH_MAX];
     snprintf(store_path, PATH_MAX, "%sthumbs", ScanCtx.index.path);
     mkdir(store_path, S_IWUSR | S_IRUSR | S_IXUSR);
-    ScanCtx.index.store = store_create(store_path);
+    ScanCtx.index.store = store_create(store_path, STORE_SIZE_TN);
 
     scan_print_header();
 
@@ -70,9 +190,18 @@ void sist2_scan(scan_args_t *args) {
 
         DIR *dir = opendir(args->incremental);
         if (dir == NULL) {
-            perror("opendir");
-            return;
+            LOG_FATALF("main.c", "Could not open original index for incremental scan: %s", strerror(errno))
         }
+
+        char descriptor_path[PATH_MAX];
+        snprintf(descriptor_path, PATH_MAX, "%s/descriptor.json", args->incremental);
+        index_descriptor_t original_desc = read_index_descriptor(descriptor_path);
+
+        if (strcmp(original_desc.version, Version) != 0) {
+            LOG_FATALF("main.c", "Version mismatch! Index is %s but executable is %s/%s", original_desc.version,
+                       Version, INDEX_VERSION_EXTERNAL)
+        }
+
         struct dirent *de;
         while ((de = readdir(dir)) != NULL) {
             if (strncmp(de->d_name, "_index_", sizeof("_index_") - 1) == 0) {
@@ -83,18 +212,22 @@ void sist2_scan(scan_args_t *args) {
         }
         closedir(dir);
 
-        printf("Loaded %d items in to mtime table.", g_hash_table_size(ScanCtx.original_table));
+        LOG_INFOF("main.c", "Loaded %d items in to mtime table.", g_hash_table_size(ScanCtx.original_table))
     }
 
+    ScanCtx.pool = tpool_create(args->threads, thread_cleanup, TRUE);
+    tpool_start(ScanCtx.pool);
     walk_directory_tree(ScanCtx.index.desc.root);
     tpool_wait(ScanCtx.pool);
     tpool_destroy(ScanCtx.pool);
 
+    generate_stats(&ScanCtx.index, args->treemap_threshold, ScanCtx.index.path);
+
     if (args->incremental != NULL) {
         char dst_path[PATH_MAX];
         snprintf(store_path, PATH_MAX, "%sthumbs", args->incremental);
         snprintf(dst_path, PATH_MAX, "%s_index_original", ScanCtx.index.path);
-        store_t *source = store_create(store_path);
+        store_t *source = store_create(store_path, STORE_SIZE_TN);
 
         DIR *dir = opendir(args->incremental);
         if (dir == NULL) {
@@ -111,6 +244,13 @@ void sist2_scan(scan_args_t *args) {
         }
         closedir(dir);
         store_destroy(source);
+
+        snprintf(store_path, PATH_MAX, "%stags", args->incremental);
+        snprintf(dst_path, PATH_MAX, "%stags", ScanCtx.index.path);
+        mkdir(store_path, S_IWUSR | S_IRUSR | S_IXUSR);
+        store_t *source_tags = store_create(store_path, STORE_SIZE_TAG);
+        store_copy(source_tags, dst_path);
+        store_destroy(source_tags);
     }
 
     store_destroy(ScanCtx.index.store);
@@ -119,6 +259,7 @@ void sist2_scan(scan_args_t *args) {
 void sist2_index(index_args_t *args) {
 
     IndexCtx.es_url = args->es_url;
+    IndexCtx.batch_size = args->batch_size;
 
     if (!args->print) {
         elastic_init(args->force_reset);
@@ -128,17 +269,25 @@ void sist2_index(index_args_t *args) {
     snprintf(descriptor_path, PATH_MAX, "%s/descriptor.json", args->index_path);
 
     index_descriptor_t desc = read_index_descriptor(descriptor_path);
-    if (strcmp(desc.version, Version) != 0) {
-        fprintf(stderr, "Version mismatch! Index is v%s but executable is v%s\n", desc.version, Version);
-        return;
+
+    LOG_DEBUGF("main.c", "descriptor version %s (%s)", desc.version, desc.type)
+
+    if (strcmp(desc.version, Version) != 0 && strcmp(desc.version, INDEX_VERSION_EXTERNAL) != 0) {
+        LOG_FATALF("main.c", "Version mismatch! Index is %s but executable is %s/%s", desc.version, Version,
+                   INDEX_VERSION_EXTERNAL)
     }
 
     DIR *dir = opendir(args->index_path);
     if (dir == NULL) {
-        perror("opendir");
-        return;
+        LOG_FATALF("main.c", "Could not open index %s: %s", args->index_path, strerror(errno))
     }
 
+    char path_tmp[PATH_MAX];
+    snprintf(path_tmp, sizeof(path_tmp), "%s/tags", args->index_path);
+    mkdir(path_tmp, S_IWUSR | S_IRUSR | S_IXUSR);
+    IndexCtx.tag_store = store_create(path_tmp, STORE_SIZE_TAG);
+    IndexCtx.tags = store_read_all(IndexCtx.tag_store);
+
     index_func f;
     if (args->print) {
         f = print_json;
@@ -146,25 +295,62 @@ void sist2_index(index_args_t *args) {
         f = index_json;
     }
 
+    void (*cleanup)();
+    if (args->print) {
+        cleanup = NULL;
+    } else {
+        cleanup = elastic_cleanup;
+    }
+
+    IndexCtx.pool = tpool_create(args->threads, cleanup, FALSE);
+    tpool_start(IndexCtx.pool);
+
     struct dirent *de;
     while ((de = readdir(dir)) != NULL) {
         if (strncmp(de->d_name, "_index_", sizeof("_index_") - 1) == 0) {
             char file_path[PATH_MAX];
             snprintf(file_path, PATH_MAX, "%s/%s", args->index_path, de->d_name);
-            read_index(file_path, desc.uuid, f);
+            read_index(file_path, desc.uuid, desc.type, f);
         }
     }
+    closedir(dir);
+
+    tpool_wait(IndexCtx.pool);
 
     if (!args->print) {
-        elastic_flush();
-        destroy_indexer();
+        finish_indexer(args->script, desc.uuid);
     }
+    tpool_destroy(IndexCtx.pool);
+
+    store_destroy(IndexCtx.tag_store);
+    g_hash_table_remove_all(IndexCtx.tags);
+    g_hash_table_destroy(IndexCtx.tags);
+}
+
+void sist2_exec_script(exec_args_t *args) {
+
+    LogCtx.verbose = TRUE;
+
+    char descriptor_path[PATH_MAX];
+    snprintf(descriptor_path, PATH_MAX, "%s/descriptor.json", args->index_path);
+    index_descriptor_t desc = read_index_descriptor(descriptor_path);
+
+    IndexCtx.es_url = args->es_url;
+
+    LOG_DEBUGF("main.c", "descriptor version %s (%s)", desc.version, desc.type)
+
+    execute_update_script(args->script, desc.uuid);
+    free(args->script);
 }
 
 void sist2_web(web_args_t *args) {
 
     WebCtx.es_url = args->es_url;
     WebCtx.index_count = args->index_count;
+    WebCtx.auth_user = args->auth_user;
+    WebCtx.auth_pass = args->auth_pass;
+    WebCtx.auth_enabled = args->auth_enabled;
+    WebCtx.tag_auth_enabled = args->tag_auth_enabled;
 
     for (int i = 0; i < args->index_count; i++) {
         char *abs_path = abspath(args->indices[i]);
@@ -174,7 +360,11 @@ void sist2_web(web_args_t *args) {
         char path_tmp[PATH_MAX];
 
         snprintf(path_tmp, PATH_MAX, "%sthumbs", abs_path);
-        WebCtx.indices[i].store = store_create(path_tmp);
+        WebCtx.indices[i].store = store_create(path_tmp, STORE_SIZE_TN);
+
+        snprintf(path_tmp, PATH_MAX, "%stags", abs_path);
+        mkdir(path_tmp, S_IWUSR | S_IRUSR | S_IXUSR);
+        WebCtx.indices[i].tag_store = store_create(path_tmp, STORE_SIZE_TAG);
 
         snprintf(path_tmp, PATH_MAX, "%sdescriptor.json", abs_path);
         WebCtx.indices[i].desc = read_index_descriptor(path_tmp);
@@ -184,46 +374,76 @@ void sist2_web(web_args_t *args) {
         free(abs_path);
     }
 
-    serve(args->bind, args->port);
+    serve(args->listen_address);
 }
 
 
 int main(int argc, const char *argv[]) {
-
-    global_init();
+    setlocale(LC_ALL, "");
 
     scan_args_t *scan_args = scan_args_create();
     index_args_t *index_args = index_args_create();
     web_args_t *web_args = web_args_create();
+    exec_args_t *exec_args = exec_args_create();
 
-    char * common_es_url;
+    int arg_version = 0;
+
+    char *common_es_url = NULL;
+    char *common_script_path = NULL;
+    int common_threads = 0;
 
     struct argparse_option options[] = {
             OPT_HELP(),
 
+            OPT_BOOLEAN('v', "version", &arg_version, "Show version and exit"),
+            OPT_BOOLEAN(0, "verbose", &LogCtx.verbose, "Turn on logging"),
+            OPT_BOOLEAN(0, "very-verbose", &LogCtx.very_verbose, "Turn on debug messages"),
+
             OPT_GROUP("Scan options"),
-            OPT_INTEGER('t', "threads", &scan_args->threads, "Number of threads. DEFAULT=1"),
+            OPT_INTEGER('t', "threads", &common_threads, "Number of threads. DEFAULT=1"),
             OPT_FLOAT('q', "quality", &scan_args->quality,
-                      "Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best. DEFAULT=15"),
-            OPT_INTEGER(0, "size", &scan_args->size, "Thumbnail size, in pixels. DEFAULT=200"),
+                      "Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best. DEFAULT=5"),
+            OPT_INTEGER(0, "size", &scan_args->size,
+                        "Thumbnail size, in pixels. Use negative value to disable. DEFAULT=500"),
             OPT_INTEGER(0, "content-size", &scan_args->content_size,
-                        "Number of bytes to be extracted from text documents. DEFAULT=4096"),
+                        "Number of bytes to be extracted from text documents. Use negative value to disable. DEFAULT=32768"),
             OPT_STRING(0, "incremental", &scan_args->incremental,
                        "Reuse an existing index and only scan modified files."),
             OPT_STRING('o', "output", &scan_args->output, "Output directory. DEFAULT=index.sist2/"),
             OPT_STRING(0, "rewrite-url", &scan_args->rewrite_url, "Serve files from this url instead of from disk."),
             OPT_STRING(0, "name", &scan_args->name, "Index display name. DEFAULT: (name of the directory)"),
+            OPT_INTEGER(0, "depth", &scan_args->depth, "Scan up to DEPTH subdirectories deep. "
+                                                       "Use 0 to only scan files in PATH. DEFAULT: -1"),
+            OPT_STRING(0, "archive", &scan_args->archive, "Archive file mode (skip|list|shallow|recurse). "
+                                                          "skip: Don't parse, list: only get file names as text, "
+                                                          "shallow: Don't parse archives inside archives. DEFAULT: recurse"),
+            OPT_STRING(0, "ocr", &scan_args->tesseract_lang, "Tesseract language (use tesseract --list-langs to see "
+                                                             "which are installed on your machine)"),
+            OPT_STRING('e', "exclude", &scan_args->exclude_regex, "Files that match this regex will not be scanned"),
+            OPT_BOOLEAN(0, "fast", &scan_args->fast, "Only index file names & mime type"),
+            OPT_STRING(0, "treemap-threshold", &scan_args->treemap_threshold_str, "Relative size threshold for treemap "
+                                                                                  "(see USAGE.md). DEFAULT: 0.0005"),
+            OPT_INTEGER(0, "mem-buffer", &scan_args->max_memory_buffer,
+                        "Maximum memory buffer size per thread in MB for files inside archives "
+                        "(see USAGE.md). DEFAULT: 2000"),
 
             OPT_GROUP("Index options"),
-            OPT_STRING(0, "es-url", &common_es_url, "Elasticsearch url. DEFAULT=http://localhost:9200"),
+            OPT_INTEGER('t', "threads", &common_threads, "Number of threads. DEFAULT=1"),
+            OPT_STRING(0, "es-url", &common_es_url, "Elasticsearch url with port. DEFAULT=http://localhost:9200"),
             OPT_BOOLEAN('p', "print", &index_args->print, "Just print JSON documents to stdout."),
+            OPT_STRING(0, "script-file", &common_script_path, "Path to user script."),
+            OPT_INTEGER(0, "batch-size", &index_args->batch_size, "Index batch size. DEFAULT: 100"),
             OPT_BOOLEAN('f', "force-reset", &index_args->force_reset, "Reset Elasticsearch mappings and settings. "
-                                                              "(You must use this option the first time you use the index command)"),
+                                                                      "(You must use this option the first time you use the index command)"),
 
             OPT_GROUP("Web options"),
             OPT_STRING(0, "es-url", &common_es_url, "Elasticsearch url. DEFAULT=http://localhost:9200"),
-            OPT_STRING(0, "bind", &web_args->bind, "Listen on this address. DEFAULT=localhost"),
-            OPT_STRING(0, "port", &web_args->port, "Listen on this port. DEFAULT=4090"),
+            OPT_STRING(0, "bind", &web_args->listen_address, "Listen on this address. DEFAULT=localhost:4090"),
+            OPT_STRING(0, "auth", &web_args->credentials, "Basic auth in user:password format"),
+            OPT_STRING(0, "tag-auth", &web_args->tag_credentials, "Basic auth in user:password format for tagging"),
+
+            OPT_GROUP("Exec-script options"),
+            OPT_STRING(0, "script-file", &common_script_path, "Path to user script."),
 
             OPT_END(),
     };
@@ -233,17 +453,31 @@ int main(int argc, const char *argv[]) {
     argparse_describe(&argparse, DESCRIPTION, EPILOG);
     argc = argparse_parse(&argparse, argc, argv);
 
+    if (arg_version) {
+        printf(Version);
+        goto end;
+    }
+
+    if (LogCtx.very_verbose != 0) {
+        LogCtx.verbose = 1;
+    }
+
     web_args->es_url = common_es_url;
     index_args->es_url = common_es_url;
+    exec_args->es_url = common_es_url;
+    index_args->script_path = common_script_path;
+    exec_args->script_path = common_script_path;
+    index_args->threads = common_threads;
+    scan_args->threads = common_threads;
 
     if (argc == 0) {
         argparse_usage(&argparse);
-        return 1;
+        goto end;
     } else if (strcmp(argv[0], "scan") == 0) {
 
         int err = scan_args_validate(scan_args, argc, argv);
         if (err != 0) {
-            return err;
+            goto end;
         }
         sist2_scan(scan_args);
 
@@ -251,7 +485,7 @@ int main(int argc, const char *argv[]) {
 
         int err = index_args_validate(index_args, argc, argv);
         if (err != 0) {
-            return err;
+            goto end;
         }
         sist2_index(index_args);
 
@@ -259,15 +493,30 @@ int main(int argc, const char *argv[]) {
 
         int err = web_args_validate(web_args, argc, argv);
         if (err != 0) {
-            return err;
+            goto end;
         }
         sist2_web(web_args);
 
+    }  else if (strcmp(argv[0], "exec-script") == 0) {
+
+        int err = exec_args_validate(exec_args, argc, argv);
+        if (err != 0) {
+            goto end;
+        }
+        sist2_exec_script(exec_args);
+
     } else {
         fprintf(stderr, "Invalid command: '%s'\n", argv[0]);
         argparse_usage(&argparse);
-        return 1;
+        goto end;
     }
     printf("\n");
+
+    end:
+    scan_args_destroy(scan_args);
+    index_args_destroy(index_args);
+    web_args_destroy(web_args);
+    exec_args_destroy(exec_args);
+
     return 0;
 }

src/parsing/font.c

@@ -1,211 +0,0 @@
-#include "font.h"
-
-#include "ft2build.h"
-#include "freetype/freetype.h"
-
-#include "src/ctx.h"
-
-__thread FT_Library library = NULL;
-
-
-typedef struct text_dimensions {
-    unsigned int width;
-    unsigned int height;
-    unsigned int baseline;
-} text_dimensions_t;
-
-typedef struct glyph {
-    unsigned int top;
-    unsigned int height;
-    unsigned int width;
-    unsigned int descent;
-    unsigned int ascent;
-    unsigned int advance_width;
-    unsigned char *pixmap;
-} glyph_t;
-
-
-__always_inline
-int kerning_offset(char c, char pc, FT_Face face) {
-    FT_Vector kerning;
-    FT_Get_Kerning(face, c, pc, FT_KERNING_DEFAULT, &kerning);
-
-    return (int) (kerning.x / 64);
-}
-
-__always_inline
-glyph_t ft_glyph_to_glyph(FT_GlyphSlot slot) {
-    glyph_t glyph;
-
-    glyph.pixmap = slot->bitmap.buffer;
-
-    glyph.width = slot->bitmap.width;
-    glyph.height = slot->bitmap.rows;
-    glyph.top = slot->bitmap_top;
-    glyph.advance_width = slot->advance.x / 64;
-
-    glyph.descent = MAX(0, glyph.height - glyph.top);
-    glyph.ascent = MAX(0, MAX(glyph.top, glyph.height) - glyph.descent);
-
-    return glyph;
-}
-
-__always_inline
-glyph_t get_glyph(char character, FT_Face face) {
-}
-
-text_dimensions_t text_dimension(char *text, FT_Face face) {
-    text_dimensions_t dimensions;
-
-    dimensions.width = 0;
-
-    int num_chars = (int) strlen(text);
-
-    unsigned int max_ascent = 0;
-    unsigned int max_descent = 0;
-
-    char pc = 0;
-    for (int i = 0; i < num_chars; i++) {
-        char c = text[i];
-
-        FT_Load_Char(face, c, 0);
-        glyph_t glyph = ft_glyph_to_glyph(face->glyph);
-
-        max_descent = MAX(max_descent, glyph.descent);
-        max_ascent = MAX(max_ascent, glyph.ascent);
-
-        int kerning_x = kerning_offset(c, pc, face);
-        dimensions.width += MAX(glyph.advance_width, glyph.width) + kerning_x;
-
-        pc = c;
-    }
-
-    dimensions.height = max_ascent + max_descent;
-    dimensions.baseline = max_descent;
-
-    return dimensions;
-}
-
-void draw_glyph(glyph_t *glyph, int x, int y, struct text_dimensions text_info, unsigned char *bitmap) {
-    unsigned int src = 0;
-    unsigned int dst = y * text_info.width + x;
-    unsigned int row_offset = text_info.width - glyph->width;
-    unsigned int buf_len = text_info.width * text_info.height;
-
-    for (unsigned int sy = 0; sy < glyph->height; sy++) {
-        for (unsigned int sx = 0; sx < glyph->width; sx++) {
-            if (dst < buf_len) {
-                bitmap[dst] |= glyph->pixmap[src];
-            }
-            src++;
-            dst++;
-        }
-        dst += row_offset;
-    }
-}
-
-void bmp_format(dyn_buffer_t *buf, text_dimensions_t dimensions, const unsigned char *bitmap) {
-
-    dyn_buffer_write_short(buf, 0x4D42); // Magic
-    dyn_buffer_write_int(buf, 0); // Size placeholder
-    dyn_buffer_write_int(buf, 0x5157); //Reserved
-    dyn_buffer_write_int(buf, 14 + 40 + 256 * 4); // pixels offset
-
-    dyn_buffer_write_int(buf, 40); // DIB size
-    dyn_buffer_write_int(buf, (int) dimensions.width);
-    dyn_buffer_write_int(buf, (int) dimensions.height);
-    dyn_buffer_write_short(buf, 1); // Color planes
-    dyn_buffer_write_short(buf, 8); // bits per pixel
-    dyn_buffer_write_int(buf, 0); // compression
-    dyn_buffer_write_int(buf, 0); // Ignored
-    dyn_buffer_write_int(buf, 3800); // hres
-    dyn_buffer_write_int(buf, 3800); // vres
-    dyn_buffer_write_int(buf, 256); // Color count
-    dyn_buffer_write_int(buf, 0); // Ignored
-
-    // RGBA32 Color table (Grayscale)
-    for (int i = 255; i >= 0; i--) {
-        dyn_buffer_write_int(buf, i + (i << 8) + (i << 16));
-    }
-
-    // Pixel array: write from bottom to top, with rows padded to multiples of 4-bytes
-    for (int y = (int) dimensions.height - 1; y >= 0; y--) {
-        for (unsigned int x = 0; x < dimensions.width; x++) {
-            dyn_buffer_write_char(buf, (char) bitmap[y * dimensions.width + x]);
-        }
-        while (buf->cur % 4 != 0) {
-            dyn_buffer_write_char(buf, 0);
-        }
-    }
-
-    // Size
-    *(int *) ((char *) buf->buf + 2) = buf->cur;
-}
-
-void parse_font(const char *buf, size_t buf_len, document_t *doc) {
-    if (library == NULL) {
-        FT_Init_FreeType(&library);
-    }
-
-    FT_Face face;
-    FT_Error err = FT_New_Memory_Face(library, (unsigned char *) buf, buf_len, 0, &face);
-    if (err != 0) {
-        return;
-    }
-
-    char font_name[1024];
-
-    if (face->style_name == NULL || *(face->style_name) == '?') {
-        strcpy(font_name, face->family_name);
-    } else {
-        snprintf(font_name, sizeof(font_name), "%s %s", face->family_name, face->style_name);
-    }
-
-    meta_line_t *meta_name = malloc(sizeof(meta_line_t) + strlen(font_name));
-    meta_name->key = MetaFontName;
-    strcpy(meta_name->strval, font_name);
-    APPEND_META(doc, meta_name)
-
-    int pixel = 64;
-    int num_chars = (int) strlen(font_name);
-
-    err = FT_Set_Pixel_Sizes(face, 0, pixel);
-    if (err != 0) {
-        return;
-    }
-
-    text_dimensions_t dimensions = text_dimension(font_name, face);
-    unsigned char *bitmap = calloc(dimensions.width * dimensions.height, 1);
-
-    FT_Vector pen;
-    pen.x = 0;
-
-    char pc = 0;
-    for (int i = 0; i < num_chars; i++) {
-        char c = font_name[i];
-
-        err = FT_Load_Char(face, c, FT_LOAD_NO_HINTING | FT_LOAD_RENDER);
-        if (err != 0) {
-            continue;
-        }
-        glyph_t glyph = ft_glyph_to_glyph(face->glyph);
-
-        pen.x += kerning_offset(c, pc, face);
-        pen.y = dimensions.height - glyph.ascent - dimensions.baseline;
-
-        draw_glyph(&glyph, pen.x, pen.y, dimensions, bitmap);
-
-        pen.x += glyph.advance_width;
-        pc = c;
-    }
-
-    dyn_buffer_t bmp_data = dyn_buffer_create();
-    bmp_format(&bmp_data, dimensions, bitmap);
-
-    store_write(ScanCtx.index.store, (char *) doc->uuid, sizeof(doc->uuid), (char *) bmp_data.buf, bmp_data.cur);
-
-    dyn_buffer_destroy(&bmp_data);
-    free(bitmap);
-
-    FT_Done_Face(face);
-}

src/parsing/font.h

@@ -1,9 +0,0 @@
-#ifndef SIST2_FONT_H
-#define SIST2_FONT_H
-
-#include "src/sist.h"
-
-
-void parse_font(const char * buf, size_t buf_len, document_t *doc);
-
-#endif

src/parsing/media.c

@@ -1,273 +0,0 @@
-#include "src/sist.h"
-#include "src/ctx.h"
-
-AVCodecContext *alloc_jpeg_encoder(int dstW, int dstH, float qscale) {
-
-    AVCodec *jpeg_codec = avcodec_find_encoder(AV_CODEC_ID_MJPEG);
-    AVCodecContext *jpeg = avcodec_alloc_context3(jpeg_codec);
-    jpeg->width = dstW;
-    jpeg->height = dstH;
-    jpeg->time_base.den = 1000000;
-    jpeg->time_base.num = 1;
-    jpeg->i_quant_factor = qscale;
-
-    jpeg->pix_fmt = AV_PIX_FMT_YUVJ420P;
-    int ret = avcodec_open2(jpeg, jpeg_codec, NULL);
-
-    if (ret != 0) {
-        printf("Could not open jpeg encoder: %s!\n", av_err2str(ret));
-        return NULL;
-    }
-
-    return jpeg;
-}
-
-AVFrame *scale_frame(const AVCodecContext *decoder, const AVFrame *frame, int size) {
-    AVFrame *scaled_frame = av_frame_alloc();
-
-    int dstW;
-    int dstH;
-    if (frame->width <= size && frame->height <= size) {
-        dstW = frame->width;
-        dstH = frame->height;
-    } else {
-        double ratio = (double) frame->width / frame->height;
-        if (frame->width > frame->height) {
-            dstW = size;
-            dstH = (int) (size / ratio);
-        } else {
-            dstW = (int) (size * ratio);
-            dstH = size;
-        }
-    }
-
-    struct SwsContext *ctx = sws_getContext(
-            decoder->width, decoder->height, decoder->pix_fmt,
-            dstW, dstH, AV_PIX_FMT_YUVJ420P,
-            SWS_FAST_BILINEAR, 0, 0, 0
-    );
-
-    int dst_buf_len = avpicture_get_size(AV_PIX_FMT_YUVJ420P, dstW, dstH);
-    uint8_t *dst_buf = (uint8_t *) av_malloc(dst_buf_len);
-
-    avpicture_fill((AVPicture *) scaled_frame, dst_buf, AV_PIX_FMT_YUVJ420P, dstW, dstH);
-
-    sws_scale(ctx,
-              (const uint8_t *const *) frame->data, frame->linesize,
-              0, decoder->height,
-              scaled_frame->data, scaled_frame->linesize
-    );
-
-    scaled_frame->width = dstW;
-    scaled_frame->height = dstH;
-    scaled_frame->format = AV_PIX_FMT_YUV420P;
-
-    sws_freeContext(ctx);
-
-    return scaled_frame;
-}
-
-AVFrame *read_frame(AVFormatContext *pFormatCtx, AVCodecContext *decoder, int stream_idx) {
-    AVFrame *frame = av_frame_alloc();
-
-    AVPacket avPacket;
-    av_init_packet(&avPacket);
-
-    int receive_ret = -EAGAIN;
-    while (receive_ret == -EAGAIN) {
-        // Get video frame
-        while (1) {
-            int read_frame_ret = av_read_frame(pFormatCtx, &avPacket);
-
-            if (read_frame_ret != 0) {
-                if (read_frame_ret != AVERROR_EOF) {
-                    fprintf(stderr, "Error reading frame: %s\n", av_err2str(read_frame_ret));
-                }
-                av_frame_free(&frame);
-                av_packet_unref(&avPacket);
-                return NULL;
-            }
-
-            //Ignore audio/other frames
-            if (avPacket.stream_index != stream_idx) {
-                av_packet_unref(&avPacket);
-                continue;
-            }
-            break;
-        }
-
-        // Feed it to decoder
-        int decode_ret = avcodec_send_packet(decoder, &avPacket);
-        if (decode_ret != 0) {
-            printf("Error decoding frame: %s\n", av_err2str(decode_ret));
-        }
-        av_packet_unref(&avPacket);
-        receive_ret = avcodec_receive_frame(decoder, frame);
-    }
-    return frame;
-}
-
-void append_audio_meta(AVFormatContext *pFormatCtx, document_t *doc) {
-
-    AVDictionaryEntry *tag = NULL;
-    while ((tag = av_dict_get(pFormatCtx->metadata, "", tag, AV_DICT_IGNORE_SUFFIX))) {
-        char *key = tag->key;
-        for (; *key; ++key) *key = (char) tolower(*key);
-
-        if (strcmp(tag->key, "artist") == 0) {
-            size_t len = strlen(tag->value);
-            meta_line_t *meta_tag = malloc(sizeof(meta_line_t) + len);
-            meta_tag->key = MetaArtist;
-            memcpy(meta_tag->strval, tag->value, len);
-            APPEND_META(doc, meta_tag)
-        } else if (strcmp(tag->key, "genre") == 0) {
-            size_t len = strlen(tag->value);
-            meta_line_t *meta_tag = malloc(sizeof(meta_line_t) + len);
-            meta_tag->key = MetaGenre;
-            memcpy(meta_tag->strval, tag->value, len);
-            APPEND_META(doc, meta_tag)
-        } else if (strcmp(tag->key, "title") == 0) {
-            size_t len = strlen(tag->value);
-            meta_line_t *meta_tag = malloc(sizeof(meta_line_t) + len);
-            meta_tag->key = MetaTitle;
-            memcpy(meta_tag->strval, tag->value, len);
-            APPEND_META(doc, meta_tag)
-        } else if (strcmp(tag->key, "album_artist") == 0) {
-            size_t len = strlen(tag->value);
-            meta_line_t *meta_tag = malloc(sizeof(meta_line_t) + len);
-            meta_tag->key = MetaAlbumArtist;
-            memcpy(meta_tag->strval, tag->value, len);
-            APPEND_META(doc, meta_tag)
-        } else if (strcmp(tag->key, "album") == 0) {
-            size_t len = strlen(tag->value);
-            meta_line_t *meta_tag = malloc(sizeof(meta_line_t) + len);
-            meta_tag->key = MetaAlbum;
-            memcpy(meta_tag->strval, tag->value, len);
-            APPEND_META(doc, meta_tag)
-        }
-    }
-}
-
-void parse_media(const char *filepath, document_t *doc) {
-
-    int video_stream = -1;
-
-    AVFormatContext *pFormatCtx = avformat_alloc_context();
-    if (pFormatCtx == NULL) {
-        fprintf(stderr, "Could not allocate AVFormatContext! %s \n", filepath);
-        return;
-    }
-    int res = avformat_open_input(&pFormatCtx, filepath, NULL, NULL);
-    if (res < 0) {
-        printf("ERR%s %s\n", filepath, av_err2str(res));
-        return;
-    }
-
-    avformat_find_stream_info(pFormatCtx, NULL);
-
-    for (int i = (int) pFormatCtx->nb_streams - 1; i >= 0; i--) {
-        AVStream *stream = pFormatCtx->streams[i];
-
-        if (stream->codecpar->codec_type == AVMEDIA_TYPE_AUDIO) {
-            meta_line_t *meta_audio = malloc(sizeof(meta_line_t));
-            meta_audio->key = MetaMediaAudioCodec;
-            meta_audio->intval = stream->codecpar->codec_id;
-            APPEND_META(doc, meta_audio)
-
-            append_audio_meta(pFormatCtx, doc);
-
-        } else if (stream->codecpar->codec_type == AVMEDIA_TYPE_VIDEO) {
-
-            meta_line_t *meta_vid = malloc(sizeof(meta_line_t));
-            meta_vid->key = MetaMediaVideoCodec;
-            meta_vid->intval = stream->codecpar->codec_id;
-            APPEND_META(doc, meta_vid)
-
-            meta_line_t *meta_w = malloc(sizeof(meta_line_t));
-            meta_w->key = MetaWidth;
-            meta_w->intval = stream->codecpar->width;
-            APPEND_META(doc, meta_w)
-
-            meta_line_t *meta_h = malloc(sizeof(meta_line_t));
-            meta_h->key = MetaHeight;
-            meta_h->intval = stream->codecpar->height;
-            APPEND_META(doc, meta_h)
-
-            video_stream = i;
-        }
-    }
-
-    if (video_stream != -1) {
-        AVStream *stream = pFormatCtx->streams[video_stream];
-
-        if (stream->nb_frames > 1) {
-            //This is a video (not a still image)
-            meta_line_t *meta_duration = malloc(sizeof(meta_line_t));
-            meta_duration->key = MetaMediaDuration;
-            meta_duration->longval = pFormatCtx->duration / AV_TIME_BASE;
-            APPEND_META(doc, meta_duration)
-
-            meta_line_t *meta_bitrate = malloc(sizeof(meta_line_t));
-            meta_bitrate->key = MetaMediaBitrate;
-            meta_bitrate->intval = pFormatCtx->bit_rate;
-            APPEND_META(doc, meta_bitrate)
-        }
-
-        if (stream->codecpar->width <= 20 || stream->codecpar->height <= 20) {
-            avformat_close_input(&pFormatCtx);
-            avformat_free_context(pFormatCtx);
-            return;
-        }
-
-        // Decoder
-        AVCodec *video_codec = avcodec_find_decoder(stream->codecpar->codec_id);
-        AVCodecContext *decoder = avcodec_alloc_context3(video_codec);
-        avcodec_parameters_to_context(decoder, stream->codecpar);
-        avcodec_open2(decoder, video_codec, NULL);
-
-        //Seek
-        if (stream->nb_frames > 1 && stream->codecpar->codec_id != AV_CODEC_ID_GIF) {
-            int seek_ret = 0;
-            for (int i = 20; i >= 0; i--) {
-                seek_ret = av_seek_frame(pFormatCtx, video_stream,
-                                         stream->duration * 0.10, 0);
-                if (seek_ret == 0) {
-                    break;
-                }
-            }
-        }
-
-        AVFrame *frame = read_frame(pFormatCtx, decoder, video_stream);
-        if (frame == NULL) {
-            avcodec_free_context(&decoder);
-            avformat_close_input(&pFormatCtx);
-            avformat_free_context(pFormatCtx);
-            return;
-        }
-
-        // Scale frame
-        AVFrame *scaled_frame = scale_frame(decoder, frame, ScanCtx.tn_size);
-
-        // Encode frame to jpeg
-        AVCodecContext *jpeg_encoder = alloc_jpeg_encoder(scaled_frame->width, scaled_frame->height, ScanCtx.tn_qscale);
-        avcodec_send_frame(jpeg_encoder, scaled_frame);
-
-        AVPacket jpeg_packet;
-        av_init_packet(&jpeg_packet);
-        avcodec_receive_packet(jpeg_encoder, &jpeg_packet);
-
-        // Save thumbnail
-        store_write(ScanCtx.index.store, (char *) doc->uuid, sizeof(doc->uuid), (char *) jpeg_packet.data, jpeg_packet.size);
-
-        av_packet_unref(&jpeg_packet);
-        av_frame_free(&frame);
-        av_free(*scaled_frame->data);
-        av_frame_free(&scaled_frame);
-        avcodec_free_context(&jpeg_encoder);
-        avcodec_free_context(&decoder);
-    }
-
-    avformat_close_input(&pFormatCtx);
-    avformat_free_context(pFormatCtx);
-}
-

src/parsing/media.h

@@ -1,11 +0,0 @@
-#ifndef SIST2_MEDIA_H
-#define SIST2_MEDIA_H
-
-
-#include "src/sist.h"
-
-#define MIN_VIDEO_SIZE 1024 * 64
-
-void parse_media(const char * filepath, document_t *doc);
-
-#endif

src/parsing/mime.c

@@ -1,10 +1,12 @@
 #include "mime.h"
 
 unsigned int mime_get_mime_by_ext(GHashTable *ext_table, const char * ext) {
-    char lower[64];
+    char lower[8];
     char *p = lower;
-    while ((*ext)) {
+    int cnt = 0;
+    while ((*ext) != '\0' && cnt + 1 < sizeof(lower)) {
         *p++ = (char)tolower(*ext++);
+        cnt++;
     }
     *p = '\0';
     return (size_t) g_hash_table_lookup(ext_table, lower);

src/parsing/mime.h

@@ -1,14 +1,14 @@
 #ifndef SIST2_MIME_H
 #define SIST2_MIME_H
 
-#include "src/sist.h"
+#include "../sist.h"
 
-#define MAJOR_MIME(mime_id) (mime_id & 0x0FFF0000) >> 16
+#define MAJOR_MIME(mime_id) (mime_id & 0x000F0000) >> 16
 
 #define MIME_EMPTY 1
 
 #define DONT_PARSE 0x80000000
-#define SHOULD_PARSE(mime_id) (mime_id & DONT_PARSE) != DONT_PARSE
+#define SHOULD_PARSE(mime_id) (ScanCtx.fast == 0 && (mime_id & DONT_PARSE) != DONT_PARSE && mime_id != 0)
 
 #define PDF_MASK 0x40000000
 #define IS_PDF(mime_id) (mime_id & PDF_MASK) == PDF_MASK
@@ -16,6 +16,24 @@
 #define FONT_MASK 0x20000000
 #define IS_FONT(mime_id) (mime_id & FONT_MASK) == FONT_MASK
 
+#define ARC_MASK 0x10000000
+#define IS_ARC(mime_id) (mime_id & ARC_MASK) == ARC_MASK
+
+#define ARC_FILTER_MASK 0x08000000
+#define IS_ARC_FILTER(mime_id) (mime_id & ARC_FILTER_MASK) == ARC_FILTER_MASK
+
+#define DOC_MASK 0x04000000
+#define IS_DOC(mime_id) (mime_id & DOC_MASK) == DOC_MASK
+
+#define MOBI_MASK 0x02000000
+#define IS_MOBI(mime_id) (mime_id & MOBI_MASK) == MOBI_MASK
+
+#define MARKUP_MASK 0x01000000
+#define IS_MARKUP(mime_id) (mime_id & MARKUP_MASK) == MARKUP_MASK
+
+#define RAW_MASK 0x00800000
+#define IS_RAW(mime_id) (mime_id & RAW_MASK) == RAW_MASK
+
 enum major_mime {
     MimeInvalid = 0,
     MimeModel = 1,

src/parsing/parse.c

@@ -1,34 +1,41 @@
+#include "parse.h"
+
 #include "src/sist.h"
 #include "src/ctx.h"
+#include "mime.h"
+#include "src/io/serialize.h"
 
-__thread magic_t Magic;
+#include <magic.h>
 
-void *read_all(parse_job_t *job, const char *buf, int bytes_read, int *fd) {
 
-    void *full_buf;
+#define MIN_VIDEO_SIZE 1024 * 64
+#define MIN_IMAGE_SIZE 1024 * 2
 
-    if (job->info.st_size <= bytes_read) {
-        full_buf = malloc(job->info.st_size);
-        memcpy(full_buf, buf, job->info.st_size);
-    } else {
-        if (*fd == -1) {
-            *fd = open(job->filepath, O_RDONLY);
-            if (*fd == -1) {
-                perror("open");
-                printf("%s\n", job->filepath);
-                free(job);
-                return NULL;
-            }
-        }
-        full_buf = malloc(job->info.st_size);
-        memcpy(full_buf, buf, bytes_read);
-        int ret = read(*fd, full_buf + bytes_read, job->info.st_size - bytes_read);
-        if (ret == -1) {
-            perror("read");
+int fs_read(struct vfile *f, void *buf, size_t size) {
+
+    if (f->fd == -1) {
+        f->fd = open(f->filepath, O_RDONLY);
+        if (f->fd == -1) {
+            LOG_ERRORF(f->filepath, "open(): [%d] %s", errno, strerror(errno))
+            return -1;
         }
     }
 
-    return full_buf;
+    return read(f->fd, buf, size);
+}
+
+#define CLOSE_FILE(f) if (f.close != NULL) {f.close(&f);};
+
+void fs_close(struct vfile *f) {
+    if (f->fd != -1) {
+        close(f->fd);
+    }
+}
+
+void fs_reset(struct vfile *f) {
+    if (f->fd != -1) {
+        lseek(f->fd, 0, SEEK_SET);
+    }
 }
 
 void parse(void *arg) {
@@ -36,91 +43,131 @@ void parse(void *arg) {
     parse_job_t *job = arg;
     document_t doc;
 
-    if (incremental_get(ScanCtx.original_table, job->info.st_ino) == job->info.st_mtim.tv_sec) {
-        incremental_mark_file_for_copy(ScanCtx.copy_table, job->info.st_ino);
-        free(job);
+    int inc_ts = incremental_get(ScanCtx.original_table, job->vfile.info.st_ino);
+    if (inc_ts != 0 && inc_ts == job->vfile.info.st_mtim.tv_sec) {
+        incremental_mark_file_for_copy(ScanCtx.copy_table, job->vfile.info.st_ino);
         return;
     }
 
-    if (Magic == NULL) {
-        Magic = magic_open(MAGIC_MIME_TYPE);
-        magic_load(Magic, NULL);
-    }
-
     doc.filepath = job->filepath;
     doc.ext = (short) job->ext;
     doc.base = (short) job->base;
     doc.meta_head = NULL;
     doc.meta_tail = NULL;
     doc.mime = 0;
-    doc.size = job->info.st_size;
-    doc.ino = job->info.st_ino;
-    doc.mtime = job->info.st_mtim.tv_sec;
+    doc.size = job->vfile.info.st_size;
+    doc.ino = job->vfile.info.st_ino;
+    doc.mtime = job->vfile.info.st_mtim.tv_sec;
 
-    uuid_generate_time_safe(doc.uuid);
-    char *buf[PARSE_BUF_SIZE];
+    uuid_generate(doc.uuid);
+    char *buf[MAGIC_BUF_SIZE];
 
-    if (job->info.st_size == 0) {
+    if (LogCtx.very_verbose) {
+        char uuid_str[UUID_STR_LEN];
+        uuid_unparse(doc.uuid, uuid_str);
+        LOG_DEBUGF(job->filepath, "Starting parse job {%s}", uuid_str)
+    }
+
+    if (job->vfile.info.st_size == 0) {
         doc.mime = MIME_EMPTY;
-    } else if (*(job->filepath + job->ext) != '\0') {
+    } else if (*(job->filepath + job->ext) != '\0' && (job->ext - job->base != 1)) {
         doc.mime = mime_get_mime_by_ext(ScanCtx.ext_table, job->filepath + job->ext);
     }
 
-    int fd = -1;
     int bytes_read = 0;
 
-    if (doc.mime == 0) {
+    if (doc.mime == 0 && !ScanCtx.fast) {
         // Get mime type with libmagic
-        fd = open(job->filepath, O_RDONLY);
-        if (fd == -1) {
-            perror("open");
-            free(job);
+        if (!job->vfile.is_fs_file) {
+            LOG_WARNING(job->filepath, "Guessing mime type with libmagic inside archive files is not currently supported");
+            goto abort;
+        }
+
+        bytes_read = job->vfile.read(&job->vfile, buf, MAGIC_BUF_SIZE);
+        if (bytes_read < 0) {
+
+            if (job->vfile.is_fs_file) {
+                LOG_ERRORF(job->filepath, "read(): [%d] %s", errno, strerror(errno))
+            } else {
+                LOG_ERRORF(job->filepath, "(virtual) read(): [%d] %s", bytes_read, archive_error_string(job->vfile.arc))
+            }
+
+            CLOSE_FILE(job->vfile)
             return;
         }
 
-        bytes_read = read(fd, buf, PARSE_BUF_SIZE);
+        magic_t magic = magic_open(MAGIC_MIME_TYPE);
+        magic_load(magic, NULL);
 
-        const char *magic_mime_str = magic_buffer(Magic, buf, bytes_read);
+        const char *magic_mime_str = magic_buffer(magic, buf, bytes_read);
         if (magic_mime_str != NULL) {
             doc.mime = mime_get_mime_by_string(ScanCtx.mime_table, magic_mime_str);
+
+            LOG_DEBUGF(job->filepath, "libmagic: %s", magic_mime_str);
+
             if (doc.mime == 0) {
-                fprintf(stderr, "Couldn't find mime %s, %s!\n", magic_mime_str, job->filepath + job->base);
+                LOG_WARNINGF(job->filepath, "Couldn't find mime %s", magic_mime_str);
             }
         }
+
+        job->vfile.reset(&job->vfile);
+
+        magic_close(magic);
     }
 
     int mmime = MAJOR_MIME(doc.mime);
 
     if (!(SHOULD_PARSE(doc.mime))) {
 
-    } else if ((mmime == MimeVideo && doc.size >= MIN_VIDEO_SIZE) || mmime == MimeAudio || mmime == MimeImage) {
-//        parse_media(job->filepath, &doc);
+    } else if (IS_RAW(doc.mime)) {
+        parse_raw(&ScanCtx.raw_ctx, &job->vfile, &doc);
+    } else if ((mmime == MimeVideo && doc.size >= MIN_VIDEO_SIZE) ||
+               (mmime == MimeImage && doc.size >= MIN_IMAGE_SIZE) || mmime == MimeAudio) {
+
+        parse_media(&ScanCtx.media_ctx, &job->vfile, &doc);
 
     } else if (IS_PDF(doc.mime)) {
-        void *pdf_buf = read_all(job, (char *) buf, bytes_read, &fd);
-        parse_pdf(pdf_buf, doc.size, &doc);
+        parse_ebook(&ScanCtx.ebook_ctx, &job->vfile, mime_get_mime_text(doc.mime), &doc);
 
-        if (pdf_buf != buf) {
-            free(pdf_buf);
+    } else if (mmime == MimeText && ScanCtx.text_ctx.content_size > 0) {
+        if (IS_MARKUP(doc.mime)) {
+            parse_markup(&ScanCtx.text_ctx, &job->vfile, &doc);
+        } else {
+            parse_text(&ScanCtx.text_ctx, &job->vfile, &doc);
         }
 
-    } else if (mmime == MimeText && ScanCtx.content_size > 0) {
-//        parse_text(bytes_read, &fd, (char *) buf, &doc);
-
     } else if (IS_FONT(doc.mime)) {
-//        void *font_buf = read_all(job, (char *) buf, bytes_read, &fd);
-//        parse_font(font_buf, doc.size, &doc);
-//
-//        if (font_buf != buf) {
-//            free(font_buf);
-//        }
+        parse_font(&ScanCtx.font_ctx, &job->vfile, &doc);
+
+    } else if (
+            ScanCtx.arc_ctx.mode != ARC_MODE_SKIP && (
+                    IS_ARC(doc.mime) ||
+                    (IS_ARC_FILTER(doc.mime) && should_parse_filtered_file(doc.filepath, doc.ext))
+            )) {
+        parse_archive(&ScanCtx.arc_ctx, &job->vfile, &doc);
+    } else if ((ScanCtx.ooxml_ctx.content_size > 0 || ScanCtx.media_ctx.tn_size > 0) && IS_DOC(doc.mime)) {
+        parse_ooxml(&ScanCtx.ooxml_ctx, &job->vfile, &doc);
+    } else if (is_cbr(&ScanCtx.comic_ctx, doc.mime) || is_cbz(&ScanCtx.comic_ctx, doc.mime)) {
+        parse_comic(&ScanCtx.comic_ctx, &job->vfile, &doc);
+    } else if (IS_MOBI(doc.mime)) {
+        parse_mobi(&ScanCtx.mobi_ctx, &job->vfile, &doc);
+    }
+
+    abort:
+
+    //Parent meta
+    if (!uuid_is_null(job->parent)) {
+        meta_line_t *meta_parent = malloc(sizeof(meta_line_t) + UUID_STR_LEN + 1);
+        meta_parent->key = MetaParent;
+        uuid_unparse(job->parent, meta_parent->str_val);
+        APPEND_META((&doc), meta_parent)
     }
 
     write_document(&doc);
 
-    if (fd != -1) {
-        close(fd);
-    }
-
-    free(job);
+    CLOSE_FILE(job->vfile)
+}
+
+void cleanup_parse() {
+    // noop
 }

src/parsing/parse.h

@@ -1,10 +1,16 @@
 #ifndef SIST2_PARSE_H
 #define SIST2_PARSE_H
 
-#include "src/sist.h"
+#include "../sist.h"
 
-#define PARSE_BUF_SIZE 4096
+#define MAGIC_BUF_SIZE 4096 * 6
+
+int fs_read(struct vfile *f, void *buf, size_t size);
+void fs_close(struct vfile *f);
+void fs_reset(struct vfile *f);
 
 void parse(void *arg);
 
+void cleanup_parse();
+
 #endif

src/parsing/pdf.c

@@ -1,142 +0,0 @@
-#include "pdf.h"
-#include "src/ctx.h"
-
-__always_inline
-fz_page *render_cover(fz_context *ctx, document_t *doc, fz_document *fzdoc) {
-
-    fz_page *cover = fz_load_page(ctx, fzdoc, 0);
-    fz_rect bounds = fz_bound_page(ctx, cover);
-
-    float scale;
-    float w = (float) bounds.x1 - bounds.x0;
-    float h = (float) bounds.y1 - bounds.y0;
-    if (w > h) {
-        scale = (float) ScanCtx.tn_size / w;
-    } else {
-        scale = (float) ScanCtx.tn_size / h;
-    }
-    fz_matrix m = fz_scale(scale, scale);
-
-    bounds = fz_transform_rect(bounds, m);
-    fz_irect bbox = fz_round_rect(bounds);
-    fz_pixmap *pixmap = fz_new_pixmap_with_bbox(ctx, ctx->colorspace->rgb, bbox, NULL, 0);
-
-    fz_clear_pixmap_with_value(ctx, pixmap, 0xFF);
-    fz_device *dev = fz_new_draw_device(ctx, m, pixmap);
-
-    pthread_mutex_lock(&ScanCtx.mupdf_mu);
-    fz_run_page(ctx, cover, dev, fz_identity, NULL);
-    pthread_mutex_unlock(&ScanCtx.mupdf_mu);
-
-    fz_drop_device(ctx, dev);
-
-    fz_buffer *fzbuf = fz_new_buffer_from_pixmap_as_png(ctx, pixmap, fz_default_color_params);
-    unsigned char *tn_buf;
-    size_t tn_len = fz_buffer_storage(ctx, fzbuf, &tn_buf);
-
-    store_write(ScanCtx.index.store, (char *) doc->uuid, sizeof(doc->uuid), (char *) tn_buf, tn_len);
-
-    fz_drop_pixmap(ctx, pixmap);
-    fz_drop_buffer(ctx, fzbuf);
-
-    return cover;
-}
-
-void fz_noop_callback(__attribute__((unused)) void *user, __attribute__((unused)) const char *message) {}
-
-
-void parse_pdf(void *buf, size_t buf_len, document_t *doc) {
-
-    static int mu_is_initialized = 0;
-    if (!mu_is_initialized) {
-        pthread_mutex_init(&ScanCtx.mupdf_mu, NULL);
-        mu_is_initialized = 1;
-    }
-    fz_context *ctx = fz_new_context(NULL, NULL, FZ_STORE_UNLIMITED);
-    fz_stream *stream = NULL;
-    fz_document *fzdoc = NULL;
-
-    fz_var(stream);
-    fz_var(fzdoc);
-
-    fz_try(ctx)
-    {
-        fz_disable_icc(ctx);
-        fz_register_document_handlers(ctx);
-
-        //disable warnings
-        ctx->warn.print = fz_noop_callback;
-        ctx->error.print = fz_noop_callback;
-
-        stream = fz_open_memory(ctx, buf, buf_len);
-        fzdoc = fz_open_document_with_stream(ctx, mime_get_mime_text(doc->mime), stream);
-
-        int page_count = fz_count_pages(ctx, fzdoc);
-        fz_page *cover = render_cover(ctx, doc, fzdoc);
-
-        fz_stext_options opts;
-
-        text_buffer_t text_buf = text_buffer_create(ScanCtx.content_size);
-
-        for (int current_page = 0; current_page < page_count; current_page++) {
-            fz_page *page;
-            if (current_page == 0) {
-                page = cover;
-            } else {
-                page = fz_load_page(ctx, fzdoc, current_page);
-            }
-
-            fz_stext_page *stext = fz_new_stext_page(ctx, fz_bound_page(ctx, page));
-            fz_device *dev = fz_new_stext_device(ctx, stext, &opts);
-
-            pthread_mutex_lock(&ScanCtx.mupdf_mu);
-            fz_run_page_contents(ctx, page, dev, fz_identity, NULL);
-            pthread_mutex_unlock(&ScanCtx.mupdf_mu);
-
-            fz_drop_device(ctx, dev);
-
-            fz_stext_block *block = stext->first_block;
-            while (block != NULL) {
-
-                if (block->type != FZ_STEXT_BLOCK_TEXT) {
-                    block = block->next;
-                    continue;
-                }
-
-                fz_stext_line *line = block->u.t.first_line;
-                while (line != NULL) {
-                    fz_stext_char *c = line->first_char;
-                    while (c != NULL) {
-                        if (text_buffer_append_char(&text_buf, c->c) == TEXT_BUF_FULL) {
-                            fz_drop_page(ctx, page);
-                            fz_drop_stext_page(ctx, stext);
-                            goto write_loop_end;
-                        }
-                        c = c->next;
-                    }
-                    line = line->next;
-                }
-                block = block->next;
-            }
-            fz_drop_page(ctx, page);
-            fz_drop_stext_page(ctx, stext);
-        }
-        write_loop_end:;
-        text_buffer_terminate_string(&text_buf);
-
-        meta_line_t *meta_content = malloc(sizeof(meta_line_t) + text_buf.dyn_buffer.cur);
-        meta_content->key = MetaContent;
-        memcpy(meta_content->strval, text_buf.dyn_buffer.buf, text_buf.dyn_buffer.cur);
-        text_buffer_destroy(&text_buf);
-        APPEND_META(doc, meta_content)
-    }
-    fz_always(ctx)
-    {
-        fz_drop_stream(ctx, stream);
-        fz_drop_document(ctx, fzdoc);
-        fz_drop_context(ctx);
-    } fz_catch(ctx) {
-        fprintf(stderr, "Error %s %s\n", doc->filepath, ctx->error.message);
-    }
-}
-

src/parsing/pdf.h

@@ -1,9 +0,0 @@
-#ifndef SIST2_PDF_H
-#define SIST2_PDF_H
-
-#include "src/sist.h"
-
-
-void parse_pdf(void *buf, size_t buf_len, document_t *doc);
-
-#endif

src/parsing/text.c

@@ -1,43 +0,0 @@
-#include "text.h"
-#include "src/ctx.h"
-
-void parse_text(int bytes_read, int *fd, char *buf, document_t *doc) {
-
-    char *intermediate_buf;
-    int intermediate_buf_len;
-
-    if (bytes_read == doc->size || bytes_read >= ScanCtx.content_size) {
-        int to_copy = MIN(bytes_read, ScanCtx.content_size);
-        intermediate_buf = malloc(to_copy);
-        intermediate_buf_len = to_copy;
-        memcpy(intermediate_buf, buf, to_copy);
-
-    } else {
-        if (*fd == -1) {
-            *fd = open(doc->filepath, O_RDONLY);
-        }
-
-        int to_read = MIN(ScanCtx.content_size, doc->size) - bytes_read;
-
-        intermediate_buf = malloc(to_read + bytes_read);
-        intermediate_buf_len = to_read + bytes_read;
-        if (bytes_read != 0) {
-            memcpy(intermediate_buf, buf, bytes_read);
-        }
-
-        read(*fd, intermediate_buf + bytes_read, to_read);
-    }
-
-    text_buffer_t text_buf = text_buffer_create(ScanCtx.content_size);
-    for (int i = 0; i < intermediate_buf_len; i++) {
-        text_buffer_append_char(&text_buf, *(intermediate_buf + i));
-    }
-    text_buffer_terminate_string(&text_buf);
-
-    meta_line_t *meta = malloc(sizeof(meta_line_t) + text_buf.dyn_buffer.cur);
-    meta->key = MetaContent;
-    strcpy(meta->strval, text_buf.dyn_buffer.buf);
-    text_buffer_destroy(&text_buf);
-    free(intermediate_buf);
-    APPEND_META(doc, meta)
-}

src/parsing/text.h

@@ -1,8 +0,0 @@
-#ifndef SIST2_TEXT_H
-#define SIST2_TEXT_H
-
-#include "src/sist.h"
-
-void parse_text(int bytes_read, int *fd, char *buf, document_t *doc);
-
-#endif

src/sist.h

@@ -1,56 +1,51 @@
 #ifndef SIST_H
 #define SIST_H
 
-#define UUID_STR_LEN 37
+#ifndef	FALSE
+#define	FALSE	(0)
+#define BOOL int
+#endif
 
-#include <glib-2.0/glib.h>
-#include <unistd.h>
+#ifndef	TRUE
+#define	TRUE	(!FALSE)
+#endif
+
+#undef	MAX
+#define MAX(a, b)  (((a) > (b)) ? (a) : (b))
+
+#undef	MIN
+#define MIN(a, b)  (((a) < (b)) ? (a) : (b))
+
+#ifndef PATH_MAX
+#define PATH_MAX 4096
+#endif
+
+#undef ABS
+#define ABS(a)	   (((a) < 0) ? -(a) : (a))
+
+#define UUID_STR_LEN 37
+#define UNUSED(x) __attribute__((__unused__))  x
+
+#include "util.h"
+#include "log.h"
+#include "types.h"
+
+#include "libscan/scan.h"
+
+#include <cjson/cJSON.h>
+
+#include <string.h>
 #include <stdio.h>
 #include <stdlib.h>
-#include <string.h>
+#include <stdarg.h>
+
+#include <unistd.h>
 #include <fcntl.h>
-#include <ftw.h>
-#include <uuid.h>
-#include <magic.h>
-#include <libavformat/avformat.h>
-#include <libswscale/swscale.h>
-#include <libswresample/swresample.h>
-#include <libavcodec/avcodec.h>
-#include <ctype.h>
-#include <mupdf/fitz.h>
-#include <mupdf/pdf.h>
-#include "argparse/argparse.h"
-#include <time.h>
-#include <limits.h>
-#include <pthread.h>
+#include <dirent.h>
 #include <sys/stat.h>
-#include <wordexp.h>
-#include <onion/onion.h>
-#include <onion/handler.h>
-#include <onion/block.h>
-#include <onion/shortcuts.h>
-#include <curl/curl.h>
+#include <sys/types.h>
+#include <errno.h>
+#include <ctype.h>
 
 
-#include "cJSON/cJSON.h"
-
-#include "types.h"
-#include "tpool.h"
-#include "util.h"
-#include "src/index/elastic.h"
-#include "io/store.h"
-#include "io/serialize.h"
-#include "io/walk.h"
-#include "parsing/parse.h"
-#include "parsing/mime.h"
-#include "parsing/text.h"
-#include "parsing/pdf.h"
-#include "parsing/media.h"
-#include "parsing/font.h"
-#include "index/web.h"
-#include "web/serve.h"
-#include "cli.h"
-
-;
-
 #endif

src/static/css/autocomplete.min.css

@@ -0,0 +1,4 @@
+.autocomplete-suggestions { text-align: left; cursor: default; border: 1px solid #ccc; border-top: 0; background: #fff; box-shadow: -1px 1px 3px rgba(0,0,0,.1); position: absolute; display: none; z-index: 9999; max-height: 254px; overflow: hidden; overflow-y: auto; box-sizing: border-box; }
+.autocomplete-suggestion { position: relative; padding: 0 .6em; line-height: 23px; white-space: nowrap; overflow: hidden; text-overflow: ellipsis; font-size: 1.02em; color: #333; }
+.autocomplete-suggestion b { font-weight: normal; color: #1f8dd6; }
+.autocomplete-suggestion.selected { background: #f0f0f0; }

src/static/css/bricklayer.min.css

@@ -0,0 +1 @@
+.bricklayer{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-box-align:start;-webkit-align-items:flex-start;-ms-flex-align:start;align-items:flex-start;-webkit-box-pack:center;-webkit-justify-content:center;-ms-flex-pack:center;justify-content:center;-webkit-flex-wrap:wrap;-ms-flex-wrap:wrap;flex-wrap:wrap}.bricklayer-column-sizer{width:100%;display:none}@media screen and (min-width:640px){.bricklayer-column-sizer{width:50%}}@media screen and (min-width:980px){.bricklayer-column-sizer{width:33.333%}}@media screen and (min-width:1200px){.bricklayer-column-sizer{width:25%}}.bricklayer-column{-webkit-box-flex:1;-webkit-flex:1;-ms-flex:1;flex:1;padding-left:5px;padding-right:5px}

src/static/css/dark.css

@@ -0,0 +1,532 @@
+*:focus {
+    outline: 0;
+}
+
+.info-icon {
+    width: 1rem;
+    margin-right: 0.2rem;
+    cursor: pointer;
+    line-height: 1rem;
+    height: 1rem;
+    background-image: url(data:image/svg+xml;base64,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);
+    filter: brightness(65%);
+}
+
+.info-icon:hover {
+    color: inherit;
+}
+
+.modal-title {
+    max-width: calc(100% - 2rem);
+    overflow: hidden;
+    text-overflow: ellipsis;
+}
+
+
+.path-row {
+    display: -ms-flexbox;
+    display: flex;
+    -ms-flex-align: start;
+    align-items: flex-start;
+}
+
+.tag-container {
+    margin-left: 0.3rem;
+}
+
+.path-line {
+    color: #BBB;
+    text-overflow: ellipsis;
+    overflow: hidden;
+    white-space: nowrap;
+}
+
+a {
+    color: #00BCD4;
+}
+
+body {
+    overflow-y: scroll;
+    background: black;
+}
+
+.progress {
+    margin-top: 1em;
+}
+
+.card, .modal-content {
+    margin-top: 1em;
+    background: #212121;
+    color: #e0e0e0;
+    border-radius: 1px;
+    border: none;
+}
+
+.table {
+    color: #e0e0e0;
+}
+
+.table td, .table th {
+    border: none;
+}
+
+.table thead th {
+    border-bottom: 1px solid #646464;
+}
+
+.modal-header .close {
+    color: #e0e0e0;
+    text-shadow: none;
+}
+
+.modal-header {
+    border-bottom: 1px solid #646464;
+}
+
+.sub-document {
+    background: #37474F !important;
+}
+
+.list-group-item.sub-document {
+    border-top: 1px solid #646464 !important;
+}
+
+.sub-document .text-muted {
+    color: #8a949c !important;
+}
+
+
+.list-group-item {
+    background: #212121;
+    color: #e0e0e0;
+
+    border-top: 1px solid #424242;
+    border-bottom: none;
+    border-left: none;
+    border-right: none;
+    padding: .25rem 0.5rem;
+}
+
+.list-group-item:first-child {
+    border-top: none;
+}
+
+.navbar-brand {
+    font-size: 1.75rem;
+    padding: 0;
+    color: #f5f5f5;
+}
+
+.navbar {
+    background: #546b7a;
+}
+
+a:hover, .btn:hover {
+    color: #fff;
+}
+
+.navbar span {
+    color: #eee;
+}
+
+.document {
+    padding: 0.3rem;
+}
+
+.card-text:last-child {
+    margin-top: -1px;
+}
+
+.document p {
+    margin-bottom: 0;
+}
+
+.document:hover p {
+    text-decoration: underline;
+}
+
+.badge-video {
+    color: #FFFFFF;
+    background-color: #F27761;
+}
+
+.badge-image {
+    color: #FFFFFF;
+    background-color: #AA99C9;
+}
+
+.badge-audio {
+    color: #FFFFFF;
+    background-color: #00ADEF;
+}
+
+.badge-resolution {
+    color: #212529;
+    background-color: #B0BEC5;
+}
+
+.badge-text {
+    color: #FFFFFF;
+    background-color: #FAAB3C;
+}
+
+.add-tag-button {
+    cursor: pointer;
+    color: #212529;
+    background-color: #e0e0e0;
+}
+
+.card-img-overlay {
+    pointer-events: none;
+    padding: 0.75rem;
+
+    bottom: unset;
+    top: 0;
+    left: unset;
+    right: unset;
+}
+
+.file-title {
+    width: 100%;
+    line-height: 1rem;
+    height: 1.1rem;
+    font-size: 10pt;
+    white-space: nowrap;
+    text-overflow: ellipsis;
+    overflow: hidden;
+    color: #00BCD4;
+}
+
+.badge {
+    margin-right: 3px;
+}
+
+.badge-delete {
+    margin-right: -2px;
+    margin-left: 2px;
+    margin-top: -1px;
+    font-family: monospace;
+    font-size: 90%;
+    background: rgba(0, 0, 0, 0.2);
+    padding: 0.1em 0.4em;
+    color: white;
+    cursor: pointer;
+}
+
+.badge-user {
+    color: #212529;
+    background-color: #e0e0e0;
+}
+
+.fit {
+    display: block;
+    min-width: 64px;
+    max-width: 100%;
+    max-height: 240px;
+    margin: 0 auto 0;
+    padding: 3px 3px 0;
+    width: auto;
+    height: auto;
+}
+
+.fit-sm {
+    display: block;
+    max-width: 64px;
+    max-height: 64px;
+    margin: 0 auto;
+    width: auto;
+    height: auto;
+}
+
+.audio-fit {
+    height: 39px;
+    vertical-align: bottom;
+    display: inline;
+    width: 100%;
+}
+
+@media screen and (min-width: 1800px) {
+    .container {
+        max-width: 1550px;
+    }
+}
+
+mark {
+    background: rgba(251, 191, 41, 0.25);
+    border-radius: 0;
+    padding: 1px 0;
+    color: inherit;
+}
+
+.content-div mark {
+    background: rgba(251, 191, 41, 0.40);
+    color: white;
+}
+
+
+.content-div {
+    font-family: SFMono-Regular, Menlo, Monaco, Consolas, "Liberation Mono", "Courier New", monospace;
+    font-size: 13px;
+    padding: 1em;
+    background-color: #37474F;
+    border: 1px solid #616161;
+    border-radius: 4px;
+    margin: 3px;
+    white-space: normal;
+    color: rgb(224, 224, 224);
+    overflow: hidden;
+}
+
+.irs-single, .irs-from, .irs-to {
+    font-size: 13px;
+    background-color: #00BCD4;
+}
+
+.irs-slider {
+    cursor: col-resize;
+}
+
+.irs {
+    margin-top: 1em;
+    margin-bottom: 1em;
+}
+
+.custom-select {
+    overflow: auto;
+    background-color: #37474F;
+    border: 1px solid #616161;
+    color: #bdbdbd;
+}
+
+.custom-select:focus {
+    border-color: #757575;
+    outline: 0;
+    box-shadow: 0 0 0 .2rem rgba(0, 123, 255, .25);
+}
+
+option {
+    outline: none;
+}
+
+.form-control {
+    background-color: #37474F;
+    border: 1px solid #616161;
+    color: #fff;
+}
+
+.form-control:focus {
+    background-color: #546E7A;
+    color: #fff;
+}
+
+.input-group-text {
+    background: #263238;
+    border: 1px solid #616161;
+    color: #dbdbdb;
+}
+
+::placeholder {
+    color: #BDBDBD !important;
+    opacity: 1;
+}
+
+
+.inspire-tree .selected > .wholerow, .inspire-tree .selected > .title-wrap:hover + .wholerow {
+    background: none;
+}
+
+.inspire-tree .icon-expand::before, .inspire-tree .icon-collapse::before {
+    background-color: black;
+}
+
+.inspire-tree .title {
+    color: #eee;
+}
+
+.inspire-tree {
+    font-weight: 400;
+    font-size: 14px;
+    font-family: Helvetica, Nueue, Verdana, sans-serif;
+    max-height: 350px;
+    overflow: auto;
+}
+
+.page-indicator {
+    line-height: 1rem;
+    padding: 0.5rem;
+    background: #212121;
+    color: #eee;
+    margin-top: 1em;
+}
+
+.btn-xs {
+    padding: .1rem .3rem;
+    font-size: .875rem;
+    border-radius: .2rem;
+}
+
+.btn {
+    color: #eee;
+}
+
+.nav-tabs .nav-link {
+    color: #e0e0e0;
+}
+
+.nav-tabs .nav-item.show .nav-link, .nav-tabs .nav-link.active {
+    background-color: #212121;
+    border-color: #616161 #616161 #212121;
+    color: #e0e0e0;
+}
+
+.nav-tabs .nav-link:focus, .nav-tabs .nav-link:focus {
+    border-color: #616161 #616161 #212121;
+    color: #e0e0e0;
+}
+
+.nav-tabs .nav-link:focus, .nav-tabs .nav-link:hover {
+    border-color: #e0e0e0 #e0e0e0 #212121;
+    color: #e0e0e0;
+}
+
+.nav-tabs {
+    border-bottom: #616161;
+}
+
+.nav {
+    margin-top: 0.5rem;
+}
+
+@media (max-width: 800px) {
+    #treeTabs {
+        flex-basis: inherit;
+        flex-grow: inherit;
+    }
+}
+
+.list-group {
+    margin-top: 1em;
+}
+
+.wrapper-sm {
+    min-width: 64px;
+}
+
+.media-expanded {
+    display: inherit;
+}
+
+.media-expanded .fit {
+    max-height: 250px;
+}
+
+@media (max-width: 650px) {
+    .media-expanded .fit {
+        max-height: none;
+    }
+
+    .tagline {
+        display: none;
+    }
+}
+
+.version {
+    color: #00BCD4;
+    margin-left: -18px;
+    margin-top: -14px;
+    font-size: 11px;
+}
+
+@media (min-width: 800px) {
+    .small-btn {
+        display: none;
+    }
+
+    .large-btn {
+        display: inherit;
+    }
+}
+
+@media (max-width: 801px) {
+    .small-btn {
+        display: inherit;
+    }
+
+    .large-btn {
+        display: none;
+    }
+}
+
+#searchBar {
+    border-right: none;
+}
+
+#pathTree .title {
+    cursor: pointer;
+}
+
+svg {
+    fill: white;
+}
+
+.play {
+    position: absolute;
+    width: 50px;
+    height: 50px;
+    left: 50%;
+    top: 50%;
+    transform: translate(-50%, -50%);
+    pointer-events: none;
+}
+
+.play svg {
+    fill: rgba(255, 255, 255, 0.7);
+}
+
+.img-wrapper:hover svg {
+    fill: rgba(255, 255, 255, 1);
+}
+
+.pointer {
+    cursor: pointer;
+}
+
+.stats-card {
+    text-align: center;
+    margin-top: 1em;
+    padding: 1em;
+
+    box-shadow: 0 .125rem .25rem rgba(0, 0, 0, .08) !important;
+    border-radius: 0;
+    border: none;
+
+    background: #212121;
+}
+
+.graph {
+    display: inline-block;
+    width: 40%;
+}
+
+.full-screen {
+    position: absolute;
+    left: 0;
+    width: 100%;
+}
+
+.stats-btn {
+    float: right;
+    margin-bottom: 10px;
+}
+
+#graphs-card svg text {
+    fill: #eee;
+}
+
+.wholerow {
+    outline: none !important;
+}
+
+.stat > .card-body {
+    padding: 0.7em 1.25em;
+}

src/static/css/jquery.toast.min.css

@@ -0,0 +1 @@
+.jq-toast-wrap,.jq-toast-wrap *{margin:0;padding:0}.jq-toast-wrap{display:block;position:fixed;width:250px;pointer-events:none!important;letter-spacing:normal;z-index:9000!important}.jq-toast-wrap.bottom-left{bottom:20px;left:20px}.jq-toast-wrap.bottom-right{bottom:20px;right:40px}.jq-toast-wrap.top-left{top:20px;left:20px}.jq-toast-wrap.top-right{top:20px;right:40px}.jq-toast-single{display:block;width:100%;padding:10px;margin:0 0 5px;border-radius:4px;font-size:12px;font-family:arial,sans-serif;line-height:17px;position:relative;pointer-events:all!important;background-color:#444;color:#fff}.jq-toast-single h2{font-family:arial,sans-serif;font-size:14px;margin:0 0 7px;background:0 0;color:inherit;line-height:inherit;letter-spacing:normal}.jq-toast-single a{color:#eee;text-decoration:none;font-weight:700;border-bottom:1px solid #fff;padding-bottom:3px;font-size:12px}.jq-toast-single ul{margin:0 0 0 15px;background:0 0;padding:0}.jq-toast-single ul li{list-style-type:disc!important;line-height:17px;background:0 0;margin:0;padding:0;letter-spacing:normal}.close-jq-toast-single{position:absolute;top:3px;right:7px;font-size:14px;cursor:pointer}.jq-toast-loader{display:block;position:absolute;top:-2px;height:5px;width:0;left:0;border-radius:5px;background:red}.jq-toast-loaded{width:100%}.jq-has-icon{padding:10px 10px 10px 50px;background-repeat:no-repeat;background-position:10px}.jq-icon-info{background-image:url(data:image/png;base64,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);background-color:#31708f;color:#d9edf7;border-color:#bce8f1}.jq-icon-warning{background-image:url(data:image/png;base64,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);background-color:#8a6d3b;color:#fcf8e3;border-color:#faebcc}.jq-icon-error{background-image:url(data:image/png;base64,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);background-color:#a94442;color:#f2dede;border-color:#ebccd1}.jq-icon-success{background-image:url(data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABgAAAAYCAYAAADgdz34AAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAAJcEhZcwAADsMAAA7DAcdvqGQAAADsSURBVEhLY2AYBfQMgf///3P8+/evAIgvA/FsIF+BavYDDWMBGroaSMMBiE8VC7AZDrIFaMFnii3AZTjUgsUUWUDA8OdAH6iQbQEhw4HyGsPEcKBXBIC4ARhex4G4BsjmweU1soIFaGg/WtoFZRIZdEvIMhxkCCjXIVsATV6gFGACs4Rsw0EGgIIH3QJYJgHSARQZDrWAB+jawzgs+Q2UO49D7jnRSRGoEFRILcdmEMWGI0cm0JJ2QpYA1RDvcmzJEWhABhD/pqrL0S0CWuABKgnRki9lLseS7g2AlqwHWQSKH4oKLrILpRGhEQCw2LiRUIa4lwAAAABJRU5ErkJggg==);color:#dff0d8;background-color:#3c763d;border-color:#d6e9c6}

src/static/css/light.css

@@ -0,0 +1,412 @@
+*:focus {
+    outline: 0;
+}
+
+.info-icon {
+    width: 1rem;
+    margin-right: 0.2rem;
+    cursor: pointer;
+    line-height: 1rem;
+    height: 1rem;
+    background-image: url(data:image/svg+xml;base64,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);
+    filter: brightness(45%);
+}
+
+.info-icon:hover {
+    color: inherit;
+}
+
+.modal-title {
+    max-width: calc(100% - 2rem);
+    overflow: hidden;
+    text-overflow: ellipsis;
+}
+
+.path-row {
+    display: -ms-flexbox;
+    display: flex;
+    -ms-flex-align: start;
+    align-items: flex-start;
+}
+
+.tag-container {
+    margin-left: 0.3rem;
+}
+
+.path-line {
+    color: #444;
+    text-overflow: ellipsis;
+    overflow: hidden;
+    white-space: nowrap;
+}
+
+body {
+    overflow-y: scroll;
+}
+
+.progress {
+    margin-top: 1em;
+}
+
+.card {
+    margin-top: 1em;
+    box-shadow: 0 .125rem .25rem rgba(0, 0, 0, .08) !important;
+    border-radius: 0;
+    border: none;
+}
+
+.sub-document {
+    background: #AB47BC1F !important;
+}
+
+.navbar-brand {
+    font-size: 1.75rem;
+    padding: 0;
+}
+
+.navbar {
+    background: #F7F7F7;
+    border-bottom: solid 1px #dfdfdf;
+}
+
+.document {
+   padding: 0.3rem;
+}
+
+.card-text:last-child {
+    margin-top: -1px;
+}
+
+.document p {
+    margin-bottom: 0;
+}
+
+.document:hover p {
+    text-decoration: underline;
+}
+
+.badge-video {
+    color: #FFFFFF;
+    background-color: #F27761;
+}
+
+.badge-image {
+    color: #FFFFFF;
+    background-color: #AA99C9;
+}
+
+.badge-audio {
+    color: #FFFFFF;
+    background-color: #00ADEF;
+}
+
+.badge-resolution {
+    color: #212529;
+    background-color: #FFC107;
+}
+
+.badge-user {
+    color: #212529;
+    background-color: #e0e0e0;
+}
+
+.badge {
+    margin-right: 3px;
+}
+
+.badge-delete {
+    margin-right: -2px;
+    margin-left: 2px;
+    margin-top: -1px;
+    font-family: monospace;
+    font-size: 90%;
+    background: rgba(0,0,0,0.2);
+    padding: 0.1em 0.4em;
+    color: white;
+    cursor: pointer;
+}
+
+.badge-text {
+    color: #FFFFFF;
+    background-color: #FAAB3C;
+}
+
+.add-tag-button {
+    cursor: pointer;
+    color: #212529;
+    background-color: #e0e0e0;
+}
+
+.card-img-overlay {
+    pointer-events: none;
+    padding: 0.75rem;
+
+    bottom: unset;
+    top: 0;
+    left: unset;
+    right: unset;
+}
+
+.file-title {
+    width: 100%;
+    line-height: 1rem;
+    height: 1.1rem;
+    font-size: 10pt;
+    white-space: nowrap;
+    text-overflow: ellipsis;
+    overflow: hidden;
+}
+
+
+.fit {
+    display: block;
+    min-width: 64px;
+    max-width: 100%;
+    max-height: 240px;
+    margin: 0 auto 0;
+    padding: 3px 3px 0 3px;
+    width: auto;
+    height: auto;
+}
+
+.fit-sm {
+    display: block;
+    max-width: 64px;
+    max-height: 64px;
+    margin: 0 auto 0;
+    width: auto;
+    height: auto;
+}
+
+.audio-fit {
+    height: 39px;
+    vertical-align: bottom;
+    display: inline;
+    width: 100%;
+}
+
+.bricklayer {
+    /*max-width: 100%;*/
+}
+
+@media screen and (max-width: 1200px) {
+    .bricklayer-column {
+        max-width: 100%;
+    }
+}
+
+@media screen and (min-width: 1500px) {
+    .container {
+        max-width: 1440px;
+    }
+
+    .bricklayer-column-sizer {
+        width: 20% !important;
+    }
+
+    .bricklayer-column {
+        max-width: 20%;
+    }
+}
+
+@media screen and (min-width: 1800px) {
+    .container {
+        max-width: 1550px;
+    }
+}
+
+mark {
+    background: #fff217;
+    border-radius: 0;
+    padding: 1px 0;
+    color: inherit;
+}
+
+.content-div {
+    font-family: SFMono-Regular, Menlo, Monaco, Consolas, "Liberation Mono", "Courier New", monospace;
+    font-size: 13px;
+    padding: 1em;
+    background-color: #f5f5f5;
+    border: 1px solid #ccc;
+    border-radius: 4px;
+    margin: 3px;
+    white-space: normal;
+    color: #000;
+    overflow: hidden;
+}
+
+.irs-single, .irs-from, .irs-to {
+    font-size: 13px;
+}
+
+.irs-slider {
+    cursor: col-resize;
+}
+
+.custom-select {
+    overflow: auto;
+}
+
+.irs {
+    margin-top: 1em;
+    margin-bottom: 1em;
+}
+
+.inspire-tree .selected > .wholerow, .inspire-tree .selected > .title-wrap:hover + .wholerow {
+    background: none;
+}
+
+.inspire-tree {
+    font-weight: 400;
+    font-size: 14px;
+    font-family: Helvetica, Nueue, Verdana, sans-serif;
+    max-height: 350px;
+    overflow: auto;
+}
+
+.page-indicator {
+    line-height: 1rem;
+    padding: 0.5rem;
+    background: #f8f9fa;
+    margin-top: 1em;
+}
+
+.btn-xs {
+    padding: .1rem .3rem;
+    font-size: .875rem;
+    border-radius: .2rem;
+}
+
+.nav {
+    margin-top: 0.5rem;
+}
+
+@media (max-width: 800px) {
+    #treeTabs {
+        flex-basis: inherit;
+        flex-grow: inherit;
+    }
+}
+
+.list-group {
+    margin-top: 1em;
+}
+
+.list-group-item {
+    padding: .25rem 0.5rem;
+}
+
+.wrapper-sm {
+    min-width: 64px;
+}
+
+.media-expanded {
+    display: inherit;
+}
+
+.media-expanded .fit {
+    max-height: 250px;
+}
+
+@media (max-width: 650px) {
+    .media-expanded .fit {
+        max-height: none;
+    }
+
+    .tagline {
+        display: none;
+    }
+}
+
+.version {
+    color: #007bff;
+    margin-left: -18px;
+    margin-top: -14px;
+    font-size: 11px;
+}
+
+@media (min-width: 800px) {
+    .small-btn {
+        display: none;
+    }
+
+    .large-btn {
+        display: inherit;
+    }
+}
+
+@media (max-width: 801px) {
+    .small-btn {
+        display: inherit;
+    }
+
+    .large-btn {
+        display: none;
+    }
+}
+
+#searchBar {
+    border-right: none;
+}
+
+#pathTree .title {
+    cursor: pointer;
+}
+
+.play {
+    position: absolute;
+    width: 50px;
+    height: 50px;
+    left: 50%;
+    top: 50%;
+    transform: translate(-50%, -50%);
+    pointer-events: none;
+}
+
+.play svg {
+    fill: rgba(0, 0, 0, 0.7);
+}
+
+.img-wrapper:hover svg {
+    fill: rgba(0, 0, 0, 1);
+}
+
+.pointer {
+    cursor: pointer;
+}
+
+.stats-card {
+    text-align: center;
+    margin-top: 1em;
+    padding: 1em;
+
+    box-shadow: 0 .125rem .25rem rgba(0, 0, 0, .08) !important;
+    border-radius: 0;
+    border: none;
+
+    background: #fff;
+}
+
+.graph {
+    display: inline-block;
+    width: 40%;
+}
+
+.full-screen {
+    position: absolute;
+    left: 0;
+    width: 100%;
+}
+
+.stats-btn {
+    float: right;
+    margin-bottom: 10px;
+}
+
+.wholerow {
+    outline: none !important;
+}
+
+.stat > .card-body {
+    padding: 0.7em 1.25em;
+}

src/static/css/lity.min.css

@@ -0,0 +1,3 @@
+/*! Lity - v2.4.0 - 2019-08-10
+* http://sorgalla.com/lity/
+* Copyright (c) 2015-2019 Jan Sorgalla; Licensed MIT */.lity{z-index:9990;position:fixed;top:0;right:0;bottom:0;left:0;white-space:nowrap;background:#0b0b0b;background:rgba(0,0,0,0.9);outline:none !important;opacity:0;-webkit-transition:opacity .3s ease;-o-transition:opacity .3s ease;transition:opacity .3s ease}.lity.lity-opened{opacity:1}.lity.lity-closed{opacity:0}.lity *{-webkit-box-sizing:border-box;-moz-box-sizing:border-box;box-sizing:border-box}.lity-wrap{z-index:9990;position:fixed;top:0;right:0;bottom:0;left:0;text-align:center;outline:none !important}.lity-wrap:before{content:'';display:inline-block;height:100%;vertical-align:middle;margin-right:-0.25em}.lity-loader{z-index:9991;color:#fff;position:absolute;top:50%;margin-top:-0.8em;width:100%;text-align:center;font-size:14px;font-family:Arial,Helvetica,sans-serif;opacity:0;-webkit-transition:opacity .3s ease;-o-transition:opacity .3s ease;transition:opacity .3s ease}.lity-loading .lity-loader{opacity:1}.lity-container{z-index:9992;position:relative;text-align:left;vertical-align:middle;display:inline-block;white-space:normal;max-width:100%;max-height:100%;outline:none !important}.lity-content{z-index:9993;width:100%;-webkit-transform:scale(1);-ms-transform:scale(1);-o-transform:scale(1);transform:scale(1);-webkit-transition:-webkit-transform .3s ease;transition:-webkit-transform .3s ease;-o-transition:-o-transform .3s ease;transition:transform .3s ease;transition:transform .3s ease, -webkit-transform .3s ease, -o-transform .3s ease}.lity-loading .lity-content,.lity-closed .lity-content{-webkit-transform:scale(.8);-ms-transform:scale(.8);-o-transform:scale(.8);transform:scale(.8)}.lity-content:after{content:'';position:absolute;left:0;top:0;bottom:0;display:block;right:0;width:auto;height:auto;z-index:-1;-webkit-box-shadow:0 0 8px rgba(0,0,0,0.6);box-shadow:0 0 8px rgba(0,0,0,0.6)}.lity-close{z-index:9994;width:35px;height:35px;position:fixed;right:0;top:0;-webkit-appearance:none;cursor:pointer;text-decoration:none;text-align:center;padding:0;color:#fff;font-style:normal;font-size:35px;font-family:Arial,Baskerville,monospace;line-height:35px;text-shadow:0 1px 2px rgba(0,0,0,0.6);border:0;background:none;outline:none;-webkit-box-shadow:none;box-shadow:none}.lity-close::-moz-focus-inner{border:0;padding:0}.lity-close:hover,.lity-close:focus,.lity-close:active,.lity-close:visited{text-decoration:none;text-align:center;padding:0;color:#fff;font-style:normal;font-size:35px;font-family:Arial,Baskerville,monospace;line-height:35px;text-shadow:0 1px 2px rgba(0,0,0,0.6);border:0;background:none;outline:none;-webkit-box-shadow:none;box-shadow:none}.lity-close:active{top:1px}.lity-image img{max-width:100%;display:block;line-height:0;border:0}.lity-iframe .lity-container,.lity-youtube .lity-container,.lity-vimeo .lity-container,.lity-facebookvideo .lity-container,.lity-googlemaps .lity-container{width:100%;max-width:964px}.lity-iframe-container{width:100%;height:0;padding-top:56.25%;overflow:hidden;pointer-events:all;-webkit-transform:translateZ(0);transform:translateZ(0);-webkit-overflow-scrolling:touch}.lity-iframe-container iframe{position:absolute;display:block;top:0;left:0;width:100%;height:100%;-webkit-box-shadow:0 0 8px rgba(0,0,0,0.6);box-shadow:0 0 8px rgba(0,0,0,0.6);background:#000}.lity-hide{display:none}

src/static/js/bricklayer.min.js

@@ -0,0 +1 @@
+!function t(e,n,r){function o(s,u){if(!n[s]){if(!e[s]){var l="function"==typeof require&&require;if(!u&&l)return l(s,!0);if(i)return i(s,!0);var a=new Error("Cannot find module '"+s+"'");throw a.code="MODULE_NOT_FOUND",a}var p=n[s]={exports:{}};e[s][0].call(p.exports,function(t){var n=e[s][1][t];return o(n?n:t)},p,p.exports,t,e,n,r)}return n[s].exports}for(var i="function"==typeof require&&require,s=0;s<r.length;s++)o(r[s]);return o}({1:[function(t,e,n){var r,o=this&&this.__extends||function(t,e){function n(){this.constructor=t}for(var r in e)e.hasOwnProperty(r)&&(t[r]=e[r]);t.prototype=null===e?Object.create(e):(n.prototype=e.prototype,new n)};!function(t){function e(t){return[].slice.call(t)}function n(t,e,n){if(window.CustomEvent)var r=new CustomEvent(e,{detail:n});else{var r=document.createEvent("CustomEvent");r.initCustomEvent(e,!0,!0,n)}return t.dispatchEvent(r)}var r={rulerClassName:"bricklayer-column-sizer",columnClassName:"bricklayer-column"},i=function(){function t(t){this.element=document.createElement("div"),this.element.className=t}return t.prototype.destroy=function(){this.element.parentNode.removeChild(this.element)},t}(),s=function(t){function e(){t.apply(this,arguments)}return o(e,t),e.prototype.getWidth=function(){this.element.setAttribute("style","\n        display: block;\n        visibility: hidden !important;\n        top: -1000px !important;\n      ");var t=this.element.offsetWidth;return this.element.removeAttribute("style"),t},e}(i),u=function(t){function e(){t.apply(this,arguments)}return o(e,t),e}(i),l=function(){function t(t,e){void 0===e&&(e=r),this.element=t,this.options=e,this.build(),this.buildResponsive()}return t.prototype.append=function(t){var n=this;if(Array.isArray(t))return void t.forEach(function(t){return n.append(t)});var r=this.findMinHeightColumn();this.elements=e(this.elements).concat([t]),this.applyPosition("append",r,t)},t.prototype.prepend=function(t){var n=this;if(Array.isArray(t))return void t.forEach(function(t){return n.prepend(t)});var r=this.findMinHeightColumn();this.elements=[t].concat(e(this.elements)),this.applyPosition("prepend",r,t)},t.prototype.on=function(t,e){return this.element.addEventListener("bricklayer."+t,e),this},t.prototype.redraw=function(){var t=this.columnCount;this.checkColumnCount(!1),this.reorderElements(t),n(this.element,"bricklayer.redraw",{columnCount:t})},t.prototype.destroy=function(){var t=this;this.ruler.destroy(),e(this.elements).forEach(function(e){return t.element.appendChild(e)}),e(this.getColumns()).forEach(function(t){return t.parentNode.removeChild(t)}),n(this.element,"bricklayer.destroy",{})},t.prototype.build=function(){this.ruler=new s(this.options.rulerClassName),this.elements=this.getElementsInOrder(),this.element.insertBefore(this.ruler.element,this.element.firstChild)},t.prototype.buildResponsive=function(){var t=this;window.addEventListener("resize",function(e){return t.checkColumnCount()}),this.checkColumnCount(),this.on("breakpoint",function(e){return t.reorderElements(e.detail.columnCount)}),this.columnCount>=1&&this.reorderElements(this.columnCount)},t.prototype.getColumns=function(){return this.element.querySelectorAll(":scope > ."+this.options.columnClassName)},t.prototype.findMinHeightColumn=function(){var t=e(this.getColumns()),n=t.map(function(t){return t.offsetHeight}),r=Math.min.apply(null,n);return t[n.indexOf(r)]},t.prototype.getElementsInOrder=function(){return this.element.querySelectorAll(":scope > *:not(."+this.options.columnClassName+"):not(."+this.options.rulerClassName+")")},t.prototype.checkColumnCount=function(t){void 0===t&&(t=!0);var e=this.getColumnCount();this.columnCount!==e&&(t&&n(this.element,"bricklayer.breakpoint",{columnCount:e}),this.columnCount=e)},t.prototype.reorderElements=function(t){var n=this;void 0===t&&(t=1),(t==1/0||1>t)&&(t=1);for(var r=e(this.elements).map(function(t){var e=t.parentNode?t.parentNode.removeChild(t):t;return e}),o=this.getColumns(),i=0;i<o.length;i++)o[i].parentNode.removeChild(o[i]);for(var i=0;t>i;i++){var s=new u(this.options.columnClassName).element;this.element.appendChild(s)}r.forEach(function(t){var e=n.findMinHeightColumn();e.appendChild(t)})},t.prototype.getColumnCount=function(){var t=this.element.offsetWidth,e=this.ruler.getWidth();return Math.round(t/e)},t.prototype.applyPosition=function(t,e,r){var o=this,i=function(i){var s=i+t.charAt(0).toUpperCase()+t.substr(1);n(o.element,"bricklayer."+s,{item:r,column:e})};switch(i("before"),t){case"append":e.appendChild(r);break;case"prepend":e.insertBefore(r,e.firstChild)}i("after")},t}();t.Container=l}(r||(r={})),function(t,n){"function"==typeof define&&define.amd?define(function(){return n()}):"undefined"!=typeof window&&t===window?t.Bricklayer=n():"object"==typeof e&&e.exports&&(e.exports=n())}("undefined"!=typeof window?window:this,function(){return r.Container})},{}]},{},[1]);

src/static/js/dom.js

@@ -0,0 +1,664 @@
+/**
+ * Enable gif loading on hover
+ */
+function gifOver(thumbnail, hit) {
+    let callee = arguments.callee;
+
+    thumbnail.addEventListener("mouseover", function () {
+
+        thumbnail.mouseStayedOver = true;
+
+        window.setTimeout(function () {
+            if (thumbnail.mouseStayedOver) {
+                thumbnail.removeEventListener('mouseover', callee, false);
+
+                //Load gif
+                thumbnail.setAttribute("src", "f/" + hit["_id"]);
+            }
+        }, 600);
+
+    });
+
+    thumbnail.addEventListener("mouseout", function () {
+        //Reset timer
+        thumbnail.mouseStayedOver = false;
+        thumbnail.setAttribute("src", `t/${hit["_source"]["index"]}/${hit["_id"]}`);
+    })
+}
+
+function getContentHighlight(hit) {
+    if (hit.hasOwnProperty("highlight")) {
+        if (hit["highlight"].hasOwnProperty("content")) {
+            return hit["highlight"]["content"][0];
+
+        } else if (hit["highlight"].hasOwnProperty("content.nGram")) {
+            return hit["highlight"]["content.nGram"][0];
+        }
+    }
+
+    return undefined;
+}
+
+function getPathHighlight(hit) {
+    if (hit.hasOwnProperty("highlight")) {
+        if (hit["highlight"].hasOwnProperty("path.text")) {
+            return hit["highlight"]["path.text"][0];
+        } else if (hit["highlight"].hasOwnProperty("path.nGram")) {
+            return hit["highlight"]["path.nGram"][0];
+        }
+    }
+
+    return undefined;
+}
+
+function applyNameToTitle(hit, title, extension) {
+    if (hit.hasOwnProperty("highlight")) {
+        if (hit["highlight"].hasOwnProperty("name")) {
+            title.insertAdjacentHTML('afterbegin', hit["highlight"]["name"] + extension);
+            return;
+        } else if (hit["highlight"].hasOwnProperty("name.nGram")) {
+            title.insertAdjacentHTML('afterbegin', hit["highlight"]["name.nGram"] + extension);
+            return;
+        }
+    }
+
+    title.appendChild(document.createTextNode(hit["_source"]["name"] + extension));
+}
+
+function shouldPlayVideo(hit) {
+    const videoc = hit["_source"]["videoc"];
+    const mime = hit["_source"]["mime"];
+
+    return mime &&
+        mime.startsWith("video/") &&
+        !("parent" in hit["_source"]) &&
+        hit["_source"]["extension"] !== "mkv" &&
+        hit["_source"]["extension"] !== "avi" &&
+        videoc !== "hevc" &&
+        videoc !== "mpeg2video" &&
+        videoc !== "wmv3";
+}
+
+function shouldDisplayRawImage(hit) {
+    const mime = hit["_source"]["mime"];
+
+    return mime &&
+        mime.startsWith("image/") &&
+        hit["_source"]["mime"] &&
+        !hit["_source"]["parent"] &&
+        hit["_source"]["videoc"] !== "tiff" &&
+        hit["_source"]["videoc"] !== "raw" &&
+        hit["_source"]["videoc"] !== "ppm";
+}
+
+function makePlaceholder(w, h, small) {
+    let calc;
+    if (small) {
+        calc = w > h
+            ? (64 / w / h) >= 100
+                ? (64 * w / h)
+                : 64
+            : 64;
+    } else {
+        calc = w > h
+            ? (175 / w / h) >= 272
+                ? (175 * w / h)
+                : 175
+            : 175;
+    }
+
+    const el = document.createElement("div");
+    el.setAttribute("style", `height: ${calc}px`);
+    return el;
+}
+
+function ext(hit) {
+    return hit["_source"].hasOwnProperty("extension") && hit["_source"]["extension"] !== "" ? "." + hit["_source"]["extension"] : "";
+}
+
+function makeTitle(hit) {
+    let title = document.createElement("div");
+    title.setAttribute("class", "file-title");
+    let extension = ext(hit);
+
+    applyNameToTitle(hit, title, extension);
+
+    title.setAttribute("title", hit["_source"]["path"] + "/" + hit["_source"]["name"] + extension);
+    return title;
+}
+
+function getTags(hit, mimeCategory) {
+
+    let tags = [];
+    switch (mimeCategory) {
+        case "video":
+        case "image":
+            if (hit["_source"].hasOwnProperty("videoc") && hit["_source"]["videoc"]) {
+                const formatTag = document.createElement("span");
+                formatTag.setAttribute("class", "badge badge-pill badge-video");
+                formatTag.appendChild(document.createTextNode(hit["_source"]["videoc"].replace(" ", "")));
+                tags.push(formatTag);
+            }
+            break;
+        case "audio": {
+            if (hit["_source"].hasOwnProperty("audioc") && hit["_source"]["audioc"]) {
+                let formatTag = document.createElement("span");
+                formatTag.setAttribute("class", "badge badge-pill badge-audio");
+                formatTag.appendChild(document.createTextNode(hit["_source"]["audioc"]));
+                tags.push(formatTag);
+            }
+        }
+            break;
+    }
+    // User tags
+    if (hit["_source"].hasOwnProperty("tag")) {
+        hit["_source"]["tag"].forEach(tag => {
+            tags.push(makeUserTag(tag, hit));
+        })
+    }
+
+    return tags
+}
+
+function makeUserTag(tag, hit) {
+    const userTag = document.createElement("span");
+    userTag.setAttribute("class", "badge badge-pill badge-user");
+    userTag.setAttribute("title", tag.split("#")[0])
+
+    const tokens = tag.split("#");
+
+    if (tokens.length > 1) {
+        const bg = "#" + tokens[1];
+        const fg = lum(tokens[1]) > 50 ? "#000" : "#fff";
+        userTag.setAttribute("style", `background-color: ${bg}; color: ${fg}`);
+    }
+
+    const deleteButton = document.createElement("span");
+    deleteButton.setAttribute("class", "badge badge-pill badge-delete")
+    deleteButton.setAttribute("title", "Delete tag")
+    deleteButton.appendChild(document.createTextNode("X"));
+    deleteButton.addEventListener("click", () => {
+        deleteTag(tag, hit).then(() => {
+            userTag.remove();
+        });
+    });
+    userTag.addEventListener("mouseenter", () => userTag.appendChild(deleteButton));
+    userTag.addEventListener("mouseleave", () => deleteButton.remove());
+
+    const name = tokens[0].split(".")[tokens[0].split(".").length - 1];
+    userTag.appendChild(document.createTextNode(name));
+
+    return userTag;
+}
+
+function infoButtonCb(hit) {
+    return () => {
+        getDocumentInfo(hit["_id"]).then(doc => {
+            $("#modal-title").text(doc["name"] + ext(hit));
+
+            const tbody = $("<tbody>");
+            $("#modal-body").empty()
+                .append($("<table class='table table-sm'>")
+                    .append($("<thead>")
+                        .append($("<tr>")
+                            .append($("<th>").text("Field"))
+                            .append($("<th>").text("Value"))
+                        )
+                    )
+                    .append(tbody)
+                );
+
+            const displayFields = new Set([
+                "mime", "size", "mtime", "path", "title", "width", "height", "duration", "audioc", "videoc",
+                "bitrate", "artist", "album", "album_artist", "genre", "title", "font_name", "tag", "author",
+                "modified_by"
+            ]);
+            Object.keys(doc)
+                .filter(key => key.startsWith("_keyword.") || key.startsWith("_text.") || displayFields.has(key) || key.startsWith("exif_"))
+                .forEach(key => {
+                    tbody.append($("<tr>")
+                        .append($("<td>").text(key))
+                        .append($("<td>").text(doc[key]))
+                    );
+                });
+            if (doc.hasOwnProperty("content") && doc["content"]) {
+                $("#modal-body").append($("<div class='content-div'>").text(doc["content"]))
+            }
+
+            $("#modal").modal();
+        });
+    }
+}
+
+function createDocCard(hit) {
+    let docCard = document.createElement("div");
+    docCard.setAttribute("class", "card");
+
+    let docCardBody = document.createElement("div");
+    docCardBody.setAttribute("class", "card-body document");
+
+    //Title
+    let title = makeTitle(hit);
+    let isSubDocument = false;
+
+    let link = document.createElement("a");
+    link.setAttribute("href", "f/" + hit["_id"]);
+    link.setAttribute("target", "_blank");
+    link.style.maxWidth = "calc(100% - 1.2rem)";
+    link.appendChild(title);
+
+    if (hit["_source"].hasOwnProperty("parent")) {
+        docCard.classList.add("sub-document");
+        isSubDocument = true;
+    }
+
+    let tagContainer = document.createElement("div");
+    tagContainer.setAttribute("class", "card-text");
+
+    if (hit["_source"].hasOwnProperty("mime") && hit["_source"]["mime"] !== null) {
+
+        let thumbnailOverlay = null;
+        let imgWrapper = document.createElement("div");
+        imgWrapper.setAttribute("style", "position: relative");
+        imgWrapper.setAttribute("class", "img-wrapper");
+
+        let mimeCategory = hit["_source"]["mime"].split("/")[0];
+
+        //Thumbnail
+        let thumbnail = makeThumbnail(mimeCategory, hit, imgWrapper, false);
+
+        //Thumbnail overlay
+        switch (mimeCategory) {
+            case "image":
+                thumbnailOverlay = document.createElement("div");
+                thumbnailOverlay.setAttribute("class", "card-img-overlay");
+
+                //Resolution
+                if (hit["_source"].hasOwnProperty("width") && hit["_source"]["width"] > 32 && hit["_source"]["height"] > 32) {
+                    let resolutionBadge = document.createElement("span");
+                    resolutionBadge.setAttribute("class", "badge badge-resolution");
+                    if (hit["_source"].hasOwnProperty("width")) {
+                        resolutionBadge.appendChild(document.createTextNode(hit["_source"]["width"] + "x" + hit["_source"]["height"]));
+                    }
+                    thumbnailOverlay.appendChild(resolutionBadge);
+                }
+
+                // Hover
+                if (thumbnail && hit["_source"]["videoc"] === "gif" && !isSubDocument) {
+                    gifOver(thumbnail, hit);
+                }
+                break;
+
+            case "video":
+                //Duration
+                if (hit["_source"].hasOwnProperty("duration")) {
+                    thumbnailOverlay = document.createElement("div");
+                    thumbnailOverlay.setAttribute("class", "card-img-overlay");
+                    const durationBadge = document.createElement("span");
+                    durationBadge.setAttribute("class", "badge badge-resolution");
+                    durationBadge.appendChild(document.createTextNode(humanTime(hit["_source"]["duration"])));
+                    thumbnailOverlay.appendChild(durationBadge);
+                }
+        }
+
+        // Tags
+        let tags = getTags(hit, mimeCategory);
+        for (let i = 0; i < tags.length; i++) {
+            tagContainer.appendChild(tags[i]);
+        }
+
+        //Content
+        let contentHl = getContentHighlight(hit);
+        if (contentHl !== undefined) {
+            const contentDiv = document.createElement("div");
+            contentDiv.setAttribute("class", "content-div");
+            contentDiv.insertAdjacentHTML('afterbegin', contentHl);
+            docCard.appendChild(contentDiv);
+        }
+
+        if (thumbnail !== null) {
+            docCard.appendChild(imgWrapper);
+        }
+
+        //Audio
+        if (mimeCategory === "audio" && hit["_source"].hasOwnProperty("audioc") && !isSubDocument) {
+
+            let audio = document.createElement("audio");
+            audio.setAttribute("preload", "none");
+            audio.setAttribute("class", "audio-fit fit");
+            audio.setAttribute("controls", "");
+            audio.setAttribute("type", hit["_source"]["mime"]);
+            audio.setAttribute("src", "f/" + hit["_id"]);
+
+            docCard.appendChild(audio)
+        }
+
+        if (thumbnailOverlay !== null) {
+            imgWrapper.appendChild(thumbnailOverlay);
+        }
+    }
+
+    //Size tag
+    let sizeTag = document.createElement("small");
+    sizeTag.appendChild(document.createTextNode(humanFileSize(hit["_source"]["size"])));
+    sizeTag.setAttribute("class", "text-muted");
+    tagContainer.appendChild(sizeTag);
+
+    const titleWrapper = document.createElement("div");
+    titleWrapper.style.display = "flex";
+
+    const infoButton = makeInfoButton(hit);
+
+    titleWrapper.appendChild(infoButton);
+    titleWrapper.appendChild(link);
+
+    docCardBody.appendChild(titleWrapper);
+    docCard.appendChild(docCardBody);
+
+    docCardBody.appendChild(tagContainer);
+
+    attachTagContainerEventListener(tagContainer, hit);
+    return docCard;
+}
+
+function attachTagContainerEventListener(tagContainer, hit) {
+    const sizeTag = Array.from(tagContainer.children).find(child => child.tagName === "SMALL");
+
+    const addTagButton = document.createElement("span");
+    addTagButton.setAttribute("class", "badge badge-pill add-tag-button");
+    addTagButton.appendChild(document.createTextNode("+Add"));
+
+    tagContainer.addEventListener("mouseenter", () => tagContainer.insertBefore(addTagButton, sizeTag));
+    tagContainer.addEventListener("mouseleave", () => addTagButton.remove());
+
+    addTagButton.addEventListener("click", () => {
+        tagBar.value = "";
+        currentDocToTag = hit;
+        currentTagCallback = tag => {
+            tagContainer.insertBefore(makeUserTag(tag, hit), sizeTag);
+        }
+        $("#tagModal").modal("show");
+        tagBar.focus();
+    });
+}
+
+function makeThumbnail(mimeCategory, hit, imgWrapper, small) {
+
+    if (!hit["_source"].hasOwnProperty("thumbnail")) {
+        return null;
+    }
+
+    let thumbnail = document.createElement("img");
+    if (small) {
+        thumbnail.setAttribute("class", "fit-sm");
+    } else {
+        thumbnail.setAttribute("class", "card-img-top fit");
+    }
+    thumbnail.setAttribute("src", `t/${hit["_source"]["index"]}/${hit["_id"]}`);
+
+    if (shouldDisplayRawImage(hit)) {
+        thumbnail.addEventListener("click", () => {
+            const l = lity(`f/${hit["_id"]}#.jpg`);
+            window.addEventListener("scroll", () => l.close());
+        });
+
+        thumbnail.classList.add("pointer");
+    } else if (shouldPlayVideo(hit)) {
+        thumbnail.addEventListener("click", () => lity(`f/${hit["_id"]}#.mp4`));
+
+        thumbnail.classList.add("pointer");
+
+        if (!small) {
+            const playOverlay = document.createElement("div");
+            playOverlay.innerHTML = '<svg viewBox="0 0 494.942 494.942" xmlns="http://www.w3.org/2000/svg"><path d="m35.353 0 424.236 247.471-424.236 247.471z"/></svg>';
+            playOverlay.classList.add("play");
+            imgWrapper.prepend(playOverlay);
+        }
+    }
+
+    const placeholder = makePlaceholder(hit["_source"]["width"], hit["_source"]["height"], small);
+    imgWrapper.appendChild(placeholder);
+
+    thumbnail.addEventListener("error", () => {
+        imgWrapper.remove();
+    });
+    thumbnail.addEventListener("load", () => {
+        placeholder.remove();
+        imgWrapper.appendChild(thumbnail);
+    });
+
+    return thumbnail;
+}
+
+function makeInfoButton(hit) {
+    const infoButton = document.createElement("span");
+    infoButton.setAttribute("class", "info-icon");
+    infoButton.addEventListener("click", infoButtonCb(hit));
+    return infoButton;
+}
+
+function createDocLine(hit) {
+
+    const mime = hit["_source"]["mime"];
+    let mimeCategory = mime ? mime.split("/")[0] : null;
+    let tags = getTags(hit, mimeCategory);
+
+    let imgWrapper = document.createElement("div");
+    imgWrapper.setAttribute("class", "align-self-start mr-1 wrapper-sm");
+
+    let media = document.createElement("div");
+    media.setAttribute("class", "media");
+
+    const line = document.createElement("div");
+    line.setAttribute("class", "list-group-item flex-column align-items-start");
+
+    if (hit["_source"].hasOwnProperty("parent")) {
+        line.classList.add("sub-document");
+    }
+
+    const infoButton = makeInfoButton(hit);
+
+    const title = makeTitle(hit);
+
+    let link = document.createElement("a");
+    link.setAttribute("href", "f/" + hit["_id"]);
+    link.setAttribute("target", "_blank");
+    link.style.maxWidth = "calc(100% - 1.2rem)";
+    link.appendChild(title);
+
+    const titleDiv = document.createElement("div");
+
+    const titleWrapper = document.createElement("div");
+    titleWrapper.style.display = "flex";
+    titleWrapper.appendChild(infoButton);
+    titleWrapper.appendChild(link);
+
+    titleDiv.appendChild(titleWrapper);
+
+    line.appendChild(media);
+
+    let thumbnail = makeThumbnail(mimeCategory, hit, imgWrapper, true);
+    if (thumbnail) {
+        media.appendChild(imgWrapper);
+        titleDiv.style.maxWidth = "calc(100% - 64px)";
+    } else {
+        titleDiv.style.maxWidth = "100%";
+    }
+    media.appendChild(titleDiv);
+
+    // Content
+    let contentHl = getContentHighlight(hit);
+    if (contentHl !== undefined) {
+        const contentDiv = document.createElement("div");
+        contentDiv.setAttribute("class", "content-div");
+        contentDiv.insertAdjacentHTML("afterbegin", contentHl);
+        titleDiv.appendChild(contentDiv);
+    }
+
+    let pathLine = document.createElement("div");
+    pathLine.setAttribute("class", "path-row");
+
+    let path = document.createElement("div");
+    path.setAttribute("class", "path-line");
+    path.setAttribute("title", hit["_source"]["path"] + "/");
+
+    const pathHighlight = getPathHighlight(hit);
+    if (pathHighlight) {
+        path.insertAdjacentHTML("afterbegin", pathHighlight + "/");
+    } else {
+        path.appendChild(document.createTextNode(hit["_source"]["path"] + "/"));
+    }
+
+    let tagContainer = document.createElement("div");
+    tagContainer.setAttribute("class", "tag-container");
+
+    for (let i = 0; i < tags.length; i++) {
+        tagContainer.appendChild(tags[i]);
+    }
+
+    //Size tag
+    let sizeTag = document.createElement("small");
+    sizeTag.appendChild(document.createTextNode(humanFileSize(hit["_source"]["size"])));
+    sizeTag.setAttribute("class", "text-muted");
+    tagContainer.appendChild(sizeTag);
+
+    titleDiv.appendChild(pathLine);
+    pathLine.appendChild(path);
+    pathLine.appendChild(tagContainer);
+
+    attachTagContainerEventListener(tagContainer, hit);
+
+    return line;
+}
+
+function makePreloader() {
+    const elem = document.createElement("div");
+    elem.setAttribute("class", "progress");
+    const bar = document.createElement("div");
+    bar.setAttribute("class", "progress-bar progress-bar-striped progress-bar-animated");
+    bar.setAttribute("style", "width: 100%");
+    elem.appendChild(bar);
+
+    return elem;
+}
+
+function makePageIndicator(searchResult) {
+    let pageIndicator = document.createElement("div");
+    pageIndicator.setAttribute("class", "page-indicator font-weight-light");
+    const totalHits = searchResult["aggregations"]["total_count"]["value"];
+    pageIndicator.appendChild(document.createTextNode(docCount + " / " + totalHits));
+    return pageIndicator;
+}
+
+
+function makeStatsCard(searchResult) {
+
+    let statsCard = document.createElement("div");
+    statsCard.setAttribute("class", "card stat");
+    let statsCardBody = document.createElement("div");
+    statsCardBody.setAttribute("class", "card-body");
+
+    // Stats
+    let stat = document.createElement("span");
+    const totalHits = searchResult["aggregations"]["total_count"]["value"];
+    stat.appendChild(document.createTextNode(totalHits + " results in " + searchResult["took"] + "ms"));
+
+    statsCardBody.appendChild(stat);
+
+    // Display mode
+    const resultMode = document.createElement("div");
+    resultMode.setAttribute("class", "btn-group btn-group-toggle");
+    resultMode.setAttribute("data-toggle", "buttons");
+    resultMode.style.cssFloat = "right";
+
+    const listMode = document.createElement("label");
+    listMode.setAttribute("class", "btn btn-primary");
+    listMode.setAttribute("title", "List mode");
+    listMode.innerHTML = '<svg width="20px" height="20px" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M80 368H16a16 16 0 0 0-16 16v64a16 16 0 0 0 16 16h64a16 16 0 0 0 16-16v-64a16 16 0 0 0-16-16zm0-320H16A16 16 0 0 0 0 64v64a16 16 0 0 0 16 16h64a16 16 0 0 0 16-16V64a16 16 0 0 0-16-16zm0 160H16a16 16 0 0 0-16 16v64a16 16 0 0 0 16 16h64a16 16 0 0 0 16-16v-64a16 16 0 0 0-16-16zm416 176H176a16 16 0 0 0-16 16v32a16 16 0 0 0 16 16h320a16 16 0 0 0 16-16v-32a16 16 0 0 0-16-16zm0-320H176a16 16 0 0 0-16 16v32a16 16 0 0 0 16 16h320a16 16 0 0 0 16-16V80a16 16 0 0 0-16-16zm0 160H176a16 16 0 0 0-16 16v32a16 16 0 0 0 16 16h320a16 16 0 0 0 16-16v-32a16 16 0 0 0-16-16z"></path></svg>';
+
+    const gridMode = document.createElement("label");
+    gridMode.setAttribute("class", "btn btn-primary");
+    gridMode.setAttribute("title", "Grid mode");
+    gridMode.innerHTML = '<svg width="20px" height="20px" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M149.333 56v80c0 13.255-10.745 24-24 24H24c-13.255 0-24-10.745-24-24V56c0-13.255 10.745-24 24-24h101.333c13.255 0 24 10.745 24 24zm181.334 240v-80c0-13.255-10.745-24-24-24H205.333c-13.255 0-24 10.745-24 24v80c0 13.255 10.745 24 24 24h101.333c13.256 0 24.001-10.745 24.001-24zm32-240v80c0 13.255 10.745 24 24 24H488c13.255 0 24-10.745 24-24V56c0-13.255-10.745-24-24-24H386.667c-13.255 0-24 10.745-24 24zm-32 80V56c0-13.255-10.745-24-24-24H205.333c-13.255 0-24 10.745-24 24v80c0 13.255 10.745 24 24 24h101.333c13.256 0 24.001-10.745 24.001-24zm-205.334 56H24c-13.255 0-24 10.745-24 24v80c0 13.255 10.745 24 24 24h101.333c13.255 0 24-10.745 24-24v-80c0-13.255-10.745-24-24-24zM0 376v80c0 13.255 10.745 24 24 24h101.333c13.255 0 24-10.745 24-24v-80c0-13.255-10.745-24-24-24H24c-13.255 0-24 10.745-24 24zm386.667-56H488c13.255 0 24-10.745 24-24v-80c0-13.255-10.745-24-24-24H386.667c-13.255 0-24 10.745-24 24v80c0 13.255 10.745 24 24 24zm0 160H488c13.255 0 24-10.745 24-24v-80c0-13.255-10.745-24-24-24H386.667c-13.255 0-24 10.745-24 24v80c0 13.255 10.745 24 24 24zM181.333 376v80c0 13.255 10.745 24 24 24h101.333c13.255 0 24-10.745 24-24v-80c0-13.255-10.745-24-24-24H205.333c-13.255 0-24 10.745-24 24z"></path></svg>';
+
+    resultMode.appendChild(gridMode);
+    resultMode.appendChild(listMode);
+
+    if (CONF.options.display === "grid") {
+        gridMode.classList.add("active")
+    } else {
+        listMode.classList.add("active")
+    }
+
+    gridMode.addEventListener("click", () => {
+        CONF.options.display = "grid";
+        CONF.save();
+        searchDebounced();
+    });
+    listMode.addEventListener("click", () => {
+        CONF.options.display = "list";
+        CONF.save();
+        searchDebounced();
+    });
+
+    statsCardBody.appendChild(resultMode);
+
+    // Sort mode
+    const sortMode = document.createElement("div");
+    sortMode.setAttribute("class", "dropdown");
+    sortMode.style.cssFloat = "right";
+    sortMode.style.marginRight = "10px";
+
+    const sortModeBtn = document.createElement("button");
+    sortModeBtn.setAttribute("class", "btn btn-md btn-primary dropdown-toggle");
+    sortModeBtn.setAttribute("id", "sortModeBtn");
+    sortModeBtn.setAttribute("type", "button");
+    sortModeBtn.setAttribute("data-toggle", "dropdown");
+    sortModeBtn.setAttribute("aria-haspopup", "true");
+    sortModeBtn.setAttribute("aria-expanded", "false");
+    sortModeBtn.setAttribute("title", "Sort options");
+    sortModeBtn.innerHTML = '<svg aria-hidden="true" width="20px" height="20px" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 320 512"><path fill="currentColor" d="M41 288h238c21.4 0 32.1 25.9 17 41L177 448c-9.4 9.4-24.6 9.4-33.9 0L24 329c-15.1-15.1-4.4-41 17-41zm255-105L177 64c-9.4-9.4-24.6-9.4-33.9 0L24 183c-15.1 15.1-4.4 41 17 41h238c21.4 0 32.1-25.9 17-41z"></path></svg>';
+
+    const sortModeMenu = document.createElement("div");
+    sortModeMenu.setAttribute("class", "dropdown-menu");
+    sortModeMenu.setAttribute("aria-labelledby", "sortModeBtn");
+
+    Object.keys(SORT_MODES).forEach(mode => {
+        const item = document.createElement("div");
+        item.setAttribute("class", "dropdown-item");
+        item.appendChild(document.createTextNode(SORT_MODES[mode].text));
+        sortModeMenu.appendChild(item);
+
+        item.onclick = function () {
+            CONF.options.sort = mode;
+            CONF.save();
+            searchDebounced();
+        };
+    });
+
+    sortMode.appendChild(sortModeBtn);
+    sortMode.appendChild(sortModeMenu);
+
+    statsCardBody.appendChild(sortMode);
+
+    if (totalHits !== 0) {
+        let sizeStat = document.createElement("div");
+        sizeStat.appendChild(document.createTextNode(humanFileSize(searchResult["aggregations"]["total_size"]["value"])));
+        statsCardBody.appendChild(sizeStat);
+    }
+
+    statsCard.appendChild(statsCardBody);
+
+    return statsCard;
+}
+
+function makeResultContainer() {
+    let resultContainer = document.createElement("div");
+
+    if (CONF.options.display === "grid") {
+        resultContainer.setAttribute("class", "bricklayer");
+    } else {
+        resultContainer.setAttribute("class", "list-group");
+    }
+    return resultContainer;
+}

src/static/js/search.js

@@ -0,0 +1,871 @@
+const SIZE = 60;
+let mimeMap = [];
+let tagMap = [];
+let mimeTree;
+let tagTree;
+
+let searchBar = document.getElementById("searchBar");
+let pathBar = document.getElementById("pathBar");
+let tagBar = document.getElementById("tagBar");
+let currentDocToTag = null;
+let currentTagCallback = null;
+let lastDoc = null;
+let reachedEnd = false;
+let docCount = 0;
+let coolingDown = false;
+let searchBusy = true;
+let selectedIndices = [];
+let indexMap = {};
+
+let size_min = 0;
+let size_max = 10000000000000;
+
+let date_min = null;
+let date_max = null;
+
+SORT_MODES = {
+    score: {
+        text: "Relevance",
+        mode: [
+            {_score: {order: "desc"}},
+            {_tie: {order: "asc"}}
+        ],
+        key: hit => hit["_score"]
+    },
+    date_asc: {
+        text: "Date (Ascending)", mode: [
+            {mtime: {order: "asc"}},
+            {_tie: {order: "asc"}}
+        ],
+        key: hit => hit["_source"]["mtime"]
+    },
+    date_desc: {
+        text: "Date (Descending)", mode: [
+            {mtime: {order: "desc"}},
+            {_tie: {order: "asc"}}
+        ],
+        key: hit => hit["_source"]["mtime"]
+    },
+    size_asc: {
+        text: "Size (Ascending)", mode: [
+            {size: {order: "asc"}},
+            {_tie: {order: "asc"}}
+        ],
+        key: hit => hit["_source"]["size"]
+    },
+    size_desc: {
+        text: "Size (Descending)", mode: [
+            {size: {order: "desc"}},
+            {_tie: {order: "asc"}}
+        ],
+        key: hit => hit["_source"]["size"]
+    },
+};
+
+function showEsError() {
+    $.toast({
+        heading: "Elasticsearch connection error",
+        text: "sist2 web module encountered an error while connecting " +
+            "to Elasticsearch. See server logs for more information.",
+        stack: false,
+        bgColor: "#a94442",
+        textColor: "#f2dede",
+        position: 'bottom-right',
+        hideAfter: false
+    });
+}
+
+window.onload = () => {
+    CONF.load();
+    new autoComplete({
+        selector: '#pathBar',
+        minChars: 1,
+        delay: 400,
+        renderItem: function (item) {
+            return '<div class="autocomplete-suggestion" data-val="' + item + '">' + item + '</div>';
+        },
+        source: async function (term, suggest) {
+
+            if (!CONF.options.suggestPath) {
+                return []
+            }
+
+            term = term.toLowerCase();
+
+            const choices = await getPathChoices();
+
+            let matches = [];
+            for (let i = 0; i < choices.length; i++) {
+                if (~choices[i].toLowerCase().indexOf(term)) {
+                    matches.push(choices[i]);
+                }
+            }
+            suggest(matches.sort());
+        },
+        onSelect: function () {
+            searchDebounced();
+        }
+    });
+    searchBar.addEventListener("keyup", searchDebounced);
+    pathBar.addEventListener("keyup", e => {
+        if (e.key === "Enter") {
+            searchDebounced();
+        }
+    });
+    new autoComplete({
+        selector: '#tagBar',
+        minChars: 1,
+        delay: 200,
+        renderItem: function (item) {
+            return '<div class="autocomplete-suggestion" data-val="' + item + '">' + item.split("#")[0] + '</div>';
+        },
+        source: async function (term, suggest) {
+            term = term.toLowerCase();
+
+            const choices = await getTagChoices();
+
+            let matches = [];
+            for (let i = 0; i < choices.length; i++) {
+                if (~choices[i].toLowerCase().indexOf(term)) {
+                    matches.push(choices[i]);
+                }
+            }
+            suggest(matches.sort());
+        },
+        onSelect: function (e, item) {
+            const name = item.split("#")[0];
+            const color = "#" + item.split("#")[1];
+            $("#tag-color").val(color);
+            $("#tag-color").trigger("keyup", color);
+            tagBar.value = name;
+            e.preventDefault();
+        }
+    });
+    [tagBar, document.getElementById("tag-color")].forEach(elem => {
+        elem.addEventListener("keyup", e => {
+            if (e.key === "Enter" && tagBar.value.length > 0) {
+                const tag = tagBar.value + document.getElementById("tag-color").value;
+                saveTag(tag, currentDocToTag).then(() => currentTagCallback(tag));
+            }
+        });
+    })
+    $("#tag-color").colorpicker({
+        format: "hex",
+        sliders: {
+            saturation: {
+                selector: '.colorpicker-saturation',
+                callLeft: 'setSaturationRatio',
+                callTop: 'setValueRatio'
+            },
+            hue: {
+                selector: '.colorpicker-hue',
+                maxLeft: 0,
+                callLeft: false,
+                callTop: 'setHueRatio'
+            }
+        }
+    });
+};
+
+function saveTag(tag, hit) {
+    const relPath = hit["_source"]["path"] + "/" + hit["_source"]["name"] + ext(hit);
+
+    return $.jsonPost("/tag/" + hit["_source"]["index"], {
+        delete: false,
+        name: tag,
+        doc_id: hit["_id"],
+        relpath: relPath
+    }).then(() => {
+        tagBar.blur();
+        $("#tagModal").modal("hide");
+        $.toast({
+            heading: "Tag added",
+            text: "Tag saved to index storage and updated in ElasticSearch",
+            stack: 3,
+            bgColor: "#00a4bc",
+            textColor: "#fff",
+            position: 'bottom-right',
+            hideAfter: 3000,
+            loaderBg: "#08c7e8",
+        });
+    })
+}
+
+function deleteTag(tag, hit) {
+    const relPath = hit["_source"]["path"] + "/" + hit["_source"]["name"] + ext(hit);
+
+    return $.jsonPost("/tag/" + hit["_source"]["index"], {
+        delete: true,
+        name: tag,
+        doc_id: hit["_id"],
+        relpath: relPath
+    }).then(() => {
+        $.toast({
+            heading: "Tag deleted",
+            text: "Tag deleted index storage and updated in ElasticSearch",
+            stack: 3,
+            bgColor: "#00a4bc",
+            textColor: "#fff",
+            position: 'bottom-right',
+            hideAfter: 3000,
+            loaderBg: "#08c7e8",
+        });
+    })
+}
+
+function toggleFuzzy() {
+    searchDebounced();
+}
+
+$.get("i").then(resp => {
+
+    const urlIndices = (new URLSearchParams(location.search)).get("i");
+    resp["indices"].forEach(idx => {
+        indexMap[idx.id] = idx.name;
+        const opt = $("<option>")
+            .attr("value", idx.id)
+            .append(idx.name);
+
+        if (urlIndices) {
+            if (urlIndices.split(",").indexOf(idx.name) !== -1) {
+                opt.attr("selected", true);
+                selectedIndices.push(idx.id);
+            }
+        } else if (!idx.name.includes("(nsfw)")) {
+            opt.attr("selected", true);
+            selectedIndices.push(idx.id);
+        }
+        $("#indices").append(opt);
+    });
+
+    createPathTree("#pathTree");
+});
+
+function getDocumentInfo(id) {
+    return $.getJSON("d/" + id).fail(showEsError)
+}
+
+function handleTreeClick(tree) {
+    return (node, e) => {
+        if (e !== "checked") {
+            return
+        }
+
+        if (node.id === "any") {
+            if (!node.itree.state.checked) {
+                tree.deselectDeep();
+            }
+        } else {
+            tree.node("any").deselect();
+        }
+
+        searchDebounced();
+    }
+}
+
+$.jsonPost("es", {
+    aggs: {
+        mimeTypes: {
+            terms: {
+                field: "mime",
+                size: 10000
+            }
+        }
+    },
+    size: 0,
+}).then(resp => {
+    resp["aggregations"]["mimeTypes"]["buckets"].sort((a, b) => a.key > b.key).forEach(bucket => {
+        let tmp = bucket["key"].split("/");
+        let category = tmp[0];
+        let mime = tmp[1];
+
+        let category_exists = false;
+
+        let child = {
+            "id": bucket["key"],
+            "text": `${mime} (${bucket["doc_count"]})`
+        };
+
+        mimeMap.forEach(node => {
+            if (node.text === category) {
+                node.children.push(child);
+                category_exists = true;
+            }
+        });
+
+        if (!category_exists) {
+            mimeMap.push({"text": category, children: [child]});
+        }
+    });
+    mimeMap.push({"text": "All", "id": "any"});
+
+    mimeTree = new InspireTree({
+        selection: {
+            mode: 'checkbox'
+        },
+        data: mimeMap
+    });
+    new InspireTreeDOM(mimeTree, {
+        target: '#mimeTree'
+    });
+    mimeTree.on("node.state.changed", handleTreeClick(mimeTree));
+    mimeTree.deselect();
+    mimeTree.node("any").select();
+});
+
+// Tags tree
+$.jsonPost("es", {
+    aggs: {
+        tags: {
+            terms: {
+                field: "tag",
+                size: 10000
+            }
+        }
+    },
+    size: 0,
+}).then(resp => {
+    resp["aggregations"]["tags"]["buckets"]
+        .sort((a, b) => a["key"].localeCompare(b["key"]))
+        .forEach(bucket => {
+            addTag(tagMap, bucket["key"], bucket["key"], bucket["doc_count"])
+        });
+
+    tagMap.push({"text": "All", "id": "any"});
+    tagTree = new InspireTree({
+        selection: {
+            mode: 'checkbox'
+        },
+        checkbox: {
+            autoCheckChildren: false
+        },
+        data: tagMap
+    });
+    new InspireTreeDOM(tagTree, {
+        target: '#tagTree'
+    });
+    tagTree.on("node.state.changed", handleTreeClick(tagTree));
+    tagTree.node("any").select();
+    searchBusy = false;
+});
+
+function addTag(map, tag, id, count) {
+    // let tags = tag.split("#")[0].split(".");
+    let tags = tag.split(".");
+
+    let child = {
+        id: id,
+        values: [id],
+        count: count,
+        text: tags.length !== 1 ? tags[0] : `${tags[0].split("#")[0]} (${count})`,
+        name: tags[0],
+        children: [],
+        isLeaf: tags.length === 1,
+        //Overwrite base functions
+        blur: function () {
+        },
+        select: function () {
+            this.state("selected", true);
+            return this.check()
+        },
+        deselect: function () {
+            this.state("selected", false);
+            return this.uncheck()
+        },
+        uncheck: function () {
+            if (!this.isLeaf) {
+                return;
+            }
+
+            baseStateChange('checked', false, 'unchecked', this, false);
+            this.state('indeterminate', false);
+
+            if (this.hasParent()) {
+                this.getParent().refreshIndeterminateState();
+            }
+
+            this._tree.end();
+            return this;
+        },
+        check: function () {
+            if (!this.isLeaf) {
+                return;
+            }
+
+            baseStateChange('checked', true, 'checked', this, false);
+
+            if (this.hasParent()) {
+                this.getParent().refreshIndeterminateState();
+            }
+
+            this._tree.end();
+            return this;
+        }
+    };
+
+    let found = false;
+    map.forEach(node => {
+        if (node.name.split("#")[0] === child.name.split("#")[0]) {
+            found = true;
+            if (tags.length !== 1) {
+                addTag(node.children, tags.slice(1).join("."), id, count);
+            } else {
+                // Same name, different color
+                node.count += count;
+                node.text = `${tags[0].split("#")[0]} (${node.count})`;
+                node.values.push(id);
+            }
+        }
+    });
+    if (!found) {
+        if (tags.length !== 1) {
+            addTag(child.children, tags.slice(1).join("."), id, count);
+            map.push(child);
+        } else {
+            map.push(child);
+        }
+    }
+}
+
+function insertHits(resultContainer, hits) {
+    for (let i = 0; i < hits.length; i++) {
+
+        if (CONF.options.display === "grid") {
+            resultContainer._brick.append(createDocCard(hits[i]));
+        } else {
+            resultContainer.appendChild(createDocLine(hits[i]));
+        }
+        docCount++;
+    }
+}
+
+window.addEventListener("scroll", function () {
+    if (!searchBusy) {
+        let threshold = 400;
+
+        if ((window.innerHeight + window.scrollY) >= document.body.offsetHeight - threshold) {
+            if (!reachedEnd) {
+                coolingDown = true;
+                search(lastDoc);
+            }
+        }
+    }
+});
+
+function getSelectedNodes(tree) {
+    let selectedNodes = [];
+
+    let selected = tree.selected();
+
+    for (let i = 0; i < selected.length; i++) {
+
+        if (selected[i].id === "any") {
+            return ["any"]
+        }
+
+        //Only get children
+        if (selected[i].text.indexOf("(") !== -1) {
+            if (selected[i].values) {
+                selectedNodes.push(selected[i].values);
+            } else {
+                selectedNodes.push(selected[i].id);
+            }
+        }
+    }
+
+    return selectedNodes
+}
+
+function search(after = null) {
+    lastDoc = null;
+
+    if (searchBusy) {
+        return;
+    }
+    searchBusy = true;
+
+    let searchResults = document.getElementById("searchResults");
+    //Clear old search results
+    let preload;
+    if (!after) {
+        while (searchResults.firstChild) {
+            searchResults.removeChild(searchResults.firstChild);
+        }
+        preload = makePreloader();
+        searchResults.appendChild(preload);
+    }
+
+    let query = searchBar.value;
+    let empty = query === "";
+    let condition = empty ? "should" : "must";
+    let filters = [
+        {range: {size: {gte: size_min, lte: size_max}}},
+        {terms: {index: selectedIndices}}
+    ];
+    let fields = [
+        "name^8",
+        "content^3",
+        "album^8", "artist^8", "title^8", "genre^2", "album_artist^8",
+        "font_name^6"
+    ];
+
+    if (CONF.options.searchInPath) {
+        fields.push("path.text^5");
+    }
+
+    if ($("#fuzzyToggle").prop("checked")) {
+        fields.push("content.nGram");
+        if (CONF.options.searchInPath) {
+            fields.push("path.nGram");
+        }
+        fields.push("name.nGram^3");
+    }
+
+    let path = pathBar.value.replace(/\/$/, "").toLowerCase(); //remove trailing slashes
+    if (path !== "") {
+        filters.push({term: {path: path}})
+    }
+    let mimeTypes = getSelectedNodes(mimeTree);
+    if (!mimeTypes.includes("any")) {
+        filters.push({terms: {"mime": mimeTypes}});
+    }
+
+    let tags = getSelectedNodes(tagTree);
+    if (!tags.includes("any")) {
+        tags.forEach(tagGroup => {
+            filters.push({terms: {"tag": tagGroup}})
+        })
+    }
+
+    if (date_min && date_max) {
+        filters.push({range: {mtime: {gte: date_min, lte: date_max}}})
+    } else if (date_min) {
+        filters.push({range: {mtime: {gte: date_min}}})
+    } else if (date_max) {
+        filters.push({range: {mtime: {lte: date_max}}})
+    }
+
+    let q = {
+        "_source": {
+            excludes: ["content", "_tie"]
+        },
+        query: {
+            bool: {
+                [condition]: {
+                    simple_query_string: {
+                        query: query,
+                        fields: fields,
+                        default_operator: "and"
+                    }
+                },
+                filter: filters
+            }
+        },
+        "sort": SORT_MODES[CONF.options.sort].mode,
+        aggs:
+            {
+                total_size: {"sum": {"field": "size"}},
+                total_count: {"value_count": {"field": "size"}}
+            },
+        size: SIZE,
+    };
+
+    if (after) {
+        q.search_after = [SORT_MODES[CONF.options.sort].key(after), after["_id"]];
+    }
+
+    if (CONF.options.highlight) {
+        q.highlight = {
+            pre_tags: ["<mark>"],
+            post_tags: ["</mark>"],
+            fragment_size: CONF.options.fragmentSize,
+            number_of_fragments: 1,
+            order: "score",
+            fields: {
+                content: {},
+                // "content.nGram": {},
+                name: {},
+                "name.nGram": {},
+                font_name: {},
+            }
+        };
+        if (CONF.options.searchInPath) {
+            q.highlight.fields["path.text"] = {};
+            q.highlight.fields["path.nGram"] = {};
+        }
+    }
+
+    $.jsonPost("es", q).then(searchResult => {
+        let hits = searchResult["hits"]["hits"];
+        if (hits) {
+            lastDoc = hits[hits.length - 1];
+        }
+
+        hits.forEach(hit => {
+            hit["_source"]["name"] = strUnescape(hit["_source"]["name"]);
+            hit["_source"]["path"] = strUnescape(hit["_source"]["path"]);
+        });
+
+        if (!after) {
+            preload.remove();
+            searchResults.appendChild(makeStatsCard(searchResult));
+        } else {
+            let pageIndicator = makePageIndicator(searchResult);
+            searchResults.appendChild(pageIndicator);
+        }
+
+        //Setup page
+        let resultContainer = makeResultContainer();
+        searchResults.appendChild(resultContainer);
+
+        if (CONF.options.display === "grid") {
+            resultContainer._brick = new Bricklayer(resultContainer);
+        }
+
+        if (!after) {
+            docCount = 0;
+        }
+        reachedEnd = hits.length !== SIZE;
+        insertHits(resultContainer, hits);
+        searchBusy = false;
+    });
+}
+
+
+let searchDebounced = _.debounce(function () {
+    coolingDown = false;
+    search()
+}, 500);
+
+
+//Size slider
+$("#sizeSlider").ionRangeSlider({
+    type: "double",
+    grid: false,
+    force_edges: true,
+    min: 0,
+    max: 3684.03149864,
+    from: 0,
+    to: 3684.03149864,
+    min_interval: 5,
+    drag_interval: true,
+    prettify: function (num) {
+
+        if (num === 0) {
+            return "0 B"
+        } else if (num >= 3684) {
+            return humanFileSize(num * num * num) + "+";
+        }
+
+        return humanFileSize(num * num * num)
+    },
+    onChange: function (e) {
+        size_min = (e.from * e.from * e.from);
+        size_max = (e.to * e.to * e.to);
+
+        if (e.to >= 3684) {
+            size_max = 10000000000000;
+        }
+
+        searchDebounced();
+    }
+});
+
+//Date slider
+$.jsonPost("es", {
+    aggs: {
+        date_min: {min: {field: "mtime"}},
+        date_max: {max: {field: "mtime"}},
+    },
+    size: 0
+}).then(resp => {
+    $("#dateSlider").ionRangeSlider({
+        type: "double",
+        grid: false,
+        force_edges: true,
+        min: resp["aggregations"]["date_min"]["value"],
+        max: resp["aggregations"]["date_max"]["value"],
+        from: resp["aggregations"]["date_min"]["value"],
+        to: (Date.now() / 1000),
+        min_interval: 3600 * 24 * 7,
+        step: 3600 * 24,
+        drag_interval: true,
+        prettify: function (num) {
+            let date = (new Date(num * 1000));
+            return date.getUTCFullYear() + "-" + ("0" + (date.getUTCMonth() + 1)).slice(-2) + "-" + ("0" + date.getUTCDate()).slice(-2)
+        },
+        onFinish: function (e) {
+            date_min = e.from === e.min ? null : e.from;
+            date_max = e.to === e.max ? null : e.to;
+            searchDebounced();
+        }
+    });
+})
+
+function updateIndices() {
+    let selected = $('#indices').find('option:selected');
+    selectedIndices = [];
+    $(selected).each(function () {
+        selectedIndices.push($(this).val());
+    });
+
+    searchDebounced();
+}
+
+document.getElementById("indices").addEventListener("change", updateIndices);
+updateIndices();
+
+window.onkeyup = function (e) {
+    if (e.key === "/" || e.key === "Escape") {
+        const bar = document.getElementById("searchBar");
+        bar.scrollIntoView();
+        bar.focus();
+    }
+};
+
+function getNextDepth(node) {
+    let q = {
+        query: {
+            bool: {
+                filter: [
+                    {term: {index: node.index}},
+                    {range: {_depth: {gte: node.depth + 1, lte: node.depth + 3}}},
+                ]
+            }
+        },
+        aggs: {
+            paths: {
+                terms: {
+                    field: "path",
+                    size: 10000
+                }
+            }
+        },
+        size: 0
+    };
+
+    if (node.depth > 0) {
+        q.query.bool.must = {
+            prefix: {
+                path: node.id,
+            }
+        };
+    }
+
+    return $.jsonPost("es", q).then(resp => {
+        const buckets = resp["aggregations"]["paths"]["buckets"];
+        if (!buckets) {
+            return false;
+        }
+
+        const paths = [];
+
+        return buckets
+            .filter(bucket => bucket.key.length > node.id.length || node.id.startsWith("/"))
+            .sort((a, b) => a.key > b.key)
+            .map(bucket => {
+
+                if (paths.some(n => bucket.key.startsWith(n))) {
+                    return null;
+                }
+
+                const name = node.id.startsWith("/") ? bucket.key : bucket.key.slice(node.id.length + 1);
+
+                paths.push(bucket.key);
+
+                return {
+                    id: bucket.key,
+                    text: `${name}/ (${bucket.doc_count})`,
+                    depth: node.depth + 1,
+                    index: node.index,
+                    values: [bucket.key],
+                    children: true,
+                }
+            }).filter(x => x !== null)
+    })
+}
+
+function handlePathTreeClick(tree) {
+    return (event, node, handler) => {
+
+        if (node.depth !== 0) {
+            $("#pathBar").val(node.id);
+            $("#pathTreeModal").modal("hide");
+            searchDebounced();
+        }
+
+        handler();
+    }
+}
+
+function createPathTree(target) {
+    let pathTree = new InspireTree({
+        data: function (node, resolve, reject) {
+            return getNextDepth(node);
+        },
+        sort: "text"
+    });
+
+    selectedIndices.forEach(index => {
+        pathTree.addNode({
+            id: "/" + index,
+            values: ["/" + index],
+            text: `/[${indexMap[index]}]`,
+            index: index,
+            depth: 0,
+            children: true
+        })
+    });
+
+    new InspireTreeDOM(pathTree, {
+        target: target
+    });
+
+    pathTree.on("node.click", handlePathTreeClick(pathTree));
+}
+
+function getPathChoices() {
+    return new Promise(getPaths => {
+        $.jsonPost("es", {
+            suggest: {
+                path: {
+                    prefix: pathBar.value,
+                    completion: {
+                        field: "suggest-path",
+                        skip_duplicates: true,
+                        size: 10000
+                    }
+                }
+            }
+        }).then(resp => getPaths(resp["suggest"]["path"][0]["options"].map(opt => opt["_source"]["path"])));
+    });
+}
+
+
+function getTagChoices() {
+    return new Promise(getPaths => {
+        $.jsonPost("es", {
+            suggest: {
+                tag: {
+                    prefix: tagBar.value,
+                    completion: {
+                        field: "suggest-tag",
+                        skip_duplicates: true,
+                        size: 10000
+                    }
+                }
+            }
+        }).then(resp => {
+            const result = [];
+            resp["suggest"]["tag"][0]["options"].map(opt => opt["_source"]["tag"]).forEach(tags => {
+                tags.forEach(tag => {
+                    const t = tag.split("#")[0];
+                    if (!result.find(x => x.split("#")[0] === t)) {
+                        result.push(tag);
+                    }
+                });
+            });
+            getPaths(result);
+        });
+    });
+}

src/static/js/util.js

@@ -0,0 +1,232 @@
+/**
+ * https://stackoverflow.com/questions/10420352
+ */
+function humanFileSize(bytes) {
+    if (bytes === 0) {
+        return "0 B"
+    }
+
+    let thresh = 1000;
+    if (Math.abs(bytes) < thresh) {
+        return bytes + ' B';
+    }
+    let units = ['k', 'M', 'G', 'T', 'P', 'E', 'Z', 'Y'];
+    let u = -1;
+    do {
+        bytes /= thresh;
+        ++u;
+    } while (Math.abs(bytes) >= thresh && u < units.length - 1);
+
+    return bytes.toFixed(1) + units[u];
+}
+
+/**
+ * https://stackoverflow.com/questions/6312993
+ */
+function humanTime(sec_num) {
+    sec_num = Math.floor(sec_num);
+    let hours = Math.floor(sec_num / 3600);
+    let minutes = Math.floor((sec_num - (hours * 3600)) / 60);
+    let seconds = sec_num - (hours * 3600) - (minutes * 60);
+
+    if (hours < 10) {
+        hours = "0" + hours;
+    }
+    if (minutes < 10) {
+        minutes = "0" + minutes;
+    }
+    if (seconds < 10) {
+        seconds = "0" + seconds;
+    }
+    return hours + ":" + minutes + ":" + seconds;
+}
+
+function debounce(func, wait) {
+    let timeout;
+    return function () {
+        let context = this, args = arguments;
+        let later = function () {
+            timeout = null;
+            func.apply(context, args);
+        };
+        clearTimeout(timeout);
+        timeout = setTimeout(later, wait);
+        func.apply(context, args);
+    };
+}
+
+function lum(c) {
+    c = c.substring(1);
+    let rgb = parseInt(c, 16);
+    let r = (rgb >> 16) & 0xff;
+    let g = (rgb >> 8) & 0xff;
+    let b = (rgb >> 0) & 0xff;
+
+    return 0.2126 * r + 0.7152 * g + 0.0722 * b;
+}
+
+function strUnescape(str) {
+    let result = "";
+
+    for (let i = 0; i < str.length; i++) {
+        const c = str[i];
+        const next = str[i+1];
+
+        if (c === ']') {
+            if (next === ']') {
+                result += c;
+                i += 1;
+            } else {
+                result += String.fromCharCode(parseInt(str.slice(i, i + 2), 16));
+                i += 2;
+            }
+        } else {
+            result += c;
+        }
+    }
+    return result;
+}
+
+const CONF = new Settings();
+
+const _defaults = {
+    display: "grid",
+    fuzzy: true,
+    highlight: true,
+    sort: "score",
+    searchInPath: false,
+    treemapType: "cascaded",
+    treemapTiling: "squarify",
+    treemapGroupingDepth: 3,
+    treemapColor: "PuBuGn",
+    treemapSize: "large",
+    suggestPath: true,
+    fragmentSize: 100,
+    columns: 5
+};
+
+function loadSettings() {
+    CONF.load();
+
+    $("#settingDisplay").val(CONF.options.display);
+    $("#settingFuzzy").prop("checked", CONF.options.fuzzy);
+    $("#settingHighlight").prop("checked", CONF.options.highlight);
+    $("#settingSearchInPath").prop("checked", CONF.options.searchInPath);
+    $("#settingTreemapTiling").val(CONF.options.treemapTiling);
+    $("#settingTreemapGroupingDepth").val(CONF.options.treemapGroupingDepth);
+    $("#settingTreemapColor").val(CONF.options.treemapColor);
+    $("#settingTreemapSize").val(CONF.options.treemapSize);
+    $("#settingTreemapType").val(CONF.options.treemapType);
+    $("#settingSuggestPath").prop("checked", CONF.options.suggestPath);
+    $("#settingFragmentSize").val(CONF.options.fragmentSize);
+    $("#settingColumns").val(CONF.options.columns);
+}
+
+function Settings() {
+    this.options = {};
+
+    this._onUpdate = function () {
+        $("#fuzzyToggle").prop("checked", this.options.fuzzy);
+        updateColumnStyle();
+    };
+
+    this.load = function () {
+        const raw = window.localStorage.getItem("options");
+        if (raw === null) {
+            this.options = _defaults;
+        } else {
+            const j = JSON.parse(raw);
+            if (!j || Object.keys(_defaults).some(k => !j.hasOwnProperty(k))) {
+                this.options = _defaults;
+            } else {
+                this.options = j;
+            }
+        }
+
+        this._onUpdate();
+    };
+
+    this.save = function () {
+        window.localStorage.setItem("options", JSON.stringify(this.options));
+        this._onUpdate();
+    }
+}
+
+function updateSettings() {
+    CONF.options.display = $("#settingDisplay").val();
+    CONF.options.fuzzy = $("#settingFuzzy").prop("checked");
+    CONF.options.highlight = $("#settingHighlight").prop("checked");
+    CONF.options.searchInPath = $("#settingSearchInPath").prop("checked");
+    CONF.options.treemapTiling = $("#settingTreemapTiling").val();
+    CONF.options.treemapGroupingDepth = $("#settingTreemapGroupingDepth").val();
+    CONF.options.treemapColor = $("#settingTreemapColor").val();
+    CONF.options.treemapSize = $("#settingTreemapSize").val();
+    CONF.options.treemapType = $("#settingTreemapType").val();
+    CONF.options.suggestPath = $("#settingSuggestPath").prop("checked");
+    CONF.options.fragmentSize = $("#settingFragmentSize").val();
+    CONF.options.columns = $("#settingColumns").val();
+    CONF.save();
+
+    if (typeof searchDebounced !== "undefined") {
+        searchDebounced();
+    }
+
+    if (typeof updateStats !== "undefined") {
+        updateStats();
+    }
+
+    $.toast({
+        heading: "Settings updated",
+        text: "Settings saved to browser storage",
+        stack: 3,
+        bgColor: "#00a4bc",
+        textColor: "#fff",
+        position: 'bottom-right',
+        hideAfter: 3000,
+        loaderBg: "#08c7e8",
+    });
+}
+
+jQuery["jsonPost"] = function (url, data) {
+    return jQuery.ajax({
+        url: url,
+        type: "post",
+        data: JSON.stringify(data),
+        contentType: "application/json"
+    }).fail(err => {
+        showEsError();
+        console.log(err);
+    });
+};
+
+function toggleTheme() {
+    if (!document.cookie.includes("sist")) {
+        document.cookie = "sist=dark;SameSite=Strict";
+    } else {
+        document.cookie = "sist=; Max-Age=-99999999;";
+    }
+    window.location.reload();
+}
+
+function updateColumnStyle() {
+    const style = document.getElementById("style");
+    if (style) {
+        style.innerHTML =
+        `
+@media screen and (min-width: 1500px) {
+    .container {
+            max-width: 1440px;
+        }
+
+    .bricklayer-column-sizer {
+            width: ${100 / CONF.options.columns}% !important;
+        }
+
+    .bricklayer-column {
+            max-width: ${100 / CONF.options.columns}%;
+        }
+    }
+}
+        `
+    }
+}

src/static/search.html

@@ -0,0 +1,345 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="utf-8">
+    <title>sist2 - Search</title>
+    <meta name='viewport' content='width=device-width, initial-scale=1.0, maximum-scale=1.0, user-scalable=no'/>
+
+    <link href="css" rel="stylesheet" type="text/css">
+    <style id="style"></style>
+</head>
+<body>
+
+<nav class="navbar navbar-expand-lg">
+    <a class="navbar-brand" href="/">sist2</a>
+    <span class="badge badge-pill version">2.7.4</span>
+    <span class="tagline">Lightning-fast file system indexer and search tool </span>
+    <a class="btn ml-auto" href="/stats">Stats</a>
+    <button class="btn" type="button" data-toggle="modal" data-target="#settings" onclick="loadSettings()">Settings
+    </button>
+    <button class="btn" title="Toggle theme" onclick="toggleTheme()">Theme</button>
+</nav>
+
+<div class="container">
+    <div class="card">
+        <div class="card-body">
+            <div class="input-group">
+                <div class="input-group-prepend">
+                    <div class="input-group-text">
+                        <span title="Toggle fuzzy searching" onclick="document.getElementById('fuzzyToggle').click()">Fuzzy&nbsp</span>
+                        <input title="Toggle fuzzy searching" type="checkbox" id="fuzzyToggle"
+                               onclick="toggleFuzzy()" checked>
+                    </div>
+                </div>
+                <input id="searchBar" type="search" class="form-control" placeholder="Search">
+                <div class="input-group-append">
+                    <button class="btn btn-outline-secondary small-btn" type="button" data-toggle="modal"
+                            data-target="#help">?
+                    </button>
+                    <button class="btn btn-outline-secondary large-btn" type="button" data-toggle="modal"
+                            data-target="#help">Help
+                    </button>
+                </div>
+            </div>
+
+            <div class="row">
+                <div class="col">
+                    <input title="File size" id="sizeSlider" name="size" width="100%">
+                </div>
+
+                <div class="col">
+                    <div class="input-group" style="margin-bottom: 0.5em; margin-top: 1em">
+                        <div class="input-group-prepend">
+                            <button id="pathBarHelper" class="btn btn-outline-secondary" data-toggle="modal"
+                                    data-target="#pathTreeModal">
+                                <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 576 512" width="20px">
+                                    <path d="M288 224h224a32 32 0 0 0 32-32V64a32 32 0 0 0-32-32H400L368 0h-80a32 32 0 0 0-32 32v64H64V8a8 8 0 0 0-8-8H40a8 8 0 0 0-8 8v392a16 16 0 0 0 16 16h208v64a32 32 0 0 0 32 32h224a32 32 0 0 0 32-32V352a32 32 0 0 0-32-32H400l-32-32h-80a32 32 0 0 0-32 32v64H64V128h192v64a32 32 0 0 0 32 32zm0 96h66.74l32 32H512v128H288zm0-288h66.74l32 32H512v128H288z"/>
+                                </svg>
+                            </button>
+                        </div>
+                        <input id="pathBar" type="search" class="form-control" placeholder="Filter path">
+                    </div>
+                </div>
+            </div>
+
+            <input title="Date filter" id="dateSlider" name="size" width="100%">
+
+            <div class="row">
+                <div class="col">
+                    <label for="indices">Search in indices</label>
+                    <select class="custom-select" id="indices" multiple size="6"></select>
+                </div>
+
+                <div class="col" id="treeTabs">
+                    <ul class="nav nav-tabs" role="tablist">
+                        <li class="nav-item">
+                            <a class="nav-link active" data-toggle="tab" href="#mime" role="tab" aria-controls="home"
+                               aria-selected="true">Mime Types</a>
+                        </li>
+                        <li class="nav-item">
+                            <a class="nav-link" data-toggle="tab" href="#tag" role="tab" aria-controls="profile"
+                               aria-selected="false" title="User-defined tags">Tags</a>
+                        </li>
+                    </ul>
+                    <div class="tab-content" id="myTabContent">
+                        <div class="tab-pane fade show active" id="mime" role="tabpanel" aria-labelledby="home-tab">
+                            <div id="mimeTree" class="tree"></div>
+                        </div>
+                        <div class="tab-pane fade" id="tag" role="tabpanel" aria-labelledby="profile-tab">
+                            <div id="tagTree" class="tree"></div>
+                        </div>
+                    </div>
+                </div>
+
+            </div>
+        </div>
+    </div>
+
+    <div class="modal" id="modal" tabindex="-1" role="dialog" aria-labelledby="modal-title" aria-hidden="true">
+        <div class="modal-dialog modal-lg modal-dialog-centered" role="document">
+            <div class="modal-content">
+                <div class="modal-header">
+                    <h5 class="modal-title" id="modal-title"></h5>
+                    <button type="button" class="close" data-dismiss="modal" aria-label="Close">
+                        <span aria-hidden="true">&times;</span>
+                    </button>
+                </div>
+                <div class="modal-body" id="modal-body"></div>
+            </div>
+        </div>
+    </div>
+
+    <div class="modal" id="help" tabindex="-1" role="dialog" aria-labelledby="modal-title" aria-hidden="true">
+        <div class="modal-dialog modal-lg modal-dialog-centered" role="document">
+            <div class="modal-content">
+                <div class="modal-header">
+                    <h5 class="modal-title">Search help</h5>
+                    <button type="button" class="close" data-dismiss="modal" aria-label="Close">
+                        <span aria-hidden="true">&times;</span>
+                    </button>
+                </div>
+                <div class="modal-body">
+
+                    <table class="table">
+                        <tbody>
+                        <tr>
+                            <td><code>+</code></td>
+                            <td>signifies AND operation</td>
+                        </tr>
+                        <tr>
+                            <td><code>|</code></td>
+                            <td>signifies OR operation</td>
+                        </tr>
+                        <tr>
+                            <td><code>-</code></td>
+                            <td>negates a single token</td>
+                        </tr>
+                        <tr>
+                            <td><code>""</code></td>
+                            <td>wraps a number of tokens to signify a phrase for searching</td>
+                        </tr>
+                        <tr>
+                            <td><code>*</code></td>
+                            <td>at the end of a term signifies a prefix query</td>
+                        </tr>
+                        <tr>
+                            <td><code>(</code> and <code>)</code></td>
+                            <td>signify precedence</td>
+                        </tr>
+                        <tr>
+                            <td><code>~N</code></td>
+                            <td>after a word signifies edit distance (fuzziness)</td>
+                        </tr>
+                        <tr>
+                            <td><code>~N</code></td>
+                            <td>after a phrase signifies slop amount</td>
+                        </tr>
+                        </tbody>
+                    </table>
+
+                    <p>For example: <code>"fried eggs" +(eggplant | potato) -frittata</code> will match the phrase
+                        <i>fried eggs</i> and either <i>eggplant</i> or <i>potato</i>, but will ignore results
+                        containing <i>frittata</i>.</p>
+
+                    <p>When neither <code>+</code> or <code>|</code> is specified, the default operator is
+                        <code>+</code> (and).</p>
+                    <p>When the <b>Fuzzy</b> option is checked, partial matches are also returned.</p>
+                    <br>
+                    <p>For more information, see <a target="_blank"
+                                                    href="//www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-simple-query-string-query.html">Elasticsearch
+                        documentation</a></p>
+                </div>
+            </div>
+        </div>
+    </div>
+
+    <div class="modal" id="settings" tabindex="-1" role="dialog" aria-labelledby="modal-title" aria-hidden="true">
+        <div class="modal-dialog modal-dialog-centered" role="document">
+            <div class="modal-content">
+                <div class="modal-header">
+                    <h5 class="modal-title">Settings</h5>
+                    <button type="button" class="close" data-dismiss="modal" aria-label="Close">
+                        <span aria-hidden="true">&times;</span>
+                    </button>
+                </div>
+                <div class="modal-body">
+                    <div class="custom-control custom-checkbox">
+                        <input type="checkbox" class="custom-control-input" id="settingHighlight">
+                        <label class="custom-control-label" for="settingHighlight">Enable highlighting</label>
+                    </div>
+
+                    <div class="custom-control custom-checkbox">
+                        <input type="checkbox" class="custom-control-input" id="settingFuzzy">
+                        <label class="custom-control-label" for="settingFuzzy">Set fuzzy search by default</label>
+                    </div>
+
+                    <div class="custom-control custom-checkbox">
+                        <input type="checkbox" class="custom-control-input" id="settingSearchInPath">
+                        <label class="custom-control-label" for="settingSearchInPath">Enable matching query against
+                            document path</label>
+                    </div>
+
+                    <div class="custom-control custom-checkbox">
+                        <input type="checkbox" class="custom-control-input" id="settingSuggestPath">
+                        <label class="custom-control-label" for="settingSuggestPath">Enable auto-complete in path filter
+                            bar</label>
+                    </div>
+
+                    <br/>
+                    <div class="form-group">
+                        <input type="number" class="form-control" id="settingFragmentSize">
+                        <label for="settingFragmentSize">Highlight context size in characters</label>
+                    </div>
+
+                    <label for="settingDisplay">Display</label>
+                    <select id="settingDisplay" class="form-control form-control-sm">
+                        <option value="grid">Grid</option>
+                        <option value="list">List</option>
+                    </select>
+
+                    <div class="form-group">
+                        <label for="settingColumns">Maximum column count</label>
+                        <select id="settingColumns" class="form-control form-control-sm">
+                            <option value="3">3</option>
+                            <option value="4">4</option>
+                            <option value="5">5</option>
+                            <option value="6">6</option>
+                            <option value="7">7</option>
+                            <option value="8">8</option>
+                            <option value="9">9</option>
+                            <option value="13">13</option>
+                        </select>
+                    </div>
+
+                    <hr/>
+                    <h4>Stats</h4>
+
+                    <div class="form-group">
+                        <label for="settingTreemapType">Treemap type</label>
+                        <select id="settingTreemapType" class="form-control form-control-sm">
+                            <option value="cascaded">Cascaded</option>
+                            <option value="flat">Flat (compact)</option>
+                        </select>
+                    </div>
+
+                    <div class="form-group">
+                        <label for="settingTreemapTiling">Treemap tiling</label>
+                        <select id="settingTreemapTiling" class="form-control form-control-sm">
+                            <option value="binary">Binary</option>
+                            <option value="squarify">Squarify</option>
+                            <option value="slice">Slice</option>
+                            <option value="dice">Dice</option>
+                            <option value="sliceDice">Slide & Dice</option>
+                        </select>
+                    </div>
+
+                    <div class="form-group">
+                        <label for="settingTreemapGroupingDepth">Treemap color grouping depth (flat)</label>
+                        <input type="number" class="form-control" id="settingTreemapGroupingDepth" min="1" max="10">
+                    </div>
+
+                    <div class="form-group">
+                        <label for="settingTreemapColor">Treemap color (cascaded)</label>
+                        <select id="settingTreemapColor" class="form-control form-control-sm">
+                            <option value="PuBuGn">Purple-Blue-Green</option>
+                            <option value="PuRd">Purple-Red</option>
+                            <option value="PuBu">Purple-Blue</option>
+                            <option value="YlOrBr">Yellow-Orange-Brown</option>
+                            <option value="YlOrRd">Yellow-Orange-Red</option>
+                            <option value="YlGn">Yellow-Green</option>
+                            <option value="YlGnBu">Yellow-Green-Blue</option>
+                            <option value="Plasma">Plasma</option>
+                            <option value="Magma">Magma</option>
+                            <option value="Inferno">Inferno</option>
+                            <option value="Viridis">Viridis</option>
+                            <option value="Turbo">Turbo</option>
+                        </select>
+                    </div>
+
+                    <div class="form-group">
+                        <label for="settingTreemapSize">Treemap size</label>
+                        <select id="settingTreemapSize" class="form-control form-control-sm">
+                            <option value="small">Small</option>
+                            <option value="medium">Medium</option>
+                            <option value="large">Large</option>
+                            <option value="x-large">X-Large</option>
+                            <option value="xx-large">XX-Large</option>
+                        </select>
+                    </div>
+
+                    <br>
+                    <button class="btn btn-primary ml-auto" onclick="updateSettings()">Update settings</button>
+                </div>
+            </div>
+        </div>
+    </div>
+
+    <div class="modal" id="pathTreeModal" tabindex="-1" role="dialog" aria-labelledby="modal-title" aria-hidden="true">
+        <div class="modal-dialog modal-lg" role="document">
+            <div class="modal-content">
+                <div class="modal-header">
+                    <h5 class="modal-title">Select path</h5>
+                    <button type="button" class="close" data-dismiss="modal" aria-label="Close">
+                        <span aria-hidden="true">&times;</span>
+                    </button>
+                </div>
+                <div class="modal-body">
+                    <div id="pathTree" class="tree"></div>
+                </div>
+            </div>
+        </div>
+    </div>
+
+    <div class="modal" id="tagModal" tabindex="-1" role="dialog" aria-labelledby="modal-title" aria-hidden="true">
+        <div class="modal-dialog modal-dialog-centered" role="document">
+            <div class="modal-content">
+                <div class="modal-header">
+                    <h5 class="modal-title">Add tag</h5>
+                    <button type="button" class="close" data-dismiss="modal" aria-label="Close">
+                        <span aria-hidden="true">&times;</span>
+                    </button>
+                </div>
+
+                <div class="modal-body">
+                    <div class="form-group">
+                        <div class="row">
+                            <div class="col col-8">
+                                <input type="text" id="tagBar" class="form-control">
+                            </div>
+                            <div class="col col-4">
+                                <input type="text" id="tag-color" value="" class="form-control"/>
+                            </div>
+                        </div>
+                    </div>
+                </div>
+            </div>
+        </div>
+    </div>
+
+    <div id="searchResults"></div>
+</div>
+
+<script src="jslib" type="text/javascript"></script>
+<script src="jssearch" type="text/javascript"></script>
+</body>
+</html>

src/static/stats.html

@@ -0,0 +1,813 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="utf-8">
+    <title>sist2 - Stats</title>
+    <meta name='viewport' content='width=device-width, initial-scale=1.0, maximum-scale=1.0, user-scalable=no'/>
+    <link href="css" rel="stylesheet" type="text/css">
+</head>
+<body>
+
+<nav class="navbar navbar-expand-lg">
+    <a class="navbar-brand" href="/">sist2</a>
+    <span class="badge badge-pill version">2.7.4</span>
+    <span class="tagline">Lightning-fast file system indexer and search tool </span>
+    <a style="margin-left: auto" class="btn" href="/">Back</a>
+    <button class="btn" type="button" data-toggle="modal" data-target="#settings"
+            onclick="loadSettings()">Settings
+    </button>
+    <button class="btn" title="Toggle theme" onclick="toggleTheme()">Theme</button>
+</nav>
+
+<div class="container pb-3">
+    <div class="card">
+        <div class="card-body">
+
+            <label for="indices">Index</label>
+            <select id="indices" onchange="updateStats()"></select>
+        </div>
+    </div>
+
+    <div id="treemap-card" class="stats-card">
+        <button class="btn stats-btn" onclick="fullScreen('treemap-card')">Enlarge</button>
+        <button class="btn stats-btn" onclick="exportTreemap()">Export</button>
+        <svg id="treemap"></svg>
+    </div>
+
+    <div id="graphs-card" class="stats-card">
+        <button class="btn stats-btn" onclick="fullScreen('graphs-card')">Enlarge</button>
+        <div class="graph">
+            <svg id="agg_mime_size"></svg>
+        </div>
+        <div class="graph">
+            <svg id="agg_mime_count"></svg>
+        </div>
+        <div class="graph">
+            <svg id="date_histogram"></svg>
+        </div>
+        <div class="graph">
+            <svg id="size_histogram"></svg>
+        </div>
+    </div>
+</div>
+
+<div class="modal" id="settings" tabindex="-1" role="dialog" aria-labelledby="modal-title" aria-hidden="true">
+    <div class="modal-dialog modal-dialog-centered" role="document">
+        <div class="modal-content">
+            <div class="modal-header">
+                <h5 class="modal-title">Settings</h5>
+                <button type="button" class="close" data-dismiss="modal" aria-label="Close">
+                    <span aria-hidden="true">&times;</span>
+                </button>
+            </div>
+            <div class="modal-body">
+                <div class="custom-control custom-checkbox">
+                    <input type="checkbox" class="custom-control-input" id="settingHighlight">
+                    <label class="custom-control-label" for="settingHighlight">Enable highlighting</label>
+                </div>
+
+                <div class="custom-control custom-checkbox">
+                    <input type="checkbox" class="custom-control-input" id="settingFuzzy">
+                    <label class="custom-control-label" for="settingFuzzy">Set fuzzy search by default</label>
+                </div>
+
+                <div class="custom-control custom-checkbox">
+                    <input type="checkbox" class="custom-control-input" id="settingSearchInPath">
+                    <label class="custom-control-label" for="settingSearchInPath">Enable matching query against document
+                        path</label>
+                </div>
+
+                <div class="custom-control custom-checkbox">
+                    <input type="checkbox" class="custom-control-input" id="settingSuggestPath">
+                    <label class="custom-control-label" for="settingSuggestPath">Enable auto-complete in path filter bar</label>
+                </div>
+
+                <br/>
+                <div class="form-group">
+                    <input type="number" class="form-control" id="settingFragmentSize">
+                    <label for="settingFragmentSize">Highlight context size in characters</label>
+                </div>
+
+                <label for="settingDisplay">Display</label>
+                <select id="settingDisplay" class="form-control form-control-sm">
+                    <option value="grid">Grid</option>
+                    <option value="list">List</option>
+                </select>
+
+                <div class="form-group">
+                    <label for="settingColumns">Maximum column count</label>
+                    <select id="settingColumns" class="form-control form-control-sm">
+                        <option value="3">3</option>
+                        <option value="4">4</option>
+                        <option value="5">5</option>
+                        <option value="6">6</option>
+                        <option value="7">7</option>
+                        <option value="8">8</option>
+                        <option value="9">9</option>
+                    </select>
+                </div>
+
+                <hr/>
+                <h4>Stats</h4>
+
+                <div class="form-group">
+                    <label for="settingTreemapType">Treemap type</label>
+                    <select id="settingTreemapType" class="form-control form-control-sm">
+                        <option value="cascaded">Cascaded</option>
+                        <option value="flat">Flat (compact)</option>
+                    </select>
+                </div>
+
+                <div class="form-group">
+                    <label for="settingTreemapTiling">Treemap tiling</label>
+                    <select id="settingTreemapTiling" class="form-control form-control-sm">
+                        <option value="binary">Binary</option>
+                        <option value="squarify">Squarify</option>
+                        <option value="slice">Slice</option>
+                        <option value="dice">Dice</option>
+                        <option value="sliceDice">Slide & Dice</option>
+                    </select>
+                </div>
+
+                <div class="form-group">
+                    <label for="settingTreemapGroupingDepth">Treemap color grouping depth (flat)</label>
+                    <input type="number" class="form-control" id="settingTreemapGroupingDepth" min="1" max="10">
+                </div>
+
+                <div class="form-group">
+                    <label for="settingTreemapColor">Treemap color (cascaded)</label>
+                    <select id="settingTreemapColor" class="form-control form-control-sm">
+                        <option value="PuBuGn">Purple-Blue-Green</option>
+                        <option value="PuRd">Purple-Red</option>
+                        <option value="PuBu">Purple-Blue</option>
+                        <option value="YlOrBr">Yellow-Orange-Brown</option>
+                        <option value="YlOrRd">Yellow-Orange-Red</option>
+                        <option value="YlGn">Yellow-Green</option>
+                        <option value="YlGnBu">Yellow-Green-Blue</option>
+                        <option value="Plasma">Plasma</option>
+                        <option value="Magma">Magma</option>
+                        <option value="Inferno">Inferno</option>
+                        <option value="Viridis">Viridis</option>
+                        <option value="Turbo">Turbo</option>
+                    </select>
+                </div>
+
+                <div class="form-group">
+                    <label for="settingTreemapSize">Treemap size</label>
+                    <select id="settingTreemapSize" class="form-control form-control-sm">
+                        <option value="small">Small</option>
+                        <option value="medium">Medium</option>
+                        <option value="large">Large</option>
+                        <option value="x-large">X-Large</option>
+                        <option value="xx-large">XX-Large</option>
+                    </select>
+                </div>
+
+                <br>
+                <button class="btn btn-primary float-right" onclick="updateSettings()">Update settings</button>
+            </div>
+        </div>
+    </div>
+</div>
+
+<script src="jslib" type="text/javascript"></script>
+<script>
+let width;
+let height;
+let indexMap = {};
+
+const barHeight = 20;
+const ordinalColor = d3.scaleOrdinal(d3.schemeCategory10);
+
+const formatSI = d3.format("~s");
+
+
+const TILING_MODES = {
+    "squarify": d3.treemapSquarify,
+    "binary": d3.treemapBinary,
+    "sliceDice": d3.treemapSliceDice,
+    "slice": d3.treemapSlice,
+    "dice": d3.treemapDice,
+};
+
+const COLORS = {
+    "PuBuGn": d3.interpolatePuBuGn,
+    "PuRd": d3.interpolatePuRd,
+    "PuBu": d3.interpolatePuBu,
+    "YlOrBr": d3.interpolateYlOrBr,
+    "YlOrRd": d3.interpolateYlOrRd,
+    "YlGn": d3.interpolateYlGn,
+    "YlGnBu": d3.interpolateYlGnBu,
+    "Plasma": d3.interpolatePlasma,
+    "Magma": d3.interpolateMagma,
+    "Inferno": d3.interpolateInferno,
+    "Viridis": d3.interpolateViridis,
+    "Turbo": d3.interpolateTurbo,
+};
+
+const SIZES = {
+    "small": [800, 600],
+    "medium": [1300, 750],
+    "large": [1900, 900],
+    "x-large": [2800, 1700],
+    "xx-large": [3600, 2000],
+};
+
+const fillOpacity = document.cookie.includes("sist") ? 0.9 : 0.6;
+
+const uids = {};
+
+function uid(name) {
+    let id = uids[name] || 0;
+    uids[name] = id + 1;
+    return name + id;
+}
+
+const burrow = function (table, addSelfDir) {
+    const root = {};
+    table.forEach(row => {
+        let layer = root;
+
+        row.taxonomy.forEach(key => {
+            layer[key] = key in layer ? layer[key] : {};
+            layer = layer[key];
+        });
+        if (Object.keys(layer).length === 0) {
+            layer["$size$"] = row.size;
+        } else if (addSelfDir) {
+            layer["."] = {
+                "$size$": row.size,
+            };
+        }
+    });
+
+    const descend = function (obj, depth) {
+        return Object.keys(obj).filter(k => k !== "$size$").map(k => {
+            const child = {
+                name: k,
+                depth: depth,
+                value: 0,
+                children: descend(obj[k], depth + 1)
+            };
+            if ("$size$" in obj[k]) {
+                child.value = obj[k]["$size$"];
+            }
+            return child;
+        });
+    };
+
+    return {
+        name: `[${indexMap[$("#indices").val()]}]`,
+        children: descend(root, 1),
+        value: 0,
+        depth: 0,
+    }
+};
+
+function flatTreemap(data, svg) {
+    const root = d3.treemap()
+        .tile(TILING_MODES[CONF.options.treemapTiling])
+        .size([width, height])
+        .padding(1)
+        .round(true)(
+            d3.hierarchy(data)
+                .sum(d => d.value)
+                .sort((a, b) => b.value - a.value)
+        );
+
+    const leaf = svg.selectAll("g")
+        .data(root.leaves())
+        .join("g")
+        .attr("transform", d => `translate(${d.x0},${d.y0})`);
+
+    leaf.append("title")
+        .text(d => `${d.ancestors().reverse().map(d => d.data.name).join("/")}\n${humanFileSize(d.value)}`);
+
+    leaf.append("rect")
+        .attr("id", d => (d.leafUid = uid("leaf")))
+        .attr("fill", d => {
+            while (d.depth > CONF.options.treemapGroupingDepth) d = d.parent;
+            return ordinalColor(d.data.name);
+        })
+        .attr("fill-opacity", fillOpacity)
+        .attr("width", d => d.x1 - d.x0)
+        .attr("height", d => d.y1 - d.y0);
+
+    leaf.append("clipPath")
+        .attr("id", d => (d.clipUid = uid("clip")))
+        .append("use")
+        .attr("href", d => `#${d.leafUid}`);
+
+    leaf.append("text")
+        .attr("clip-path", d => `url(#${d.clipUid})`)
+        .selectAll("tspan")
+        .data(d => {
+            if (d.data.name === ".") {
+                d = d.parent;
+            }
+            return [d.data.name, humanFileSize(d.value)]
+        })
+        .join("tspan")
+        .attr("x", 2)
+        .attr("y", (d, i, nodes) => `${i === 0 ? 1.1 : 2.3}em`)
+        .text(d => d);
+}
+
+function cascade(root, offset) {
+    const x = new Map;
+    const y = new Map;
+    return root.eachAfter(d => {
+        if (d.children && d.children.length !== 0) {
+            x.set(d, 1 + d3.max(d.children, c => c.x1 === d.x1 - offset ? x.get(c) : NaN));
+            y.set(d, 1 + d3.max(d.children, c => c.y1 === d.y1 - offset ? y.get(c) : NaN));
+        } else {
+            x.set(d, 0);
+            y.set(d, 0);
+        }
+    }).eachBefore(d => {
+        d.x1 -= 2 * offset * x.get(d);
+        d.y1 -= 2 * offset * y.get(d);
+    });
+}
+
+function cascadeTreemap(data, svg) {
+
+    const root = cascade(
+        d3.treemap()
+            .size([width, height])
+            .tile(TILING_MODES[CONF.options.treemapTiling])
+            .paddingOuter(3)
+            .paddingTop(16)
+            .paddingInner(1)
+            .round(true)(
+                d3.hierarchy(data)
+                    .sum(d => d.value)
+                    .sort((a, b) => b.value - a.value)
+            ),
+        3 // treemap.paddingOuter
+    );
+
+    const maxDepth = Math.max(...root.descendants().map(d => d.depth));
+    const color = d3.scaleSequential([maxDepth, -1], COLORS[CONF.options.treemapColor]);
+
+    svg.append("filter")
+        .attr("id", "shadow")
+        .append("feDropShadow")
+        .attr("flood-opacity", 0.3)
+        .attr("dx", 0)
+        .attr("stdDeviation", 3);
+
+    const node = svg.selectAll("g")
+        .data(
+            d3.nest()
+                .key(d => d.depth).sortKeys(d3.ascending)
+                .entries(root.descendants())
+        )
+        .join("g")
+        .attr("filter", "url(#shadow)")
+        .selectAll("g")
+        .data(d => d.values)
+        .join("g")
+        .attr("transform", d => `translate(${d.x0},${d.y0})`);
+
+    node.append("title")
+        .text(d => `${d.ancestors().reverse().splice(1).map(d => d.data.name).join("/")}\n${humanFileSize(d.value)}`);
+
+    node.append("rect")
+        .attr("id", d => (d.nodeUid = uid("node")))
+        .attr("fill", d => color(d.depth))
+        .attr("width", d => d.x1 - d.x0)
+        .attr("height", d => d.y1 - d.y0);
+
+    node.append("clipPath")
+        .attr("id", d => (d.clipUid = uid("clip")))
+        .append("use")
+        .attr("href", d => `#${d.nodeUid}`);
+
+    node.append("text")
+        .attr("fill", d => d3.hsl(color(d.depth)).l > .5 ? "#333" : "#eee")
+        .attr("clip-path", d => `url(#${d.clipUid})`)
+        .selectAll("tspan")
+        .data(d => [d.data.name, humanFileSize(d.value)])
+        .join("tspan")
+        .text(d => d);
+
+    node.filter(d => d.children).selectAll("tspan")
+        .attr("dx", 3)
+        .attr("y", 13);
+
+    node.filter(d => !d.children).selectAll("tspan")
+        .attr("x", 3)
+        .attr("y", (d, i, nodes) => `${i === 0 ? 1.1 : 2.3}em`);
+}
+
+
+function mimeBarSize(data, svg) {
+
+    const margin = {
+        top: 50,
+        right: 0,
+        bottom: 10,
+        left: Math.max(
+            d3.max(data.sort((a, b) => b.count - a.count).slice(0, 15), d => d.mime.length) * 6,
+            d3.max(data.sort((a, b) => b.size - a.size).slice(0, 15), d => d.mime.length) * 6,
+        )
+    };
+
+    data.forEach(d => {
+        d.name = d.mime;
+        d.value = Number(d.size);
+    });
+    data = data.sort((a, b) => b.value - a.value).slice(0, 15);
+
+    const width = 550;
+    const height = Math.ceil((data.length + 0.1) * barHeight) + margin.top + margin.bottom;
+
+    svg.selectAll("*").remove();
+    svg.attr("viewBox", [0, 0, width, height]);
+
+    const y = d3.scaleBand()
+        .domain(d3.range(data.length))
+        .rangeRound([margin.top, height - margin.bottom]);
+
+    const x = d3.scaleLinear()
+        .domain([0, d3.max(data, d => d.value)])
+        .range([margin.left, width - margin.right]);
+
+    svg.append("g")
+        .attr("fill-opacity", fillOpacity)
+        .selectAll("rect")
+        .data(data)
+        .join("rect")
+        .attr("fill", d => ordinalColor(d.name))
+        .attr("x", x(0))
+        .attr("y", (d, i) => y(i))
+        .attr("width", d => x(d.value) - x(0))
+        .attr("height", y.bandwidth())
+        .append("title")
+        .text(d => formatSI(d.value));
+
+    svg.append("g")
+        .attr("transform", `translate(0,${margin.top})`)
+        .call(d3.axisTop(x).ticks(width / 80, data.format).tickFormat(formatSI))
+        .call(g => g.select(".domain").remove());
+
+    svg.append("g")
+        .attr("transform", `translate(${margin.left},0)`)
+        .call(d3.axisLeft(y).tickFormat(i => data[i].name).tickSizeOuter(0));
+
+    svg.append("text")
+        .attr("x", (width / 2))
+        .attr("y", (margin.top / 2))
+        .attr("text-anchor", "middle")
+        .style("font-size", "16px")
+        .text("Size distribution by MIME type");
+}
+
+function mimeBarCount(data, svg) {
+
+    const margin = {
+        top: 50,
+        right: 0,
+        bottom: 10,
+        left: Math.max(
+            d3.max(data.sort((a, b) => b.count - a.count).slice(0, 15), d => d.mime.length) * 6,
+            d3.max(data.sort((a, b) => b.size - a.size).slice(0, 15), d => d.mime.length) * 6,
+        )
+    };
+
+    data.forEach(d => {
+        d.name = d.mime;
+        d.value = Number(d.count);
+    });
+
+    data = data.sort((a, b) => b.value - a.value).slice(0, 15);
+
+    const width = 550;
+    const height = Math.ceil((data.length + 0.1) * barHeight) + margin.top + margin.bottom;
+
+    svg.selectAll("*").remove();
+    svg.attr("viewBox", [0, 0, width, height]);
+
+    const y = d3.scaleBand()
+        .domain(d3.range(data.length))
+        .rangeRound([margin.top, height - margin.bottom]);
+
+    const x = d3.scaleLinear()
+        .domain([0, d3.max(data, d => d.value)])
+        .range([margin.left, width - margin.right]);
+
+    svg.append("g")
+        .attr("fill-opacity", fillOpacity)
+        .selectAll("rect")
+        .data(data)
+        .join("rect")
+        .attr("fill", d => ordinalColor(d.name))
+        .attr("x", x(0))
+        .attr("y", (d, i) => y(i))
+        .attr("width", d => x(d.value) - x(0))
+        .attr("height", y.bandwidth())
+        .append("title")
+        .text(d => d3.format(",")(d.value));
+
+    svg.append("g")
+        .attr("transform", `translate(0,${margin.top})`)
+        .call(d3.axisTop(x).ticks(width / 80, data.format).tickFormat(formatSI))
+        .call(g => g.select(".domain").remove());
+
+    svg.append("g")
+        .attr("transform", `translate(${margin.left},0)`)
+        .call(d3.axisLeft(y).tickFormat(i => data[i].name).tickSizeOuter(0));
+
+    svg.append("text")
+        .attr("x", (width / 2))
+        .attr("y", (margin.top / 2))
+        .attr("text-anchor", "middle")
+        .style("font-size", "16px")
+        .text("File count distribution by MIME type");
+}
+
+function dateHistogram(data, svg) {
+
+    let bins = data.map(d => {
+        return {
+            length: Number(d.count),
+            x0: Number(d.bucket),
+            x1: Number(d.bucket) + 2629800
+        }
+    });
+    bins.sort((a, b) => a.length - b.length);
+
+    const margin = {
+        top: 50,
+        right: 20,
+        bottom: 70,
+        left: 40
+    };
+
+    const thresh = d3.quantile(bins, 0.9, d => d.length);
+    bins = bins.filter(d => d.length > thresh);
+
+    const width = 550;
+    const height = 450;
+
+    svg.selectAll("*").remove();
+    svg.attr("viewBox", [0, 0, width, height]);
+
+    const y = d3.scaleLinear()
+        .domain([0, d3.max(bins, d => d.length)]).nice()
+        .range([height - margin.bottom, margin.top]);
+
+    const x = d3.scaleLinear()
+        .domain(d3.extent(bins, d => d.x0)).nice()
+        .range([margin.left, width - margin.right]);
+
+    svg.append("g")
+        .attr("fill", "steelblue")
+        .selectAll("rect")
+        .data(bins)
+        .join("rect")
+        .attr("x", d => x(d.x0) + 1)
+        .attr("width", d => Math.max(1, x(d.x1) - x(d.x0) - 1))
+        .attr("y", d => y(d.length))
+        .attr("height", d => y(0) - y(d.length))
+        .call(g => g
+            .append("title")
+            .text(d => d.length)
+        );
+
+    svg.append("g")
+        .attr("transform", `translate(0,${height - margin.bottom})`)
+        .call(
+            d3.axisBottom(x)
+                .ticks(width / 30)
+                .tickSizeOuter(0)
+                .tickFormat(t => d3.timeFormat("%Y-%m-%d")(d3.utcParse("%s")(t)))
+        )
+        .call(g => g
+            .selectAll("text")
+            .style("text-anchor", "end")
+            .attr("dx", "-.8em")
+            .attr("dy", ".15em")
+            .attr("transform", "rotate(-65)")
+        )
+        .call(g => g.append("text")
+            .attr("x", width - margin.right)
+            .attr("y", -4)
+            .attr("fill", "currentColor")
+            .attr("font-weight", "bold")
+            .attr("text-anchor", "end")
+            .text("mtime")
+        );
+
+    svg.append("g")
+        .attr("transform", `translate(${margin.left},0)`)
+        .call(
+            d3.axisLeft(y)
+                .ticks(height / 40)
+                .tickFormat(t => formatSI(t))
+        )
+        .call(g => g.select(".domain").remove())
+        .call(g => g.select(".tick:last-of-type text").clone()
+            .attr("x", 4)
+            .attr("text-anchor", "start")
+            .attr("font-weight", "bold")
+            .text("File count"));
+
+    svg.append("text")
+        .attr("x", (width / 2))
+        .attr("y", (margin.top / 2))
+        .attr("text-anchor", "middle")
+        .style("font-size", "16px")
+        .text("File modification time distribution");
+}
+
+function sizeHistogram(data, svg) {
+
+    let bins = data.map(d => {
+        return {
+            length: Number(d.count),
+            x0: Number(d.bucket),
+            x1: Number(d.bucket) + (5 * 1024 * 1024)
+        }
+    });
+    bins = bins.sort((a, b) => b.length - a.length).slice(0, 25);
+
+    const margin = {
+        top: 50,
+        right: 20,
+        bottom: 70,
+        left: 40
+    };
+
+    const width = 550;
+    const height = 450;
+
+    svg.selectAll("*").remove();
+    svg.attr("viewBox", [0, 0, width, height]);
+
+    const y = d3.scaleLinear()
+        .domain([0, d3.max(bins, d => d.length)])
+        .range([height - margin.bottom, margin.top]);
+
+    const x = d3.scaleLinear()
+        .domain(d3.extent(bins, d => d.x0)).nice()
+        .range([margin.left, width - margin.right]);
+
+    svg.append("g")
+        .attr("fill", "steelblue")
+        .selectAll("rect")
+        .data(bins)
+        .join("rect")
+        .attr("x", d => x(d.x0) + 1)
+        .attr("width", d => Math.max(1, x(d.x1) - x(d.x0) - 1))
+        .attr("y", d => y(d.length))
+        .attr("height", d => y(0) - y(d.length))
+        .call(g => g
+            .append("title")
+            .text(d => d.length)
+        );
+
+    svg.append("g")
+        .attr("transform", `translate(0,${height - margin.bottom})`)
+        .call(
+            d3.axisBottom(x)
+                .ticks(width / 30)
+                .tickSizeOuter(0)
+                .tickFormat(formatSI)
+        )
+        .call(g => g
+            .selectAll("text")
+            .style("text-anchor", "end")
+            .attr("dx", "-.8em")
+            .attr("dy", ".15em")
+            .attr("transform", "rotate(-65)")
+        )
+        .call(g => g.append("text")
+            .attr("x", width - margin.right)
+            .attr("y", -4)
+            .attr("fill", "currentColor")
+            .attr("font-weight", "bold")
+            .attr("text-anchor", "end")
+            .text("size (bytes)")
+        );
+
+    svg.append("g")
+        .attr("transform", `translate(${margin.left},0)`)
+        .call(
+            d3.axisLeft(y)
+                .ticks(height / 40)
+                .tickFormat(t => formatSI(t))
+        )
+        .call(g => g.select(".domain").remove())
+        .call(g => g.select(".tick:last-of-type text").clone()
+            .attr("x", 4)
+            .attr("text-anchor", "start")
+            .attr("font-weight", "bold")
+            .text("File count"));
+
+    svg.append("text")
+        .attr("x", (width / 2))
+        .attr("y", (margin.top / 2))
+        .attr("text-anchor", "middle")
+        .style("font-size", "16px")
+        .text("File size distribution");
+}
+
+
+function updateStats() {
+    width = SIZES[CONF.options.treemapSize][0];
+    height = SIZES[CONF.options.treemapSize][1];
+
+    const treemapSvg = d3.select("#treemap");
+    const mimeSvgSize = d3.select("#agg_mime_size");
+    const mimeSvgCount = d3.select("#agg_mime_count");
+    const dateHistogramSvg = d3.select("#date_histogram");
+    const sizeHistogramSvg = d3.select("#size_histogram");
+
+    const indexId = $("#indices").val();
+
+    d3.csv(`/s/${indexId}/1`).then(tabularData => {
+        tabularData.forEach(row => {
+            row.taxonomy = row.path.split("/");
+            row.size = Number(row.size);
+        });
+
+        if (CONF.options.treemapType === "cascaded") {
+            const data = burrow(tabularData, false);
+            cascadeTreemap(data, treemapSvg);
+        } else {
+            const data = burrow(tabularData.sort((a, b) => b.taxonomy.length - a.taxonomy.length), true);
+            flatTreemap(data, treemapSvg);
+        }
+    });
+
+    d3.csv(`/s/${indexId}/2`).then(tabularData => {
+        mimeBarSize(tabularData.slice(), mimeSvgSize);
+        mimeBarCount(tabularData.slice(), mimeSvgCount);
+    });
+
+    d3.csv(`/s/${indexId}/3`).then(tabularData => {
+        sizeHistogram(tabularData, sizeHistogramSvg);
+    });
+
+    d3.csv(`/s/${indexId}/4`).then(tabularData => {
+        dateHistogram(tabularData, dateHistogramSvg);
+    });
+
+    treemapSvg.selectAll("*").remove();
+    treemapSvg.attr("viewBox", [0, 0, width, height])
+        .attr("xmlns", "http://www.w3.org/2000/svg")
+        .attr("xmlns:xlink", "http://www.w3.org/1999/xlink")
+        .attr("version", "1.1")
+        .style("overflow", "visible")
+        .style("font", "10px sans-serif");
+
+}
+
+window.onload = function () {
+    CONF.load();
+
+    $.jsonPost("i").then(resp => {
+        const select = $("#indices");
+
+        const urlIndices = (new URLSearchParams(location.search)).get("i");
+        resp["indices"].forEach(idx => {
+            indexMap[idx.id] = idx.name;
+            select.append($("<option>")
+                .attr("value", idx.id)
+                .append(idx.name));
+
+            if (urlIndices && urlIndices.split(",").indexOf(idx.name) !== -1) {
+                select.select(idx.name);
+            }
+        });
+
+        updateStats();
+    });
+};
+
+function fullScreen(selector) {
+    const card = document.getElementById(selector);
+    card.classList.toggle("full-screen");
+}
+
+function exportTreemap() {
+    domtoimage.toBlob(document.getElementById("treemap"), {width: width, height: height})
+        .then(function (blob) {
+            let a = document.createElement("a");
+            let url = URL.createObjectURL(blob);
+
+            a.href = url;
+            a.download = `${indexMap[$("#indices").val()]}_treemap.png`;
+            document.body.appendChild(a);
+            a.click();
+            setTimeout(function() {
+                document.body.removeChild(a);
+                window.URL.revokeObjectURL(url);
+            }, 0);
+        });
+}
+</script>
+</body>
+</html>

src/stats.c

@@ -0,0 +1,353 @@
+#include "sist.h"
+#include "io/serialize.h"
+#include "ctx.h"
+
+#include <glib.h>
+
+static GHashTable *FlatTree;
+static GHashTable *BufferTable;
+
+static GHashTable *AggMime;
+static GHashTable *AggSize;
+static GHashTable *AggDate;
+
+#define SIZE_BUCKET (long)(5 * 1024 * 1024)
+#define DATE_BUCKET (long)(2629800)
+
+static long TotalSize = 0;
+static long DocumentCount = 0;
+
+typedef struct {
+    long size;
+    long count;
+} agg_t;
+
+void fill_tables(cJSON *document, UNUSED(const char uuid_str[UUID_STR_LEN])) {
+
+    if (cJSON_GetObjectItem(document, "parent") != NULL) {
+        return;
+    }
+
+    const char *json_path = cJSON_GetObjectItem(document, "path")->valuestring;
+    char *path = malloc(strlen(json_path) + 1);
+    strcpy(path, json_path);
+
+    const char *json_mime = cJSON_GetObjectItem(document, "mime")->valuestring;
+    char *mime;
+    if (json_mime == NULL) {
+        mime = NULL;
+    } else {
+        mime = malloc(strlen(json_mime) + 1);
+        strcpy(mime, json_mime);
+    }
+
+    long size = (long) cJSON_GetObjectItem(document, "size")->valuedouble;
+    int mtime = cJSON_GetObjectItem(document, "mtime")->valueint;
+
+    // treemap
+    void *existing_path = g_hash_table_lookup(FlatTree, path);
+    if (existing_path == NULL) {
+        g_hash_table_insert(FlatTree, path, (gpointer) size);
+    } else {
+        g_hash_table_replace(FlatTree, path, (gpointer) ((long) existing_path + size));
+    }
+
+    // mime agg
+    if (mime != NULL) {
+        agg_t *orig_agg = g_hash_table_lookup(AggMime, mime);
+        if (orig_agg == NULL) {
+            agg_t *agg = malloc(sizeof(agg_t));
+            agg->size = size;
+            agg->count = 1;
+            g_hash_table_insert(AggMime, mime, agg);
+        } else {
+            orig_agg->size += size;
+            orig_agg->count += 1;
+            free(mime);
+        }
+    }
+
+    // size agg
+    long size_bucket = size - (size % SIZE_BUCKET);
+    agg_t *orig_agg = g_hash_table_lookup(AggSize, (gpointer) size_bucket);
+    if (orig_agg == NULL) {
+        agg_t *agg = malloc(sizeof(agg_t));
+        agg->size = size;
+        agg->count = 1;
+        g_hash_table_insert(AggSize, (gpointer) size_bucket, agg);
+    } else {
+        orig_agg->count += 1;
+        orig_agg->size += size;
+    }
+
+    // date agg
+    long date_bucket = mtime - (mtime % DATE_BUCKET);
+    orig_agg = g_hash_table_lookup(AggDate, (gpointer) date_bucket);
+    if (orig_agg == NULL) {
+        agg_t *agg = malloc(sizeof(agg_t));
+        agg->size = size;
+        agg->count = 1;
+        g_hash_table_insert(AggDate, (gpointer) date_bucket, agg);
+    } else {
+        orig_agg->count += 1;
+        orig_agg->size += size;
+    }
+
+    TotalSize += size;
+    DocumentCount += 1;
+}
+
+void read_index_into_tables(index_t *index) {
+    DIR *dir = opendir(index->path);
+    struct dirent *de;
+    while ((de = readdir(dir)) != NULL) {
+        if (strncmp(de->d_name, "_index_", sizeof("_index_") - 1) == 0) {
+            char file_path[PATH_MAX];
+            snprintf(file_path, PATH_MAX, "%s/%s", index->path, de->d_name);
+            read_index(file_path, index->desc.uuid, index->desc.type, fill_tables);
+        }
+    }
+    closedir(dir);
+}
+
+static size_t rfind(const char *str, int c) {
+    for (int i = (int)strlen(str); i >= 0; i--) {
+        if (str[i] == c) {
+            return i;
+        }
+    }
+    return -1;
+}
+
+int merge_up(double thresh) {
+    long min_size = (long) (thresh * (double) TotalSize);
+
+    int count = 0;
+    GHashTableIter iter;
+    g_hash_table_iter_init(&iter, FlatTree);
+
+    void *key;
+    void *value;
+
+    while (g_hash_table_iter_next(&iter, &key, &value)) {
+        long size = (long) value;
+
+        if (size < min_size) {
+            int stop = rfind(key, '/');
+            if (stop == -1) {
+                stop = 0;
+            }
+            char *parent = malloc(stop + 1);
+            strncpy(parent, key, stop);
+            *(parent + stop) = '\0';
+
+            void *existing_parent = g_hash_table_lookup(FlatTree, parent);
+            if (existing_parent == NULL) {
+                void *existing_parent2_key;
+                void *existing_parent2_val;
+                int found = g_hash_table_lookup_extended(BufferTable, parent, &existing_parent2_key,
+                                                         &existing_parent2_val);
+                if (!found) {
+                    g_hash_table_insert(BufferTable, parent, value);
+                } else {
+                    g_hash_table_replace(BufferTable, parent, (gpointer) ((long) existing_parent2_val + size));
+                    free(existing_parent2_key);
+                }
+            } else {
+                g_hash_table_replace(FlatTree, parent, (gpointer) ((long) existing_parent + size));
+            }
+
+            g_hash_table_iter_remove(&iter);
+
+            count += 1;
+        }
+    }
+
+    g_hash_table_iter_init(&iter, BufferTable);
+    while (g_hash_table_iter_next(&iter, &key, &value)) {
+        g_hash_table_insert(FlatTree, key, value);
+        g_hash_table_iter_remove(&iter);
+    }
+
+    int size = g_hash_table_size(FlatTree);
+
+    LOG_DEBUGF("stats.c", "Merge up iteration (%d merged, %d in tree)", count, size)
+    return count;
+}
+
+/**
+ * Assumes out is at at least PATH_MAX *4
+ */
+void csv_escape(char *dst, const char *str) {
+
+    const char *ptr = str;
+    char *out = dst;
+
+    if (rfind(str, ',') == -1 && rfind(str, '"') == -1) {
+        strcpy(dst, str);
+        return;
+    }
+
+    *out++ = '"';
+    char c;
+    while ((c = *ptr++) != 0) {
+        if (c == '"') {
+            *out++ = '"';
+            *out++ = '"';
+        } else {
+            *out++ = c;
+        }
+    }
+    *out++ = '"';
+    *out = '\0';
+}
+
+int open_or_exit(const char *path) {
+    int fd = open(path, O_CREAT | O_WRONLY, S_IRUSR | S_IWUSR);
+    if (fd < 0) {
+        LOG_FATALF("stats.c", "Error while creating file: %s [%d]\n", strerror(errno), errno)
+    }
+    return fd;
+}
+
+#define TREEMAP_CSV_HEADER "path,size"
+#define MIME_AGG_CSV_HEADER "mime,size,count"
+#define SIZE_AGG_CSV_HEADER "bucket,size,count"
+#define DATE_AGG_CSV_HEADER "bucket,size,count"
+
+void write_treemap_csv(double thresh, const char *out_path) {
+
+    void *key;
+    void *value;
+
+    long min_size = (long) (thresh * (double) TotalSize);
+
+    int fd = open_or_exit(out_path);
+    int ret = write(fd, TREEMAP_CSV_HEADER, sizeof(TREEMAP_CSV_HEADER) - 1);
+    if (ret == -1) {
+        LOG_FATALF("stats.c", "Write error: %s", strerror(errno))
+    }
+
+    GHashTableIter iter;
+    g_hash_table_iter_init(&iter, FlatTree);
+    while (g_hash_table_iter_next(&iter, &key, &value)) {
+        long size = (long) value;
+
+        if (size >= min_size) {
+            char path_buf[PATH_MAX * 4];
+            char buf[PATH_MAX * 4 + 16];
+
+            csv_escape(path_buf, key);
+            size_t written = sprintf(buf, "\n%s,%ld", path_buf, (long) value);
+            ret = write(fd, buf, written);
+            if (ret == -1) {
+                LOG_FATALF("stats.c", "Write error: %s", strerror(errno))
+            }
+        }
+    }
+    close(fd);
+}
+
+void write_agg_csv_str(const char *out_path, const char *header, GHashTable *table) {
+    void *key;
+    void *value;
+    char buf[4096];
+
+    int fd = open_or_exit(out_path);
+    int ret = write(fd, header, strlen(header));
+    if (ret == -1) {
+        LOG_FATALF("stats.c", "Write error: %s", strerror(errno))
+    }
+
+    GHashTableIter iter;
+    g_hash_table_iter_init(&iter, table);
+    while (g_hash_table_iter_next(&iter, &key, &value)) {
+        agg_t *agg = value;
+
+        size_t written = sprintf(buf, "\n%s,%ld,%ld", (const char*)key, agg->size, agg->count);
+        ret = write(fd, buf, written);
+        if (ret == -1) {
+            LOG_FATALF("stats.c", "Write error: %s", strerror(errno))
+        }
+    }
+
+    close(fd);
+}
+
+void write_agg_csv_long(const char *out_path, const char *header, GHashTable *table) {
+    void *key;
+    void *value;
+    char buf[4096];
+
+    int fd = open_or_exit(out_path);
+    int ret = write(fd, header, strlen(header));
+    if (ret == -1) {
+        LOG_FATALF("stats.c", "Write error: %s", strerror(errno))
+    }
+
+    GHashTableIter iter;
+    g_hash_table_iter_init(&iter, table);
+    while (g_hash_table_iter_next(&iter, &key, &value)) {
+        agg_t *agg = value;
+        size_t written = sprintf(buf, "\n%ld,%ld,%ld", (long)key, agg->size, agg->count);
+        ret = write(fd, buf, written);
+        if (ret == -1) {
+            LOG_FATALF("stats.c", "Write error: %s", strerror(errno))
+        }
+    }
+
+    close(fd);
+}
+
+int generate_stats(index_t *index, const double threshold, const char *out_prefix) {
+
+    FlatTree = g_hash_table_new_full(g_str_hash, g_str_equal, free, NULL);
+    BufferTable = g_hash_table_new(g_str_hash, g_str_equal);
+
+    AggMime = g_hash_table_new_full(g_str_hash, g_str_equal, free, free);
+    AggSize = g_hash_table_new_full(g_direct_hash, g_direct_equal, NULL, free);
+    AggDate = g_hash_table_new_full(g_direct_hash, g_direct_equal, NULL, free);
+
+    LOG_INFO("stats.c", "Generating stats...")
+
+    read_index_into_tables(index);
+
+    LOG_DEBUG("stats.c", "Read index into tables")
+    LOG_DEBUGF("stats.c", "Total size is %ld", TotalSize)
+    LOG_DEBUGF("stats.c", "Document count is %ld", DocumentCount)
+    LOG_DEBUGF("stats.c", "Merging small directories upwards with a threshold of %f%%", threshold * 100)
+
+    while (merge_up(threshold) > 100) {}
+
+    char tmp[PATH_MAX];
+
+    strncpy(tmp, out_prefix, sizeof(tmp));
+    strcat(tmp, "treemap.csv");
+    write_treemap_csv(threshold, tmp);
+
+    strncpy(tmp, out_prefix, sizeof(tmp));
+    strcat(tmp, "mime_agg.csv");
+    write_agg_csv_str(tmp, MIME_AGG_CSV_HEADER, AggMime);
+
+    strncpy(tmp, out_prefix, sizeof(tmp));
+    strcat(tmp, "size_agg.csv");
+    write_agg_csv_long(tmp, SIZE_AGG_CSV_HEADER, AggSize);
+
+    strncpy(tmp, out_prefix, sizeof(tmp));
+    strcat(tmp, "date_agg.csv");
+    write_agg_csv_long(tmp, DATE_AGG_CSV_HEADER, AggDate);
+
+    g_hash_table_remove_all(FlatTree);
+    g_hash_table_destroy(FlatTree);
+    g_hash_table_destroy(BufferTable);
+
+    g_hash_table_remove_all(AggMime);
+    g_hash_table_destroy(AggMime);
+    g_hash_table_remove_all(AggSize);
+    g_hash_table_destroy(AggSize);
+    g_hash_table_remove_all(AggDate);
+    g_hash_table_destroy(AggDate);
+
+    return 0;
+}
+

src/stats.h

@@ -0,0 +1,6 @@
+#ifndef SIST2_STATS_H
+#define SIST2_STATS_H
+
+int generate_stats(index_t *index, double threshold, const char* out_prefix);
+
+#endif

src/tpool.c

@@ -1,5 +1,9 @@
 #include "tpool.h"
 #include "ctx.h"
+#include "sist.h"
+#include <pthread.h>
+
+#define MAX_QUEUE_SIZE 10000
 
 typedef void (*thread_func_t)(void *arg);
 
@@ -24,7 +28,9 @@ typedef struct tpool {
     int work_cnt;
     int done_cnt;
 
+    int free_arg;
     int stop;
+
     void (*cleanup_func)();
 } tpool_t;
 
@@ -76,6 +82,10 @@ int tpool_add_work(tpool_t *pool, thread_func_t func, void *arg) {
         return 0;
     }
 
+    while ((pool->work_cnt - pool->done_cnt) >= MAX_QUEUE_SIZE) {
+        usleep(100000);
+    }
+
     pthread_mutex_lock(&(pool->work_mutex));
     if (pool->work_head == NULL) {
         pool->work_head = work;
@@ -100,7 +110,7 @@ static void *tpool_worker(void *arg) {
     tpool_t *pool = arg;
 
     while (1) {
-        pthread_mutex_lock(&(pool->work_mutex));
+        pthread_mutex_lock(&pool->work_mutex);
         if (pool->stop) {
             break;
         }
@@ -113,14 +123,23 @@ static void *tpool_worker(void *arg) {
         pthread_mutex_unlock(&(pool->work_mutex));
 
         if (work != NULL) {
+            if (pool->stop) {
+                break;
+            }
+
             work->func(work->arg);
+            if (pool->free_arg) {
+                free(work->arg);
+            }
             free(work);
         }
 
         pthread_mutex_lock(&(pool->work_mutex));
-        pool->done_cnt++;
+        if (work != NULL) {
+            pool->done_cnt++;
+        }
 
-        progress_bar_print((double)pool->done_cnt / pool->work_cnt, ScanCtx.stat_tn_size, ScanCtx.stat_index_size);
+        progress_bar_print((double) pool->done_cnt / pool->work_cnt, ScanCtx.stat_tn_size, ScanCtx.stat_index_size);
 
         if (pool->work_head == NULL) {
             pthread_cond_signal(&(pool->working_cond));
@@ -128,7 +147,10 @@ static void *tpool_worker(void *arg) {
         pthread_mutex_unlock(&(pool->work_mutex));
     }
 
-    pool->cleanup_func();
+    if (pool->cleanup_func != NULL) {
+        LOG_INFO("tpool.c", "Executing cleanup function")
+        pool->cleanup_func();
+    }
 
     pthread_cond_signal(&(pool->working_cond));
     pthread_mutex_unlock(&(pool->work_mutex));
@@ -136,17 +158,24 @@ static void *tpool_worker(void *arg) {
 }
 
 void tpool_wait(tpool_t *pool) {
+    LOG_INFO("tpool.c", "Waiting for worker threads to finish")
     pthread_mutex_lock(&(pool->work_mutex));
     while (1) {
         if (pool->done_cnt < pool->work_cnt) {
             pthread_cond_wait(&(pool->working_cond), &(pool->work_mutex));
         } else {
-            pool->stop = 1;
-            break;
+            usleep(500000);
+            if (pool->done_cnt == pool->work_cnt) {
+                pool->stop = 1;
+                usleep(1000000);
+                break;
+            }
         }
-        progress_bar_print(100.0, ScanCtx.stat_tn_size, ScanCtx.stat_index_size);
     }
+    progress_bar_print(1.0, ScanCtx.stat_tn_size, ScanCtx.stat_index_size);
     pthread_mutex_unlock(&(pool->work_mutex));
+
+    LOG_INFO("tpool.c", "Worker threads finished")
 }
 
 void tpool_destroy(tpool_t *pool) {
@@ -154,6 +183,8 @@ void tpool_destroy(tpool_t *pool) {
         return;
     }
 
+    LOG_INFO("tpool.c", "Destroying thread pool")
+
     pthread_mutex_lock(&(pool->work_mutex));
     tpool_work_t *work = pool->work_head;
     while (work != NULL) {
@@ -167,9 +198,14 @@ void tpool_destroy(tpool_t *pool) {
 
     for (size_t i = 0; i < pool->thread_cnt; i++) {
         pthread_t thread = pool->threads[i];
-        pthread_cancel(thread);
+        if (thread != 0) {
+            void *_;
+            pthread_join(thread, &_);
+        }
     }
 
+    LOG_INFO("tpool.c", "Final cleanup")
+
     pthread_mutex_destroy(&(pool->work_mutex));
     pthread_cond_destroy(&(pool->has_work_cond));
     pthread_cond_destroy(&(pool->working_cond));
@@ -182,15 +218,16 @@ void tpool_destroy(tpool_t *pool) {
  * Create a thread pool
  * @param thread_cnt Worker threads count
  */
-tpool_t *tpool_create(size_t thread_cnt, void cleanup_func()) {
+tpool_t *tpool_create(size_t thread_cnt, void cleanup_func(), int free_arg) {
 
     tpool_t *pool = malloc(sizeof(tpool_t));
     pool->thread_cnt = thread_cnt;
-    pool->work_cnt =0;
-    pool->done_cnt =0;
+    pool->work_cnt = 0;
+    pool->done_cnt = 0;
     pool->stop = 0;
+    pool->free_arg = free_arg;
     pool->cleanup_func = cleanup_func;
-    pool->threads = malloc(sizeof(pthread_t) * thread_cnt);
+    pool->threads = calloc(sizeof(pthread_t), thread_cnt);
 
     pthread_mutex_init(&(pool->work_mutex), NULL);
 
@@ -200,11 +237,14 @@ tpool_t *tpool_create(size_t thread_cnt, void cleanup_func()) {
     pool->work_head = NULL;
     pool->work_tail = NULL;
 
-    for (size_t i = 0; i < thread_cnt; i++) {
-        pthread_t thread = pool->threads[i];
-        pthread_create(&thread, NULL, tpool_worker, pool);
-        pthread_detach(thread);
-    }
-
     return pool;
 }
+
+void tpool_start(tpool_t *pool) {
+
+    LOG_INFOF("tpool.c", "Starting thread pool with %d threads", pool->thread_cnt)
+
+    for (size_t i = 0; i < pool->thread_cnt; i++) {
+        pthread_create(&pool->threads[i], NULL, tpool_worker, pool);
+    }
+}

src/tpool.h

@@ -8,7 +8,8 @@ typedef struct tpool tpool_t;
 
 typedef void (*thread_func_t)(void *arg);
 
-tpool_t *tpool_create(size_t num, void (*cleanup_func)());
+tpool_t *tpool_create(size_t num, void (*cleanup_func)(), int free_arg);
+void tpool_start(tpool_t *pool);
 void tpool_destroy(tpool_t *tm);
 
 int tpool_add_work(tpool_t *pool, thread_func_t func, void *arg);

src/types.h

@@ -1,87 +1,26 @@
 #ifndef SIST2_TYPES_H
 #define SIST2_TYPES_H
 
-
-#define META_INT_MASK 0xF0
-#define META_STR_MASK 0xE0
-#define META_LONG_MASK 0xD0
-#define IS_META_INT(key) (key & META_INT_MASK) == META_INT_MASK
-#define IS_META_LONG(key) (key & META_LONG_MASK) == META_LONG_MASK
-#define IS_META_STR(meta) (meta->key & META_STR_MASK) == META_STR_MASK
-
-// This is written to file as a 8bit char!
-enum metakey {
-    MetaContent = 1 | META_STR_MASK,
-    MetaWidth = 2 | META_INT_MASK,
-    MetaHeight = 3 | META_INT_MASK,
-    MetaMediaDuration = 4 | META_LONG_MASK,
-    MetaMediaAudioCodec = 5 | META_INT_MASK,
-    MetaMediaVideoCodec = 6 | META_INT_MASK,
-    MetaMediaBitrate = 7 | META_LONG_MASK,
-    MetaArtist = 8 | META_STR_MASK,
-    MetaAlbum = 9 | META_STR_MASK,
-    MetaAlbumArtist = 10 | META_STR_MASK,
-    MetaGenre = 11 | META_STR_MASK,
-    MetaTitle = 12 | META_STR_MASK,
-    MetaFontName = 13 | META_STR_MASK,
-};
+#define INDEX_TYPE_BIN "binary"
+#define INDEX_TYPE_JSON "json"
+#define INDEX_VERSION_EXTERNAL "_external_v1"
 
 typedef struct index_descriptor {
     char uuid[UUID_STR_LEN];
-    char version[6];
+    char version[64];
     long timestamp;
     char root[PATH_MAX];
-    char rewrite_url[8196];
+    char rewrite_url[8192];
     short root_len;
     char name[1024];
+    char type[64];
 } index_descriptor_t;
 
 typedef struct index_t {
     struct index_descriptor desc;
     struct store_t *store;
+    struct store_t *tag_store;
     char path[PATH_MAX];
 } index_t;
 
-typedef struct meta_line {
-    struct meta_line *next;
-    enum metakey key;
-    union {
-        unsigned long longval;
-        int intval;
-        char strval[0];
-    };
-} meta_line_t;
-
-
-typedef struct document {
-    unsigned char uuid[16];
-    unsigned long ino;
-    unsigned long size;
-    unsigned int mime;
-    int mtime;
-    short base;
-    short ext;
-    meta_line_t *meta_head;
-    meta_line_t *meta_tail;
-    char *filepath;
-} document_t;
-
-typedef struct parse_job_t {
-    int base;
-    int ext;
-    struct stat info;
-    char filepath[1];
-} parse_job_t;
-
-
-#define APPEND_META(doc, meta) \
-    meta->next = NULL;\
-    if (doc->meta_head == NULL) {\
-        doc->meta_head = meta;\
-        doc->meta_tail = doc->meta_head;\
-    } else {\
-        doc->meta_tail->next = meta;\
-        doc->meta_tail = meta;\
-    }
-
 #endif

src/util.c

@@ -1,31 +1,84 @@
-#define _GNU_SOURCE
 #include "util.h"
+#include "src/ctx.h"
 
+#include <wordexp.h>
+#include <glib.h>
 
 #define PBSTR "========================================"
 #define PBWIDTH 40
 
-char *abspath(const char *path) {
-    wordexp_t w;
-    wordexp(path, &w, 0);
+dyn_buffer_t url_escape(char *str) {
 
-    char *abs = canonicalize_file_name(w.we_wordv[0]);
+    dyn_buffer_t text = dyn_buffer_create();
+
+    char *ptr = str;
+    while (*ptr) {
+        if (*ptr == '#') {
+            dyn_buffer_write(&text, "%23", 3);
+            ptr++;
+        }
+
+        dyn_buffer_write_char(&text, *ptr++);
+    }
+    dyn_buffer_write_char(&text, '\0');
+
+    return text;
+}
+
+char *abspath(const char *path) {
+
+    char *expanded = expandpath(path);
+
+    char *abs = realpath(expanded, NULL);
+    free(expanded);
     if (abs == NULL) {
         return NULL;
     }
-    abs = realloc(abs, strlen(abs) + 2);
-    strcat(abs, "/");
+    if (strlen(abs) > 1) {
+        abs = realloc(abs, strlen(abs) + 2);
+        strcat(abs, "/");
+    }
 
-    wordfree(&w);
     return abs;
 }
 
-char *expandpath(const char *path) {
-    wordexp_t w;
-    wordexp(path, &w, 0);
+void shell_escape(char *dst, const char *src) {
+    const char *ptr = src;
+    char *out = dst;
+    while ((*ptr)) {
+        char c = *ptr++;
 
-    char * expanded = malloc(strlen(w.we_wordv[0]) + 2);
-    strcpy(expanded, w.we_wordv[0]);
+        if (c == '&' || c == '\n' || c == '|' || c == ';' || c == '<' ||
+            c == '>' || c == '(' || c == ')' || c == '{' || c == '}') {
+            *out++ = '\\';
+        }
+        *out++ = c;
+    }
+    *out = 0;
+}
+
+char *expandpath(const char *path) {
+    char tmp[PATH_MAX * 2];
+
+    shell_escape(tmp, path);
+
+    wordexp_t w;
+    wordexp(tmp, &w, 0);
+
+    if (w.we_wordv == NULL) {
+        return NULL;
+    }
+
+    *tmp = '\0';
+    for (int i = 0; i < w.we_wordc; i++) {
+        strcat(tmp, w.we_wordv[i]);
+        if (i != w.we_wordc - 1) {
+            strcat(tmp, " ");
+        }
+    }
+
+    char *expanded = malloc(strlen(tmp) + 2);
+    strcpy(expanded, tmp);
     strcat(expanded, "/");
 
     wordfree(&w);
@@ -76,4 +129,117 @@ GHashTable *incremental_get_table() {
     return file_table;
 }
 
+const char *find_file_in_paths(const char *paths[], const char *filename) {
 
+    for (int i = 0; paths[i] != NULL; i++) {
+
+        char *apath = abspath(paths[i]);
+        if (apath == NULL) {
+            continue;
+        }
+
+        char path[PATH_MAX];
+        snprintf(path, sizeof(path), "%s%s", apath, filename);
+
+        LOG_DEBUGF("util.c", "Looking for '%s' in folder '%s'", filename, apath)
+        free(apath);
+
+        struct stat info;
+        int ret = stat(path, &info);
+        if (ret != -1) {
+            return paths[i];
+        }
+    }
+
+    return NULL;
+}
+
+#define ESCAPE_CHAR ']'
+
+void str_escape(char *dst, const char *str) {
+    const size_t len = strlen(str);
+
+    char buf[16384];
+    memset(buf + len, 0, 8);
+    strcpy(buf, str);
+
+    char *cur = dst;
+    const char *ptr = buf;
+    const char *oldPtr = ptr;
+
+    utf8_int32_t c;
+    char tmp[16];
+
+    do {
+        ptr = (char *) utf8codepoint(ptr, &c);
+        *(int *) tmp = 0x00000000;
+        size_t code_len = (ptr - oldPtr);
+        memcpy(tmp, oldPtr, code_len);
+        oldPtr = ptr;
+
+        if (!utf8_validchr2(tmp)) {
+            for (int i = 0; i < code_len; i++) {
+                if (tmp[i] == 0) {
+                    break;
+                }
+
+                cur += sprintf(cur, "%c%02X", ESCAPE_CHAR, (unsigned char) tmp[i]);
+            }
+            continue;
+        }
+
+        if (c == ESCAPE_CHAR) {
+            *cur++ = ESCAPE_CHAR;
+            *cur++ = ESCAPE_CHAR;
+            continue;
+        }
+
+        if (((utf8_int32_t) 0xffffff80 & c) == 0) {
+            *(cur++) = (char) c;
+        } else if (((utf8_int32_t) 0xfffff800 & c) == 0) {
+            *(cur++) = 0xc0 | (char) (c >> 6);
+            *(cur++) = 0x80 | (char) (c & 0x3f);
+        } else if (((utf8_int32_t) 0xffff0000 & c) == 0) {
+            *(cur++) = 0xe0 | (char) (c >> 12);
+            *(cur++) = 0x80 | (char) ((c >> 6) & 0x3f);
+            *(cur++) = 0x80 | (char) (c & 0x3f);
+        } else {
+            *(cur++) = 0xf0 | (char) (c >> 18);
+            *(cur++) = 0x80 | (char) ((c >> 12) & 0x3f);
+            *(cur++) = 0x80 | (char) ((c >> 6) & 0x3f);
+            *(cur++) = 0x80 | (char) (c & 0x3f);
+        }
+
+    } while (*ptr != '\0');
+
+    *cur = '\0';
+}
+
+void str_unescape(char *dst, const char *str) {
+    char *cur = dst;
+    const char *ptr = str;
+
+    char tmp[3];
+    tmp[2] = '\0';
+
+    while (*ptr != 0) {
+        char c = *ptr++;
+
+        if (c == ESCAPE_CHAR) {
+            char next = *ptr;
+
+            if (next == ESCAPE_CHAR) {
+                *cur++ = (char) c;
+                ptr += 1;
+            } else {
+                tmp[0] = *(ptr);
+                tmp[1] = *(ptr + 1);
+                *cur++ = (char) strtol(tmp, NULL, 16);
+                ptr += 2;
+            }
+        } else {
+            *cur++ = c;
+        }
+    }
+    *cur = '\0';
+}

src/util.h

@@ -2,164 +2,32 @@
 #define SIST2_UTIL_H
 
 #include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
 
-#define TEXT_BUF_FULL -1
-#define INITIAL_BUF_SIZE 1024 * 16
-#define SHOULD_IGNORE_CHAR(c) c < '0' || c > 'z'
+#include <glib.h>
 
-typedef struct dyn_buffer {
-    char *buf;
-    size_t cur;
-    size_t size;
-} dyn_buffer_t;
+#include "third-party/utf8.h/utf8.h"
+#include "libscan/scan.h"
 
-#include "sist.h"
 
-typedef struct text_buffer {
-    size_t max_size;
-    int last_char_was_whitespace;
-    dyn_buffer_t dyn_buffer;
-} text_buffer_t;
+char *abspath(const char *path);
 
-__always_inline
-dyn_buffer_t dyn_buffer_create() {
-    dyn_buffer_t buf;
-
-    buf.size = INITIAL_BUF_SIZE;
-    buf.cur = 0;
-    buf.buf = malloc(INITIAL_BUF_SIZE);
-
-    return buf;
-}
-
-__always_inline
-void grow_buffer(dyn_buffer_t *buf, size_t size) {
-    if (buf->cur + size > buf->size) {
-        do {
-            buf->size *= 2;
-        } while (buf->cur + size > buf->size);
-
-        buf->buf = realloc(buf->buf, buf->size);
-    }
-}
-
-__always_inline
-void grow_buffer_small(dyn_buffer_t *buf) {
-    if (buf->cur + sizeof(long) > buf->size) {
-        buf->size *= 2;
-        buf->buf = realloc(buf->buf, buf->size);
-    }
-}
-
-__always_inline
-void dyn_buffer_write(dyn_buffer_t *buf, void *data, size_t size) {
-    grow_buffer(buf, size);
-
-    memcpy(buf->buf + buf->cur, data, size);
-    buf->cur += size;
-}
-
-__always_inline
-void dyn_buffer_write_char(dyn_buffer_t *buf, char c) {
-    grow_buffer_small(buf);
-
-    *(buf->buf + buf->cur) = c;
-    buf->cur += sizeof(c);
-}
-
-__always_inline
-void dyn_buffer_write_str(dyn_buffer_t *buf, char *str) {
-    dyn_buffer_write(buf, str, strlen(str));
-    dyn_buffer_write_char(buf, '\0');
-}
-
-__always_inline
-void dyn_buffer_write_int(dyn_buffer_t *buf, int d) {
-    grow_buffer_small(buf);
-
-    *(int *) (buf->buf + buf->cur) = d;
-    buf->cur += sizeof(int);
-}
-
-__always_inline
-void dyn_buffer_write_short(dyn_buffer_t *buf, short s) {
-    grow_buffer_small(buf);
-
-    *(short *) (buf->buf + buf->cur) = s;
-    buf->cur += sizeof(short);
-}
-
-__always_inline
-void dyn_buffer_write_long(dyn_buffer_t *buf, unsigned long l) {
-    grow_buffer_small(buf);
-
-    *(unsigned long *) (buf->buf + buf->cur) = l;
-    buf->cur += sizeof(unsigned long);
-}
-
-__always_inline
-void dyn_buffer_destroy(dyn_buffer_t *buf) {
-    free(buf->buf);
-}
-
-__always_inline
-void text_buffer_destroy(text_buffer_t *buf) {
-    dyn_buffer_destroy(&buf->dyn_buffer);
-}
-
-__always_inline
-text_buffer_t text_buffer_create(int max_size) {
-    text_buffer_t text_buf;
-
-    text_buf.dyn_buffer = dyn_buffer_create();
-    text_buf.max_size = max_size;
-    text_buf.last_char_was_whitespace = FALSE;
-
-    return text_buf;
-}
-
-__always_inline
-void text_buffer_terminate_string(text_buffer_t *buf) {
-    dyn_buffer_write_char(&buf->dyn_buffer, '\0');
-}
-
-__always_inline
-int text_buffer_append_char(text_buffer_t *buf, int c) {
-
-    if (SHOULD_IGNORE_CHAR(c)) {
-        if (!buf->last_char_was_whitespace) {
-            dyn_buffer_write_char(&buf->dyn_buffer, ' ');
-            buf->last_char_was_whitespace = TRUE;
-
-            if (buf->dyn_buffer.cur >= buf->max_size) {
-                return TEXT_BUF_FULL;
-            }
-        }
-    } else {
-        buf->last_char_was_whitespace = FALSE;
-        dyn_buffer_write_char(&buf->dyn_buffer, (char) c);
-
-        if (buf->dyn_buffer.cur >= buf->max_size) {
-            return TEXT_BUF_FULL;
-        }
-    }
-
-    return 0;
-}
-
-char *abspath(const char * path);
 char *expandpath(const char *path);
 
+dyn_buffer_t url_escape(char *str);
+
 void progress_bar_print(double percentage, size_t tn_size, size_t index_size);
 
+GHashTable *incremental_get_table();
 
 __always_inline
-void incremental_put(GHashTable *table, unsigned long inode_no, int mtime) {
+static void incremental_put(GHashTable *table, unsigned long inode_no, int mtime) {
     g_hash_table_insert(table, (gpointer) inode_no, GINT_TO_POINTER(mtime));
 }
 
 __always_inline
-int incremental_get(GHashTable *table, unsigned long inode_no) {
+static int incremental_get(GHashTable *table, unsigned long inode_no) {
     if (table != NULL) {
         return GPOINTER_TO_INT(g_hash_table_lookup(table, (gpointer) inode_no));
     } else {
@@ -168,12 +36,16 @@ int incremental_get(GHashTable *table, unsigned long inode_no) {
 }
 
 __always_inline
-int incremental_mark_file_for_copy(GHashTable *table, unsigned long inode_no) {
-    g_hash_table_insert(table, GINT_TO_POINTER(inode_no), GINT_TO_POINTER(1));
+static int incremental_mark_file_for_copy(GHashTable *table, unsigned long inode_no) {
+    return g_hash_table_insert(table, GINT_TO_POINTER(inode_no), GINT_TO_POINTER(1));
 }
 
-GHashTable *incremental_get_table();
+
+const char *find_file_in_paths(const char **paths, const char *filename);
 
 
+void str_escape(char *dst, const char *str);
+
+void str_unescape(char *dst, const char *str);
 
 #endif