Add support for auth0

Bump express from 4.17.1 to 4.18.2 in /sist2-vue

.dockerignore

@@ -0,0 +1,36 @@
+.idea/
+*.sist2
+docs/
+test_i/
+test_i_inc/
+Testing/
+.drone.yml
+**/cmake_install.cmake
+**/CMakeCache.txt
+**/CMakeFiles/
+.cmake
+LICENSE
+Makefile
+**/*.md
+**/*.cbp
+VERSION
+**/node_modules/
+.git/
+sist2-*-linux-debug
+sist2-*-linux
+sist2_debug
+sist2
+**/libscan-test-files
+**/scan_ub_test
+**/scan_a_test
+**/scan_test
+**/ext_ffmpeg
+**/ext_libmobi
+**/ext_libwpd
+**/core
+*.a
+tmp_scan/
+Dockerfile
+Dockerfile.arm64
+docker-compose.yml
+state.db

.drone.yml

@@ -0,0 +1,88 @@
+kind: pipeline
+type: docker
+name: amd64
+
+platform:
+  os: linux
+  arch: amd64
+
+steps:
+  - name: build
+    image: simon987/sist2-build
+    commands:
+      - ./scripts/build.sh
+  - name: scp files
+    image: appleboy/drone-scp
+    settings:
+      host:
+        from_secret: SSH_HOST
+      port:
+        from_secret: SSH_PORT
+      user:
+        from_secret: SSH_USER
+      key:
+        from_secret: SSH_KEY
+      target: /files/sist2/${DRONE_REPO_OWNER}_${DRONE_REPO_NAME}/${DRONE_BRANCH}_${DRONE_BUILD_NUMBER}_${DRONE_COMMIT}/
+      source:
+        - ./VERSION
+        - ./sist2-x64-linux
+        - ./sist2-x64-linux-debug
+  - name: docker
+    image: plugins/docker
+    settings:
+      username:
+        from_secret: DOCKER_USER
+      password:
+        from_secret: DOCKER_PASSWORD
+      repo: simon987/sist2
+      context: ./
+      dockerfile: ./Dockerfile
+      auto_tag: true
+      auto_tag_suffix: x64-linux
+      when:
+        event:
+          - tag
+
+---
+kind: pipeline
+type: docker
+name: arm64
+
+platform:
+  arch: arm64
+
+steps:
+  - name: build
+    image: simon987/sist2-build-arm64
+    commands:
+      - ./scripts/build_arm64.sh
+  - name: scp files
+    image: appleboy/drone-scp
+    settings:
+      host:
+        from_secret: SSH_HOST
+      port:
+        from_secret: SSH_PORT
+      user:
+        from_secret: SSH_USER
+      key:
+        from_secret: SSH_KEY
+      target: /files/sist2/${DRONE_REPO_OWNER}_${DRONE_REPO_NAME}/arm_${DRONE_BRANCH}_${DRONE_BUILD_NUMBER}_${DRONE_COMMIT}/
+      source:
+        - ./sist2-arm64-linux
+        - ./sist2-arm64-linux-debug
+  - name: docker
+    image: plugins/docker
+    settings:
+      username:
+        from_secret: DOCKER_USER
+      password:
+        from_secret: DOCKER_PASSWORD
+      repo: simon987/sist2
+      context: ./
+      dockerfile: ./Dockerfile.arm64
+      auto_tag: true
+      auto_tag_suffix: arm64-linux
+      when:
+        event:
+          - tag

.gitattributes

@@ -1,3 +1,3 @@
 CMakeModules/* linguist-vendored
-web/js/*.min.js linguist-vendored
-web/css/*.min.css linguist-vendored
+**/*_generated.c linguist-vendored
+**/*_generated.h linguist-vendored

.github/ISSUE_TEMPLATE/bug_report.md

@@ -0,0 +1,40 @@
+---
+name: "🐞 Bug Report"
+about: Submit a bug report 
+title: ''
+labels: bug
+assignees: ''
+
+---
+
+**Device Information (please complete the following information):**
+ - OS: `[e.g., Ubuntu 20.04, WSL2]`
+ - Deployment: `[Linux, Linux ARM64 or Docker]`
+ - Browser *(if relevant)*: `[e.g., chrome, safari]`
+ - SIST2 Version: `[e.g., v2.9.0]`
+ - Elasticsearch Version *(if relevant)* : ``
+
+**Command with arguments** 
+<!-- `ex: "scan ~/Documents -o ./i2 --threads 3 -q 1.0` -->
+
+**Describe the bug**
+<!-- A clear and concise description of what the bug is. -->
+
+**Steps To Reproduce**
+Please be specific!
+1. Go to '...'
+2. Click on '....'
+3. etc.
+
+**Expected behavior**
+<!-- A clear and concise description of what you expected to happen. -->
+
+**Actual Behavior**
+<!-- A clear and concise description of what actually happens. -->
+
+**Screenshots**
+<!-- If applicable, add screenshots to help explain your problem. -->
+
+**Additional context**
+<!-- Add any other context about the problem here. If applicable, please include why you think the bug is occurring and/or troubleshooting you have already performed. -->
+<!-- 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. -->

.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,5 @@
+blank_issues_enabled: false
+contact_links:
+  - name: SIST2 Documentation
+    url: https://github.com/simon987/sist2/blob/master/docs/USAGE.md
+    about: Check out the SIST2 documentation for answers to common questions

.github/ISSUE_TEMPLATE/feature_request.md

@@ -0,0 +1,18 @@
+---
+name: "🚀 Feature Request"
+about: Suggest an idea for SIST2
+title: ''
+assignees: ''
+
+---
+**Which SIST2 component is your Feature Request related to?**
+<!-- e.g., Scan, Index, or Web? -->
+
+**Is your feature request related to a problem? Please describe.**
+<!-- A clear and concise description of what the problem is. e.g., "I'm always frustrated when [...]" -->
+
+**What would you like to see happen?**
+<!-- A clear and concise description of what you want to happen. -->
+
+**Additional context**
+<!-- Add any other context or screenshots about the feature request here. -->

.github/ISSUE_TEMPLATE/issue-template.md

@@ -0,0 +1,18 @@
+---
+name: Issue template
+about: General
+title: ''
+labels: ''
+assignees: ''
+
+---
+
+sist2 version:
+
+Platform (Linux or Docker, x86-64 or arm64):
+
+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

@@ -1,6 +1,5 @@
 .idea
 thumbs
-test
 *.cbp
 CMakeCache.txt
 CMakeFiles
@@ -10,8 +9,27 @@ Makefile
 *.out
 LOG
 sist2*
-index.sist2/
-bundle.css
+!sist2-vue/
+!sist2-admin
+!sist2_admin
+!sist2.py
+*.sist2/
+bundle*.css
 bundle.js
 *.a
 vgcore.*
+build/
+third-party/argparse
+*.idx/
+VERSION
+git_hash.h
+Testing/
+test_i
+test_i_inc
+node_modules/
+.cmake/
+i_inc/
+state.db
+*.pyc
+!sist2-admin/frontend/dist
+*.js.map

.gitmodules

@@ -1,18 +1,12 @@
-[submodule "argparse"]
-	path = 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
+[submodule "third-party/argparse"]
+	path = third-party/argparse
+	url = https://github.com/simon987/argparse
+[submodule "third-party/libscan/third-party/utf8.h"]
+	path = third-party/libscan/third-party/utf8.h
+	url = https://github.com/sheredom/utf8.h
+[submodule "third-party/libscan/third-party/antiword"]
+	path = third-party/libscan/third-party/antiword
+	url = https://github.com/simon987/antiword
+[submodule "third-party/libscan/third-party/libmobi"]
+	path = third-party/libscan/third-party/libmobi
+	url = https://github.com/bfabiszewski/libmobi

CMakeLists.txt

@@ -1,127 +1,157 @@
 cmake_minimum_required(VERSION 3.7)
+
+project(sist2)
 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)
+option(SIST_FAST "Enable more optimisation flags" off)
+option(SIST_FAKE_STORE "Disable IO operations of LMDB stores for debugging purposes" 0)
 
-add_executable(
-        sist2
+add_compile_definitions(
+        "SIST_PLATFORM=${SIST_PLATFORM}"
+)
+
+if (SIST_DEBUG)
+    add_compile_definitions(
+            "SIST_DEBUG=${SIST_DEBUG}"
+    )
+endif()
+
+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
+        src/cli.c src/cli.h
+        src/stats.c src/stats.h src/ctx.c
+        src/parsing/sidecar.c src/parsing/sidecar.h
+
+        src/auth0/auth0_c_api.h src/auth0/auth0_c_api.cpp
 
         # argparse
-        argparse/argparse.h argparse/argparse.c
+        third-party/argparse/argparse.h third-party/argparse/argparse.c
+        )
+set_target_properties(sist2 PROPERTIES LINKER_LANGUAGE C)
 
-        # cJSON
-        cJSON/cJSON.h cJSON/cJSON.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(LibMagic REQUIRED)
-find_package(FFmpeg REQUIRED)
-find_package(OpenSSL REQUIRED)
-find_package(Freetype REQUIRED)
+pkg_search_module(GLIB REQUIRED glib-2.0)
 
-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/
+find_package(lmdb CONFIG REQUIRED)
+find_package(cJSON CONFIG REQUIRED)
+find_package(unofficial-mongoose CONFIG REQUIRED)
+find_package(CURL CONFIG REQUIRED)
+find_library(MAGIC_LIB
+        NAMES libmagic.so.1 magic
+        PATHS /usr/lib/x86_64-linux-gnu/ /usr/lib/aarch64-linux-gnu/
 )
 
-target_compile_options(sist2
-        PRIVATE
-        -O3
-#        -march=native
-        -fno-stack-protector
-        -fomit-frame-pointer
-        )
 
-TARGET_LINK_LIBRARIES(
+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}/
+        ${GLIB_INCLUDE_DIRS}
+)
+
+target_compile_options(
+        sist2
+        PRIVATE
+        -fPIC
+)
+
+if (SIST_DEBUG)
+    target_compile_options(
+            sist2
+            PRIVATE
+            -g
+            -fstack-protector
+            -fno-omit-frame-pointer
+            -fsanitize=address
+            -fno-inline
+#            -O2
+    )
+    target_link_options(
+            sist2
+            PRIVATE
+            -fsanitize=address
+            -static-libasan
+    )
+    set_target_properties(
+            sist2
+            PROPERTIES
+            OUTPUT_NAME sist2_debug
+    )
+elseif (SIST_FAST)
+    target_compile_options(
+            sist2
+            PRIVATE
+
+            -Ofast
+            -march=native
+            -fno-stack-protector
+            -fomit-frame-pointer
+            -freciprocal-math
+    )
+else ()
+    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
+        ${GLIB_LDFLAGS}
+        unofficial::mongoose::mongoose
+        CURL::libcurl
 
         pthread
-        curl
-        m
-        bz2
-        magic
+
+        c
+
+        scan
+
+        ${MAGIC_LIB}
 )
 
 add_custom_target(
         before_sist2
         COMMAND ${CMAKE_CURRENT_SOURCE_DIR}/scripts/before_build.sh
 )
+
 add_dependencies(sist2 before_sist2)

Dockerfile

@@ -0,0 +1,46 @@
+FROM simon987/sist2-build as build
+MAINTAINER simon987 <me@simon987.net>
+
+WORKDIR /build/
+
+COPY scripts scripts
+COPY schema schema
+COPY CMakeLists.txt .
+COPY third-party third-party
+COPY src src
+COPY sist2-vue sist2-vue
+
+RUN cmake -DSIST_PLATFORM=x64_linux -DSIST_DEBUG=off -DBUILD_TESTS=off -DCMAKE_TOOLCHAIN_FILE=/vcpkg/scripts/buildsystems/vcpkg.cmake .
+RUN make -j$(nproc)
+RUN strip sist2 || mv sist2_debug sist2
+
+FROM --platform="linux/amd64" ubuntu@sha256:965fbcae990b0467ed5657caceaec165018ef44a4d2d46c7cdea80a9dff0d1ea
+
+WORKDIR /root
+
+ENV LANG C.UTF-8
+ENV LC_ALL C.UTF-8
+
+ENTRYPOINT ["/root/sist2"]
+
+RUN apt update && DEBIAN_FRONTEND=noninteractive apt install -y curl libasan5 libmagic1 python3  \
+    python3-pip git tesseract-ocr libpq-dev && rm -rf /var/lib/apt/lists/*
+
+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/osd.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/osd.traineddata &&\
+    curl -o /usr/share/tessdata/spa.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/spa.traineddata &&\
+    curl -o /usr/share/tessdata/chi_sim.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/chi_sim.traineddata
+
+# sist2
+COPY --from=build /build/sist2 /root/sist2
+
+# sist2-admin
+COPY sist2-admin/requirements.txt sist2-admin/
+RUN python3 -m pip install --no-cache -r sist2-admin/requirements.txt
+COPY sist2-admin/ sist2-admin/

Dockerfile.arm64

@@ -0,0 +1,28 @@
+FROM simon987/sist2-build-arm64 as build
+MAINTAINER simon987 <me@simon987.net>
+
+WORKDIR /build/
+ADD . /build/
+RUN cmake -DSIST_PLATFORM=arm64_linux -DSIST_DEBUG=off -DBUILD_TESTS=off -DCMAKE_TOOLCHAIN_FILE=/vcpkg/scripts/buildsystems/vcpkg.cmake .
+RUN make -j$(nproc)
+RUN strip sist2
+
+FROM --platform="linux/arm64/v8" ubuntu:20.04
+
+RUN apt update && apt install -y curl libasan5 && rm -rf /var/lib/apt/lists/*
+
+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
+
+ENV LANG C.UTF-8
+ENV LC_ALL C.UTF-8
+
+ENTRYPOINT ["/root/sist2"]
+
+COPY --from=build /build/sist2 /root/sist2

README.md

@@ -1,5 +1,8 @@
 ![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/api/badges/simon987/sist2/status.svg)](https://files.simon987.net/.gate/sist2/simon987_sist2/)
+
+**Demo**: [sist2.simon987.net](https://sist2.simon987.net/)
 
 # sist2
 
@@ -7,84 +10,152 @@ sist2 (Simple incremental search tool)
 
 *Warning: sist2 is in early development*
 
+![search panel](docs/sist2.png)
+
 ## Features
 
-* Fast, low memory usage
+* Fast, low memory usage, multi-threaded
+* Mobile-friendly Web interface
 * Portable (all its features are packaged in a single executable)
-* Extracts text from common file types\*
-* Generates thumbnails\*
+* 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)
 
-\* See [format support](#format-support)
+![stats](docs/stats.png)
 
 ## Getting Started
 
-1. Have an [Elasticsearch](https://www.elastic.co/downloads/elasticsearch) instance running
-1. Download the [latest sist2 release](https://github.com/simon987/sist2/releases)
+1. Have an Elasticsearch (>= 6.8.X, ideally >=7.14.0) instance running
+    1. Download [from official website](https://www.elastic.co/downloads/elasticsearch)
+    1. *(or)* Run using docker:
+        ```bash
+        docker run -d -p 9200:9200 -e "discovery.type=single-node" elasticsearch:7.14.0
+        ```
+    1. *(or)* Run using docker-compose:
+        ```yaml
+          elasticsearch:
+            image: docker.elastic.co/elasticsearch/elasticsearch:7.14.0
+            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). 
+Select the file corresponding to your CPU architecture and mark the binary as executable with `chmod +x` *
+    2. *(or)* Download a [development snapshot](https://files.simon987.net/.gate/sist2/simon987_sist2/) *(Not
+       recommended!)*
+    3. *(or)* `docker pull simon987/sist2:2.12.1-x64-linux`
 
-*Windows users*: `sist2` runs under [WSL](https://en.wikipedia.org/wiki/Windows_Subsystem_for_Linux)
-
-*Mac users*: See [#1](https://github.com/simon987/sist2/issues/1)
+1. See [Usage guide](docs/USAGE.md)
 
+\* *Windows users*: **sist2** runs under [WSL](https://en.wikipedia.org/wiki/Windows_Subsystem_for_Linux)
 
 ## Example usage
 
-![demo](demo.gif)
+See [Usage guide](docs/USAGE.md) for more details
 
-See help page `sist2 --help` for more details.
-
-**Scan a directory**
-```bash
-sist2 scan ~/Documents -o ./orig_idx/
-sist2 scan --threads 4 --content-size 16384 /mnt/Pictures
-sist2 scan --incremental ./orig_idx/ -o ./updated_idx/ ~/Documents
-```
-
-**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
-```
+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`
 
 ## 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* |
-ttf,ttc,cff,woff,fnt,otf | Freetype2 | - | yes, `bmp` | Name & style |
-`text/plain` | *(none)* | yes | no | - |
-docx, xlsx, pptx |  | *planned* | no | *planned* |
+| File type                                                                 | Library                                                                      | Content  | Thumbnail   | Metadata                                                                                                                               |
+|:--------------------------------------------------------------------------|:-----------------------------------------------------------------------------|:---------|:------------|:---------------------------------------------------------------------------------------------------------------------------------------|
+| pdf,xps,fb2,epub                                                          | MuPDF                                                                        | text+ocr | yes         | author, title                                                                                                                          |
+| cbz,cbr                                                                   | [libscan](https://github.com/simon987/sist2/tree/master/third-party/libscan) | -        | yes         | -                                                                                                                                      |
+| `audio/*`                                                                 | ffmpeg                                                                       | -        | yes         | ID3 tags                                                                                                                               |
+| `video/*`                                                                 | ffmpeg                                                                       | -        | yes         | title, comment, artist                                                                                                                 |
+| `image/*`                                                                 | ffmpeg                                                                       | ocr      | yes         | [Common EXIF tags](https://github.com/simon987/sist2/blob/efdde2734eca9b14a54f84568863b7ffd59bdba3/src/parsing/media.c#L190), GPS tags |
+| raw, rw2, dng, cr2, crw, dcr, k25, kdc, mrw, pef, xf3, arw, sr2, srf, erf | LibRaw                                                                       | no       | yes         | Common EXIF tags, GPS tags                                                                                                             |
+| ttf,ttc,cff,woff,fnt,otf                                                  | Freetype2                                                                    | -        | yes, `bmp`  | Name & style                                                                                                                           |
+| `text/plain`                                                              | [libscan](https://github.com/simon987/sist2/tree/master/third-party/libscan) | yes      | no          | -                                                                                                                                      |
+| html, xml                                                                 | [libscan](https://github.com/simon987/sist2/tree/master/third-party/libscan) | yes      | no          | -                                                                                                                                      |
+| tar, zip, rar, 7z, ar ...                                                 | Libarchive                                                                   | yes\*    | -           | no                                                                                                                                     |
+| docx, xlsx, pptx                                                          | [libscan](https://github.com/simon987/sist2/tree/master/third-party/libscan) | yes      | if embedded | creator, modified_by, title                                                                                                            |
+| doc (MS Word 97-2003)                                                     | antiword                                                                     | yes      | yes         | author, title                                                                                                                          |
+| mobi, azw, azw3                                                           | libmobi                                                                      | yes      | no          | author, title                                                                                                                          |
+| wpd (WordPerfect)                                                         | libwpd                                                                       | yes      | no          | *planned*                                                                                                                              |
+| json, jsonl, ndjson                                                       | [libscan](https://github.com/simon987/sist2/tree/master/third-party/libscan) | yes      | -           | -                                                                                                                                      |
 
+\* *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 ebook (pdf,xps,fb2,epub) or image file types with the
+`--ocr-lang <lang>` option in combination with `--ocr-images` and/or `--ocr-ebooks`.
+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).
+
+The `simon987/sist2` image comes with common languages
+(hin, jpn, eng, fra, rus, spa) pre-installed.
+
+You can use the `+` separator to specify multiple languages. The language
+name must be identical to the `*.traineddata` file installed on your system 
+(use `chi_sim` rather than `chi-sim`).
+
+Examples:
+
+```bash
+sist2 scan --ocr-ebooks --ocr-lang jpn ~/Books/Manga/
+sist2 scan --ocr-images --ocr-lang eng ~/Images/Screenshots/
+sist2 scan --ocr-ebooks --ocr-images --ocr-lang eng+chi_sim ~/Chinese-Bilingual/
+```
 
 ## Build from source
 
-You can compile **sist2** by yourself if you don't want to use the pre-compiled
-binaries.
+You can compile **sist2** by yourself if you don't want to use the pre-compiled binaries
+
+### With docker (recommended)
+
+```bash
+git clone --recursive https://github.com/simon987/sist2/
+cd sist2
+docker build . -f ./Dockerfile -t my-sist2-image
+docker run --rm --entrypoint cat my-sist2-image /root/sist2 > sist2-x64-linux
+```
+
+### On a linux computer
 
 1. Install compile-time dependencies
 
-    *(Debian)*
+   ```bash
+   apt install gcc g++ python3 yasm ragel automake autotools-dev wget libtool libssl-dev curl zip unzip tar xorg-dev libglu1-mesa-dev libxcursor-dev libxml2-dev libxinerama-dev gettext nasm git
+   ```
+
+1. Apply vcpkg patches, as per [sist2-build](https://github.com/simon987/sist2-build) Dockerfile
+
+1. Install vcpkg dependencies
+
     ```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
-   
-2. Build
+    vcpkg install curl[core,openssl]
+    vcpkg install lmdb cjson glib brotli libarchive[core,bzip2,libxml2,lz4,lzma,lzo] pthread tesseract libxml2 libmupdf gtest mongoose libmagic libraw jasper lcms gumbo
+    ```
+
+1. 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 -DSIST_DEBUG=off -DCMAKE_TOOLCHAIN_FILE=<VCPKG_ROOT>/scripts/buildsystems/vcpkg.cmake .
     make
-    ```
+    ```

contrib/systemd/Makefile

@@ -0,0 +1,7 @@
+install:
+	install sist2-update-all.sh /usr/bin/sist2-update-all.sh
+	install sist2-update-files.sh /usr/bin/sist2-update-files.sh
+	install sist2-update-nextcloud.sh /usr/bin/sist2-update-nextcloud.sh
+	install sist2-update.service /etc/systemd/system/sist2-update.service
+	install sist2-update.timer /etc/systemd/system/sist2-update.timer
+	systemctl daemon-reload

contrib/systemd/README.md

@@ -0,0 +1,31 @@
+# Systemd integration example
+
+This example contains my (yatli) personal configuration for sist2 auto-updating.
+The following indices are involved in this configuration:
+
+| Index     | Path             | Description                                |
+|-----------|------------------|--------------------------------------------|
+| files     | /zpool/files     | Main file repository                       |
+| nextcloud | /zpool/nextcloud | Externally synchronized to a cloud account |
+
+The systemd integration achieves automatic sist2 scanning & indexing everyday at 3:00AM.
+
+### Tailoring the configuration for yourself
+
+`sist2-update-all.sh` calls update scripts for each sist2 index. Add or remove
+update scripts accordingly to suit your need. Each update script (e.g.
+`sist2-update-files.sh`) has important parameters laid down at the beginning so
+make sure to edit them to point to your files and index locations.
+
+### Installation
+
+```bash
+# install the services and scripts
+sudo make install
+# enable & start the timer
+sudo systemctl enable sist2-update.timer
+sudo systemctl start sist2-update.timer
+# verify that the timer has been enabled
+systemctl list-timers --all
+```
+

contrib/systemd/sist2-update-all.sh

@@ -0,0 +1,9 @@
+#!/bin/bash
+set -e
+__dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+echo "Update index: Files"
+source ${__dir}/sist2-update-files.sh
+echo "Update index: Nextcloud"
+source ${__dir}/sist2-update-nextcloud.sh
+echo "Done. Restarting sist2."
+docker restart sist2-sist2-1

contrib/systemd/sist2-update-files.sh

@@ -0,0 +1,34 @@
+#!/bin/bash
+set -e
+DATE=$(date +%Y_%m_%d)
+CONTENT=/zpool/files
+ORIG=/mnt/ssd/sist-index/files.idx
+NEW=/mnt/ssd/sist-index/files_$DATE.idx
+EXCLUDE='ZArchives|TorrentStore|TorrentDownload|624f0c59-1fef-44f6-95e9-7483296f2833|ubuntu-full-2021-12-07'
+NAME=Files
+#REWRITE_URL="http://localhost:33333/activate?collection=$NAME&path="
+REWRITE_URL=""
+
+sist2 scan \
+  --threads 14 \
+  --mem-throttle 32768 \
+  --quality 1.0 \
+  --name $NAME \
+  --ocr-lang=eng+chi_sim \
+  --ocr-ebooks \
+  --ocr-images \
+  --exclude=$EXCLUDE \
+  --rewrite-url=$REWRITE_URL \
+  --incremental=$ORIG \
+  --output=$NEW \
+  $CONTENT
+echo ">>> Scan complete"
+rm -rf $ORIG
+mv $NEW $ORIG 
+
+unset http_proxy
+unset https_proxy
+unset HTTP_PROXY
+unset HTTPS_PROXY
+sist2 index $ORIG --incremental-index
+echo ">>> Index complete"

contrib/systemd/sist2-update-nextcloud.sh

@@ -0,0 +1,33 @@
+#!/bin/bash
+set -e
+DATE=$(date +%Y_%m_%d)
+CONTENT=/zpool/nextcloud/v-yadli
+ORIG=/mnt/ssd/sist-index/nextcloud.idx
+NEW=/mnt/ssd/sist-index/nextcloud_$DATE.idx
+EXCLUDE='Yatao|.*263418493\\/Image\\/.*'
+NAME=NextCloud
+# REWRITE_URL="http://localhost:33333/activate?collection=$NAME&path="
+REWRITE_URL=""
+
+sist2 scan \
+  --threads 14 \
+  --mem-throttle 32768 \
+  --quality 1.0 \
+  --name $NAME \
+  --ocr-lang=eng+chi_sim \
+  --ocr-ebooks \
+  --ocr-images \
+  --exclude=$EXCLUDE \
+  --rewrite-url=$REWRITE_URL \
+  --incremental=$ORIG \
+  --output=$NEW \
+  $CONTENT
+echo ">>> Scan complete"
+rm -rf $ORIG
+mv $NEW $ORIG 
+
+unset http_proxy
+unset https_proxy
+unset HTTP_PROXY
+unset HTTPS_PROXY
+sist2 index $ORIG --incremental-index

contrib/systemd/sist2-update.service

@@ -0,0 +1,6 @@
+[Unit]
+Description=sist2-update
+
+[Service]
+User=yatli
+ExecStart=/bin/bash /usr/bin/sist2-update-all.sh

contrib/systemd/sist2-update.timer

@@ -0,0 +1,10 @@
+[Unit]
+Description=sist2-update
+
+[Timer]
+OnCalendar=*-*-* 3:00:00
+Persistent=true
+Unit=sist2-update.service
+
+[Install]
+WantedBy=timers.target

docker-compose.yml

@@ -0,0 +1,24 @@
+version: "3"
+
+services:
+  elasticsearch:
+    image: elasticsearch:7.14.0
+    container_name: sist2-es
+    environment:
+      - "discovery.type=single-node"
+      - "ES_JAVA_OPTS=-Xms2g -Xmx2g"
+  sist2-admin:
+    build:
+      context: .
+    container_name: sist2-admin
+    volumes:
+      - /mnt/array/sist2-admin-data/:/sist2-admin/
+      - /:/host
+    ports:
+      # NOTE: Don't export this port publicly!
+      - 8080:8080
+      - 4090:4090
+    working_dir: /root/sist2-admin/
+    entrypoint: python3
+    command:
+      - /root/sist2-admin/sist2_admin/app.py

docs/USAGE.md

@@ -0,0 +1,377 @@
+# 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)
+* [elasticsearch](#elasticsearch)
+* [exec-script](#exec-script)
+* [tagging](#tagging)
+* [sidecar files](#sidecar-files)
+
+```
+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
+    --mem-throttle=<int>              Total memory threshold in MiB for scan throttling. DEFAULT=0
+    -q, --thumbnail-quality=<flt>     Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best. DEFAULT=1
+    --thumbnail-size=<int>            Thumbnail size, in pixels. DEFAULT=500
+    --thumbnail-count=<int>           Number of thumbnails to generate. Set a value > 1 to create video previews, set to 0 to disable thumbnails. DEFAULT=1
+    --content-size=<int>              Number of bytes to be extracted from text documents. Set to 0 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
+    --archive-passphrase=<str>        Passphrase for encrypted archive files
+    --ocr-lang=<str>                  Tesseract language (use 'tesseract --list-langs' to see which are installed on your machine)
+    --ocr-images                      Enable OCR'ing of image files.
+    --ocr-ebooks                      Enable OCR'ing of ebook files.
+    -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 MiB for files inside archives (see USAGE.md). DEFAULT: 2000
+    --read-subtitles                  Read subtitles from media files.
+    --fast-epub                       Faster but less accurate EPUB parsing (no thumbnails, metadata)
+    --checksums                       Calculate file checksums when scanning.
+    --list-file=<str>                 Specify a list of newline-delimited paths to be scanned instead of normal directory traversal. Use '-' to read from stdin.
+
+Index options
+    -t, --threads=<int>               Number of threads. DEFAULT=1
+    --es-url=<str>                    Elasticsearch url with port. DEFAULT=http://localhost:9200
+    --es-index=<str>                  Elasticsearch index name. DEFAULT=sist2
+    -p, --print                       Just print JSON documents to stdout.
+    --incremental-index               Conduct incremental indexing, assumes that the old index is already digested by Elasticsearch.
+    --script-file=<str>               Path to user script.
+    --mappings-file=<str>             Path to Elasticsearch mappings.
+    --settings-file=<str>             Path to Elasticsearch settings.
+    --async-script                    Execute user script asynchronously.
+    --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
+    --es-index=<str>                  Elasticsearch index name. DEFAULT=sist2
+    --bind=<str>                      Listen on this address. DEFAULT=localhost:4090
+    --auth=<str>                      Basic auth in user:password format
+    --auth0-audience=<str>            API audience/identifier
+    --auth0-domain=<str>              Application domain
+    --auth0-client-id=<str>           Application client ID
+    --auth0-public-key-file=<str>     Path to Auth0 public key file extracted from <domain>/pem
+    --tag-auth=<str>                  Basic auth in user:password format for tagging
+    --tagline=<str>                   Tagline in navbar
+    --dev                             Serve html & js files from disk (for development)
+    --lang=<str>                      Default UI language. Can be changed by the user
+
+Exec-script options
+    --es-url=<str>                    Elasticsearch url. DEFAULT=http://localhost:9200
+    --es-index=<str>                  Elasticsearch index name. DEFAULT=sist2
+    --script-file=<str>               Path to user script.
+    --async-script                    Execute user script asynchronously.
+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-CimInstance Win32_ComputerSystem).NumberOfLogicalProcessors` in Windows!**
+* `--mem-throttle`
+    Total memory threshold in MiB for scan throttling. Worker threads will not start a new parse job
+    until the total memory usage of sist2 is below this threshold. Set to 0 to disable. DEFAULT=0
+* `-q, --thumbnail-quality` 
+    Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best.
+* `--thumbnail-size` 
+    Thumbnail size in pixels.
+* `--thumbnail-count`
+    Maximum number of thumbnails to generate. When set to a value >= 2, thumbnails for video previews
+    will be generated. The actual number of thumbnails generated depends on the length of the video (maximum 1 image 
+    every ~7s). Set to 0 to completely disable thumbnails.
+* `--content-size` 
+    Number of bytes of text to be extracted from the content of files (plain text, PDFs etc.).
+    Repeated whitespace and special characters do not count toward this limit.
+    Set to 0 to completely disable content parsing.
+* `--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-lang`, `--ocr-ebooks`, `--ocr-images` 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 MiB (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 -`
+* `--read-subtitles` When enabled, will attempt to read the subtitles stream from media files.
+* `--fast-epub` Much faster but less accurate EPUB parsing. When enabled, sist2 will use a simple HTML parser to read epub files instead of the MuPDF library. No thumbnails are generated and author/title metadata are not parsed.
+* `--checksums` Calculate file checksums (SHA1) when scanning files. This option does not cause any additional read 
+  operations. Checksums are not calculated for all file types, unless the file is inside an archive. When enabled, duplicate
+  files are hidden in the web UI (this behaviour can be toggled in the Configuration page).
+
+### 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 `ndjson` type index structure looks like this:
+```
+documents.idx/
+├── descriptor.json
+├── _index_main.ndjson.zst
+├── treemap.csv
+├── agg_mime.csv
+├── agg_date.csv
+├── add_size.csv
+├── thumbs/
+|   ├── data.mdb
+|   └── lock.mdb
+├── tags/
+|   ├── data.mdb
+|   └── lock.mdb
+└── meta/
+    ├── data.mdb
+    └── lock.mdb
+```
+
+The `_index_*.ndjson.zst` files contain the document data in JSON format, in a compressed newline-delemited file.
+
+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.
+
+*thumbs/*:
+
+LMDB key-value store. Keys are **binary** 16-byte md5 hash* (`_id` field)
+and values are raw image bytes.
+
+*\* Hash is calculated from the full path of the file, including the extension, relative to the index root*
+
+
+## Index
+### Index options
+ * `--es-url` 
+ Elasticsearch url and port. If you are using docker, make sure that both containers are on the
+ same network.
+ * `--es-index` 
+    Elasticsearch index name. DEFAULT=sist2
+ * `-p, --print` 
+    Print index in JSON format to stdout.
+ * `--incremental-index`
+   Conduct incremental indexing. Assumes that the old index is already ingested in Elasticsearch.
+   Only the new changes since the last scan will be sent.
+ * `--script-file` 
+    Path to user script. See [Scripting](scripting.md).
+ * `--mappings-file`
+    Path to custom Elasticsearch mappings. If none is specified, [the bundled mappings](https://github.com/simon987/sist2/tree/master/schema) will be used.
+ * `--settings-file`
+    Path to custom Elasticsearch settings. *(See above)*
+ * `--async-script` 
+    Use `wait_for_completion=false` elasticsearch option while executing user script.
+     (See [Elasticsearch documentation](https://www.elastic.co/guide/en/elasticsearch/reference/current/tasks.html))
+ * `--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.
+ * `-t, --threads` Number of threads to use. Ideally, choose a number equal to the number of logical cores of the machine hosting Elasticsearch.
+    
+### 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.
+ * `--es-index` 
+    Elasticsearch index name. DEFAULT=sist2
+ * `--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.
+ * `--tagline=<str>` When specified, will replace the default tagline in the navbar.
+ * `--dev` Serve html & js files from disk (for development, used to modify frontend files without having to recompile)
+ * `--lang=<str>` Set the default web UI language (See #180 for a list of supported languages, default
+   is `en`). The user can change the language in the configuration page
+ * `--auth0-audience`, `--auth0-domain`, `--auth0-client-id`, `--auth0-public-key-file` See [Authentication with Auth0](auth0.md)
+
+### 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.
+
+# Elasticsearch
+
+Elasticsearch versions >=6.8.0, 7.X.X and 8.X.X are supported by sist2. 
+
+Using a version >=7.14.0 is recommended to enable the following features:
+
+- Bug fix for large documents (See #198)
+
+When using a legacy version of ES, a notice will be displayed next to the sist2 version in the web UI.
+If you don't care about the features above, you can ignore it or disable it in the configuration page.
+
+## 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.*
+
+
+### Automatic tagging
+
+See [scripting](scripting.md) documentation.
+
+# Sidecar files
+
+When scanning, sist2 will read metadata from `.s2meta` JSON files and overwrite the 
+original document's indexed metadata (does not modify the actual file). Sidecar metadata files will also work inside archives.
+Sidecar files themselves are not saved in the index.
+
+This feature is useful to leverage third-party applications such as speech-to-text or
+OCR to add additional metadata to a file.
+
+**Example**
+
+```
+~/Documents/
+├── Video.mp4
+└── Video.mp4.s2meta
+```
+
+The sidecar file must have exactly the same file path and the `.s2meta` suffix.
+
+`Video.mp4.s2meta`:
+```json
+{
+  "content": "This sidecar file will overwrite some metadata fields of Video.mp4",
+  "author": "Some author",
+  "duration": 12345,
+  "bitrate": 67890,
+  "some_arbitrary_field": [1,2,3]
+}
+```
+
+```
+sist2 scan ~/Documents -o ./docs.idx
+sist2 index ./docs.idx
+```
+
+*NOTE*: It is technically possible to overwrite the `tag` value using sidecar files, however,
+it is not currently possible to restore both manual tags and sidecar tags without user scripts
+while reindexing.

docs/auth0.md

@@ -0,0 +1,19 @@
+
+# Authentication with Auth0
+
+1. Create a new Auth0 application (Single page app)
+2. Create a new Auth0 API:
+   1. Choose `RS256` signing algorithm
+   2. Set identifier (audience) to `https://sist2`
+3. Download the Auth0 certificate from https://<domain>.auth0.com/pem (you can find the domain Applications->Basic information)
+4. Extract the public key from the certificate using `openssl x509 -pubkey -noout -in cert.pem > pubkey.txt`
+5. Start the sist2 web server
+
+Example options:
+```bash
+sist2 web \
+  --auth0-client-id XXX \
+  --auth0-audience https://sist2 \
+  --auth0-domain YYY.auth0.com \
+  --auth0-public-key-file /ZZZ/pubkey.txt
+```

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

@@ -1,34 +1,74 @@
 {
   "properties": {
+    "_tie": {
+      "type": "keyword",
+      "doc_values": true
+    },
+    "checksum": {
+      "type": "keyword",
+      "index": false
+    },
+    "_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"
     },
+    "parent": {
+      "type": "keyword",
+      "index": false
+    },
+    "thumbnail": {
+      "type": "integer",
+      "index": false
+    },
     "videoc": {
-      "type": "keyword"
+      "type": "keyword",
+      "index": false
     },
     "audioc": {
-      "type": "keyword"
+      "type": "keyword",
+      "index": false
     },
     "duration": {
-      "type": "float"
+      "type": "integer",
+      "index": false
     },
     "width": {
-      "type": "integer"
+      "type": "integer",
+      "index": false
     },
     "height": {
-      "type": "integer"
+      "type": "integer",
+      "index": false
+    },
+    "pages": {
+      "type": "integer",
+      "index": false
     },
     "mtime": {
-      "type": "integer"
+      "type": "date",
+      "format": "epoch_millis"
     },
     "size": {
       "type": "long"
@@ -39,6 +79,7 @@
     "name": {
       "analyzer": "content_analyzer",
       "type": "text",
+      "fielddata": true,
       "fields": {
         "nGram": {
           "type": "text",
@@ -70,6 +111,23 @@
       "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 +138,70 @@
           "analyzer": "my_nGram"
         }
       }
+    },
+    "tag": {
+      "type": "text",
+      "fielddata": true,
+      "analyzer": "tag_analyzer",
+      "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"
+    },
+    "exif_gps_longitude_ref": {
+      "type": "keyword",
+      "index": false
+    },
+    "exif_gps_longitude_dms": {
+      "type": "keyword",
+      "index": false
+    },
+    "exif_gps_longitude_dec": {
+      "type": "keyword",
+      "index": false
+    },
+    "exif_gps_latitude_ref": {
+      "type": "keyword",
+      "index": false
+    },
+    "exif_gps_latitude_dms": {
+      "type": "keyword",
+      "index": false
+    },
+    "exif_gps_latitude_dec": {
+      "type": "keyword",
+      "index": false
+    },
+    "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

@@ -1,15 +1,22 @@
 {
   "index": {
-    "refresh_interval": "-1",
-    "codec": "best_compression"
+    "refresh_interval": "30s",
+    "codec": "best_compression",
+    "number_of_replicas": 0,
+    "highlight.max_analyzed_offset": 1000000
   },
   "analysis": {
     "tokenizer": {
       "path_tokenizer": {
-        "type": "path_hierarchy"
+        "type": "path_hierarchy",
+        "delimiter": "/"
+      },
+      "tag_tokenizer": {
+        "type": "path_hierarchy",
+        "delimiter": "."
       },
       "my_nGram_tokenizer": {
-        "type": "nGram",
+        "type": "ngram",
         "min_gram": 3,
         "max_gram": 3
       }
@@ -21,16 +28,30 @@
           "lowercase"
         ]
       },
+      "tag_analyzer": {
+        "tokenizer": "tag_tokenizer",
+        "filter": [
+          "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"
         ]
       }
     }

schema/settings_legacy.json

@@ -0,0 +1,58 @@
+{
+  "index": {
+    "refresh_interval": "30s",
+    "codec": "best_compression",
+    "number_of_replicas": 0
+  },
+  "analysis": {
+    "tokenizer": {
+      "path_tokenizer": {
+        "type": "path_hierarchy",
+        "delimiter": "/"
+      },
+      "tag_tokenizer": {
+        "type": "path_hierarchy",
+        "delimiter": "."
+      },
+      "my_nGram_tokenizer": {
+        "type": "nGram",
+        "min_gram": 3,
+        "max_gram": 3
+      }
+    },
+    "analyzer": {
+      "path_analyzer": {
+        "tokenizer": "path_tokenizer",
+        "filter": [
+          "lowercase"
+        ]
+      },
+      "tag_analyzer": {
+        "tokenizer": "tag_tokenizer",
+        "filter": [
+          "lowercase"
+        ]
+      },
+      "case_insensitive_kw_analyzer": {
+        "tokenizer": "keyword",
+        "filter": [
+          "lowercase"
+        ]
+      },
+      "my_nGram": {
+        "tokenizer": "my_nGram_tokenizer",
+        "filter": [
+          "lowercase",
+          "asciifolding"
+        ]
+      },
+      "content_analyzer": {
+        "tokenizer": "standard",
+        "filter": [
+          "lowercase",
+          "asciifolding"
+        ]
+      }
+    }
+  }
+}

scripts/before_build.sh

@@ -1,15 +1,10 @@
-#!/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 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
-
 python3 scripts/mime.py > src/parsing/mime_generated.c
 python3 scripts/serve_static.py > src/web/static_generated.c
 python3 scripts/index_static.py > src/index/static_generated.c
+python3 scripts/magic_static.py > src/magic_generated.c
+
+printf "static const char *const Sist2CommitHash = \"%s\";\n" $(git rev-parse HEAD) > src/git_hash.h

scripts/build.sh

@@ -0,0 +1,17 @@
+#!/usr/bin/env bash
+
+VCPKG_ROOT="/vcpkg"
+
+git submodule update --init --recursive
+
+rm -rf CMakeFiles CMakeCache.txt
+cmake -DSIST_PLATFORM=x64_linux -DSIST_DEBUG=off -DBUILD_TESTS=off -DCMAKE_TOOLCHAIN_FILE="${VCPKG_ROOT}/scripts/buildsystems/vcpkg.cmake" .
+make -j $(nproc)
+strip sist2
+./sist2 -v > VERSION
+mv sist2 sist2-x64-linux
+
+rm -rf CMakeFiles CMakeCache.txt
+cmake -DSIST_PLATFORM=x64_linux -DSIST_DEBUG=on -DBUILD_TESTS=off -DCMAKE_TOOLCHAIN_FILE="${VCPKG_ROOT}/scripts/buildsystems/vcpkg.cmake" .
+make -j  $(nproc)
+mv sist2_debug sist2-x64-linux-debug

scripts/build_arm64.sh

@@ -0,0 +1,17 @@
+#!/usr/bin/env bash
+
+VCPKG_ROOT="/vcpkg"
+
+git submodule update --init --recursive
+
+rm -rf CMakeFiles CMakeCache.txt
+cmake -DSIST_PLATFORM=arm64_linux -DSIST_DEBUG=off -DBUILD_TESTS=off -DCMAKE_TOOLCHAIN_FILE="${VCPKG_ROOT}/scripts/buildsystems/vcpkg.cmake" .
+make -j $(nproc)
+strip sist2
+mv sist2 sist2-arm64-linux
+
+rm -rf CMakeFiles CMakeCache.txt
+cmake -DSIST_PLATFORM=arm64_linux -DSIST_DEBUG=on -DBUILD_TESTS=off -DCMAKE_TOOLCHAIN_FILE="${VCPKG_ROOT}/scripts/buildsystems/vcpkg.cmake" .
+make -j $(nproc)
+strip sist2
+mv sist2_debug sist2-arm64-linux-debug

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 +1,10 @@
+import json
+
 files = [
     "schema/mappings.json",
     "schema/settings.json",
+    "schema/settings_legacy.json",
+    "schema/pipeline.json",
 ]
 
 
@@ -9,6 +13,7 @@ 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/magic_static.py

@@ -0,0 +1,8 @@
+
+try:
+    with open("/usr/lib/file/magic.mgc", "rb") as f:
+        data = f.read()
+except:
+    data = bytes([])
+
+print("char magic_database_buffer[%d] = {%s};" % (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
@@ -17,8 +21,8 @@ application/inf, inf
 application/java-archive, jar
 application/java, class
 application/javascript,
-application/x-archive, a
 application/json, json
+application/ndjson, jsonl|ndjson
 application/marc, mrc
 application/mbedlet, mbd
 application/mime, aps
@@ -27,7 +31,9 @@ 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 +49,10 @@ 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 +64,73 @@ 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/wordperfect6.0, w60
-application/wordperfect6.1, w61
-application/wordperfect, wp|wp5|wp6|wpd
+application/warc, warc
+application/winhelp, hlp
+application/wordperfect, wp|wp5|wp6|wpd|w60|w61
 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 +140,67 @@ 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-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 +208,17 @@ 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 +227,22 @@ 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 +250,10 @@ 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 +267,24 @@ 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 +293,9 @@ 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 +303,15 @@ 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 +325,34 @@ 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|make|log|markdown|yaml
+application/vnd.coffeescript, coffee
 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 +361,7 @@ 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 +376,31 @@ 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 +415,36 @@ 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
+sist2/sidecar, s2meta

scripts/mime.py

@@ -3,6 +3,7 @@ noparse = set()
 ext_in_hash = set()
 
 major_mime = {
+    "sist2": 0,
     "model": 1,
     "example": 2,
     "message": 3,
@@ -12,18 +13,19 @@ 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 +34,68 @@ 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
 
 
@@ -46,8 +110,24 @@ 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":
+        cnt -= 1
         return "1"
+    elif mime == "sist2/sidecar":
+        cnt -= 1
+        return "2"
     return mime_id
 
 
@@ -55,7 +135,7 @@ 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 +147,11 @@ 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 +1,10 @@
 files = [
-    "web/css/bundle.css",
-    "web/js/bundle.js",
-    "web/img/bg-bars.png",
-    "web/img/sprite-skin-flat.png",
-    "web/search.html",
+    "sist2-vue/src/assets/favicon.ico",
+    "sist2-vue/dist/css/chunk-vendors.css",
+    "sist2-vue/dist/css/index.css",
+    "sist2-vue/dist/js/chunk-vendors.js",
+    "sist2-vue/dist/js/index.js",
+    "sist2-vue/dist/index.html",
 ]
 
 
@@ -12,6 +13,10 @@ def clean(filepath):
 
 
 for file in files:
-    with open(file, "rb") as f:
-        data = f.read()
+    try:
+        with open(file, "rb") as f:
+            data = f.read()
+    except:
+        data = bytes([])
+
     print("char %s[%d] = {%s};" % (clean(file), len(data), ",".join(str(int(b)) for b in data)))

scripts/start_dev_es.sh

@@ -0,0 +1,3 @@
+docker run --rm -it --name "sist2-dev-es"\
+       	-p 9200:9200 -e "discovery.type=single-node" \
+	-e "ES_JAVA_OPTS=-Xms8g -Xmx8g" elasticsearch:7.14.0

scripts/start_dev_es_6.sh

@@ -0,0 +1,3 @@
+docker run --rm -it --name "sist2-dev-es-6"\
+       	-p 9202:9200 -e "discovery.type=single-node" \
+	-e "ES_JAVA_OPTS=-Xms8g -Xmx8g" elasticsearch:6.8.0

scripts/start_dev_es_8.sh

@@ -0,0 +1,3 @@
+docker run --rm -it --name "sist2-dev-es"\
+       	-p 9200:9200 -p 9300:9300 -e "discovery.type=single-node" \
+	-e "ES_JAVA_OPTS=-Xms8g -Xmx8g" elasticsearch:8.1.2

sist2-admin/frontend/babel.config.js

@@ -0,0 +1,5 @@
+module.exports = {
+  presets: [
+    '@vue/cli-plugin-babel/preset'
+  ]
+}

sist2-admin/frontend/dist/css/app.css

@@ -0,0 +1 @@
+.navbar[data-v-27bc1d68]{box-shadow:0 .125rem .25rem rgba(0,0,0,.08)!important;border-radius:0}.theme-black .navbar[data-v-27bc1d68]{background:rgba(84,107,122,.18823529411764706);border-bottom:none}.navbar-brand[data-v-27bc1d68]{color:#222!important;font-size:1.75rem;padding:0}.navbar-brand[data-v-27bc1d68]:hover{color:#000!important}.version[data-v-27bc1d68]{color:#222!important;margin-left:-18px;margin-top:-14px;font-size:11px;font-family:monospace}.btn-link[data-v-27bc1d68]{color:#222}body,html{height:100%}#app{-webkit-font-smoothing:antialiased;-moz-osx-font-smoothing:grayscale;color:#2c3e50;padding-bottom:1em;min-height:100%}.info-icon{width:1rem;margin-right:.2rem;cursor:pointer;line-height:1rem;height:1rem;background-image:url(data:image/svg+xml;base64,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);filter:brightness(45%);display:block}.tabs{margin-top:10px}.modal-title{text-overflow:ellipsis;overflow:hidden;white-space:nowrap}@media screen and (min-width:1500px){.container{max-width:1440px}}label{margin-top:.5rem;margin-bottom:0}.shrink[data-v-9b017c42]{flex-grow:inherit}#task-history[data-v-46960281]{font-family:monospace;font-size:12px}#log-tail-output span{display:block}span.DEBUG{color:#9e9e9e}span.WARNING{color:#ffb300}span.INFO{color:#039be5}span.ERROR,span.FATAL{color:#f4511e}span.ADMIN{color:#ee05ff}#log-tail-output{font-size:13px;font-family:monospace;padding:6px;background-color:#f5f5f5;border:1px solid #ccc;border-radius:4px;margin:3px;white-space:pre;color:#000;overflow:hidden}

sist2-admin/frontend/dist/index.html

@@ -0,0 +1 @@
+<!DOCTYPE html><html lang=""><head><meta charset="utf-8"><meta http-equiv="X-UA-Compatible" content="IE=edge"><meta name="viewport" content="width=device-width,initial-scale=1"><link rel="icon" href="favicon.ico"><title>sist2-admin</title><link href="css/app.css" rel="preload" as="style"><link href="css/chunk-vendors.css" rel="preload" as="style"><link href="js/app.js" rel="preload" as="script"><link href="js/chunk-vendors.js" rel="preload" as="script"><link href="css/chunk-vendors.css" rel="stylesheet"><link href="css/app.css" rel="stylesheet"></head><body><noscript><strong>We're sorry but sist2-admin-vue doesn't work properly without JavaScript enabled. Please enable it to continue.</strong></noscript><div id="app"></div><script src="js/chunk-vendors.js"></script><script src="js/app.js"></script></body></html>

sist2-admin/frontend/package.json

@@ -0,0 +1,52 @@
+{
+  "name": "sist2-admin-vue",
+  "version": "0.1.0",
+  "private": true,
+  "scripts": {
+    "serve": "vue-cli-service serve",
+    "build": "vue-cli-service build",
+    "watch": "vue-cli-service build --watch"
+  },
+  "dependencies": {
+    "axios": "^0.27.2",
+    "bootstrap-vue": "^2.21.2",
+    "core-js": "^3.6.5",
+    "moment": "^2.29.3",
+    "socket.io-client": "^4.5.1",
+    "vue": "^2.6.14",
+    "vue-i18n": "^8.24.4",
+    "vue-router": "^3.5.4",
+    "vuex": "^3.4.0"
+  },
+  "devDependencies": {
+    "@vue/cli-plugin-babel": "~5.0.8",
+    "@vue/cli-plugin-eslint": "~5.0.8",
+    "@vue/cli-plugin-router": "~5.0.8",
+    "@vue/cli-plugin-vuex": "~5.0.8",
+    "@vue/cli-service": "~5.0.8",
+    "babel-eslint": "^10.1.0",
+    "bootstrap": "^4.5.2",
+    "eslint": "^6.7.2",
+    "eslint-plugin-vue": "^6.2.2",
+    "vue-template-compiler": "^2.6.11"
+  },
+  "eslintConfig": {
+    "root": true,
+    "env": {
+      "node": true
+    },
+    "extends": [
+      "plugin:vue/essential",
+      "eslint:recommended"
+    ],
+    "parserOptions": {
+      "parser": "babel-eslint"
+    },
+    "rules": {}
+  },
+  "browserslist": [
+    "> 1%",
+    "last 2 versions",
+    "not dead"
+  ]
+}

sist2-admin/frontend/public/index.html

@@ -0,0 +1,17 @@
+<!DOCTYPE html>
+<html lang="">
+  <head>
+    <meta charset="utf-8">
+    <meta http-equiv="X-UA-Compatible" content="IE=edge">
+    <meta name="viewport" content="width=device-width,initial-scale=1.0">
+    <link rel="icon" href="<%= BASE_URL %>favicon.ico">
+    <title>sist2-admin</title>
+  </head>
+  <body>
+    <noscript>
+      <strong>We're sorry but <%= htmlWebpackPlugin.options.title %> doesn't work properly without JavaScript enabled. Please enable it to continue.</strong>
+    </noscript>
+    <div id="app"></div>
+    <!-- built files will be auto injected -->
+  </body>
+</html>

sist2-admin/frontend/src/App.vue

@@ -0,0 +1,98 @@
+<template>
+  <div id="app">
+    <NavBar></NavBar>
+    <b-container class="pt-4">
+      <b-alert show dismissible variant="info">
+        This is a beta version of sist2-admin. Please submit bug reports, usability issues and feature requests
+        to the <a href="https://github.com/simon987/sist2/issues/new/choose" target="_blank">issue tracker on Github</a>. Thank you!
+      </b-alert>
+      <router-view/>
+    </b-container>
+  </div>
+</template>
+
+<script>
+import NavBar from "@/components/NavBar";
+import Sist2AdminApi from "@/Sist2AdminApi";
+
+export default {
+  components: {NavBar},
+  data() {
+    return {
+      socket: null
+    }
+  },
+  mounted() {
+    Sist2AdminApi.getSist2AdminInfo()
+        .then(resp => this.$store.commit("setSist2AdminInfo", resp.data));
+    this.$store.dispatch("loadBrowserSettings");
+    this.connectNotifications();
+    // this.socket.onclose = this.connectNotifications;
+  },
+  methods: {
+    connectNotifications() {
+      this.socket = new WebSocket(`ws://${window.location.host}/notifications`);
+      this.socket.onopen = () => {
+        this.socket.send("Hello from client");
+      }
+
+      this.socket.onmessage = e => {
+        const notification = JSON.parse(e.data);
+        if (notification.message) {
+          notification.messageString = this.$t(notification.message).toString();
+        }
+
+        this.$store.dispatch("notify", notification)
+      }
+    }
+  }
+}
+</script>
+
+<style>
+html, body {
+  height: 100%;
+}
+
+#app {
+  /*font-family: Avenir, Helvetica, Arial, sans-serif;*/
+  -webkit-font-smoothing: antialiased;
+  -moz-osx-font-smoothing: grayscale;
+  /*text-align: center;*/
+  color: #2c3e50;
+  padding-bottom: 1em;
+  min-height: 100%;
+}
+
+.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%);
+  display: block;
+}
+
+.tabs {
+  margin-top: 10px;
+}
+
+.modal-title {
+  text-overflow: ellipsis;
+  overflow: hidden;
+  white-space: nowrap;
+}
+
+@media screen and (min-width: 1500px) {
+  .container {
+    max-width: 1440px;
+  }
+}
+
+label {
+  margin-top: 0.5rem;
+  margin-bottom: 0;
+}
+</style>

sist2-admin/frontend/src/Sist2AdminApi.js

@@ -0,0 +1,117 @@
+import axios from "axios";
+
+class Sist2AdminApi {
+
+    constructor() {
+        this.baseUrl = window.location.protocol + "//" + window.location.host;
+    }
+
+    getJobs() {
+        return axios.get(`${this.baseUrl}/api/job/`);
+    }
+
+    getFrontends() {
+        return axios.get(`${this.baseUrl}/api/frontend/`);
+    }
+
+    getTasks() {
+        return axios.get(`${this.baseUrl}/api/task/`);
+    }
+
+    killTask(taskId) {
+        return axios.post(`${this.baseUrl}/api/task/${taskId}/kill`)
+    }
+
+    getTaskHistory() {
+        return axios.get(`${this.baseUrl}/api/task/history`);
+    }
+
+    /**
+     * @param {string} name
+     */
+    getJob(name) {
+        return axios.get(`${this.baseUrl}/api/job/${name}`);
+    }
+
+    /**
+     * @param {string} name
+     */
+    getFrontend(name) {
+        return axios.get(`${this.baseUrl}/api/frontend/${name}`);
+    }
+
+    /**
+     * @param {string} name
+     */
+    startFrontend(name) {
+        return axios.post(`${this.baseUrl}/api/frontend/${name}/start`);
+    }
+
+    /**
+     * @param {string} name
+     */
+    stopFrontend(name) {
+        return axios.post(`${this.baseUrl}/api/frontend/${name}/stop`);
+    }
+
+    /**
+     * @param {string} name
+     * @param job
+     */
+    updateJob(name, job) {
+        return axios.put(`${this.baseUrl}/api/job/${name}`, job);
+    }
+
+    /**
+     * @param {string} name
+     * @param frontend
+     */
+    updateFrontend(name, frontend) {
+        return axios.put(`${this.baseUrl}/api/frontend/${name}`, frontend);
+    }
+
+    /**
+     * @param {string} name
+     */
+    runJob(name) {
+        return axios.get(`${this.baseUrl}/api/job/${name}/run`);
+    }
+
+    /**
+     * @param {string} name
+     */
+    deleteJob(name) {
+        return axios.delete(`${this.baseUrl}/api/job/${name}`);
+    }
+
+    /**
+     * @param {string} name
+     */
+    deleteFrontend(name) {
+        return axios.delete(`${this.baseUrl}/api/frontend/${name}`);
+    }
+
+    /**
+     * @param {string} name
+     */
+    createJob(name) {
+        return axios.post(`${this.baseUrl}/api/job/${name}`);
+    }
+
+    /**
+     * @param {string} name
+     */
+    createFrontend(name) {
+        return axios.post(`${this.baseUrl}/api/frontend/${name}`);
+    }
+
+    pingEs(url, insecure) {
+        return axios.get(`${this.baseUrl}/api/ping_es`, {params: {url, insecure}});
+    }
+
+    getSist2AdminInfo() {
+        return axios.get(`${this.baseUrl}/api/`);
+    }
+}
+
+export default new Sist2AdminApi()

sist2-admin/frontend/src/components/FrontendListItem.vue

@@ -0,0 +1,31 @@
+<template>
+  <b-list-group-item action :to="`/frontend/${frontend.name}`">
+
+    <div class="d-flex w-100 justify-content-between">
+      <h5 class="mb-1" style="display: block">
+        {{ frontend.name }}
+        <b-badge variant="light">{{ formatBindAddress(frontend.web_options.bind) }}</b-badge>
+      </h5>
+
+      <div>
+        <b-badge v-if="frontend.running" variant="success">{{$t("online")}}</b-badge>
+        <b-badge v-else variant="secondary">{{$t("offline")}}</b-badge>
+      </div>
+    </div>
+
+  </b-list-group-item>
+</template>
+
+<script>
+import {formatBindAddress} from "@/util";
+
+export default {
+  name: "FrontendListItem",
+  props: ["frontend"],
+  data() {
+    return {
+      formatBindAddress
+    }
+  }
+}
+</script>

sist2-admin/frontend/src/components/IndexOptions.vue

@@ -0,0 +1,64 @@
+<template>
+  <div>
+    <label>{{ $t("indexOptions.threads") }}</label>
+    <b-form-input v-model="options.threads" type="number" min="1" @change="update()"></b-form-input>
+
+    <label>{{ $t("webOptions.esUrl") }}</label>
+    <b-alert :variant="esTestOk ? 'success' : 'danger'" :show="showEsTestAlert" class="mt-1">
+      {{ esTestMessage }}
+    </b-alert>
+    <b-input-group>
+      <b-form-input v-model="options.es_url" @change="update()"></b-form-input>
+      <b-input-group-append>
+        <b-button variant="outline-primary" @click="testEs()">{{ $t("test") }}</b-button>
+      </b-input-group-append>
+    </b-input-group>
+
+    <label>{{ $t("indexOptions.esIndex") }}</label>
+    <b-form-input v-model="options.es_index" @change="update()"></b-form-input>
+
+    <br>
+    <b-form-checkbox v-model="options.es_insecure_ssl" :disabled="!options.es_url.startsWith('https')" @change="update()">
+      {{ $t("webOptions.esInsecure") }}
+    </b-form-checkbox>
+
+    <label>{{ $t("indexOptions.batchSize") }}</label>
+    <b-form-input v-model="options.batch_size" type="number" min="1" @change="update()"></b-form-input>
+
+    <label>{{ $t("indexOptions.script") }}</label>
+    <b-form-textarea v-model="options.script" rows="6" @change="update()"></b-form-textarea>
+  </div>
+
+</template>
+
+<script>
+import sist2AdminApi from "@/Sist2AdminApi";
+
+export default {
+  name: "IndexOptions",
+  props: ["options"],
+  data() {
+    return {
+      showEsTestAlert: false,
+      esTestOk: false,
+      esTestMessage: "",
+    }
+  },
+  methods: {
+    update() {
+      this.$emit("change", this.options);
+    },
+    testEs() {
+      sist2AdminApi.pingEs(this.options.es_url, this.options.es_insecure_ssl).then((resp) => {
+        this.showEsTestAlert = true;
+        this.esTestOk = resp.data.ok;
+        this.esTestMessage = resp.data.message;
+      });
+    }
+  },
+}
+</script>
+
+<style scoped>
+
+</style>

sist2-admin/frontend/src/components/JobCheckboxGroup.vue

@@ -0,0 +1,42 @@
+<template>
+  <div>
+    <h5>{{ $t("selectJobs") }}</h5>
+    <b-progress v-if="loading" striped animated value="100"></b-progress>
+    <b-form-group v-else>
+      <b-form-checkbox-group
+          v-if="jobs.length > 0"
+          :checked="frontend.jobs"
+          @input="frontend.jobs = $event; $emit('input')"
+      >
+        <div v-for="job in jobs" :key="job.name">
+          <b-form-checkbox :disabled="job.status !== 'indexed'" :value="job.name">[{{ job.name }}]</b-form-checkbox>
+          <br/>
+        </div>
+      </b-form-checkbox-group>
+      <div v-else>
+        <span class="text-muted">{{ $t('jobOptions.noJobAvailable') }}</span>
+        &nbsp;<router-link to="/">{{$t("create")}}</router-link>
+      </div>
+    </b-form-group>
+  </div>
+</template>
+
+<script>
+import Sist2AdminApi from "@/Sist2AdminApi";
+
+export default {
+  name: "JobCheckboxGroup",
+  props: ["frontend"],
+  mounted() {
+    Sist2AdminApi.getJobs().then(resp => {
+      this.jobs = resp.data;
+      this.loading = false;
+    });
+  },
+  data() {
+    return {
+      loading: true,
+    }
+  }
+}
+</script>

sist2-admin/frontend/src/components/JobListItem.vue

@@ -0,0 +1,56 @@
+<template>
+  <b-list-group-item class="flex-column align-items-start" action :to="`job/${job.name}`">
+
+    <div class="d-flex w-100 justify-content-between">
+      <div>
+        <h5 class="mb-1">
+          {{ job.name }}
+        </h5>
+      </div>
+      <div>
+        <b-row>
+          <b-col>
+            <small v-if="job.last_index_date">
+              {{ $t("scanned") }} {{ formatLastIndexDate(job.last_index_date) }}</small>
+            <div v-else>&nbsp;</div>
+          </b-col>
+        </b-row>
+        <b-row v-if="job.schedule_enabled">
+          <b-col>
+            <small><code>{{job.cron_expression }}</code></small>
+          </b-col>
+        </b-row>
+        <b-row v-else>
+          <b-col>
+            &nbsp;
+          </b-col>
+        </b-row>
+
+      </div>
+    </div>
+
+  </b-list-group-item>
+</template>
+
+<script>
+import moment from "moment";
+
+export default {
+  name: "JobListItem",
+  props: ["job"],
+  methods: {
+    formatLastIndexDate(dateString) {
+      if (dateString === null) {
+        return "";
+      }
+
+      const date = Date.parse(dateString);
+      return moment(date).fromNow();
+    }
+  }
+}
+</script>
+
+<style scoped>
+
+</style>

sist2-admin/frontend/src/components/JobOptions.vue

@@ -0,0 +1,51 @@
+<template>
+  <div>
+    <b-form-checkbox :checked="desktopNotificationsEnabled" @change="updateNotifications($event)">
+      {{ $t("jobOptions.desktopNotifications") }}
+    </b-form-checkbox>
+
+    <b-form-checkbox v-model="job.schedule_enabled" @change="update()">
+      {{ $t("jobOptions.scheduleEnabled") }}
+    </b-form-checkbox>
+
+    <label>{{ $t("jobOptions.cron") }}</label>
+    <b-form-input class="text-monospace" :state="cronValid" v-model="job.cron_expression" :disabled="!job.schedule_enabled" @change="update()"></b-form-input>
+  </div>
+</template>
+
+<script>
+
+export default {
+  name: "JobOptions",
+  props: ["job"],
+  data() {
+    return {
+      cronValid: undefined
+    }
+  },
+  computed: {
+    desktopNotificationsEnabled() {
+      return this.$store.state.jobDesktopNotificationMap[this.job.name];
+    }
+  },
+  methods: {
+    updateNotifications(value) {
+      this.$store.dispatch("setJobDesktopNotification", {
+        job: this.job.name,
+        enabled: value
+      })
+    },
+    update() {
+      if (this.job.schedule_enabled) {
+        this.cronValid = /((((\d+,)+\d+|(\d+([/-])\d+)|\d+|\*) ?){5,7})/.test(this.job.cron_expression);
+      } else {
+        this.cronValid = undefined;
+      }
+
+      if (this.cronValid !== false) {
+        this.$emit("change", this.job);
+      }
+    },
+  },
+}
+</script>

sist2-admin/frontend/src/components/NavBar.vue

@@ -0,0 +1,69 @@
+<template>
+  <b-navbar>
+    <b-navbar-brand to="/">
+      <Sist2Icon></Sist2Icon>
+    </b-navbar-brand>
+
+    <b-button class="ml-auto" to="/task" variant="link">{{ $t("tasks") }}</b-button>
+  </b-navbar>
+</template>
+
+<script>
+import Sist2Icon from "@/components/icons/Sist2Icon";
+
+export default {
+  name: "NavBar",
+  components: {Sist2Icon},
+  methods: {
+    tagline() {
+      return this.$store.state.sist2Info.tagline;
+    },
+    sist2Version() {
+      return this.$store.state.sist2Info.version;
+    },
+    isDebug() {
+      return this.$store.state.sist2Info.debug;
+    },
+    isLegacy() {
+      return this.$store.state.sist2Info.esVersionLegacy;
+    },
+    hideLegacy() {
+      return this.$store.state.optHideLegacy;
+    }
+  }
+}
+</script>
+
+<style scoped>
+.navbar {
+  box-shadow: 0 0.125rem 0.25rem rgb(0 0 0 / 8%) !important;
+  border-radius: 0;
+}
+
+.theme-black .navbar {
+  background: #546b7a30;
+  border-bottom: none;
+}
+
+.navbar-brand {
+  color: #222 !important;
+  font-size: 1.75rem;
+  padding: 0;
+}
+
+.navbar-brand:hover {
+  color: #000 !important;
+}
+
+.version {
+  color: #222 !important;
+  margin-left: -18px;
+  margin-top: -14px;
+  font-size: 11px;
+  font-family: monospace;
+}
+
+.btn-link {
+  color: #222;
+}
+</style>

sist2-admin/frontend/src/components/ScanOptions.vue

@@ -0,0 +1,111 @@
+<template>
+  <div>
+    <label>{{ $t("scanOptions.path") }}</label>
+    <b-form-input v-model="options.path" @change="update()"></b-form-input>
+
+    <label>{{ $t("scanOptions.threads") }}</label>
+    <b-form-input type="number" min="1" v-model="options.threads" @change="update()"></b-form-input>
+
+    <label>{{ $t("scanOptions.memThrottle") }}</label>
+    <b-form-input type="number" min="0" v-model="options.mem_throttle" @change="update()"></b-form-input>
+
+    <label>{{ $t("scanOptions.thumbnailQuality") }}</label>
+    <b-form-input type="number" min="1" max="31" v-model="options.thumbnail_quality" @change="update()"></b-form-input>
+
+    <label>{{ $t("scanOptions.thumbnailCount") }}</label>
+    <b-form-input type="number" min="0" max="1000" v-model="options.thumbnail_count" @change="update()"></b-form-input>
+
+    <label>{{ $t("scanOptions.thumbnailSize") }}</label>
+    <b-form-input type="number" min="100" v-model="options.thumbnail_size" @change="update()"></b-form-input>
+
+    <label>{{ $t("scanOptions.contentSize") }}</label>
+    <b-form-input type="number" min="0" v-model="options.content_size" @change="update()"></b-form-input>
+
+    <label>{{ $t("scanOptions.rewriteUrl") }}</label>
+    <b-form-input v-model="options.rewrite_url" @change="update()"></b-form-input>
+
+    <label>{{ $t("scanOptions.depth") }}</label>
+    <b-form-input type="number" min="0" v-model="options.depth" @change="update()"></b-form-input>
+
+    <label>{{ $t("scanOptions.archive") }}</label>
+    <b-form-select :options="['skip', 'list', 'shallow', 'recurse']" v-model="options.archive"
+                   @change="update()"></b-form-select>
+
+    <label>{{ $t("scanOptions.archivePassphrase") }}</label>
+    <b-form-input v-model="options.archive_passphrase" @change="update()"></b-form-input>
+
+    <label>{{ $t("scanOptions.ocrLang") }}</label>
+    <b-alert variant="danger" show v-if="selectedOcrLangs.length === 0 && !disableOcrLang">{{ $t("scanOptions.ocrLangAlert") }}</b-alert>
+    <b-checkbox-group :disabled="disableOcrLang" v-model="selectedOcrLangs" @input="onOcrLangChange">
+      <b-checkbox v-for="lang in ocrLangs" :key="lang" :value="lang">{{ lang }}</b-checkbox>
+    </b-checkbox-group>
+
+    <!--    <b-form-input readonly v-model="options.ocr_lang" @change="update()"></b-form-input>-->
+
+    <div style="height: 10px"></div>
+
+    <b-form-checkbox v-model="options.ocr_images" @change="update()">
+      {{ $t("scanOptions.ocrImages") }}
+    </b-form-checkbox>
+
+    <b-form-checkbox v-model="options.ocr_ebooks" @change="update()">
+      {{ $t("scanOptions.ocrEbooks") }}
+    </b-form-checkbox>
+
+    <label>{{ $t("scanOptions.exclude") }}</label>
+    <b-form-input v-model="options.exclude" @change="update()"
+                  :placeholder="$t('scanOptions.excludePlaceholder')"></b-form-input>
+
+    <div style="height: 10px"></div>
+
+    <b-form-checkbox v-model="options.fast" @change="update()">
+      {{ $t("scanOptions.fast") }}
+    </b-form-checkbox>
+
+    <b-form-checkbox v-model="options.checksums" @change="update()">
+      {{ $t("scanOptions.checksums") }}
+    </b-form-checkbox>
+
+    <b-form-checkbox v-model="options.read_subtitles" @change="update()">
+      {{ $t("scanOptions.readSubtitles") }}
+    </b-form-checkbox>
+
+    <label>{{ $t("scanOptions.memBuffer") }}</label>
+    <b-form-input type="number" min="0" v-model="options.mem_buffer" @change="update()"></b-form-input>
+
+    <label>{{ $t("scanOptions.treemapThreshold") }}</label>
+    <b-form-input type="number" min="0" v-model="options.treemap_threshold" @change="update()"></b-form-input>
+  </div>
+</template>
+
+<script>
+
+export default {
+  name: "ScanOptions",
+  props: ["options"],
+  data() {
+    return {
+      disableOcrLang: false,
+      selectedOcrLangs: []
+    }
+  },
+  computed: {
+    ocrLangs() {
+      return this.$store.state.sist2AdminInfo?.tesseract_langs || [];
+    }
+  },
+  methods: {
+    onOcrLangChange() {
+      this.options.ocr_lang = this.selectedOcrLangs.join("+");
+    },
+    update() {
+      this.disableOcrLang = this.options.ocr_images === false && this.options.ocr_ebooks === false;
+      this.$emit("change", this.options);
+    },
+  },
+  mounted() {
+    this.disableOcrLang = this.options.ocr_images === false && this.options.ocr_ebooks === false;
+    this.selectedOcrLangs = this.options.ocr_lang ? this.options.ocr_lang.split("+") : [];
+  }
+}
+</script>

sist2-admin/frontend/src/components/TaskListItem.vue

@@ -0,0 +1,57 @@
+<template>
+  <b-list-group-item>
+    <b-row style="height: 50px">
+      <b-col><h5>{{ task.display_name }}</h5></b-col>
+      <b-col class="shrink">
+        <router-link class="btn btn-link" :to="`/log/${task.id}`">{{ $t("logs") }}</router-link>
+      </b-col>
+      <b-col class="shrink">
+        <b-btn variant="link" @click="killTask(task.id)">{{ $t("kill") }}</b-btn>
+      </b-col>
+    </b-row>
+
+    <b-row>
+      <b-col>
+        <b-progress :max="task.progress.count">
+          <b-progress-bar :value="task.progress.done" :label-html="label" :striped="!task.progress.waiting"/>
+        </b-progress>
+      </b-col>
+    </b-row>
+
+  </b-list-group-item>
+</template>
+
+<script>
+import sist2AdminApi from "@/Sist2AdminApi";
+
+export default {
+  name: "TaskListItem",
+  props: ["task"],
+  computed: {
+    label() {
+
+      const count = this.task.progress.count;
+      const done = this.task.progress.done;
+
+      return `<span>${done}/${count}</span>`
+    }
+  },
+  methods: {
+    killTask(taskId) {
+      sist2AdminApi.killTask(taskId).then(() => {
+        this.$bvToast.toast(this.$t("killConfirmation"), {
+          title: this.$t("killConfirmationTitle"),
+          variant: "success",
+          toaster: "b-toaster-bottom-right"
+        });
+      });
+    }
+  }
+}
+</script>
+
+<style scoped>
+.shrink {
+  flex-grow: inherit;
+}
+</style>

sist2-admin/frontend/src/components/WebOptions.vue

@@ -0,0 +1,91 @@
+<template>
+  <div>
+    <label>{{ $t("webOptions.esUrl") }}</label>
+    <b-alert :variant="esTestOk ? 'success' : 'danger'" :show="showEsTestAlert" class="mt-1">
+      {{ esTestMessage }}
+    </b-alert>
+
+    <b-input-group>
+      <b-form-input v-model="options.es_url" @change="update()"></b-form-input>
+      <b-input-group-append>
+        <b-button variant="outline-primary" @click="testEs()">{{ $t("test") }}</b-button>
+      </b-input-group-append>
+    </b-input-group>
+
+    <b-form-checkbox v-model="options.es_insecure_ssl" :disabled="!this.options.es_url.startsWith('https')" @change="update()">
+      {{ $t("webOptions.esInsecure") }}
+    </b-form-checkbox>
+
+    <label>{{ $t("webOptions.esIndex") }}</label>
+    <b-form-input v-model="options.es_index" @change="update()"></b-form-input>
+
+    <label>{{ $t("webOptions.lang") }}</label>
+    <b-form-select v-model="options.lang" :options="['en', 'fr', 'zh-CN']" @change="update()"></b-form-select>
+
+    <label>{{ $t("webOptions.bind") }}</label>
+    <b-form-input v-model="options.bind" @change="update()"></b-form-input>
+
+    <label>{{ $t("webOptions.tagline") }}</label>
+    <b-form-textarea v-model="options.tagline" @change="update()"></b-form-textarea>
+
+    <label>{{ $t("webOptions.auth") }}</label>
+    <b-form-input v-model="options.auth" @change="update()"></b-form-input>
+
+    <label>{{ $t("webOptions.tagAuth") }}</label>
+    <b-form-input v-model="options.tag_auth" @change="update()"></b-form-input>
+
+    <br>
+    <h5>Auth0 options</h5>
+    <label>{{ $t("webOptions.auth0Audience") }}</label>
+    <b-form-input v-model="options.auth0_audience" @change="update()"></b-form-input>
+
+    <label>{{ $t("webOptions.auth0Domain") }}</label>
+    <b-form-input v-model="options.auth0_domain" @change="update()"></b-form-input>
+
+    <label>{{ $t("webOptions.auth0ClientId") }}</label>
+    <b-form-input v-model="options.auth0_client_id" @change="update()"></b-form-input>
+
+    <label>{{ $t("webOptions.auth0PublicKey") }}</label>
+    <b-textarea rows="10" v-model="options.auth0_public_key" @change="update()"></b-textarea>
+
+
+
+  </div>
+</template>
+
+<script>
+
+import sist2AdminApi from "@/Sist2AdminApi";
+
+export default {
+  name: "WebOptions",
+  props: ["options", "frontendName"],
+  data() {
+    return {
+      showEsTestAlert: false,
+      esTestOk: false,
+      esTestMessage: "",
+    }
+  },
+  methods: {
+    update() {
+      if (!this.options.es_url.startsWith("https")) {
+        this.options.es_insecure_ssl = false;
+      }
+
+      this.$emit("change", this.options);
+    },
+    testEs() {
+      sist2AdminApi.pingEs(this.options.es_url, this.options.es_insecure_ssl).then((resp) => {
+        this.showEsTestAlert = true;
+        this.esTestOk = resp.data.ok;
+        this.esTestMessage = resp.data.message;
+      });
+    }
+  }
+}
+</script>
+
+<style scoped>
+
+</style>

sist2-admin/frontend/src/components/icons/Sist2Icon.vue

@@ -0,0 +1,40 @@
+<template>
+  <svg
+      xmlns="http://www.w3.org/2000/svg"
+      width="27.868069mm"
+      height="7.6446671mm"
+      viewBox="0 0 27.868069 7.6446671"
+  >
+    <g transform="translate(-4.5018313,-4.1849793)">
+      <g
+          style="fill: currentColor;fill-opacity:1;stroke:none;stroke-width:0.26458332">
+        <path
+            d="m 6.3153296,11.829646 q -0.7717014,0 -1.8134983,-0.337619 v -0.916395 q 1.0128581,0.511252 1.803852,0.511252 0.5643067,0 0.901926,-0.236334 0.3376194,-0.236333 0.3376194,-0.63183 0,-0.3424428 -0.2845649,-0.5498376 Q 6.980922,9.4566645 6.3635609,9.3264399 L 5.9921796,9.2492698 Q 5.2301245,9.0949295 4.8732126,8.7428407 4.5211238,8.3859288 4.5211238,7.7733908 q 0,-0.7765245 0.5305447,-1.1961372 0.5305447,-0.4196126 1.5096409,-0.4196126 0.829579,0 1.6061036,0.3183268 V 7.3441319 Q 7.4101809,6.9004036 6.5854251,6.9004036 q -1.1671984,0 -1.1671984,0.7958171 0,0.2604492 0.1012858,0.4147895 0.1012858,0.1495171 0.3858507,0.2556261 0.2845649,0.1012858 0.8392253,0.2122179 l 0.3569119,0.067524 q 1.3408312,0.2652724 1.3408312,1.4614098 0,0.80064 -0.5691298,1.263661 -0.5691298,0.458197 -1.5578722,0.458197 z"
+            style="stroke-width:0.26458332"
+        />
+        <path
+            d="m 11.943927,5.3087694 q -0.144694,0 -0.144694,-0.144694 V 4.3296733 q 0,-0.144694 0.144694,-0.144694 h 0.694531 q 0.144694,0 0.144694,0.144694 v 0.8344021 q 0,0.144694 -0.144694,0.144694 z M 13.5645,11.728361 q -0.795817,0 -1.234722,-0.511253 -0.434082,-0.516075 -0.434082,-1.4469398 V 6.9823969 H 10.714028 V 6.2878656 h 2.069124 v 3.4823026 q 0,0.5884228 0.221864,0.8971028 0.221865,0.308681 0.6463,0.308681 h 1.036974 v 0.752409 z"
+            style="stroke-width:0.26458332"
+        />
+        <path
+            d="m 18.209178,11.829646 q -0.771701,0 -1.813498,-0.337619 v -0.916395 q 1.012858,0.511252 1.803852,0.511252 0.564306,0 0.901926,-0.236334 0.337619,-0.236333 0.337619,-0.63183 0,-0.3424428 -0.284565,-0.5498376 Q 18.87477,9.4566645 18.257409,9.3264399 l -0.371381,-0.07717 Q 17.123973,9.0949295 16.767061,8.7428407 16.414972,8.3859288 16.414972,7.7733908 q 0,-0.7765245 0.530545,-1.1961372 0.530545,-0.4196126 1.509641,-0.4196126 0.829579,0 1.606103,0.3183268 v 0.8681641 q -0.757232,-0.4437283 -1.581988,-0.4437283 -1.167198,0 -1.167198,0.7958171 0,0.2604492 0.101286,0.4147895 0.101286,0.1495171 0.385851,0.2556261 0.284565,0.1012858 0.839225,0.2122179 l 0.356912,0.067524 q 1.340831,0.2652724 1.340831,1.4614098 0,0.80064 -0.56913,1.263661 -0.56913,0.458197 -1.557872,0.458197 z"
+            style="stroke-width:0.26458332"
+        />
+        <path
+            d="m 25.207545,11.709068 q -0.993565,0 -1.408355,-0.40032 -0.409966,-0.405143 -0.409966,-1.3794164 V 6.9775737 H 21.947107 V 6.2878656 h 1.442117 V 4.8746874 l 0.887457,-0.3858507 v 1.7990289 h 2.016069 v 0.6897081 h -2.016069 v 2.9517579 q 0,0.5932454 0.226687,0.8344024 0.226687,0.236333 0.790994,0.236333 h 0.998388 v 0.709001 z"
+            style="stroke-width:0.26458332"
+        />
+        <path
+            d="m 27.995317,11.043476 q 0,-0.178456 0.120578,-0.299035 0.274919,-0.289388 0.651123,-0.684885 0.376205,-0.4003199 0.805464,-0.8681638 0.327973,-0.356912 0.491959,-0.5353679 0.16881,-0.1832791 0.255626,-0.2845649 0.09164,-0.1012858 0.178456,-0.2073948 0.255626,-0.3086805 0.405144,-0.5257215 0.15434,-0.2170411 0.250803,-0.4292589 0.168809,-0.3762045 0.168809,-0.7524089 0,-0.5980686 -0.352089,-0.935688 -0.356911,-0.3424425 -0.979096,-0.3424425 -0.863341,0 -1.938899,0.6414768 V 4.8361023 q 0.491959,-0.2363335 0.979096,-0.3569119 0.47749,-0.1205783 0.945334,-0.1205783 0.501606,0 0.940511,0.1350477 0.438905,0.1350478 0.766878,0.4244358 0.289388,0.2556261 0.463021,0.6270074 0.173633,0.3665582 0.173633,0.829579 0,0.4726671 -0.212218,0.9501574 -0.106109,0.2411567 -0.274919,0.4726671 -0.163986,0.2266873 -0.424435,0.540191 Q 31.270225,8.501684 31.077299,8.718725 30.884374,8.9357661 30.628748,9.2106847 30.445469,9.4084332 30.286305,9.5675966 30.131965,9.72676 29.958332,9.9003928 29.7847,10.069203 29.558012,10.300713 29.336148,10.5274 29.012998,10.869843 h 3.356901 v 0.819932 h -4.374582 z"
+            style="stroke-width:0.26458332"
+        />
+      </g>
+    </g>
+  </svg>
+</template>
+
+<script>
+export default {
+  name: "Sist2Icon"
+}
+</script>

sist2-admin/frontend/src/i18n/messages.js

@@ -0,0 +1,109 @@
+export default {
+    en: {
+        start: "Start",
+        stop: "Stop",
+        go: "Go",
+        online: "online",
+        offline: "offline",
+        delete: "Delete",
+        runNow: "Index now",
+        create: "Create",
+        test: "Test",
+
+        jobTitle: "job configuration",
+        tasks: "Tasks",
+        runningTasks: "Running tasks",
+        frontends: "Frontends",
+        jobDisabled: "There is no valid index for this job",
+        status: "Status",
+
+        taskHistory: "Task history",
+        taskName: "Task name",
+        taskStarted: "Started",
+        taskDuration: "Duration",
+        taskStatus: "Status",
+        logs: "Logs",
+        kill: "Kill",
+        killConfirmation: "SIGTERM signal sent to sist2 process",
+        killConfirmationTitle: "Confirmation",
+        follow: "Follow",
+        wholeFile: "Whole file",
+        logLevel: "Log level",
+        logMode: "Follow mode",
+        logFile: "Reading log file",
+
+        jobs: "Jobs",
+        newJobName: "New job name",
+        newJobHelp: "Create a new job to get started!",
+        newFrontendName: "New frontend name",
+        scanned: "last scan",
+        autoStart: "Start automatically",
+
+        runJobConfirmationTitle: "Task queued",
+        runJobConfirmation: "Check the Tasks page to monitor the status.",
+
+        extraQueryArgs: "Extra query arguments when launching from sist2-admin",
+        customUrl: "Custom URL when launching from sist2-admin",
+
+        selectJobs: "Select jobs",
+        webOptions: {
+            title: "Web options",
+            esUrl: "Elasticsearch URL",
+            esIndex: "Elasticsearch index name",
+            esInsecure: "Do not verify SSL connections to Elasticsearch.",
+            lang: "UI Language",
+            bind: "Listen address",
+            tagline: "Tagline in navbar",
+            auth: "Basic auth in user:password format",
+            tagAuth: "Basic auth in user:password format for tagging",
+        },
+        scanOptions: {
+            title: "Scanning options",
+            path: "Path",
+            threads: "Number of threads",
+            memThrottle: "Total memory threshold in MiB for scan throttling",
+            thumbnailQuality: "Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best",
+            thumbnailCount: "Number of thumbnails to generate. Set a value > 1 to create video previews, set to 0 to disable thumbnails.",
+            thumbnailSize: "Thumbnail size, in pixels",
+            contentSize: "Number of bytes to be extracted from text documents. Set to 0 to disable",
+            rewriteUrl: "Serve files from this url instead of from disk",
+            depth: "Scan up to this many subdirectories deep",
+            archive: "Archive file mode",
+            archivePassphrase: "Passphrase for encrypted archive files",
+            ocrLang: "Tesseract language",
+            ocrLangAlert: "You must select at least one language",
+            ocrEbooks: "Enable OCR'ing of ebook files",
+            ocrImages: "Enable OCR'ing of image files",
+            exclude: "Files that match this regex will not be scanned",
+            excludePlaceholder: "Exclude",
+            fast: "Only index file names & mime type",
+            checksums: "Calculate file checksums when scanning",
+            readSubtitles: "Read subtitles from media files",
+            memBuffer: "Maximum memory buffer size per thread in MiB for files inside archives",
+            treemapThreshold: "Relative size threshold for treemap"
+        },
+        indexOptions: {
+            title: "Indexing options",
+            threads: "Number of threads",
+            esUrl: "Elasticsearch URL",
+            esIndex: "Elasticsearch index name",
+            esInsecure: "Do not verify SSL connections to Elasticsearch.",
+            batchSize: "Index batch size",
+            script: "User script"
+        },
+        jobOptions: {
+            title: "Job options",
+            cron: "Job schedule",
+            scheduleEnabled: "Enable scheduled re-scan",
+            noJobAvailable: "No jobs available.",
+            desktopNotifications: "Desktop notifications"
+        },
+        frontendOptions: {
+            title: "Frontend options",
+            noJobSelectedWarning: "You must select at least one job to start this frontend"
+        },
+        notifications: {
+            indexCompleted: "Task completed for [$JOB$]"
+        }
+    }
+}

sist2-admin/frontend/src/main.js

@@ -0,0 +1,31 @@
+import Vue from 'vue'
+
+import { BootstrapVue, IconsPlugin } from 'bootstrap-vue'
+
+import "bootstrap/dist/css/bootstrap.min.css"
+import "bootstrap-vue/dist/bootstrap-vue.min.css"
+
+Vue.use(BootstrapVue);
+Vue.use(IconsPlugin);
+
+import App from './App.vue';
+import router from './router';
+import store from './store';
+import VueI18n from "vue-i18n";
+import messages from "@/i18n/messages";
+
+Vue.use(VueI18n);
+
+const i18n = new VueI18n({
+  locale: "en",
+  messages: messages
+});
+
+Vue.config.productionTip = false
+
+new Vue({
+  router,
+  store,
+  i18n,
+  render: h => h(App)
+}).$mount('#app')

sist2-admin/frontend/src/router/index.js

@@ -0,0 +1,45 @@
+import Vue from 'vue'
+import VueRouter from 'vue-router'
+import Home from '../views/Home.vue'
+import Job from "@/views/Job";
+import Tasks from "@/views/Tasks";
+import Frontend from "@/views/Frontend";
+import Tail from "@/views/Tail";
+
+Vue.use(VueRouter);
+
+const routes = [
+  {
+    path: "/",
+    name: "Home",
+    component: Home
+  },
+  {
+    path: "/job/:name",
+    name: "Job",
+    component: Job
+  },
+  {
+    path: "/task/",
+    name: "Tasks",
+    component: Tasks
+  },
+  {
+    path: "/frontend/:name",
+    name: "Frontend",
+    component: Frontend
+  },
+  {
+    path: "/log/:taskId",
+    name: "Tail",
+    component: Tail
+  },
+]
+
+const router = new VueRouter({
+  mode: "hash",
+  base: process.env.BASE_URL,
+  routes
+})
+
+export default router

sist2-admin/frontend/src/store/index.js

@@ -0,0 +1,63 @@
+import Vue from "vue";
+import Vuex from "vuex";
+
+Vue.use(Vuex);
+
+function saveBrowserSettings(state) {
+    const settings = {
+        jobDesktopNotificationMap: state.jobDesktopNotificationMap
+    };
+    localStorage.setItem("sist2-admin-settings", JSON.stringify(settings));
+
+    console.log("SAVED");
+    console.log(settings);
+}
+
+export default new Vuex.Store({
+    state: {
+        sist2AdminInfo: null,
+        jobDesktopNotificationMap: {}
+    },
+    mutations: {
+        setSist2AdminInfo: (state, payload) => state.sist2AdminInfo = payload,
+        setJobDesktopNotificationMap: (state, payload) => state.jobDesktopNotificationMap = payload,
+    },
+    actions: {
+        notify: async ({state}, notification) => {
+
+            if (!state.jobDesktopNotificationMap[notification.job]) {
+                console.log("pass");
+                return;
+            }
+
+            new Notification(notification.messageString.replace("$JOB$", notification.job));
+        },
+        setJobDesktopNotification: async ({state}, {job, enabled}) => {
+
+            if (enabled === true) {
+                const permission = await Notification.requestPermission()
+
+                if (permission !== "granted") {
+                    return false;
+                }
+            }
+
+            state.jobDesktopNotificationMap[job] = enabled;
+            saveBrowserSettings(state);
+
+            return true;
+        },
+        loadBrowserSettings({commit}) {
+            const settingString = localStorage.getItem("sist2-admin-settings");
+
+            if (!settingString) {
+                return;
+            }
+
+            const settings = JSON.parse(settingString);
+
+            commit("setJobDesktopNotificationMap", settings["jobDesktopNotificationMap"]);
+        }
+    },
+    modules: {}
+})

sist2-admin/frontend/src/util.js

@@ -0,0 +1,8 @@
+
+export function formatBindAddress(address) {
+    if (address.startsWith("0.0.0.0")) {
+        return address.slice("0.0.0.0".length)
+    }
+
+    return address
+}

sist2-admin/frontend/src/views/Frontend.vue

@@ -0,0 +1,129 @@
+<template>
+  <b-card>
+    <b-card-title>
+      {{ name }}
+      <small style="vertical-align: top">
+        <b-badge v-if="!loading && frontend.running" variant="success">{{ $t("online") }}</b-badge>
+        <b-badge v-else-if="!loading" variant="secondary">{{ $t("offline") }}</b-badge>
+      </small>
+    </b-card-title>
+
+    <div class="mb-3" v-if="!loading">
+      <b-button class="mr-1" :disabled="frontend.running || !valid" variant="success" @click="start()">{{
+          $t("start")
+        }}
+      </b-button>
+      <b-button class="mr-1" :disabled="!frontend.running" variant="danger" @click="stop()">{{
+          $t("stop")
+        }}
+      </b-button>
+      <b-button class="mr-1" :disabled="!frontend.running" variant="primary" :href="frontendUrl" target="_blank">
+        {{ $t("go") }}
+      </b-button>
+      <b-button variant="danger" @click="deleteFrontend()">{{ $t("delete") }}</b-button>
+    </div>
+
+
+    <b-progress v-if="loading" striped animated value="100"></b-progress>
+    <b-card-body v-else>
+
+      <h4>{{ $t("frontendOptions.title") }}</h4>
+      <b-card>
+        <b-form-checkbox v-model="frontend.auto_start" @change="update()">
+          {{ $t("autoStart") }}
+        </b-form-checkbox>
+
+        <label>{{ $t("extraQueryArgs") }}</label>
+        <b-form-input v-model="frontend.extra_query_args" @change="update()"></b-form-input>
+
+        <label>{{ $t("customUrl") }}</label>
+        <b-form-input v-model="frontend.custom_url" @change="update()" placeholder="http://"></b-form-input>
+
+        <br/>
+
+        <b-alert v-if="!valid" variant="warning" show>{{ $t("frontendOptions.noJobSelectedWarning") }}</b-alert>
+
+        <JobCheckboxGroup :frontend="frontend" @input="update()"></JobCheckboxGroup>
+      </b-card>
+
+      <br/>
+
+      <h4>{{ $t("webOptions.title") }}</h4>
+      <b-card>
+        <WebOptions :options="frontend.web_options" :frontend-name="$route.params.name" @change="update()"></WebOptions>
+      </b-card>
+    </b-card-body>
+
+  </b-card>
+</template>
+
+<script>
+
+import Sist2AdminApi from "@/Sist2AdminApi";
+import JobCheckboxGroup from "@/components/JobCheckboxGroup";
+import WebOptions from "@/components/WebOptions";
+
+export default {
+  name: 'Frontend',
+  components: {JobCheckboxGroup, WebOptions},
+  data() {
+    return {
+      loading: true,
+      frontend: null,
+    }
+  },
+  computed: {
+    valid() {
+      return !this.loading && this.frontend.jobs.length > 0;
+    },
+    frontendUrl() {
+      if (this.frontend.custom_url) {
+        return this.frontend.custom_url + this.args;
+      }
+
+      if (this.frontend.web_options.bind.startsWith("0.0.0.0")) {
+        return window.location.protocol + "//" + window.location.hostname + ":" + this.port + this.args;
+      }
+
+      return window.location.protocol + "//" + this.frontend.web_options.bind + this.args;
+    },
+    name() {
+      return this.$route.params.name;
+    },
+    port() {
+      return this.frontend.web_options.bind.split(":")[1]
+    },
+    args() {
+      const args = this.frontend.extra_query_args;
+      if (args !== "") {
+        return "#" + (args.startsWith("?") ? (args) : ("?" + args));
+      }
+      return "";
+    }
+  },
+  mounted() {
+    Sist2AdminApi.getFrontend(this.name).then(resp => {
+      this.frontend = resp.data;
+      this.loading = false;
+    });
+  },
+  methods: {
+    start() {
+      this.frontend.running = true;
+      Sist2AdminApi.startFrontend(this.name)
+    },
+    stop() {
+      this.frontend.running = false;
+      Sist2AdminApi.stopFrontend(this.name)
+    },
+    deleteFrontend() {
+      Sist2AdminApi.deleteFrontend(this.name).then(() => {
+        this.$router.push("/frontends");
+      });
+    },
+    update() {
+      Sist2AdminApi.updateFrontend(this.name, this.frontend);
+    },
+  }
+}
+</script>

sist2-admin/frontend/src/views/Home.vue

@@ -0,0 +1,122 @@
+<template>
+  <div>
+    <b-card>
+      <b-card-title>{{ $t("jobs") }}</b-card-title>
+      <b-row>
+        <b-col>
+          <b-input id="new-job" v-model="newJobName" :placeholder="$t('newJobName')"></b-input>
+          <b-popover
+              :show.sync="showHelp"
+              target="new-job"
+              placement="top"
+              triggers="manual"
+              variant="primary"
+              :content="$t('newJobHelp')"
+          ></b-popover>
+        </b-col>
+        <b-col>
+          <b-button variant="primary" @click="createJob()" :disabled="!jobNameValid(newJobName)">{{ $t("create") }}
+          </b-button>
+        </b-col>
+      </b-row>
+
+      <hr/>
+
+      <b-progress v-if="jobsLoading" striped animated value="100"></b-progress>
+      <b-list-group v-else>
+        <JobListItem v-for="job in jobs" :key="job.name" :job="job"></JobListItem>
+      </b-list-group>
+    </b-card>
+
+    <br/>
+
+    <b-card>
+
+      <b-card-title>{{ $t("frontends") }}</b-card-title>
+
+      <b-row>
+        <b-col>
+          <b-input v-model="newFrontendName" :placeholder="$t('newFrontendName')"></b-input>
+        </b-col>
+        <b-col>
+          <b-button variant="primary" @click="createFrontend()" :disabled="!frontendNameValid(newFrontendName)">
+            {{ $t("create") }}
+          </b-button>
+        </b-col>
+      </b-row>
+
+      <hr/>
+
+      <b-progress v-if="frontendsLoading" striped animated value="100"></b-progress>
+      <b-list-group v-else>
+        <FrontendListItem v-for="frontend in frontends"
+                          :key="frontend.name" :frontend="frontend"></FrontendListItem>
+      </b-list-group>
+
+    </b-card>
+  </div>
+</template>
+
+<script>
+import JobListItem from "@/components/JobListItem";
+import {formatBindAddress} from "@/util";
+import Sist2AdminApi from "@/Sist2AdminApi";
+import FrontendListItem from "@/components/FrontendListItem";
+
+export default {
+  name: "Jobs",
+  components: {JobListItem, FrontendListItem},
+  data() {
+    return {
+      jobsLoading: true,
+      newJobName: "",
+      jobs: [],
+
+      frontendsLoading: true,
+      frontends: [],
+      formatBindAddress,
+      newFrontendName: "",
+
+      showHelp: false
+    }
+  },
+  mounted() {
+    this.loading = true;
+    this.reload();
+  },
+  methods: {
+    jobNameValid(name) {
+      if (this.jobs.some(job => job.name === name)) {
+        return false;
+      }
+
+      return /^[a-zA-Z0-9-_,.; ]+$/.test(name);
+    },
+    frontendNameValid(name) {
+      if (this.frontends.some(job => job.name === name)) {
+        return false;
+      }
+
+      return /^[a-zA-Z0-9-_,.; ]+$/.test(name);
+    },
+    reload() {
+      Sist2AdminApi.getJobs().then(resp => {
+        this.jobs = resp.data;
+        this.jobsLoading = false;
+
+        this.showHelp = this.jobs.length === 0;
+      });
+      Sist2AdminApi.getFrontends().then(resp => {
+        this.frontends = resp.data;
+        this.frontendsLoading = false;
+      });
+    },
+    createJob() {
+      Sist2AdminApi.createJob(this.newJobName).then(this.reload);
+    },
+    createFrontend() {
+      Sist2AdminApi.createFrontend(this.newFrontendName).then(this.reload)
+    }
+  }
+}
+</script>

sist2-admin/frontend/src/views/Job.vue

@@ -0,0 +1,92 @@
+<template>
+  <b-card>
+    <b-card-title>
+      [{{ getName() }}]
+      {{ $t("jobTitle") }}
+    </b-card-title>
+
+    <div class="mb-3">
+      <b-button class="mr-1" variant="primary" @click="runJob()">{{ $t("runNow") }}</b-button>
+      <b-button variant="danger" @click="deleteJob()">{{ $t("delete") }}</b-button>
+    </div>
+
+    <div v-if="job">
+      {{ $t("status") }}: <code>{{ job.status }}</code>
+    </div>
+
+    <b-progress v-if="loading" striped animated value="100"></b-progress>
+    <b-card-body v-else>
+
+      <h4>{{ $t("jobOptions.title") }}</h4>
+      <b-card>
+        <JobOptions :job="job" @change="update"></JobOptions>
+      </b-card>
+
+      <br/>
+
+      <h4>{{ $t("scanOptions.title") }}</h4>
+      <b-card>
+        <ScanOptions :options="job.scan_options" @change="update()"></ScanOptions>
+      </b-card>
+
+      <br/>
+
+      <h4>{{ $t("indexOptions.title") }}</h4>
+      <b-card>
+        <IndexOptions :options="job.index_options" @change="update()"></IndexOptions>
+      </b-card>
+
+    </b-card-body>
+
+  </b-card>
+</template>
+
+<script>
+import ScanOptions from "@/components/ScanOptions";
+import Sist2AdminApi from "@/Sist2AdminApi";
+import IndexOptions from "@/components/IndexOptions";
+import JobOptions from "@/components/JobOptions";
+
+export default {
+  name: "Job",
+  components: {
+    IndexOptions,
+    ScanOptions,
+    JobOptions
+  },
+  data() {
+    return {
+      loading: true,
+      job: null
+    }
+  },
+  methods: {
+    getName() {
+      return this.$route.params.name;
+    },
+    update() {
+      Sist2AdminApi.updateJob(this.getName(), this.job);
+    },
+    runJob() {
+      Sist2AdminApi.runJob(this.getName()).then(() => {
+        this.$bvToast.toast(this.$t("runJobConfirmation"), {
+          title: this.$t("runJobConfirmationTitle"),
+          variant: "success",
+          toaster: "b-toaster-bottom-right"
+        });
+      });
+    },
+    deleteJob() {
+      Sist2AdminApi.deleteJob(this.getName()).then(() => {
+        this.$router.push("/");
+      })
+    }
+  },
+  mounted() {
+    Sist2AdminApi.getJob(this.getName()).then(resp => {
+      this.loading = false;
+      this.job = resp.data;
+    })
+  }
+}
+</script>

sist2-admin/frontend/src/views/Tail.vue

@@ -0,0 +1,168 @@
+<template>
+  <b-card>
+    <b-card-body>
+
+      <h4 class="mb-3">{{ taskId }} {{ $t("logs") }}</h4>
+
+      <div v-if="$store.state.sist2AdminInfo">
+        {{ $t("logFile") }}
+        <code>{{ $store.state.sist2AdminInfo.logs_folder }}/sist2-{{ taskId }}.log</code>
+        <br/>
+        <br/>
+      </div>
+
+      <b-row>
+        <b-col>
+          <span>{{ $t("logLevel") }}</span>
+          <b-select :options="levels.slice(0, -1)" v-model="logLevel" @input="connect()"></b-select>
+        </b-col>
+        <b-col>
+          <span>{{ $t("logMode") }}</span>
+          <b-select :options="modeOptions" v-model="mode" @input="connect()"></b-select>
+        </b-col>
+      </b-row>
+
+      <div id="log-tail-output" class="mt-3 ml-1"></div>
+
+    </b-card-body>
+  </b-card>
+</template>
+
+<script>
+
+export default {
+  name: "Tail",
+  data() {
+    return {
+      logLevel: "DEBUG",
+      levels: ["DEBUG", "INFO", "WARNING", "ERROR", "ADMIN", "FATAL"],
+      socket: null,
+      mode: "follow",
+      modeOptions: [
+        {
+          "text": this.$t('follow'),
+          "value": "follow"
+        },
+        {
+          "text": this.$t('wholeFile'),
+          "value": "wholeFile"
+        }
+      ]
+    }
+  },
+  computed: {
+    taskId: function () {
+      return this.$route.params.taskId;
+    }
+  },
+  methods: {
+    connect() {
+      let lineCount = 0;
+      const outputElem = document.getElementById("log-tail-output")
+      outputElem.replaceChildren();
+      if (this.socket !== null) {
+        this.socket.close();
+      }
+
+      const n = this.mode === "follow" ? 32 : 9999999999;
+      this.socket = new WebSocket(`ws://${window.location.host}/log/${this.taskId}?n=${n}`);
+      this.socket.onopen = () => {
+        this.socket.send("Hello from client");
+      }
+
+      this.socket.onmessage = e => {
+        let message;
+        try {
+          message = JSON.parse(e.data);
+        } catch {
+          console.error(e.data)
+          return;
+        }
+
+        if ("ping" in message) {
+          return;
+        }
+
+        if (message.level === undefined) {
+
+          if ("stderr" in message) {
+            message.level = "ERROR";
+            message.message = message["stderr"];
+          } else {
+            message.level = "ADMIN";
+            message.message = message["sist2-admin"];
+          }
+          message.datetime = ""
+          message.filepath = ""
+        }
+
+        if (this.levels.indexOf(message.level) < this.levels.indexOf(this.logLevel)) {
+          return;
+        }
+
+        const logLine = `${message.datetime} [${message.level} ${message.filepath}] ${message.message}`;
+
+        const span = document.createElement("span");
+        span.setAttribute("class", message.level);
+        span.appendChild(document.createTextNode(logLine));
+
+        outputElem.appendChild(span);
+        lineCount += 1;
+
+        if (this.mode === "follow" && lineCount >= n) {
+          outputElem.firstChild.remove();
+        }
+      }
+    }
+  },
+  mounted() {
+    this.connect()
+  }
+}
+
+</script>
+
+<style>
+#log-tail-output span {
+  display: block;
+}
+
+span.DEBUG {
+  color: #9E9E9E;
+}
+
+span.WARNING {
+  color: #FFB300;
+}
+
+span.INFO {
+  color: #039BE5;
+}
+
+span.ERROR {
+  color: #F4511E;
+}
+
+span.FATAL {
+  color: #F4511E;
+}
+
+span.ADMIN {
+  color: #ee05ff;
+}
+
+
+#log-tail-output {
+  font-size: 13px;
+  font-family: monospace;
+
+  padding: 6px;
+  background-color: #f5f5f5;
+  border: 1px solid #ccc;
+  border-radius: 4px;
+  margin: 3px;
+  white-space: pre;
+  color: #000;
+  overflow: hidden;
+}
+</style>

sist2-admin/frontend/src/views/Tasks.vue

@@ -0,0 +1,124 @@
+<template>
+  <div>
+
+    <b-card v-if="tasks.length > 0">
+      <h2>{{ $t("runningTasks") }}</h2>
+      <b-list-group>
+        <TaskListItem v-for="task in tasks" :key="task.id" :task="task"></TaskListItem>
+      </b-list-group>
+    </b-card>
+
+    <b-card class="mt-4">
+
+      <b-card-title>{{ $t("taskHistory") }}</b-card-title>
+
+      <br/>
+
+      <b-table
+          id="task-history"
+          :items="historyItems"
+          :fields="historyFields"
+          :current-page="historyCurrentPage"
+          :tbody-tr-class="rowClass"
+          :per-page="10"
+      >
+        <template #cell(logs)="data">
+          <router-link :to="`/log/${data.item.logs}`">{{ $t("logs") }}</router-link>
+        </template>
+
+      </b-table>
+
+      <b-pagination limit="20" v-model="historyCurrentPage" :total-rows="historyItems.length"
+                    :per-page="10"></b-pagination>
+
+    </b-card>
+  </div>
+</template>
+
+<script>
+import TaskListItem from "@/components/TaskListItem";
+import Sist2AdminApi from "@/Sist2AdminApi";
+import moment from "moment";
+
+export default {
+  name: 'Tasks',
+  components: {TaskListItem},
+  data() {
+    return {
+      loading: true,
+      tasks: [],
+      taskHistory: [],
+      timerId: null,
+      historyFields: [
+        {key: "name", label: this.$t("taskName")},
+        {key: "time", label: this.$t("taskStarted")},
+        {key: "duration", label: this.$t("taskDuration")},
+        {key: "status", label: this.$t("taskStatus")},
+        {key: "logs", label: this.$t("logs")},
+      ],
+      historyCurrentPage: 1,
+      historyItems: []
+    }
+  },
+  props: {
+    msg: String
+  },
+  mounted() {
+    this.loading = true;
+    this.update().then(() => this.loading = false);
+
+    this.timerId = window.setInterval(this.update, 1000);
+    this.updateHistory();
+  },
+  destroyed() {
+    if (this.timerId) {
+      window.clearInterval(this.timerId);
+    }
+  },
+  methods: {
+    rowClass(row) {
+      if (row.status === "failed") {
+        return "table-danger";
+      }
+      return null;
+    },
+    updateHistory() {
+      Sist2AdminApi.getTaskHistory().then(resp => {
+        this.historyItems = resp.data.map(row => ({
+          id: row.id,
+          name: row.name,
+          duration: this.taskDuration(row),
+          time: moment(row.started).format("dd, MMM Do YYYY, HH:mm:ss"),
+          logs: row.id,
+          status: row.return_code === 0 ? "ok" : "failed"
+        }));
+      });
+    },
+    update() {
+      return Sist2AdminApi.getTasks().then(resp => {
+        this.tasks = resp.data;
+      })
+    },
+    taskDuration(task) {
+      const start = moment(task.started);
+      const end = moment(task.ended);
+
+      let duration = moment.utc(end.diff(start)).format("HH[h] mm[m] ss[s]");
+
+      duration = duration.replace("00h ", "");
+      duration = duration.replace(/^00m /, "");
+      duration = duration.replace(/00s/, "<1s");
+      duration = duration.replace(/^0/, "");
+
+      return duration;
+    }
+  }
+}
+</script>
+
+<style scoped>
+#task-history {
+  font-family: monospace;
+  font-size: 12px;
+}
+</style>

sist2-admin/frontend/vue.config.js

@@ -0,0 +1,5 @@
+module.exports = {
+    publicPath: "",
+    filenameHashing: false,
+    productionSourceMap: false,
+};

sist2-admin/requirements.txt

@@ -0,0 +1,5 @@
+fastapi
+git+https://github.com/simon987/hexlib.git
+uvicorn
+websockets
+pycron

sist2-admin/sist2_admin/app.py

@@ -0,0 +1,392 @@
+import asyncio
+import os
+import signal
+from datetime import datetime
+from urllib.parse import urlparse
+
+import requests
+import uvicorn
+from fastapi import FastAPI, HTTPException
+from hexlib.db import PersistentState
+from requests import ConnectionError
+from requests.exceptions import SSLError
+from starlette.middleware.cors import CORSMiddleware
+from starlette.responses import RedirectResponse
+from starlette.staticfiles import StaticFiles
+from starlette.websockets import WebSocket
+from websockets.exceptions import ConnectionClosed
+
+import cron
+from config import LOG_FOLDER, logger, WEBSERVER_PORT, DATA_FOLDER, SIST2_BINARY
+from jobs import Sist2Job, Sist2ScanTask, TaskQueue, Sist2IndexTask, JobStatus
+from notifications import Subscribe, Notifications
+from sist2 import Sist2
+from state import PickleTable, RUNNING_FRONTENDS, TESSERACT_LANGS, DB_SCHEMA_VERSION
+from web import Sist2Frontend
+
+VERSION = "1.0"
+
+sist2 = Sist2(SIST2_BINARY, DATA_FOLDER)
+db = PersistentState(table_factory=PickleTable, dbfile=os.path.join(DATA_FOLDER, "state.db"))
+notifications = Notifications()
+task_queue = TaskQueue(sist2, db, notifications)
+
+app = FastAPI()
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_credentials=True,
+    allow_origins=["*"],
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+app.mount("/ui/", StaticFiles(directory="./frontend/dist", html=True), name="static")
+
+
+@app.get("/")
+async def home():
+    return RedirectResponse("ui")
+
+
+@app.get("/api")
+async def api():
+    return {
+        "version": VERSION,
+        "tesseract_langs": TESSERACT_LANGS,
+        "logs_folder": LOG_FOLDER
+    }
+
+
+@app.get("/api/job/{name:str}")
+async def get_job(name: str):
+    row = db["jobs"][name]
+    if row:
+        return row["job"]
+    raise HTTPException(status_code=404)
+
+
+@app.get("/api/frontend/{name:str}")
+async def get_frontend(name: str):
+    row = db["frontends"][name]
+    if row:
+        frontend = row["frontend"]
+        frontend: Sist2Frontend
+        frontend.running = frontend.name in RUNNING_FRONTENDS
+        return frontend
+    raise HTTPException(status_code=404)
+
+
+@app.get("/api/job/")
+async def get_jobs():
+    return [row["job"] for row in db["jobs"]]
+
+
+@app.put("/api/job/{name:str}")
+async def update_job(name: str, job: Sist2Job):
+    # TODO: Check etag
+
+    job.last_modified = datetime.now()
+    row = db["jobs"][name]
+    if not row:
+        raise HTTPException(status_code=404)
+
+    args_that_trigger_full_scan = [
+        "path",
+        "thumbnail_count",
+        "thumbnail_quality",
+        "thumbnail_size",
+        "content_size",
+        "depth",
+        "archive",
+        "archive_passphrase",
+        "ocr_lang",
+        "ocr_images",
+        "ocr_ebooks",
+        "fast",
+        "checksums",
+        "read_subtitles",
+    ]
+    for arg in args_that_trigger_full_scan:
+        if getattr(row["job"].scan_options, arg) != getattr(job.scan_options, arg):
+            job.do_full_scan = True
+
+    db["jobs"][name] = {"job": job}
+
+
+@app.put("/api/frontend/{name:str}")
+async def update_frontend(name: str, frontend: Sist2Frontend):
+    db["frontends"][name] = {"frontend": frontend}
+
+    # TODO: Check etag
+
+    return "ok"
+
+
+@app.get("/api/task/")
+async def get_tasks():
+    return list(map(lambda t: t.json(), task_queue.tasks()))
+
+
+@app.get("/api/task/history")
+async def task_history():
+    return list(db["task_done"].sql("ORDER BY started DESC"))
+
+
+@app.post("/api/task/{task_id:str}/kill")
+async def kill_job(task_id: str):
+    return task_queue.kill_task(task_id)
+
+
+def _run_job(job: Sist2Job):
+    job.last_modified = datetime.now()
+    if job.status == JobStatus("created"):
+        job.status = JobStatus("started")
+    db["jobs"][job.name] = {"job": job}
+
+    scan_task = Sist2ScanTask(job, f"Scan [{job.name}]")
+    index_task = Sist2IndexTask(job, f"Index [{job.name}]", depends_on=scan_task)
+
+    task_queue.submit(scan_task)
+    task_queue.submit(index_task)
+
+
+@app.get("/api/job/{name:str}/run")
+async def run_job(name: str):
+    row = db["jobs"][name]
+    if not row:
+        raise HTTPException(status_code=404)
+
+    _run_job(row["job"])
+
+    return "ok"
+
+
+@app.delete("/api/job/{name:str}")
+async def delete_job(name: str):
+    row = db["jobs"][name]
+    if row:
+        del db["jobs"][name]
+    else:
+        raise HTTPException(status_code=404)
+
+
+@app.delete("/api/frontend/{name:str}")
+async def delete_frontend(name: str):
+    if name in RUNNING_FRONTENDS:
+        os.kill(RUNNING_FRONTENDS[name], signal.SIGTERM)
+        del RUNNING_FRONTENDS[name]
+
+    row = db["frontends"][name]
+    if row:
+        del db["frontends"][name]
+    else:
+        raise HTTPException(status_code=404)
+
+
+@app.post("/api/job/{name:str}")
+async def create_job(name: str):
+    if db["jobs"][name]:
+        raise ValueError("Job with the same name already exists")
+
+    job = Sist2Job.create_default(name)
+    db["jobs"][name] = {"job": job}
+
+    return job
+
+
+@app.post("/api/frontend/{name:str}")
+async def create_frontend(name: str):
+    if db["frontend"][name]:
+        raise ValueError("Frontend with the same name already exists")
+
+    frontend = Sist2Frontend.create_default(name)
+    db["frontends"][name] = {"frontend": frontend}
+
+    return frontend
+
+
+@app.get("/api/ping_es")
+async def ping_es(url: str, insecure: bool):
+    return check_es_version(url, insecure)
+
+
+def check_es_version(es_url: str, insecure: bool):
+    try:
+        url = urlparse(es_url)
+        if url.username:
+            auth = (url.username, url.password)
+            es_url = f"{url.scheme}://{url.hostname}:{url.port}"
+        else:
+            auth = None
+        r = requests.get(es_url, verify=insecure, auth=auth)
+    except SSLError:
+        return {
+            "ok": False,
+            "message": "Invalid SSL certificate"
+        }
+    except ConnectionError as e:
+        return {
+            "ok": False,
+            "message": "Connection refused"
+        }
+    except ValueError as e:
+        return {
+            "ok": False,
+            "message": str(e)
+        }
+
+    if r.status_code == 401:
+        return {
+            "ok": False,
+            "message": "Authentication failure"
+        }
+
+    try:
+        return {
+            "ok": True,
+            "message": "Elasticsearch version " + r.json()["version"]["number"]
+        }
+    except:
+        return {
+            "ok": False,
+            "message": "Could not read version"
+        }
+
+
+def start_frontend_(frontend: Sist2Frontend):
+    frontend.web_options.indices = list(map(lambda j: db["jobs"][j]["job"].last_index, frontend.jobs))
+
+    pid = sist2.web(frontend.web_options, frontend.name)
+    RUNNING_FRONTENDS[frontend.name] = pid
+
+
+@app.post("/api/frontend/{name:str}/start")
+async def start_frontend(name: str):
+    row = db["frontends"][name]
+    if not row:
+        raise HTTPException(status_code=404)
+
+    start_frontend_(row["frontend"])
+
+
+@app.post("/api/frontend/{name:str}/stop")
+async def stop_frontend(name: str):
+    if name in RUNNING_FRONTENDS:
+        os.kill(RUNNING_FRONTENDS[name], signal.SIGTERM)
+        del RUNNING_FRONTENDS[name]
+
+
+@app.get("/api/frontend/")
+async def get_frontends():
+    res = []
+    for row in db["frontends"]:
+        frontend = row["frontend"]
+        frontend: Sist2Frontend
+        frontend.running = frontend.name in RUNNING_FRONTENDS
+        res.append(frontend)
+    return res
+
+
+def tail(filepath: str, n: int):
+    with open(filepath) as file:
+
+        reached_eof = False
+        buffer = []
+
+        line = ""
+        while True:
+            tmp = file.readline()
+            if tmp:
+                line += tmp
+
+                if line.endswith("\n"):
+
+                    if reached_eof:
+                        yield line
+                    else:
+                        if len(buffer) > n:
+                            buffer.pop(0)
+                        buffer.append(line)
+                line = ""
+            else:
+                if not reached_eof:
+                    reached_eof = True
+                    yield from buffer
+                yield None
+
+
+@app.websocket("/notifications")
+async def ws_tail_log(websocket: WebSocket):
+    await websocket.accept()
+
+    try:
+        await websocket.receive_text()
+
+        async with Subscribe(notifications) as ob:
+            async for notification in ob.notifications():
+                await websocket.send_json(notification)
+                print(notification)
+
+    except ConnectionClosed:
+        return
+
+
+@app.websocket("/log/{task_id}")
+async def ws_tail_log(websocket: WebSocket, task_id: str, n: int):
+    log_file = os.path.join(LOG_FOLDER, f"sist2-{task_id}.log")
+
+    await websocket.accept()
+
+    try:
+        await websocket.receive_text()
+    except ConnectionClosed:
+        return
+
+    while True:
+        for line in tail(log_file, n):
+
+            try:
+                if line:
+                    await websocket.send_text(line)
+                else:
+                    await websocket.send_json({"ping": ""})
+                    await asyncio.sleep(0.1)
+            except ConnectionClosed:
+                return
+
+
+def main():
+    uvicorn.run(app, port=WEBSERVER_PORT, host="0.0.0.0")
+
+
+def initialize_db():
+    db["sist2_admin"]["info"] = {"version": DB_SCHEMA_VERSION}
+
+    frontend = Sist2Frontend.create_default("default")
+    db["frontends"]["default"] = {"frontend": frontend}
+
+    logger.info("Initialized database.")
+
+
+def start_frontends():
+    for row in db["frontends"]:
+        frontend: Sist2Frontend = row["frontend"]
+        if frontend.auto_start and len(frontend.jobs) > 0:
+            start_frontend_(frontend)
+
+
+if __name__ == '__main__':
+
+    if not db["sist2_admin"]["info"]:
+        initialize_db()
+    elif db["sist2_admin"]["info"]["version"] != DB_SCHEMA_VERSION:
+        print("Database has incompatible schema version! Delete state.db to continue.")
+        exit(-1)
+
+    start_frontends()
+    cron.initialize(db, _run_job)
+
+    logger.info("Started sist2-admin. Hello!")
+
+    main()

sist2-admin/sist2_admin/config.py

@@ -0,0 +1,30 @@
+import os
+import logging
+import sys
+from logging import StreamHandler
+from logging.handlers import RotatingFileHandler
+
+MAX_LOG_SIZE = 1 * 1024 * 1024
+
+SIST2_BINARY = os.environ.get("SIST2_BINARY", "/root/sist2")
+DATA_FOLDER = os.environ.get("DATA_FOLDER", "/sist2-admin/")
+LOG_FOLDER = os.path.join(DATA_FOLDER, "logs")
+WEBSERVER_PORT = 8080
+
+os.makedirs(LOG_FOLDER, exist_ok=True)
+os.makedirs(DATA_FOLDER, exist_ok=True)
+
+logger = logging.Logger("sist2-admin")
+
+_log_file = os.path.join(LOG_FOLDER, "sist2-admin.log")
+_log_fmt = "%(asctime)s [%(levelname)s] %(message)s"
+_log_formatter = logging.Formatter(_log_fmt, datefmt='%Y-%m-%d %H:%M:%S')
+
+console_handler = StreamHandler(sys.stdout)
+console_handler.setFormatter(_log_formatter)
+
+file_handler = RotatingFileHandler(_log_file, mode="a", maxBytes=MAX_LOG_SIZE, backupCount=1)
+file_handler.setFormatter(_log_formatter)
+
+logger.addHandler(console_handler)
+logger.addHandler(file_handler)

sist2-admin/sist2_admin/cron.py

@@ -0,0 +1,33 @@
+from threading import Thread
+
+import pycron
+import time
+
+from hexlib.db import PersistentState
+
+from config import logger
+from jobs import Sist2Job
+
+
+def _check_schedule(db: PersistentState, run_job):
+    for job in (row["job"] for row in db["jobs"]):
+        job: Sist2Job
+
+        if job.schedule_enabled:
+            if pycron.is_now(job.cron_expression):
+                logger.info(f"Submit scan task to queue for [{job.name}]")
+                run_job(job)
+
+
+def _cron_thread(db, run_job):
+    time.sleep(60 - (time.time() % 60))
+    start = time.time()
+
+    while True:
+        _check_schedule(db, run_job)
+        time.sleep(60 - ((time.time() - start) % 60))
+
+
+def initialize(db, run_job):
+    t = Thread(target=_cron_thread, args=(db, run_job), daemon=True, name="timer")
+    t.start()

sist2-admin/sist2_admin/jobs.py

@@ -0,0 +1,315 @@
+import json
+import logging
+import os.path
+import shutil
+import signal
+import uuid
+from datetime import datetime
+from enum import Enum
+from hashlib import md5
+from logging import FileHandler
+from threading import Lock, Thread
+from time import sleep
+from uuid import uuid4, UUID
+
+from hexlib.db import PersistentState
+from pydantic import BaseModel, validator
+
+from config import logger, LOG_FOLDER
+from notifications import Notifications
+from sist2 import ScanOptions, IndexOptions, Sist2, Sist2Index
+from state import RUNNING_FRONTENDS
+from web import Sist2Frontend
+
+
+class JobStatus(Enum):
+    CREATED = "created"
+    STARTED = "started"
+    INDEXED = "indexed"
+    FAILED = "failed"
+
+
+class Sist2Job(BaseModel):
+    name: str
+    scan_options: ScanOptions
+    index_options: IndexOptions
+
+    cron_expression: str
+    schedule_enabled: bool = False
+
+    previous_index: str = None
+    last_index: str = None
+    last_index_date: datetime = None
+    status: JobStatus = JobStatus("created")
+    last_modified: datetime
+    etag: str = None
+    do_full_scan: bool = False
+
+    def __init__(self, **kwargs):
+        super().__init__(**kwargs)
+
+    @staticmethod
+    def create_default(name: str):
+        return Sist2Job(
+            name=name,
+            scan_options=ScanOptions(path="/"),
+            index_options=IndexOptions(),
+            last_modified=datetime.now(),
+            cron_expression="0 0 * * *"
+        )
+
+    @validator("etag", always=True)
+    def validate_etag(cls, value, values):
+        s = values["name"] + values["scan_options"].json() + values["index_options"].json() + values["cron_expression"]
+        return md5(s.encode()).hexdigest()
+
+
+class Sist2TaskProgress:
+
+    def __init__(self, done: int = 0, count: int = 0, index_size: int = 0, tn_size: int = 0, waiting: bool = False):
+        self.done = done
+        self.count = count
+        self.index_size = index_size
+        self.store_size = tn_size
+        self.waiting = waiting
+
+    def percent(self):
+        return (self.done / self.count) if self.count else 0
+
+
+class Sist2Task:
+
+    def __init__(self, job: Sist2Job, display_name: str, depends_on: uuid.UUID = None):
+        self.job = job
+        self.display_name = display_name
+
+        self.progress = Sist2TaskProgress()
+        self.id = uuid4()
+        self.pid = None
+        self.started = None
+        self.ended = None
+        self.depends_on = depends_on
+
+        self._logger = logging.Logger(name=f"{self.id}")
+        self._logger.addHandler(FileHandler(os.path.join(LOG_FOLDER, f"sist2-{self.id}.log")))
+
+    def json(self):
+        return {
+            "id": self.id,
+            "job": self.job,
+            "display_name": self.display_name,
+            "progress": self.progress,
+            "started": self.started,
+            "ended": self.ended,
+            "depends_on": self.depends_on,
+        }
+
+    def log_callback(self, log_json):
+
+        if "progress" in log_json:
+            self.progress = Sist2TaskProgress(**log_json["progress"])
+        elif self._logger:
+            self._logger.info(json.dumps(log_json))
+
+    def run(self, sist2: Sist2, db: PersistentState):
+        self.started = datetime.now()
+
+        logger.info(f"Started task {self.display_name}")
+
+
+class Sist2ScanTask(Sist2Task):
+
+    def run(self, sist2: Sist2, db: PersistentState):
+        super().run(sist2, db)
+
+        self.job.scan_options.name = self.job.name
+
+        if self.job.last_index and os.path.exists(self.job.last_index) and not self.job.do_full_scan:
+            self.job.scan_options.incremental = self.job.last_index
+        else:
+            self.job.scan_options.incremental = None
+
+        def set_pid(pid):
+            self.pid = pid
+
+        return_code = sist2.scan(self.job.scan_options, logs_cb=self.log_callback, set_pid_cb=set_pid)
+        self.ended = datetime.now()
+
+        if return_code != 0:
+            self._logger.error(json.dumps({"sist2-admin": f"Process returned non-zero exit code ({return_code})"}))
+            logger.info(f"Task {self.display_name} failed ({return_code})")
+        else:
+            index = Sist2Index(self.job.scan_options.output)
+
+            # Save latest index
+            self.job.previous_index = self.job.last_index
+
+            self.job.last_index = index.path
+            self.job.last_index_date = datetime.now()
+            self.job.do_full_scan = False
+            db["jobs"][self.job.name] = {"job": self.job}
+            self._logger.info(json.dumps({"sist2-admin": f"Save last_index={self.job.last_index}"}))
+
+        logger.info(f"Completed {self.display_name} ({return_code=})")
+
+        return return_code
+
+
+class Sist2IndexTask(Sist2Task):
+
+    def __init__(self, job: Sist2Job, display_name: str, depends_on: Sist2Task):
+        super().__init__(job, display_name, depends_on=depends_on.id)
+
+    def run(self, sist2: Sist2, db: PersistentState):
+        super().run(sist2, db)
+
+        self.job.index_options.path = self.job.scan_options.output
+
+        return_code = sist2.index(self.job.index_options, logs_cb=self.log_callback)
+        self.ended = datetime.now()
+
+        duration = self.ended - self.started
+
+        ok = return_code == 0
+
+        if ok:
+            # Remove old index
+            if self.job.previous_index is not None:
+                self._logger.info(json.dumps({"sist2-admin": f"Remove {self.job.previous_index=}"}))
+                try:
+                    shutil.rmtree(self.job.previous_index)
+                except FileNotFoundError:
+                    pass
+
+            self.restart_running_frontends(db, sist2)
+
+        # Update status
+        self.job.status = JobStatus("indexed") if ok else JobStatus("failed")
+        db["jobs"][self.job.name] = {"job": self.job}
+
+        self._logger.info(json.dumps({"sist2-admin": f"Sist2Scan task finished {return_code=}, {duration=}"}))
+
+        logger.info(f"Completed {self.display_name} ({return_code=})")
+
+        return return_code
+
+    def restart_running_frontends(self, db: PersistentState, sist2: Sist2):
+        for frontend_name, pid in RUNNING_FRONTENDS.items():
+            frontend = db["frontends"][frontend_name]["frontend"]
+            frontend: Sist2Frontend
+
+            os.kill(pid, signal.SIGTERM)
+            try:
+                os.wait()
+            except ChildProcessError:
+                pass
+
+            frontend.web_options.indices = map(lambda j: db["jobs"][j]["job"].last_index, frontend.jobs)
+
+            pid = sist2.web(frontend.web_options, frontend.name)
+            RUNNING_FRONTENDS[frontend_name] = pid
+
+            self._logger.info(json.dumps({"sist2-admin": f"Restart frontend {pid=} {frontend_name=}"}))
+
+
+class TaskQueue:
+    def __init__(self, sist2: Sist2, db: PersistentState, notifications: Notifications):
+        self._lock = Lock()
+
+        self._sist2 = sist2
+        self._db = db
+        self._notifications = notifications
+
+        self._tasks = {}
+        self._queue = []
+        self._sem = 0
+
+        self._thread = Thread(target=self._check_new_task, daemon=True)
+        self._thread.start()
+
+    def _tasks_failed(self):
+        done = set()
+
+        for row in self._db["task_done"].sql("WHERE return_code != 0"):
+            done.add(uuid.UUID(row["id"]))
+
+        return done
+
+    def _tasks_done(self):
+
+        done = set()
+
+        for row in self._db["task_done"]:
+            done.add(uuid.UUID(row["id"]))
+
+        return done
+
+    def _check_new_task(self):
+        while True:
+            with self._lock:
+                for task in list(self._queue):
+                    task: Sist2Task
+
+                    if self._sem >= 1:
+                        break
+
+                    if not task.depends_on or task.depends_on in self._tasks_done():
+                        self._queue.remove(task)
+
+                        if task.depends_on in self._tasks_failed():
+                            # The task which we depend on failed, continue
+                            continue
+
+                        self._sem += 1
+
+                        t = Thread(target=self._run_task, args=(task,))
+
+                        self._tasks[task.id] = {
+                            "task": task,
+                            "thread": t,
+                        }
+
+                        t.start()
+                    break
+            sleep(1)
+
+    def tasks(self):
+        return list(map(lambda t: t["task"], self._tasks.values()))
+
+    def kill_task(self, task_id):
+
+        task = self._tasks.get(UUID(task_id))
+
+        if task:
+            pid = task["task"].pid
+            logger.info(f"Killing task {task_id} (pid={pid})")
+            os.kill(pid, signal.SIGTERM)
+            return True
+
+        return False
+
+    def _run_task(self, task: Sist2Task):
+        task_result = task.run(self._sist2, self._db)
+
+        with self._lock:
+            del self._tasks[task.id]
+            self._sem -= 1
+
+            self._db["task_done"][task.id] = {
+                "ended": task.ended,
+                "started": task.started,
+                "name": task.display_name,
+                "return_code": task_result
+            }
+        if isinstance(task, Sist2IndexTask):
+            self._notifications.notify({
+                "message": "notifications.indexCompleted",
+                "job": task.job.name
+            })
+
+    def submit(self, task: Sist2Task):
+
+        logger.info(f"Submitted task to queue {task.display_name}")
+
+        with self._lock:
+            self._queue.append(task)

sist2-admin/sist2_admin/notifications.py

@@ -0,0 +1,40 @@
+import asyncio
+from typing import List
+
+
+class Notifications:
+    def __init__(self):
+        self._subscribers: List[Subscribe] = []
+
+    def subscribe(self, ob):
+        self._subscribers.append(ob)
+
+    def unsubscribe(self, ob):
+        self._subscribers.remove(ob)
+
+    def notify(self, notification: dict):
+        for ob in self._subscribers:
+            ob.notify(notification)
+
+
+class Subscribe:
+    def __init__(self, notifications: Notifications):
+        self._queue = []
+        self._notifications = notifications
+
+    async def __aenter__(self):
+        self._notifications.subscribe(self)
+        return self
+
+    async def __aexit__(self, exc_type, exc_val, exc_tb):
+        self._notifications.unsubscribe(self)
+
+    def notify(self, notification: dict):
+        self._queue.append(notification)
+
+    async def notifications(self):
+        while True:
+            try:
+                yield self._queue.pop(0)
+            except IndexError:
+                await asyncio.sleep(0.1)

sist2-admin/sist2_admin/sist2.py

@@ -0,0 +1,328 @@
+import datetime
+import json
+import logging
+import os.path
+import traceback
+from datetime import datetime
+from io import TextIOWrapper
+from logging import FileHandler
+from subprocess import Popen, PIPE
+from tempfile import NamedTemporaryFile
+from threading import Thread
+from typing import List
+
+from pydantic import BaseModel
+
+from config import logger, LOG_FOLDER
+
+
+class Sist2Version:
+    def __init__(self, version: str):
+        self._version = version
+
+        self.major, self.minor, self.patch = [int(x) for x in version.split(".")]
+
+    def __str__(self):
+        return f"{self.major}.{self.minor}.{self.patch}"
+
+
+class WebOptions(BaseModel):
+    indices: List[str] = []
+    es_url: str = "http://elasticsearch:9200"
+    es_insecure_ssl: bool = False
+    es_index: str = "sist2"
+    bind: str = "0.0.0.0:4090"
+    auth: str = None
+    tag_auth: str = None
+    tagline: str = "Lightning-fast file system indexer and search tool"
+    dev: bool = False
+    lang: str = "en"
+    auth0_audience: str = None
+    auth0_domain: str = None
+    auth0_client_id: str = None
+    auth0_public_key: str = None
+    auth0_public_key_file: str = None
+
+    def __init__(self, **kwargs):
+        super().__init__(**kwargs)
+
+    def args(self):
+        args = ["web", f"--es-url={self.es_url}", f"--bind={self.bind}",
+                f"--tagline={self.tagline}", f"--lang={self.lang}"]
+
+        if self.auth0_audience:
+            args.append(f"--auth0-audience={self.auth0_audience}")
+        if self.auth0_domain:
+            args.append(f"--auth0-domain={self.auth0_domain}")
+        if self.auth0_client_id:
+            args.append(f"--auth0-client-id={self.auth0_client_id}")
+        if self.auth0_public_key_file:
+            args.append(f"--auth0-public-key-file={self.auth0_public_key_file}")
+        if self.es_insecure_ssl:
+            args.append(f"--es-insecure-ssl")
+        if self.auth:
+            args.append(f"--auth={self.auth}")
+        if self.tag_auth:
+            args.append(f"--tag_auth={self.tag_auth}")
+        if self.dev:
+            args.append(f"--dev")
+
+        args.extend(self.indices)
+
+        return args
+
+
+class IndexOptions(BaseModel):
+    path: str = None
+    threads: int = 1
+    es_url: str = "http://elasticsearch:9200"
+    es_insecure_ssl: bool = False
+    es_index: str = "sist2"
+    incremental_index: bool = False
+    script: str = ""
+    script_file: str = None
+    batch_size: int = 100
+
+    def __init__(self, **kwargs):
+        super().__init__(**kwargs)
+
+    def args(self):
+
+        args = ["index", self.path, f"--threads={self.threads}", f"--es-url={self.es_url}",
+                f"--es-index={self.es_index}", f"--batch-size={self.batch_size}"]
+
+        if self.script_file:
+            args.append(f"--script-file={self.script_file}")
+        if self.es_insecure_ssl:
+            args.append(f"--es-insecure-ssl")
+        if self.incremental_index:
+            args.append(f"--incremental-index")
+
+        return args
+
+
+ARCHIVE_SKIP = "skip"
+ARCHIVE_LIST = "list"
+ARCHIVE_SHALLOW = "shallow"
+ARCHIVE_RECURSE = "recurse"
+
+
+class ScanOptions(BaseModel):
+    path: str
+    threads: int = 1
+    mem_throttle: int = 0
+    thumbnail_quality: float = 1.0
+    thumbnail_size: int = 500
+    thumbnail_count: int = 1
+    content_size: int = 32768
+    depth: int = -1
+    archive: str = ARCHIVE_RECURSE
+    archive_passphrase: str = None
+    ocr_lang: bool = None
+    ocr_images: bool = False
+    ocr_ebooks: bool = False
+    exclude: str = None
+    fast: bool = False
+    treemap_threshold: float = 0.0005
+    mem_buffer: int = 2000
+    read_subtitles: bool = False
+    fast_epub: bool = False
+    checksums: bool = False
+    incremental: str = None
+    output: str = None
+    name: str = None
+    rewrite_url: str = None
+    list_file: str = None
+
+    def __init__(self, **kwargs):
+        super().__init__(**kwargs)
+
+    def args(self):
+        args = ["scan", self.path, f"--threads={self.threads}", f"--mem-throttle={self.mem_throttle}",
+                f"--thumbnail-quality={self.thumbnail_quality}", f"--thumbnail-count={self.thumbnail_count}",
+                f"--content-size={self.content_size}", f"--output={self.output}", f"--depth={self.depth}",
+                f"--archive={self.archive}", f"--mem-buffer={self.mem_buffer}"]
+
+        if self.incremental:
+            args.append(f"--incremental={self.incremental}")
+        if self.rewrite_url:
+            args.append(f"--rewrite-url={self.rewrite_url}")
+        if self.name:
+            args.append(f"--name={self.name}")
+        if self.archive_passphrase:
+            args.append(f"--archive-passphrase={self.archive_passphrase}")
+        if self.ocr_lang:
+            args.append(f"--ocr-lang={self.ocr_lang}")
+        if self.ocr_ebooks:
+            args.append(f"--ocr-ebooks")
+        if self.ocr_images:
+            args.append(f"--ocr-images")
+        if self.exclude:
+            args.append(f"--exclude={self.exclude}")
+        if self.fast:
+            args.append(f"--fast")
+        if self.treemap_threshold:
+            args.append(f"--treemap-threshold={self.treemap_threshold}")
+        if self.read_subtitles:
+            args.append(f"--read-subtitles")
+        if self.fast_epub:
+            args.append(f"--fast-epub")
+        if self.checksums:
+            args.append(f"--checksums")
+        if self.list_file:
+            args.append(f"--list_file={self.list_file}")
+
+        return args
+
+
+class Sist2Index:
+    def __init__(self, path):
+        self.path = path
+
+        with open(os.path.join(path, "descriptor.json")) as f:
+            self._descriptor = json.load(f)
+
+    def to_json(self):
+        return {
+            "path": self.path,
+            "version": self.version(),
+            "timestamp": self.timestamp(),
+            "name": self.name()
+        }
+
+    def version(self) -> Sist2Version:
+        return Sist2Version(self._descriptor["version"])
+
+    def timestamp(self) -> datetime:
+        return datetime.fromtimestamp(self._descriptor["timestamp"])
+
+    def name(self) -> str:
+        return self._descriptor["name"]
+
+
+class Sist2:
+
+    def __init__(self, bin_path: str, data_directory: str):
+        self._bin_path = bin_path
+        self._data_dir = data_directory
+
+    def index(self, options: IndexOptions, logs_cb):
+
+        if options.script:
+            with NamedTemporaryFile("w", prefix="sist2-admin", suffix=".painless", delete=False) as f:
+                f.write(options.script)
+            options.script_file = f.name
+        else:
+            options.script_file = None
+
+        args = [
+            self._bin_path,
+            *options.args(),
+            "--json-logs",
+            "--very-verbose"
+        ]
+        proc = Popen(args, stdout=PIPE, stderr=PIPE)
+
+        t_stderr = Thread(target=self._consume_logs_stderr, args=(logs_cb, proc))
+        t_stderr.start()
+
+        self._consume_logs_stdout(logs_cb, proc)
+
+        t_stderr.join()
+
+        return proc.returncode
+
+    def scan(self, options: ScanOptions, logs_cb, set_pid_cb):
+
+        output_dir = os.path.join(
+            self._data_dir,
+            f"scan-{datetime.now()}.sist2"
+        )
+        options.output = output_dir
+
+        args = [
+            self._bin_path,
+            *options.args(),
+            "--json-logs",
+            "--very-verbose"
+        ]
+
+        logs_cb({"sist2-admin": f"Starting sist2 command with args {args}"})
+
+        proc = Popen(args, stdout=PIPE, stderr=PIPE)
+
+        set_pid_cb(proc.pid)
+
+        t_stderr = Thread(target=self._consume_logs_stderr, args=(logs_cb, proc))
+        t_stderr.start()
+
+        self._consume_logs_stdout(logs_cb, proc)
+
+        t_stderr.join()
+
+        return proc.returncode
+
+    @staticmethod
+    def _consume_logs_stderr(logs_cb, proc):
+        pipe_wrapper = TextIOWrapper(proc.stderr, encoding="utf8", errors="ignore")
+        try:
+            for line in pipe_wrapper:
+                if line.strip() == "":
+                    continue
+                logs_cb({"stderr": line})
+        finally:
+            proc.wait()
+            pipe_wrapper.close()
+
+    @staticmethod
+    def _consume_logs_stdout(logs_cb, proc):
+        pipe_wrapper = TextIOWrapper(proc.stdout, encoding="utf8", errors="ignore")
+        try:
+            for line in pipe_wrapper:
+                if line.strip() == "":
+                    continue
+                log_object = json.loads(line)
+                logs_cb(log_object)
+        except Exception as e:
+            proc.kill()
+            try:
+                print(line)
+            except NameError:
+                pass
+            print(traceback.format_exc())
+        finally:
+            pass
+            # proc.wait()
+            # pipe_wrapper.close()
+
+    def web(self, options: WebOptions, name: str):
+
+        if options.auth0_public_key:
+            with NamedTemporaryFile("w", prefix="sist2-admin", suffix=".txt", delete=False) as f:
+                f.write(options.auth0_public_key)
+            options.auth0_public_key_file = f.name
+        else:
+            options.auth0_public_key_file = None
+
+        args = [
+            self._bin_path,
+            *options.args()
+        ]
+
+        web_logger = logging.Logger(name=f"sist2-frontend-{name}")
+        web_logger.addHandler(FileHandler(os.path.join(LOG_FOLDER, f"frontend-{name}.log")))
+
+        def logs_cb(message):
+            web_logger.info(json.dumps(message))
+
+        logger.info(f"Starting frontend {' '.join(args)}")
+
+        proc = Popen(args, stdout=PIPE, stderr=PIPE)
+
+        t_stderr = Thread(target=self._consume_logs_stderr, args=(logs_cb, proc))
+        t_stderr.start()
+
+        t_stdout = Thread(target=self._consume_logs_stdout, args=(logs_cb, proc))
+        t_stdout.start()
+
+        return proc.pid

sist2-admin/sist2_admin/state.py

@@ -0,0 +1,50 @@
+from typing import Dict
+
+from hexlib.db import Table
+import pickle
+
+from tesseract import get_tesseract_langs
+
+RUNNING_FRONTENDS: Dict[str, int] = {}
+
+TESSERACT_LANGS = get_tesseract_langs()
+
+DB_SCHEMA_VERSION = "1"
+
+from pydantic import BaseModel
+
+
+def _serialize(item):
+    if isinstance(item, BaseModel):
+        return pickle.dumps(item)
+    if isinstance(item, bytes):
+        raise Exception("FIXME: bytes in PickleTable")
+    return item
+
+
+def _deserialize(item):
+    if isinstance(item, bytes):
+        return pickle.loads(item)
+    return item
+
+
+class PickleTable(Table):
+
+    def __getitem__(self, item):
+        row = super().__getitem__(item)
+        if row:
+            return dict((k, _deserialize(v)) for k, v in row.items())
+        return row
+
+    def __setitem__(self, key, value):
+        value = dict((k, _serialize(v)) for k, v in value.items())
+        super().__setitem__(key, value)
+
+    def __iter__(self):
+        for row in super().__iter__():
+            yield dict((k, _deserialize(v)) for k, v in row.items())
+
+    def sql(self, where_clause, *params):
+        for row in super().sql(where_clause, *params):
+            yield dict((k, _deserialize(v)) for k, v in row.items())
+

sist2-admin/sist2_admin/tesseract.py

@@ -0,0 +1,14 @@
+import subprocess
+
+
+def get_tesseract_langs():
+
+    res = subprocess.check_output([
+        "tesseract",
+        "--list-langs"
+    ]).decode()
+
+    languages = res.split("\n")[1:]
+
+    return list(filter(lambda lang: lang and lang != "osd", languages))
+

sist2-admin/sist2_admin/web.py

@@ -0,0 +1,28 @@
+import os.path
+from typing import List
+
+from pydantic import BaseModel
+
+from sist2 import WebOptions
+
+
+class Sist2Frontend(BaseModel):
+    name: str
+    jobs: List[str]
+    web_options: WebOptions
+    running: bool = False
+
+    auto_start: bool = False
+    extra_query_args: str = ""
+    custom_url: str = None
+
+    def get_log_path(self, log_folder: str):
+        return os.path.join(log_folder, f"frontend-{self.name}.log")
+
+    @staticmethod
+    def create_default(name: str):
+        return Sist2Frontend(
+            name=name,
+            web_options=WebOptions(),
+            jobs=[]
+        )

sist2-vue/.gitignore

@@ -0,0 +1,23 @@
+.DS_Store
+node_modules
+
+
+# local env files
+.env.local
+.env.*.local
+
+# Log files
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+pnpm-debug.log*
+
+# Editor directories and files
+.idea
+.vscode
+*.suo
+*.ntvs*
+*.njsproj
+*.sln
+*.sw?
+*.iml

sist2-vue/babel.config.js

@@ -0,0 +1,5 @@
+module.exports = {
+    "presets": [
+        "@vue/cli-plugin-babel/preset"
+    ]
+}