UI tweaks, search syntax (#25 )

hotfix
UI tweak
2025-12-12 23:18:51 +00:00 · 2020-02-16 15:24:29 -05:00 · 2020-02-15 19:33:18 -05:00 · 2020-02-15 09:30:14 -05:00 · 2020-02-14 20:58:10 -05:00 · 2020-02-14 18:28:55 -05:00
52 changed files with 1190 additions and 592 deletions
--- a/.gitignore
+++ b/.gitignore
@@ -15,3 +15,4 @@ bundle*.css
 bundle.js
 *.a
 vgcore.*
+build/
--- a/.gitmodules
+++ b/.gitmodules
@@ -31,15 +31,15 @@
 [submodule "lib/mupdf"]
 	path = lib/mupdf
 	url = git://git.ghostscript.com/mupdf.git
-[submodule "lib/zstd"]
-	path = lib/zstd
-	url = https://github.com/facebook/zstd
-[submodule "lib/lz4"]
-	path = lib/lz4
-	url = https://github.com/lz4/lz4
-[submodule "lib/libarchive"]
-	path = lib/libarchive
-	url = https://github.com/libarchive/libarchive
-[submodule "lib/libxml2"]
-	path = lib/libxml2
-	url = https://github.com/GNOME/libxml2
+[submodule "lib/tesseract"]
+	path = lib/tesseract
+	url = https://github.com/tesseract-ocr/tesseract
+[submodule "lib/leptonica"]
+	path = lib/leptonica
+	url = https://github.com/danbloomberg/leptonica
+[submodule "lib/libtiff"]
+	path = lib/libtiff
+	url = https://gitlab.com/libtiff/libtiff
+[submodule "lib/libpng"]
+	path = lib/libpng
+	url = https://github.com/glennrp/libpng
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -1,13 +1,9 @@
 cmake_minimum_required(VERSION 3.7)
 set(CMAKE_C_STANDARD 11)

-option(WITH_SIST2 "Build main executable" ON)
-option(WITH_SIST2_SCAN "Build scan executable" ON)
-
 project(sist2 C)
 list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_LIST_DIR}/CMakeModules")

-if (WITH_SIST2)
 add_executable(
        sist2
        src/main.c
@@ -45,48 +41,10 @@ if (WITH_SIST2)
        # utf8.h
        utf8.h/utf8.h
 )
-endif ()
-
-if (WITH_SIST2_SCAN)
-    add_executable(
-            sist2_scan
-            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/util.c src/util.h
-            src/ctx.h src/types.h src/parsing/font.c src/parsing/font.h
-            src/parsing/arc.h src/parsing/arc.c
-            src/parsing/doc.h src/parsing/doc.c
-            src/log.h src/log.c
-
-            # argparse
-            argparse/argparse.h argparse/argparse.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
-
-            # utf8.h
-            utf8.h/utf8.h
-    )
-endif ()

 find_package(PkgConfig REQUIRED)
 set(ENV{PKG_CONFIG_PATH} "$ENV{PKG_CONFIG_PATH}:/usr/local/lib/pkgconfig/")

-#find_package(OpenSSL REQUIRED)
 find_package(Freetype REQUIRED)

 pkg_check_modules(GLIB REQUIRED glib-2.0)
@@ -102,7 +60,6 @@ list(REMOVE_ITEM GLIB_LIBRARIES pcre)
 list(REMOVE_ITEM GOBJECT_LIBRARIES pcre)
 list(REMOVE_ITEM UUID_LIBRARIES pcre)

-if (WITH_SIST2)
 target_include_directories(
        sist2 PUBLIC
        ${GOBJECT_INCLUDE_DIRS}
@@ -116,17 +73,18 @@ if (WITH_SIST2)
        ${PROJECT_SOURCE_DIR}/lib/mupdf/include/
        ${PROJECT_SOURCE_DIR}/include/
        /usr/include/libxml2/
+        ${PROJECT_SOURCE_DIR}/lib/tesseract/include/
 )
 target_link_directories(
        sist2 PUBLIC
        ${UUID_LIBRARY_DIRS}
 )

-
 target_compile_options(sist2
        PRIVATE
        -Ofast
 #        -march=native
+        -fPIC
        -fno-stack-protector
        -fomit-frame-pointer
        )
@@ -153,7 +111,7 @@ if (WITH_SIST2)
        ${PROJECT_SOURCE_DIR}/lib/libonion_static.a

        pthread
-            curl
+
        m
        bz2
        ${PROJECT_SOURCE_DIR}/lib/libmagic.a
@@ -162,100 +120,27 @@ if (WITH_SIST2)
        freetype
        archive

-#            ${PROJECT_SOURCE_DIR}/lib/libxml2.a
        xml2
        ${PROJECT_SOURCE_DIR}/lib/libopc/libmce.a
        ${PROJECT_SOURCE_DIR}/lib/libopc/libopc.a
        ${PROJECT_SOURCE_DIR}/lib/libopc/libplib.a
+
+        ${PROJECT_SOURCE_DIR}/lib/libtesseract.a
+        ${PROJECT_SOURCE_DIR}/lib/liblept.a
+        ${PROJECT_SOURCE_DIR}/lib/libtiff.a
+        ${PROJECT_SOURCE_DIR}/lib/libpng16.a
+        stdc++
+
+        # curl
+        ${PROJECT_SOURCE_DIR}/lib/libcurl.a
+        ${PROJECT_SOURCE_DIR}/lib/libcrypto.a
+        ${PROJECT_SOURCE_DIR}/lib/libssl.a
+        dl
 )

-endif ()
-
-if (WITH_SIST2_SCAN)
-
-    set_target_properties(
-            sist2_scan
-            PROPERTIES COMPILE_DEFINITIONS SIST_SCAN_ONLY
-    )
-    set_target_properties(
-            sist2_scan
-            PROPERTIES
-            COMPILE_DEFINITIONS SIST_SCAN_ONLY
-            LINK_FLAGS -static
-    )
-    target_include_directories(
-            sist2_scan PUBLIC
-            ${LIBMAGIC_INCLUDE_DIRS}
-            ${GOBJECT_INCLUDE_DIRS}
-            ${OPENSSL_INCLUDE_DIR}
-            ${PROJECT_SOURCE_DIR}/lib/ffmpeg/
-            ${GLIB_INCLUDE_DIRS}
-            ${UUID_INCLUDE_DIRS}
-            ${FREETYPE_INCLUDE_DIRS}
-            ${PROJECT_SOURCE_DIR}/
-            ${PROJECT_SOURCE_DIR}/lmdb/libraries/liblmdb/
-            ${PROJECT_SOURCE_DIR}/lib/onion/src/
-            ${PROJECT_SOURCE_DIR}/lib/mupdf/include/
-            ${PROJECT_SOURCE_DIR}/include/
-            /usr/include/libxml2/
-    )
-    target_link_directories(
-            sist2_scan PUBLIC
-            ${UUID_LIBRARY_DIRS}
-    )
-    target_compile_options(sist2_scan
-            PRIVATE
-            -Ofast
-            #        -march=native
-            -fno-stack-protector
-            -fomit-frame-pointer
-            )
-    TARGET_LINK_LIBRARIES(
-            sist2_scan
-
-            ${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
-
-            # mupdf
-            ${PROJECT_SOURCE_DIR}/lib/libmupdf.a
-            ${PROJECT_SOURCE_DIR}/lib/libmupdf-third.a
-
-            ${PROJECT_SOURCE_DIR}/lib/libbz2.a
-            ${PROJECT_SOURCE_DIR}/lib/libmagic.a
-
-            pthread
-            m
-            ${PROJECT_SOURCE_DIR}/lib/libharfbuzz.a
-            ${PROJECT_SOURCE_DIR}/lib/libopenjp2.a
-            freetype
-
-            ${PROJECT_SOURCE_DIR}/lib/libarchive.a
-            ${PROJECT_SOURCE_DIR}/lib/liblz4.a
-            ${PROJECT_SOURCE_DIR}/lib/liblzma.a
-            ${PROJECT_SOURCE_DIR}/lib/libzstd.a
-
-            ${PROJECT_SOURCE_DIR}/lib/libxml2.a
-            ${PROJECT_SOURCE_DIR}/lib/libopc/libmce.a
-            ${PROJECT_SOURCE_DIR}/lib/libopc/libopc.a
-            ${PROJECT_SOURCE_DIR}/lib/libopc/libplib.a
-    )
-endif ()
-
 add_custom_target(
        before_sist2
        COMMAND ${CMAKE_CURRENT_SOURCE_DIR}/scripts/before_build.sh
 )

-IF (WITH_SIST2)
 add_dependencies(sist2 before_sist2)
-else ()
-    add_dependencies(sist2_scan before_sist2)
-endif ()
--- a/Docker/Dockerfile
+++ b/Docker/Dockerfile
@@ -2,8 +2,21 @@ FROM ubuntu:19.10
 MAINTAINER simon987 <me@simon987.net>

 RUN apt update
-RUN apt install -y libglib2.0-0 libcurl4 libmagic1 libharfbuzz-bin libopenjp2-7 libarchive13 liblzma5 libzstd1 liblz4-1
+RUN apt install -y libglib2.0-0 libcurl4 libmagic1 libharfbuzz-bin libopenjp2-7 libarchive13 liblzma5 libzstd1 liblz4-1 \
+ curl libtiff5 libpng16-16
+
+RUN mkdir -p /usr/share/tessdata && \
+    cd /usr/share/tessdata/ && \
+    curl -o /usr/share/tessdata/hin.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/hin.traineddata &&\
+    curl -o /usr/share/tessdata/jpn.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/jpn.traineddata &&\
+    curl -o /usr/share/tessdata/eng.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/eng.traineddata &&\
+    curl -o /usr/share/tessdata/fra.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/fra.traineddata &&\
+    curl -o /usr/share/tessdata/rus.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/rus.traineddata &&\
+    curl -o /usr/share/tessdata/spa.traineddata https://raw.githubusercontent.com/tesseract-ocr/tessdata/master/spa.traineddata && ls -lh

 ADD sist2 /root/sist2

+ENV LANG C.UTF-8
+ENV LC_ALL C.UTF-8
+
 ENTRYPOINT ["/root/sist2"]
--- a/Docker/build.sh
+++ b/Docker/build.sh
@@ -1,9 +1,15 @@
 rm ./sist2
 cp ../sist2 .
+strip sist2

 version=$(./sist2 --version)

 echo "Version ${version}"
-docker build . -t simon987/sist2:${version} -t simon987/sist2:latest
+docker build . -t simon987/sist2:${version} -t simon987/sist2:latest \
+  -t docker.pkg.github.com/simon987/sist2/sist2:latest -t docker.pkg.github.com/simon987/sist2/sist2:${version}
 docker push simon987/sist2:${version}
 docker push simon987/sist2:latest
+docker push docker.pkg.github.com/simon987/sist2/sist2:latest
+docker push docker.pkg.github.com/simon987/sist2/sist2:${version}
+
+docker run --rm -it simon987/sist2 -v
--- a/README.md
+++ b/README.md
@@ -1,5 +1,6 @@
 ![GitHub](https://img.shields.io/github/license/simon987/sist2.svg)
 [![CodeFactor](https://www.codefactor.io/repository/github/simon987/sist2/badge?s=05daa325188aac4eae32c786f3d9cf4e0593f822)](https://www.codefactor.io/repository/github/simon987/sist2)
+[![Development snapshots](https://ci.simon987.net/app/rest/builds/buildType(Sist2_Build)/statusIcon)](https://files.simon987.net/artifacts/Sist2/Build/)

 # sist2

@@ -16,16 +17,19 @@ sist2 (Simple incremental search tool)
 * Incremental scanning
 * Automatic tagging from file attributes via [user scripts](scripting/README.md)
 * Recursive scan inside archive files \*\*
+* OCR support with tesseract \*\*\*


 \* See [format support](#format-support)    
-\** See [Archive files](#archive-files)
+\*\* See [Archive files](#archive-files)    
+\*\*\* See [OCR](#ocr)    

 ## Getting Started

 1. Have an [Elasticsearch](https://www.elastic.co/downloads/elasticsearch) instance running
 1. 
    1. Download the [latest sist2 release](https://github.com/simon987/sist2/releases) *
+    1. *(or)* Download a [development snapshot](https://files.simon987.net/artifacts/Sist2/Build/) *(Not recommended!)*
    1. *(or)* `docker pull simon987/sist2:latest`
   

@@ -35,6 +39,9 @@ sist2 (Simple incremental search tool)

 ## Example usage

+
+![demo](demo.gif)
+
 See help page `sist2 --help` for more details.

 **Scan a directory**
@@ -85,10 +92,10 @@ docker stop sist2

 File type | Library | Content | Thumbnail | Metadata
 :---|:---|:---|:---|:---
-pdf,xps,cbz,fb2,epub | MuPDF | yes | yes, `png` | title |
+pdf,xps,cbz,fb2,epub | MuPDF | text+ocr | yes, `png` | title |
 `audio/*` | ffmpeg | - | yes, `jpeg` | ID3 tags |
 `video/*` | ffmpeg | - | yes, `jpeg` | title, comment, artist |
-`image/*` | ffmpeg | - | yes, `jpeg` | `EXIF:Artist`, `EXIF:ImageDescription` |
+`image/*` | ffmpeg | - | yes, `jpeg` | [Common EXIF tags](https://github.com/simon987/sist2/blob/efdde2734eca9b14a54f84568863b7ffd59bdba3/src/parsing/media.c#L190) |
 ttf,ttc,cff,woff,fnt,otf | Freetype2 | - | yes, `bmp` | Name & style |
 `text/plain` | *(none)* | yes | no | - |
 tar, zip, rar, 7z, ar ...  | Libarchive | yes\* | - | no |
@@ -111,6 +118,21 @@ scan is also supported.
 To check if a media file can be parsed without *seek*, execute `cat file.mp4 | ffprobe -`
 
 
+### OCR
+
+You can enable OCR support for pdf,xps,cbz,fb2,epub file types with the
+`--ocr <lang>` option. Download the language data files with your
+package manager (`apt install tesseract-ocr-eng`) or directly [from Github](https://github.com/tesseract-ocr/tesseract/wiki/Data-Files).
+
+The `simon987/sist2` github image comes with common languages 
+(hin, jpn, eng, fra, rus, spa) pre-installed.
+
+Examples
+```bash
+sist2 scan --ocr jpn ~/Books/Manga/
+sist2 scan --ocr eng ~/Books/Textbooks/
+```
+

 ## Build from source

@@ -124,12 +146,7 @@ binaries.
    apt install git cmake pkg-config libglib2.0-dev \
        libssl-dev uuid-dev python3 libmagic-dev libfreetype6-dev \
        libcurl-dev libbz2-dev yasm libharfbuzz-dev ragel \
-        libarchive-dev
-   ```
-    *(FreeBSD)*
-    ```bash
-   pkg install cmake gcc yasm gmake bash ffmpeg e2fsprogs-uuid\
-        autotools ragel libarchive
+        libarchive-dev libtiff5 libpng16-16 libpango1.0-dev
   ```

 2. Build
--- a/ci/build.sh
+++ b/ci/build.sh
@@ -5,4 +5,3 @@
 cmake .
 make
 strip sist2
-strip sist2_scan
--- a/lib/leptonica
+++ b/lib/leptonica
--- a/lib/libarchive
+++ b/lib/libarchive
--- a/lib/libopc/libmce.a
+++ b/lib/libopc/libmce.a
--- a/lib/libopc/libopc.a
+++ b/lib/libopc/libopc.a
--- a/lib/libpng
+++ b/lib/libpng
--- a/lib/libtiff
+++ b/lib/libtiff
--- a/lib/libxml2
+++ b/lib/libxml2
--- a/lib/lz4
+++ b/lib/lz4
--- a/lib/tesseract
+++ b/lib/tesseract
--- a/lib/zstd
+++ b/lib/zstd
--- a/mime.csv
+++ b/mime.csv
@@ -320,7 +320,7 @@ video/x-dv, dif|dv
 video/x-fli, fli
 video/x-isvideo, isu
 video/x-motion-jpeg, mjpg
-video/x-ms-asf, asf|asx
+video/x-ms-asf, asf|asx|wmv
 video/x-qtc, qtc
 video/x-sgi-movie, movie|mv
 application/x-7z-compressed, 7z
--- a/schema/mappings.json
+++ b/schema/mappings.json
@@ -1,5 +1,9 @@
 {
  "properties": {
+    "_tie": {
+      "type": "keyword",
+      "doc_values": true
+    },
    "path": {
      "type": "text",
      "analyzer": "path_analyzer",
@@ -7,25 +11,30 @@
    },
    "suggest-path": {
      "type": "completion",
-      "analyzer": "keyword"
+      "analyzer": "case_insensitive_kw_analyzer"
    },
    "mime": {
      "type": "keyword"
    },
    "videoc": {
-      "type": "keyword"
+      "type": "keyword",
+      "index": false
    },
    "audioc": {
-      "type": "keyword"
+      "type": "keyword",
+      "index": false
    },
    "duration": {
-      "type": "float"
+      "type": "float",
+      "index": false
    },
    "width": {
-      "type": "integer"
+      "type": "integer",
+      "index": false
    },
    "height": {
-      "type": "integer"
+      "type": "integer",
+      "index": false
    },
    "mtime": {
      "type": "integer"
@@ -70,6 +79,23 @@
      "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",
@@ -83,6 +109,30 @@
    },
    "tag": {
      "type": "keyword"
+    },
+    "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"
    }
  }
 }
--- a/schema/pipeline.json
+++ b/schema/pipeline.json
@@ -0,0 +1,10 @@
+{
+  "description": "Copy _id to _tie",
+  "processors": [
+    {
+      "script": {
+        "source": "ctx._tie = ctx._id;"
+      }
+    }
+  ]
+}
--- a/schema/settings.json
+++ b/schema/settings.json
@@ -21,6 +21,12 @@
          "lowercase"
        ]
      },
+      "case_insensitive_kw_analyzer": {
+        "tokenizer": "keyword",
+        "filter": [
+          "lowercase"
+        ]
+      },
      "my_nGram": {
        "tokenizer": "my_nGram_tokenizer",
        "filter": [
--- a/scripts/get_static_libs.sh
+++ b/scripts/get_static_libs.sh
@@ -5,7 +5,7 @@ THREADS=$(nproc)
 cd lib

 cd mupdf
-make USE_SYSTEM_HARFBUZZ=yes USE_SYSTEM_OPENJPEG=yes HAVE_X11=no HAVE_GLUT=no -j $THREADS
+CFLAGS=-fPIC make USE_SYSTEM_HARFBUZZ=yes USE_SYSTEM_OPENJPEG=yes HAVE_X11=no HAVE_GLUT=no -j $THREADS
 cd ..

 mv mupdf/build/release/libmupdf.a .
@@ -13,8 +13,7 @@ mv mupdf/build/release/libmupdf-third.a .

 # openjp2
 cd openjpeg
-#cmake . -DCMAKE_BUILD_TYPE=Release -DCMAKE_C_FLAGS="-O3 -march=native -DNDEBUG"
-cmake . -DCMAKE_BUILD_TYPE=Release -DCMAKE_C_FLAGS="-O3"
+cmake . -DCMAKE_BUILD_TYPE=Release -DCMAKE_C_FLAGS="-O3 -DNDEBUG -fPIC"
 make -j $THREADS
 cd ..
 mv openjpeg/bin/libopenjp2.a .
@@ -22,7 +21,7 @@ mv openjpeg/bin/libopenjp2.a .
 # harfbuzz
 cd harfbuzz
 ./autogen.sh
-./configure --disable-shared --enable-static
+CFLAGS=-fPIC ./configure --disable-shared --enable-static
 make -j $THREADS
 cd ..
 mv harfbuzz/src/.libs/libharfbuzz.a .
@@ -33,7 +32,8 @@ cd ffmpeg
 --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
+ --disable-vdpau --disable-vaapi --disable-sdl2 --disable-network\
+ --extra-cflags=-fPIC
 make -j $THREADS
 cd ..

@@ -70,43 +70,59 @@ make -j $THREADS
 cd ..
 mv libmagic/src/.libs/libmagic.a .

-# libarchive
-cd libarchive/build
+# tesseract
+cd tesseract
+mkdir build
+cd build
+cmake -DSTATIC=on -DBUILD_TRAINING_TOOLS=off -DBUILD_TESTS=off -DCMAKE_BUILD_TYPE=Release \
+  -DCMAKE_CXX_FLAGS="-fPIC" -DAUTO_OPTIMIZE=off ..
+make -j $THREADS
+cd ../..
+mv tesseract/build/libtesseract.a .
+
+# leptonica
+cd leptonica
 ./autogen.sh
-cd ..
-./configure --without-nettle --without-expat --without-xml2 --without-openssl
+CFLAGS="-fPIC" ./configure --without-zlib --without-jpeg --without-giflib \
+  --without-giflib --without-libwebp --without-libwebpmux --without-libopenjpeg \
+  --enable-static --disable-shared
 make -j $THREADS
 cd ..
-mv libarchive/.libs/libarchive.a .
+mv leptonica/src/.libs/liblept.a .

-# lz4
-cd lz4
-make -j $THREADS
-cd ..
-mv lz4/lib/liblz4.a .
-
-# lzma
-wget https://newcontinuum.dl.sourceforge.net/project/lzmautils/xz-5.2.3.tar.gz
-tar -xzf xz-5.2.3.tar.gz
-rm xz-5.2.3.tar.gz
-cd xz-5.2.3
+# tiff
+cd libtiff
 ./autogen.sh
-./configure
+CFLAGS="-fPIC" CXXFLAGS="-fPIC" CXX_FLAGS="-fPIC" ./configure --enable-static --disable-shared --disable-lzw --disable-jpeg --disable-webp \
+  --disable-lzma --disable-zstd --disable-jbig
 make -j $THREADS
 cd ..
-mv xz-5.2.3/src/liblzma/.libs/liblzma.a .
+mv libtiff/libtiff/.libs/libtiff.a .

-# zstd
-cd zstd
+# png
+cd libpng
+CFLAGS="-fPIC" ./configure --enable-static --disable-shared
 make -j $THREADS
 cd ..
-mv zstd/lib/libzstd.a .
+mv libpng/.libs/libpng16.a .

-# xml2
-cd libxml2
-./autogen.sh --without-zlib --without-lzma
-make
+# openssl...
+git clone --depth 1 -b OpenSSL_1_1_0-stable https://github.com/openssl/openssl
+cd openssl
+./config --prefix=$(pwd)/../ssl
+make depend
+make -j $THREADS
+make install
 cd ..
-mv libxml2/.libs/libxml2.a .
+mv ./openssl/libcrypto.a ./openssl/libssl.a .

+# curl
+wget -nc https://curl.haxx.se/download/curl-7.68.0.tar.gz
+tar -xzf curl-7.68.0.tar.gz
+cd curl-7.68.0
+./configure --disable-ldap --disable-ldaps --without-librtmp --disable-rtsp --disable-crypto-auth \
+  --disable-smtp --without-libidn2 --without-nghttp2 --without-brotli --enable-static --disable-shared \
+  --without-libpsl --with-ssl=$(pwd)/../ssl
+make -j $THREADS
 cd ..
+mv curl-7.68.0/lib/.libs/libcurl.a .
--- a/scripts/get_static_libs_freebsd.sh
+++ b/scripts/get_static_libs_freebsd.sh
@@ -1,58 +0,0 @@
-#!/usr/bin/env bash
-cd lib
-
-# mupdf
-cd mupdf
-HAVE_X11=no HAVE_GLUT=no gmake -j 4
-cd ..
-
-mv mupdf/build/release/libmupdf.a .
-mv mupdf/build/release/libmupdf-third.a .
-
-# openjp2
-cd openjpeg
-#cmake . -DCMAKE_BUILD_TYPE=Release -DCMAKE_C_FLAGS="-O3 -march=native -DNDEBUG"
-cmake . -DCMAKE_BUILD_TYPE=Release -DCMAKE_C_FLAGS="-O3"
-gmake -j 4
-cd ..
-mv openjpeg/bin/libopenjp2.a .
-
-# harfbuzz
-cd harfbuzz
-./autogen.sh
-./configure --disable-shared --enable-static
-gmake -j 4
-cd ..
-mv harfbuzz/src/.libs/libharfbuzz.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
-gmake -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 .
-
-#bzip2
-cd bzip2-1.0.6
-make -j 4
-cd ..
-mv bzip2-1.0.6/libbz2.a .
-
-# magic
-cd libmagic
-./autogen.sh
-./configure --enable-static --disable-shared
-make -j 4
-cd ..
-mv libmagic/src/.libs/libmagic.a .
-
-cd ..
--- a/scripts/index_static.py
+++ b/scripts/index_static.py
@@ -1,6 +1,9 @@
+import json
+
 files = [
    "schema/mappings.json",
    "schema/settings.json",
+    "schema/pipeline.json",
 ]


@@ -9,6 +12,6 @@ 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()
    print("char %s[%d] = {%s};" % (clean(file), len(data), ",".join(str(int(b)) for b in data)))
--- a/src/cli.c
+++ b/src/cli.c
@@ -1,6 +1,8 @@
 #include "cli.h"
 #include "ctx.h"

+#include <tesseract/capi.h>
+
 #define DEFAULT_OUTPUT "index.sist2/"
 #define DEFAULT_CONTENT_SIZE 32768
 #define DEFAULT_QUALITY 5
@@ -13,6 +15,13 @@
 #define DEFAULT_BIND_ADDR "localhost"
 #define DEFAULT_PORT "4090"

+const char* TESS_DATAPATHS[] = {
+        "/usr/share/tessdata/",
+        "/usr/share/tesseract-ocr/tessdata/",
+        "./",
+        NULL
+};
+

 scan_args_t *scan_args_create() {
    scan_args_t *args = calloc(sizeof(scan_args_t), 1);
@@ -35,8 +44,6 @@ void scan_args_destroy(scan_args_t *args) {
    free(args);
 }

-#ifndef SIST_SCAN_ONLY
-
 void index_args_destroy(index_args_t *args) {
    //todo
    free(args);
@@ -47,8 +54,6 @@ void web_args_destroy(web_args_t *args) {
    free(args);
 }

-#endif
-
 int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
    if (argc < 2) {
        fprintf(stderr, "Required positional argument: PATH.\n");
@@ -136,6 +141,27 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
        return 1;
    }

+    if (args->tesseract_lang != NULL) {
+        TessBaseAPI *api = TessBaseAPICreate();
+
+        char filename[128];
+        sprintf(filename, "%s.traineddata", args->tesseract_lang);
+        const char * path = find_file_in_paths(TESS_DATAPATHS, filename);
+        if (path == NULL) {
+            LOG_FATAL("cli.c", "Could not find tesseract language file!");
+        }
+
+        ret = TessBaseAPIInit3(api, path, args->tesseract_lang);
+        if (ret != 0) {
+            fprintf(stderr, "Could not initialize tesseract with lang '%s'\n", args->tesseract_lang);
+            return 1;
+        }
+        TessBaseAPIEnd(api);
+        TessBaseAPIDelete(api);
+
+        args->tesseract_path = path;
+    }
+
    LOG_DEBUGF("cli.c", "arg quality=%f", args->quality)
    LOG_DEBUGF("cli.c", "arg size=%d", args->size)
    LOG_DEBUGF("cli.c", "arg content_size=%d", args->content_size)
@@ -147,14 +173,16 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
    LOG_DEBUGF("cli.c", "arg depth=%d", args->depth)
    LOG_DEBUGF("cli.c", "arg path=%s", args->path)
    LOG_DEBUGF("cli.c", "arg archive=%s", args->archive)
+    LOG_DEBUGF("cli.c", "arg tesseract_lang=%s", args->tesseract_lang)
+    LOG_DEBUGF("cli.c", "arg tesseract_path=%s", args->tesseract_path)

    return 0;
 }

-#ifndef SIST_SCAN_ONLY
-
 int index_args_validate(index_args_t *args, int argc, const char **argv) {

+    LogCtx.verbose = 1;
+
    if (argc < 2) {
        fprintf(stderr, "Required positional argument: PATH.\n");
        return 1;
@@ -216,6 +244,8 @@ int index_args_validate(index_args_t *args, int argc, const char **argv) {

 int web_args_validate(web_args_t *args, int argc, const char **argv) {

+    LogCtx.verbose = 1;
+
    if (argc < 2) {
        fprintf(stderr, "Required positional argument: PATH.\n");
        return 1;
@@ -273,5 +303,3 @@ web_args_t *web_args_create() {
    return args;
 }

-#endif
-
--- a/src/cli.h
+++ b/src/cli.h
@@ -16,13 +16,14 @@ typedef struct scan_args {
    char *path;
    char *archive;
    archive_mode_t archive_mode;
+    char *tesseract_lang;
+    const char *tesseract_path;
 } scan_args_t;

 scan_args_t *scan_args_create();
 void scan_args_destroy(scan_args_t *args);
 int scan_args_validate(scan_args_t *args, int argc, const char **argv);

-#ifndef SIST_SCAN_ONLY
 typedef struct index_args {
    char *es_url;
    const char *index_path;
@@ -51,6 +52,5 @@ void web_args_destroy(web_args_t *args);

 int index_args_validate(index_args_t *args, int argc, const char **argv);
 int web_args_validate(web_args_t *args, int argc, const char **argv);
-#endif

 #endif
--- a/src/ctx.h
+++ b/src/ctx.h
@@ -27,6 +27,8 @@ struct {
    GHashTable *copy_table;

    pthread_mutex_t mupdf_mu;
+    char * tesseract_lang;
+    const char * tesseract_path;
 } ScanCtx;

 struct {
@@ -35,7 +37,6 @@ struct {
    int no_color;
 } LogCtx;

-#ifndef SIST_SCAN_ONLY
 struct {
    char *es_url;
    int batch_size;
@@ -47,7 +48,6 @@ struct {
    char *b64credentials;
    struct index_t indices[16];
 } WebCtx;
-#endif


 #endif
--- a/src/index/elastic.c
+++ b/src/index/elastic.c
@@ -27,13 +27,14 @@ void print_json(cJSON *document, const char uuid_str[UUID_STR_LEN]) {
    cJSON_AddStringToObject(line, "_id", uuid_str);
    cJSON_AddStringToObject(line, "_index", "sist2");
    cJSON_AddStringToObject(line, "_type", "_doc");
-    cJSON_AddItemToObject(line, "_source", document);
+    cJSON_AddItemReferenceToObject(line, "_source", document);

    char *json = cJSON_PrintUnformatted(line);

    printf("%s\n", json);

-    cJSON_free(line);
+    cJSON_free(json);
+    cJSON_Delete(line);
 }

 void index_json(cJSON *document, const char uuid_str[UUID_STR_LEN]) {
@@ -68,7 +69,7 @@ void execute_update_script(const char *script, const char index_id[UUID_STR_LEN]
    char bulk_url[4096];
    snprintf(bulk_url, 4096, "%s/sist2/_update_by_query?pretty", Indexer->es_url);
    response_t *r = web_post(bulk_url, str, "Content-Type: application/json");
-    printf("Executed user script <%d>\n", r->status_code);
+    LOG_INFOF("elastic.c", "Executed user script <%d>", r->status_code);
    cJSON *resp = cJSON_Parse(r->body);

    cJSON_free(str);
@@ -79,7 +80,7 @@ void execute_update_script(const char *script, const char index_id[UUID_STR_LEN]
    if (error != NULL) {
        char *error_str = cJSON_Print(error);

-        fprintf(stderr, "User script error: \n%s\n", error_str);
+        LOG_ERRORF("elastic.c", "User script error: \n%s", error_str);
        cJSON_free(error_str);
    }

@@ -128,29 +129,30 @@ void elastic_flush() {
    Indexer->queued = 0;

    char bulk_url[4096];
-    snprintf(bulk_url, 4096, "%s/sist2/_bulk", Indexer->es_url);
+    snprintf(bulk_url, 4096, "%s/sist2/_bulk?pipeline=tie", Indexer->es_url);
    response_t *r = web_post(bulk_url, buf, "Content-Type: application/x-ndjson");

    if (r->status_code == 0) {
-        fprintf(stderr, "Could not connect to %s, make sure that elasticsearch is running!\n", IndexCtx.es_url);
-        exit(1);
+        LOG_FATALF("elastic.c", "Could not connect to %s, make sure that elasticsearch is running!\n", IndexCtx.es_url)
    }

-    printf("Indexed %3d documents (%zukB) <%d>\n", count, buf_cur / 1024, r->status_code);
+    LOG_INFOF("elastic.c", "Indexed %d documents (%zukB) <%d>", count, buf_cur / 1024, r->status_code);

+    if (r->status_code != 200 && r->status_code != 413) {
        cJSON *ret_json = cJSON_Parse(r->body);
        if (cJSON_GetObjectItem(ret_json, "errors")->valueint != 0) {
            cJSON *err;
            cJSON_ArrayForEach(err, cJSON_GetObjectItem(ret_json, "items")) {
                if (cJSON_GetObjectItem(cJSON_GetObjectItem(err, "index"), "status")->valueint != 201) {
                    char* str = cJSON_Print(err);
-                fprintf(stderr, "%s\n", str);
+                    LOG_ERRORF("elastic.c", "%s\n", str);
                    cJSON_free(str);
                }
            }
        }

        cJSON_Delete(ret_json);
+    }

    free_response(r);
    free(buf);
@@ -198,7 +200,7 @@ void destroy_indexer(char * script, char index_id[UUID_STR_LEN]) {

    snprintf(url, sizeof(url), "%s/sist2/_refresh", IndexCtx.es_url);
    response_t *r = web_post(url, "", NULL);
-    printf("Refresh index <%d>\n", r->status_code);
+    LOG_INFOF("elastic.c", "Refresh index <%d>", r->status_code);
    free_response(r);

    if (script != NULL) {
@@ -207,12 +209,12 @@ void destroy_indexer(char * script, char index_id[UUID_STR_LEN]) {

    snprintf(url, sizeof(url), "%s/sist2/_refresh", IndexCtx.es_url);
    r = web_post(url, "", NULL);
-    printf("Refresh index <%d>\n", r->status_code);
+    LOG_INFOF("elastic.c", "Refresh index <%d>", r->status_code);
    free_response(r);

    snprintf(url, sizeof(url), "%s/sist2/_forcemerge", IndexCtx.es_url);
    r = web_post(url, "", NULL);
-    printf("Merge index <%d>\n", r->status_code);
+    LOG_INFOF("elastic.c", "Merge index <%d>", r->status_code);
    free_response(r);

    if (Indexer != NULL) {
@@ -232,32 +234,37 @@ void elastic_init(int force_reset) {

    if (!index_exists || force_reset) {
        r = web_delete(url);
-        printf("Delete index <%d>\n", r->status_code);
+        LOG_INFOF("elastic.c", "Delete index <%d>", r->status_code);
        free_response(r);

        snprintf(url, 4096, "%s/sist2", IndexCtx.es_url);
        r = web_put(url, "", NULL);
-        printf("Create index <%d>\n", r->status_code);
+        LOG_INFOF("elastic.c", "Create index <%d>", r->status_code);
        free_response(r);

        snprintf(url, 4096, "%s/sist2/_close", IndexCtx.es_url);
        r = web_post(url, "", NULL);
-        printf("Close index <%d>\n", r->status_code);
+        LOG_INFOF("elastic.c", "Close index <%d>", r->status_code);
+        free_response(r);
+
+        snprintf(url, 4096, "%s/_ingest/pipeline/tie", IndexCtx.es_url);
+        r = web_put(url, pipeline_json, "Content-Type: application/json");
+        LOG_INFOF("elastic.c", "Create pipeline <%d>", r->status_code);
        free_response(r);

        snprintf(url, 4096, "%s/sist2/_settings", IndexCtx.es_url);
        r = web_put(url, settings_json, "Content-Type: application/json");
-        printf("Update settings <%d>\n", r->status_code);
+        LOG_INFOF("elastic.c", "Update settings <%d>", r->status_code);
        free_response(r);

        snprintf(url, 4096, "%s/sist2/_mappings/_doc?include_type_name=true", IndexCtx.es_url);
        r = web_put(url, mappings_json, "Content-Type: application/json");
-        printf("Update mappings <%d>\n", r->status_code);
+        LOG_INFOF("elastic.c", "Update mappings <%d>", r->status_code);
        free_response(r);

        snprintf(url, 4096, "%s/sist2/_open", IndexCtx.es_url);
        r = web_post(url, "", NULL);
-        printf("Open index <%d>\n", r->status_code);
+        LOG_INFOF("elastic.c", "Open index <%d>", r->status_code);
        free_response(r);
    }
 }
@@ -274,3 +281,28 @@ cJSON *elastic_get_document(const char *uuid_str) {
    free_response(r);
    return json;
 }
+
+char *elastic_get_status() {
+    char url[4096];
+    snprintf(url, 4096,
+            "%s/_cluster/state/metadata/sist2?filter_path=metadata.indices.*.state", WebCtx.es_url);
+
+    response_t *r = web_get(url);
+    cJSON *json = NULL;
+    char *status = malloc(128 * sizeof(char));
+    status[0] = '\0';
+
+    if (r->status_code == 200) {
+        json = cJSON_Parse(r->body);
+        const cJSON *metadata = cJSON_GetObjectItem(json, "metadata");
+        if (metadata != NULL) {
+            const cJSON *indices = cJSON_GetObjectItem(metadata, "indices");
+            const cJSON *sist2 = cJSON_GetObjectItem(indices, "sist2");
+            const cJSON *state = cJSON_GetObjectItem(sist2, "state");
+            strcpy(status, state->valuestring);
+        }
+    }
+    free_response(r);
+    cJSON_Delete(json);
+    return status;
+}
--- a/src/index/elastic.h
+++ b/src/index/elastic.h
@@ -30,4 +30,6 @@ void elastic_init(int force_reset);

 cJSON *elastic_get_document(const char *uuid_str);

+char *elastic_get_status();
+
 #endif
--- a/src/index/static_generated.c
+++ b/src/index/static_generated.c
--- a/src/io/serialize.c
+++ b/src/io/serialize.c
@@ -34,6 +34,7 @@ void write_index_descriptor(char *path, index_descriptor_t *desc) {
    cJSON_AddStringToObject(json, "version", desc->version);
    cJSON_AddStringToObject(json, "root", desc->root);
    cJSON_AddStringToObject(json, "name", desc->name);
+    cJSON_AddStringToObject(json, "type", desc->type);
    cJSON_AddStringToObject(json, "rewrite_url", desc->rewrite_url);
    cJSON_AddNumberToObject(json, "timestamp", (double) desc->timestamp);

@@ -56,8 +57,7 @@ index_descriptor_t read_index_descriptor(char *path) {
    int fd = open(path, O_RDONLY);

    if (fd == -1) {
-        fprintf(stderr, "Invalid/corrupt index (Could not find descriptor)\n");
-        exit(1);
+        LOG_FATAL("serialize.c", "Invalid/corrupt index (Could not find descriptor)\n")
    }

    char *buf = malloc(info.st_size + 1);
@@ -75,6 +75,11 @@ index_descriptor_t read_index_descriptor(char *path) {
    descriptor.root_len = (short) strlen(descriptor.root);
    strcpy(descriptor.version, cJSON_GetObjectItem(json, "version")->valuestring);
    strcpy(descriptor.uuid, cJSON_GetObjectItem(json, "uuid")->valuestring);
+    if (cJSON_GetObjectItem(json, "type") == NULL) {
+        strcpy(descriptor.type, INDEX_TYPE_BIN);
+    } else {
+        strcpy(descriptor.type, cJSON_GetObjectItem(json, "type")->valuestring);
+    }

    cJSON_Delete(json);
    free(buf);
@@ -113,6 +118,24 @@ char *get_meta_key_text(enum metakey meta_key) {
            return "font_name";
        case MetaParent:
            return "parent";
+        case MetaExifMake:
+            return "exif_make";
+        case MetaExifSoftware:
+            return "exif_software";
+        case MetaExifExposureTime:
+            return "exif_exposure_time";
+        case MetaExifFNumber:
+            return "exif_fnumber";
+        case MetaExifFocalLength:
+            return "exif_focal_length";
+        case MetaExifUserComment:
+            return "exif_user_comment";
+        case MetaExifIsoSpeedRatings:
+            return "exif_iso_speed_ratings";
+        case MetaExifModel:
+            return "exif_model";
+        case MetaExifDateTime:
+            return "exif_datetime";
        default:
            return NULL;
    }
@@ -172,8 +195,8 @@ void thread_cleanup() {
    close(index_fd);
 }

-void read_index(const char *path, const char index_id[UUID_STR_LEN], index_func func) {

+void read_index_bin(const char *path, const char *index_id, index_func func) {
    line_t line;
    dyn_buffer_t buf = dyn_buffer_create();

@@ -191,7 +214,12 @@ void read_index(const char *path, const char index_id[UUID_STR_LEN], index_func
        char uuid_str[UUID_STR_LEN];
        uuid_unparse(line.uuid, uuid_str);

+        const char* mime_text = mime_get_mime_text(line.mime);
+        if (mime_text == NULL) {
+            cJSON_AddNullToObject(document, "mime");
+        } else {
            cJSON_AddStringToObject(document, "mime", mime_get_mime_text(line.mime));
+        }
        cJSON_AddNumberToObject(document, "size", (double) line.size);
        cJSON_AddNumberToObject(document, "mtime", line.mtime);

@@ -229,7 +257,7 @@ void read_index(const char *path, const char index_id[UUID_STR_LEN], index_func
                case MetaMediaBitrate: {
                    long value;
                    fread(&value, sizeof(long), 1, file);
-                    cJSON_AddNumberToObject(document, get_meta_key_text(key), value);
+                    cJSON_AddNumberToObject(document, get_meta_key_text(key), (double) value);
                    break;
                }
                case MetaMediaAudioCodec:
@@ -250,6 +278,15 @@ void read_index(const char *path, const char index_id[UUID_STR_LEN], index_func
                case MetaGenre:
                case MetaFontName:
                case MetaParent:
+                case MetaExifMake:
+                case MetaExifSoftware:
+                case MetaExifExposureTime:
+                case MetaExifFNumber:
+                case MetaExifFocalLength:
+                case MetaExifUserComment:
+                case MetaExifIsoSpeedRatings:
+                case MetaExifDateTime:
+                case MetaExifModel:
                case MetaTitle: {
                    buf.cur = 0;
                    while ((c = getc(file)) != 0) {
@@ -275,6 +312,89 @@ void read_index(const char *path, const char index_id[UUID_STR_LEN], index_func
    fclose(file);
 }

+const char *json_type_copy_fields[] = {
+        "mime", "name", "path", "extension", "index", "size", "mtime", "parent",
+
+        // Meta
+        "title", "content", "width", "height", "duration", "audioc", "videoc",
+        "bitrate", "artist", "album", "album_artist", "genre", "title", "font_name",
+
+        // Special
+        "tag", "_url"
+};
+
+const char *json_type_array_fields[] = {
+        "_keyword", "_text"
+};
+
+void read_index_json(const char *path, UNUSED(const char *index_id), index_func func) {
+
+    FILE *file = fopen(path, "r");
+    while (1) {
+        char *line = NULL;
+        size_t len;
+        size_t read = getline(&line, &len, file);
+        if (read == -1) {
+            if (line) {
+                free(line);
+            }
+            break;
+        }
+
+        cJSON *input = cJSON_Parse(line);
+        if (input == NULL) {
+            LOG_FATALF("serialize.c", "Could not parse JSON line: \n%s", line)
+        }
+        if (line) {
+            free(line);
+        }
+
+        cJSON *document = cJSON_CreateObject();
+        const char *uuid_str = cJSON_GetObjectItem(input, "_id")->valuestring;
+
+        for (int i = 0; i < (sizeof(json_type_copy_fields) / sizeof(json_type_copy_fields[0])); i++) {
+            cJSON *value = cJSON_GetObjectItem(input, json_type_copy_fields[i]);
+            if (value != NULL) {
+                cJSON_AddItemReferenceToObject(document, json_type_copy_fields[i], value);
+            }
+        }
+
+        for (int i = 0; i < (sizeof(json_type_array_fields) / sizeof(json_type_array_fields[0])); i++) {
+            cJSON *arr = cJSON_GetObjectItem(input, json_type_array_fields[i]);
+            if (arr != NULL) {
+                cJSON *obj;
+                cJSON_ArrayForEach(obj, arr) {
+                    char key[1024];
+                    cJSON *k = cJSON_GetObjectItem(obj, "k");
+                    cJSON *v = cJSON_GetObjectItem(obj, "v");
+                    if (k == NULL || v == NULL || !cJSON_IsString(k) || !cJSON_IsString(v)) {
+                        char *str = cJSON_Print(obj);
+                        LOG_FATALF("serialize.c", "Invalid %s member: must contain .k and .v string fields: \n%s",
+                                   json_type_array_fields[i], str)
+                    }
+                    snprintf(key, sizeof(key), "%s.%s", json_type_array_fields[i], k->valuestring);
+                    cJSON_AddStringToObject(document, key, v->valuestring);
+                }
+            }
+        }
+
+        func(document, uuid_str);
+        cJSON_Delete(document);
+        cJSON_Delete(input);
+
+    }
+    fclose(file);
+}
+
+void read_index(const char *path, const char index_id[UUID_STR_LEN], const char *type, index_func func) {
+
+    if (strcmp(type, INDEX_TYPE_BIN) == 0) {
+        read_index_bin(path, index_id, func);
+    } else if (strcmp(type, INDEX_TYPE_JSON) == 0) {
+        read_index_json(path, index_id, func);
+    }
+}
+
 void incremental_read(GHashTable *table, const char *filepath) {
    FILE *file = fopen(filepath, "rb");
    line_t line;
--- a/src/io/serialize.h
+++ b/src/io/serialize.h
@@ -11,7 +11,7 @@ void incremental_copy(store_t *store, store_t *dst_store, const char *filepath,

 void write_document(document_t *doc);

-void read_index(const char *path, const char[UUID_STR_LEN], index_func);
+void read_index(const char *path, const char[UUID_STR_LEN], const char *type, index_func);

 void incremental_read(GHashTable *table, const char *filepath);

--- a/src/log.c
+++ b/src/log.c
@@ -56,7 +56,7 @@ void sist_logf(char *filepath, int level, char *format, ...) {
        log_len += 1;
    }

-    write(STDOUT_FILENO, log_str, log_len);
+    write(STDERR_FILENO, log_str, log_len);
 }

 void sist_log(char *filepath, int level, char *str) {
--- a/src/main.c
+++ b/src/main.c
@@ -1,16 +1,12 @@
 #include "sist.h"
 #include "ctx.h"

-#ifndef SIST_SCAN_ONLY
 #define DESCRIPTION "Lightning-fast file system indexer and search tool."
-#else
-#define DESCRIPTION "Lightning-fast file system indexer and search tool. (SCAN ONLY)"
-#endif

 #define EPILOG "Made by simon987 <me@simon987.net>. Released under GPL-3.0"


-static const char *const Version = "1.1.14";
+static const char *const Version = "1.2.12";
 static const char *const usage[] = {
        "sist2 scan [OPTION]... PATH",
        "sist2 index [OPTION]... INDEX",
@@ -19,9 +15,7 @@ static const char *const usage[] = {
 };

 void global_init() {
-#ifndef SIST_SCAN_ONLY
    curl_global_init(CURL_GLOBAL_NOTHING);
-#endif
    av_log_set_level(AV_LOG_QUIET);
    opcInitLibrary();
 }
@@ -35,6 +29,7 @@ void init_dir(const char *dirpath) {
    uuid_unparse(uuid, ScanCtx.index.desc.uuid);
    time(&ScanCtx.index.desc.timestamp);
    strcpy(ScanCtx.index.desc.version, Version);
+    strcpy(ScanCtx.index.desc.type, INDEX_TYPE_BIN);

    write_index_descriptor(path, &ScanCtx.index.desc);
 }
@@ -54,7 +49,10 @@ void sist2_scan(scan_args_t *args) {
    strncpy(ScanCtx.index.path, args->output, sizeof(ScanCtx.index.path));
    strncpy(ScanCtx.index.desc.name, args->name, sizeof(ScanCtx.index.desc.name));
    strncpy(ScanCtx.index.desc.root, args->path, sizeof(ScanCtx.index.desc.root));
+    strncpy(ScanCtx.index.desc.rewrite_url, args->rewrite_url, sizeof(ScanCtx.index.desc.rewrite_url));
    ScanCtx.index.desc.root_len = (short) strlen(ScanCtx.index.desc.root);
+    ScanCtx.tesseract_lang = args->tesseract_lang;
+    ScanCtx.tesseract_path = args->tesseract_path;

    init_dir(ScanCtx.index.path);

@@ -122,8 +120,6 @@ void sist2_scan(scan_args_t *args) {
    store_destroy(ScanCtx.index.store);
 }

-#ifndef SIST_SCAN_ONLY
-
 void sist2_index(index_args_t *args) {

    IndexCtx.es_url = args->es_url;
@@ -137,8 +133,12 @@ void sist2_index(index_args_t *args) {
    snprintf(descriptor_path, PATH_MAX, "%s/descriptor.json", args->index_path);

    index_descriptor_t desc = read_index_descriptor(descriptor_path);
-    if (strcmp(desc.version, Version) != 0) {
-        fprintf(stderr, "Version mismatch! Index is v%s but executable is v%s\n", desc.version, Version);
+
+    LOG_DEBUGF("main.c", "descriptor version %s (%s)", desc.version, desc.type)
+
+    if (strcmp(desc.version, Version) != 0 && strcmp(desc.version, INDEX_VERSION_EXTERNAL) != 0) {
+        fprintf(stderr, "Version mismatch! Index is %s but executable is %s/%s\n",
+                desc.version, Version, INDEX_VERSION_EXTERNAL);
        return;
    }

@@ -160,7 +160,7 @@ void sist2_index(index_args_t *args) {
        if (strncmp(de->d_name, "_index_", sizeof("_index_") - 1) == 0) {
            char file_path[PATH_MAX];
            snprintf(file_path, PATH_MAX, "%s/%s", args->index_path, de->d_name);
-            read_index(file_path, desc.uuid, f);
+            read_index(file_path, desc.uuid, desc.type, f);
        }
    }
    closedir(dir);
@@ -198,18 +198,14 @@ void sist2_web(web_args_t *args) {
    serve(args->bind, args->port);
 }

-#endif
-

 int main(int argc, const char *argv[]) {

    global_init();

    scan_args_t *scan_args = scan_args_create();
-#ifndef SIST_SCAN_ONLY
    index_args_t *index_args = index_args_create();
    web_args_t *web_args = web_args_create();
-#endif

    int arg_version = 0;

@@ -240,8 +236,9 @@ int main(int argc, const char *argv[]) {
            OPT_STRING(0, "archive", &scan_args->archive, "Archive file mode (skip|list|shallow|recurse). "
                                                          "skip: Don't parse, list: only get file names as text, "
                                                          "shallow: Don't parse archives inside archives. DEFAULT: recurse"),
+            OPT_STRING(0, "ocr", &scan_args->tesseract_lang, "Tesseract language (use tesseract --list-langs to see "
+                                                             "which are installed on your machine)"),

-#ifndef SIST_SCAN_ONLY
            OPT_GROUP("Index options"),
            OPT_STRING(0, "es-url", &common_es_url, "Elasticsearch url with port. DEFAULT=http://localhost:9200"),
            OPT_BOOLEAN('p', "print", &index_args->print, "Just print JSON documents to stdout."),
@@ -255,7 +252,6 @@ int main(int argc, const char *argv[]) {
            OPT_STRING(0, "bind", &web_args->bind, "Listen on this address. DEFAULT=localhost"),
            OPT_STRING(0, "port", &web_args->port, "Listen on this port. DEFAULT=4090"),
            OPT_STRING(0, "auth", &web_args->credentials, "Basic auth in user:password format"),
-#endif

            OPT_END(),
    };
@@ -274,10 +270,8 @@ int main(int argc, const char *argv[]) {
        LogCtx.verbose = 1;
    }

-#ifndef SIST_SCAN_ONLY
    web_args->es_url = common_es_url;
    index_args->es_url = common_es_url;
-#endif

    if (argc == 0) {
        argparse_usage(&argparse);
@@ -292,7 +286,6 @@ int main(int argc, const char *argv[]) {

    }

-#ifndef SIST_SCAN_ONLY
    else if (strcmp(argv[0], "index") == 0) {

        int err = index_args_validate(index_args, argc, argv);
@@ -310,7 +303,6 @@ int main(int argc, const char *argv[]) {
        sist2_web(web_args);

    }
-#endif
    else {
        fprintf(stderr, "Invalid command: '%s'\n", argv[0]);
        argparse_usage(&argparse);
@@ -320,10 +312,8 @@ int main(int argc, const char *argv[]) {

    scan_args_destroy(scan_args);

-#ifndef SIST_SCAN_ONLY
    index_args_destroy(index_args);
    web_args_destroy(web_args);
-#endif

    return 0;
 }
--- a/src/parsing/doc.c
+++ b/src/parsing/doc.c
@@ -1,10 +1,20 @@
 #include "doc.h"
 #include "src/ctx.h"

-void dump_text(mceTextReader_t *reader, dyn_buffer_t *buf) {
+int dump_text(mceTextReader_t *reader, dyn_buffer_t *buf) {

    mce_skip_attributes(reader);

+    xmlErrorPtr err = xmlGetLastError();
+    if (err != NULL) {
+        if (err->level == XML_ERR_FATAL) {
+            LOG_ERRORF("doc.c", "Got fatal XML error while parsing document: %s", err->message)
+            return -1;
+        } else {
+            LOG_ERRORF("doc.c", "Got recoverable XML error while parsing document: %s", err->message)
+        }
+    }
+
    mce_start_children(reader) {
        mce_start_element(reader, NULL, _X("t")) {
            mce_skip_attributes(reader);
@@ -18,10 +28,14 @@ void dump_text(mceTextReader_t *reader, dyn_buffer_t *buf) {
        } mce_end_element(reader);

        mce_start_element(reader, NULL, NULL) {
-            dump_text(reader, buf);
+            int ret = dump_text(reader, buf);
+            if (ret != 0) {
+                return ret;
+            }
        } mce_end_element(reader);

    } mce_end_children(reader)
+    return 0;
 }

 __always_inline
@@ -52,23 +66,28 @@ int should_read_part(opcPart part) {
 }

 __always_inline
-void read_part(opcContainer *c, dyn_buffer_t *buf, opcPart part, document_t *doc) {
+int read_part(opcContainer *c, dyn_buffer_t *buf, opcPart part, document_t *doc) {

    mceTextReader_t reader;
-    int ret = opcXmlReaderOpen(c, &reader, part, NULL, "UTF-8", 0);
+    int ret = opcXmlReaderOpen(c, &reader, part, NULL, "UTF-8", XML_PARSE_NOWARNING | XML_PARSE_NOERROR | XML_PARSE_NONET);

    if (ret != OPC_ERROR_NONE) {
        LOG_ERRORF(doc->filepath, "(doc.c) opcXmlReaderOpen() returned error code %d", ret);
-        return;
+        return -1;
    }

    mce_start_document(&reader) {
        mce_start_element(&reader, NULL, NULL) {
-                dump_text(&reader, buf);
+            ret = dump_text(&reader, buf);
+            if (ret != 0) {
+                mceTextReaderCleanup(&reader);
+                return -1;
+            }
        } mce_end_element(&reader);
    } mce_end_document(&reader);

    mceTextReaderCleanup(&reader);
+    return 0;
 }

 void parse_doc(void *mem, size_t mem_len, document_t *doc) {
@@ -88,7 +107,10 @@ void parse_doc(void *mem, size_t mem_len, document_t *doc) {
    opcPart part = opcPartGetFirst(c);
    do {
        if (should_read_part(part)) {
-            read_part(c, &buf, part, doc);
+            int ret = read_part(c, &buf, part, doc);
+            if (ret != 0) {
+                break;
+            }
        }
    } while ((part = opcPartGetNext(c, part)));

--- a/src/parsing/media.c
+++ b/src/parsing/media.c
@@ -127,6 +127,7 @@ AVFrame *read_frame(AVFormatContext *pFormatCtx, AVCodecContext *decoder, int st
 #define APPEND_TAG_META(doc, tag_, keyname) \
    text_buffer_t tex = text_buffer_create(-1); \
    text_buffer_append_string0(&tex, tag_->value); \
+    text_buffer_terminate_string(&tex); \
    meta_line_t *meta_tag = malloc(sizeof(meta_line_t) + tex.dyn_buffer.cur); \
    meta_tag->key = keyname; \
    strcpy(meta_tag->strval, tex.dyn_buffer.buf); \
@@ -192,6 +193,24 @@ append_video_meta(AVFormatContext *pFormatCtx, AVFrame *frame, document_t *doc,
                APPEND_TAG_META(doc, tag, MetaArtist)
            } else if (strcmp(tag->key, "ImageDescription") == 0) {
                APPEND_TAG_META(doc, tag, MetaContent)
+            } else if (strcmp(tag->key, "Make") == 0) {
+                APPEND_TAG_META(doc, tag, MetaExifMake)
+            } else if (strcmp(tag->key, "Model") == 0) {
+                APPEND_TAG_META(doc, tag, MetaExifModel)
+            } else if (strcmp(tag->key, "Software") == 0) {
+                APPEND_TAG_META(doc, tag, MetaExifSoftware)
+            } else if (strcmp(tag->key, "FNumber") == 0) {
+                APPEND_TAG_META(doc, tag, MetaExifFNumber)
+            } else if (strcmp(tag->key, "FocalLength") == 0) {
+                APPEND_TAG_META(doc, tag, MetaExifFocalLength)
+            } else if (strcmp(tag->key, "UserComment") == 0) {
+                APPEND_TAG_META(doc, tag, MetaExifUserComment)
+            } else if (strcmp(tag->key, "ISOSpeedRatings") == 0) {
+                APPEND_TAG_META(doc, tag, MetaExifIsoSpeedRatings)
+            } else if (strcmp(tag->key, "ExposureTime") == 0) {
+                APPEND_TAG_META(doc, tag, MetaExifExposureTime)
+            } else if (strcmp(tag->key, "DateTime") == 0) {
+                APPEND_TAG_META(doc, tag, MetaExifDateTime)
            }
        }
    }
--- a/src/parsing/mime_generated.c
+++ b/src/parsing/mime_generated.c
@@ -1293,6 +1293,7 @@ g_hash_table_insert(ext_table, "isu", (gpointer)video_x_isvideo);
 g_hash_table_insert(ext_table, "mjpg", (gpointer)video_x_motion_jpeg);
 g_hash_table_insert(ext_table, "asf", (gpointer)video_x_ms_asf);
 g_hash_table_insert(ext_table, "asx", (gpointer)video_x_ms_asf);
+g_hash_table_insert(ext_table, "wmv", (gpointer)video_x_ms_asf);
 g_hash_table_insert(ext_table, "qtc", (gpointer)video_x_qtc);
 g_hash_table_insert(ext_table, "movie", (gpointer)video_x_sgi_movie);
 g_hash_table_insert(ext_table, "mv", (gpointer)video_x_sgi_movie);
--- a/src/parsing/pdf.c
+++ b/src/parsing/pdf.c
@@ -1,6 +1,11 @@
 #include "pdf.h"
 #include "src/ctx.h"

+#define MIN_OCR_SIZE 350
+#define MIN_OCR_LEN 10
+__thread text_buffer_t thread_buffer;
+
+
 fz_page *render_cover(fz_context *ctx, document_t *doc, fz_document *fzdoc) {

    int err = 0;
@@ -78,7 +83,8 @@ fz_page *render_cover(fz_context *ctx, document_t *doc, fz_document *fzdoc) {
    fz_drop_pixmap(ctx, pixmap);

    if (err != 0) {
-        LOG_WARNINGF(doc->filepath, "fz_new_buffer_from_pixmap_as_png() returned error code [%d] %s", err, ctx->error.message)
+        LOG_WARNINGF(doc->filepath, "fz_new_buffer_from_pixmap_as_png() returned error code [%d] %s", err,
+                     ctx->error.message)
        fz_drop_page(ctx, cover);
        return NULL;
    }
@@ -86,14 +92,13 @@ fz_page *render_cover(fz_context *ctx, document_t *doc, fz_document *fzdoc) {
    return cover;
 }

-void fz_err_callback(void *user, __attribute__((unused)) const char *message) {
+void fz_err_callback(void *user, UNUSED(const char *message)) {
    if (LogCtx.verbose) {
        document_t *doc = (document_t *) user;
        LOG_WARNINGF(doc->filepath, "FZ: %s", message)
    }
 }

-
 __always_inline
 void init_ctx(fz_context *ctx, document_t *doc) {
    fz_disable_icc(ctx);
@@ -124,6 +129,43 @@ int read_stext_block(fz_stext_block *block, text_buffer_t *tex) {
    return 0;
 }

+#define IS_VALID_BPP(d) (d==1 || d==2 || d==4 || d==8 || d==16 || d==24 || d==32)
+
+void fill_image(fz_context *ctx, UNUSED(fz_device *dev),
+                fz_image *img, UNUSED(fz_matrix ctm), UNUSED(float alpha),
+                UNUSED(fz_color_params color_params)) {
+
+    int l2factor = 0;
+
+    if (img->w > MIN_OCR_SIZE && img->h > MIN_OCR_SIZE && IS_VALID_BPP(img->n)) {
+
+        fz_pixmap *pix = img->get_pixmap(ctx, img, NULL, img->w, img->h, &l2factor);
+
+        if (pix->h > MIN_OCR_SIZE && img->h > MIN_OCR_SIZE && img->xres != 0) {
+            TessBaseAPI *api = TessBaseAPICreate();
+            TessBaseAPIInit3(api, ScanCtx.tesseract_path, ScanCtx.tesseract_lang);
+
+            TessBaseAPISetImage(api, pix->samples, pix->w, pix->h, pix->n, pix->stride);
+            TessBaseAPISetSourceResolution(api, pix->xres);
+
+            char *text = TessBaseAPIGetUTF8Text(api);
+            size_t len = strlen(text);
+            if (len >= MIN_OCR_LEN) {
+                text_buffer_append_string(&thread_buffer, text, len - 1);
+                LOG_DEBUGF(
+                        "pdf.c",
+                        "(OCR) %dx%d got %dB from tesseract (%s), buffer:%dB",
+                        pix->w, pix->h, len, ScanCtx.tesseract_lang, thread_buffer.dyn_buffer.cur
+                )
+            }
+
+            TessBaseAPIEnd(api);
+            TessBaseAPIDelete(api);
+        }
+        fz_drop_pixmap(ctx, pix);
+    }
+}
+
 void parse_pdf(void *buf, size_t buf_len, document_t *doc) {

    if (buf == NULL) {
@@ -210,7 +252,7 @@ void parse_pdf(void *buf, size_t buf_len, document_t *doc) {

    if (ScanCtx.content_size > 0) {
        fz_stext_options opts = {0};
-        text_buffer_t text_buf = text_buffer_create(ScanCtx.content_size);
+        thread_buffer = text_buffer_create(ScanCtx.content_size);

        for (int current_page = 0; current_page < page_count; current_page++) {
            fz_page *page = NULL;
@@ -224,7 +266,7 @@ void parse_pdf(void *buf, size_t buf_len, document_t *doc) {
                    err = ctx->error.errcode;
                if (err != 0) {
                    LOG_WARNINGF(doc->filepath, "fz_load_page() returned error code [%d] %s", err, ctx->error.message)
-                    text_buffer_destroy(&text_buf);
+                    text_buffer_destroy(&thread_buffer);
                    fz_drop_page(ctx, page);
                    fz_drop_stream(ctx, stream);
                    fz_drop_document(ctx, fzdoc);
@@ -235,6 +277,15 @@ void parse_pdf(void *buf, size_t buf_len, document_t *doc) {

            fz_stext_page *stext = fz_new_stext_page(ctx, fz_bound_page(ctx, page));
            fz_device *dev = fz_new_stext_device(ctx, stext, &opts);
+            dev->stroke_path = NULL;
+            dev->stroke_text = NULL;
+            dev->clip_text = NULL;
+            dev->clip_stroke_path = NULL;
+            dev->clip_stroke_text = NULL;
+
+            if (ScanCtx.tesseract_lang != NULL) {
+                dev->fill_image = fill_image;
+            }

            fz_var(err);
            fz_try(ctx)
@@ -249,7 +300,7 @@ void parse_pdf(void *buf, size_t buf_len, document_t *doc) {

            if (err != 0) {
                LOG_WARNINGF(doc->filepath, "fz_run_page() returned error code [%d] %s", err, ctx->error.message)
-                text_buffer_destroy(&text_buf);
+                text_buffer_destroy(&thread_buffer);
                fz_drop_page(ctx, page);
                fz_drop_stext_page(ctx, stext);
                fz_drop_stream(ctx, stream);
@@ -260,7 +311,7 @@ void parse_pdf(void *buf, size_t buf_len, document_t *doc) {

            fz_stext_block *block = stext->first_block;
            while (block != NULL) {
-                int ret = read_stext_block(block, &text_buf);
+                int ret = read_stext_block(block, &thread_buffer);
                if (ret == TEXT_BUF_FULL) {
                    break;
                }
@@ -269,18 +320,18 @@ void parse_pdf(void *buf, size_t buf_len, document_t *doc) {
            fz_drop_stext_page(ctx, stext);
            fz_drop_page(ctx, page);

-            if (text_buf.dyn_buffer.cur >= text_buf.dyn_buffer.size) {
+            if (thread_buffer.dyn_buffer.cur >= thread_buffer.dyn_buffer.size) {
                break;
            }
        }
-        text_buffer_terminate_string(&text_buf);
+        text_buffer_terminate_string(&thread_buffer);

-        meta_line_t *meta_content = malloc(sizeof(meta_line_t) + text_buf.dyn_buffer.cur);
+        meta_line_t *meta_content = malloc(sizeof(meta_line_t) + thread_buffer.dyn_buffer.cur);
        meta_content->key = MetaContent;
-        memcpy(meta_content->strval, text_buf.dyn_buffer.buf, text_buf.dyn_buffer.cur);
+        memcpy(meta_content->strval, thread_buffer.dyn_buffer.buf, thread_buffer.dyn_buffer.cur);
        APPEND_META(doc, meta_content)

-        text_buffer_destroy(&text_buf);
+        text_buffer_destroy(&thread_buffer);
    }

    fz_drop_stream(ctx, stream);
--- a/src/parsing/text.c
+++ b/src/parsing/text.c
@@ -25,6 +25,7 @@ void parse_text(int bytes_read, struct vfile *f, char *buf, document_t *doc) {
    }
    text_buffer_t tex = text_buffer_create(ScanCtx.content_size);
    text_buffer_append_string(&tex, intermediate_buf, intermediate_buf_len);
+    text_buffer_terminate_string(&tex);

    meta_line_t *meta = malloc(sizeof(meta_line_t) + tex.dyn_buffer.cur);
    meta->key = MetaContent;
--- a/src/sist.h
+++ b/src/sist.h
@@ -2,6 +2,7 @@
 #define SIST_H

 #define UUID_STR_LEN 37
+#define UNUSED(x) __attribute__((__unused__))  x

 #include <glib-2.0/glib.h>
 #include <unistd.h>
@@ -32,15 +33,15 @@
 #include <archive_entry.h>
 #include <opc/opc.h>
 #include <libxml/xmlstring.h>
+#define BOOL int
+#include <tesseract/capi.h>

-#ifndef SIST_SCAN_ONLY
 #include <onion/onion.h>
 #include <onion/handler.h>
 #include <onion/block.h>
 #include <onion/shortcuts.h>
 #include <onion/codecs.h>
 #include <curl/curl.h>
-#endif

 #include "cJSON/cJSON.h"

@@ -62,12 +63,10 @@
 #include "log.h"
 #include "utf8.h/utf8.h"

-#ifndef SIST_SCAN_ONLY
 #include "src/index/elastic.h"
 #include "index/web.h"
 #include "web/serve.h"
 #include "web/auth_basic.h"
-#endif

 ;

--- a/src/types.h
+++ b/src/types.h
@@ -2,9 +2,9 @@
 #define SIST2_TYPES_H


-#define META_INT_MASK 0xF0
-#define META_STR_MASK 0xE0
-#define META_LONG_MASK 0xD0
+#define META_INT_MASK 0x80
+#define META_STR_MASK 0x40
+#define META_LONG_MASK 0x20
 #define IS_META_INT(key) (key & META_INT_MASK) == META_INT_MASK
 #define IS_META_LONG(key) (key & META_LONG_MASK) == META_LONG_MASK
 #define IS_META_STR(meta) (meta->key & META_STR_MASK) == META_STR_MASK
@@ -31,16 +31,31 @@ enum metakey {
    MetaTitle = 12 | META_STR_MASK,
    MetaFontName = 13 | META_STR_MASK,
    MetaParent = 14 | META_STR_MASK,
+    MetaExifMake = 15 | META_STR_MASK,
+    MetaExifSoftware = 16 | META_STR_MASK,
+    MetaExifExposureTime = 17 | META_STR_MASK,
+    MetaExifFNumber = 18 | META_STR_MASK,
+    MetaExifFocalLength = 19 | META_STR_MASK,
+    MetaExifUserComment = 20 | META_STR_MASK,
+    MetaExifModel = 21 | META_STR_MASK,
+    MetaExifIsoSpeedRatings = 22 | META_STR_MASK,
+    MetaExifDateTime = 23 | META_STR_MASK,
+    //Note to self: this will break after 31 entries
 };

+#define INDEX_TYPE_BIN "binary"
+#define INDEX_TYPE_JSON "json"
+#define INDEX_VERSION_EXTERNAL "_external_v1"
+
 typedef struct index_descriptor {
    char uuid[UUID_STR_LEN];
-    char version[6];
+    char version[64];
    long timestamp;
    char root[PATH_MAX];
    char rewrite_url[8196];
    short root_len;
    char name[1024];
+    char type[64];
 } index_descriptor_t;

 typedef struct index_t {
@@ -70,8 +85,6 @@ typedef struct document {
    short ext;
    meta_line_t *meta_head;
    meta_line_t *meta_tail;
-    struct document *child_head;
-    struct document *child_tail;
    char *filepath;
 } document_t;

@@ -79,8 +92,6 @@ typedef struct vfile vfile_t;

 typedef int (*read_func_t)(struct vfile *, void *buf, size_t size);

-typedef int (*seek_func_t)(struct vfile *, size_t size, int whence);
-
 typedef void (*close_func_t)(struct vfile *);

 typedef struct vfile {
--- a/src/util.c
+++ b/src/util.c
@@ -1,4 +1,5 @@
 #include "util.h"
+#include "src/ctx.h"

 dyn_buffer_t dyn_buffer_create() {
    dyn_buffer_t buf;
@@ -90,8 +91,12 @@ text_buffer_t text_buffer_create(int max_size) {
 }

 void text_buffer_terminate_string(text_buffer_t *buf) {
+    if (*(buf->dyn_buffer.buf + buf->dyn_buffer.cur - 1) == ' ') {
+        *(buf->dyn_buffer.buf + buf->dyn_buffer.cur - 1) = '\0';
+    } else {
        dyn_buffer_write_char(&buf->dyn_buffer, '\0');
    }
+}

 __always_inline
 int utf8_validchr(const char *s) {
@@ -149,7 +154,6 @@ int text_buffer_append_string(text_buffer_t *buf, char *str, size_t len) {
        (0xe0 == (0xf0 & str[0]) && len < 3) ||
        (0xc0 == (0xe0 & str[0]) && len == 1) ||
        *(str) == 0) {
-        text_buffer_terminate_string(buf);
        return 0;
    }

@@ -158,7 +162,6 @@ int text_buffer_append_string(text_buffer_t *buf, char *str, size_t len) {
            text_buffer_append_char(buf, c);
        }
    }
-    text_buffer_terminate_string(buf);
    return 0;
 }

@@ -169,13 +172,12 @@ int text_buffer_append_string0(text_buffer_t *buf, char *str) {
            text_buffer_append_char(buf, c);
        }
    }
-    text_buffer_terminate_string(buf);
 }

 int text_buffer_append_char(text_buffer_t *buf, int c) {

-    if (SHOULD_IGNORE_CHAR(c)) {
-        if (!buf->last_char_was_whitespace) {
+    if (SHOULD_IGNORE_CHAR(c) || c == ' ') {
+        if (!buf->last_char_was_whitespace && buf->dyn_buffer.cur != 0) {
            dyn_buffer_write_char(&buf->dyn_buffer, ' ');
            buf->last_char_was_whitespace = TRUE;

@@ -320,4 +322,29 @@ GHashTable *incremental_get_table() {
    return file_table;
 }

+const char *find_file_in_paths(const char *paths[], const char *filename) {
+
+    for (int i = 0; paths[i] != NULL; i++) {
+
+        char *apath = abspath(paths[i]);
+        if (apath == NULL) {
+            continue;
+        }
+
+        char path[PATH_MAX];
+        snprintf(path, sizeof(path), "%s%s", apath, filename);
+
+        LOG_DEBUGF("util.c", "Looking for '%s' in folder '%s'", filename, apath)
+        free(apath);
+
+        struct stat info;
+        int ret = stat(path, &info);
+        if (ret != -1) {
+            return paths[i];
+        }
+    }
+
+    return NULL;
+}
+

--- a/src/util.h
+++ b/src/util.h
@@ -7,7 +7,7 @@
 #define INITIAL_BUF_SIZE 1024 * 16

 #define SHOULD_IGNORE_CHAR(c) !(SHOULD_KEEP_CHAR(c))
-#define SHOULD_KEEP_CHAR(c) (c >= (int)'!')
+#define SHOULD_KEEP_CHAR(c) ((c >= '\'' && c <= ';') || (c >= 'A' && c <= 'z') || (c > 127))


 typedef struct dyn_buffer {
@@ -74,5 +74,6 @@ int incremental_get(GHashTable *table, unsigned long inode_no);

 int incremental_mark_file_for_copy(GHashTable *table, unsigned long inode_no);

+const char *find_file_in_paths(const char **paths, const char *filename);

 #endif
--- a/src/web/serve.c
+++ b/src/web/serve.c
@@ -110,7 +110,7 @@ int thumbnail(void *p, onion_request *req, onion_response *res) {
    int written = onion_response_write(res, data, data_len);
    onion_response_flush(res);
    if (written != data_len || data_len == 0) {
-        printf("Couldn't write thumb\n");
+        LOG_DEBUG("serve.c", "Couldn't write thumbnail");
    }
    free(data);

@@ -181,7 +181,12 @@ int chunked_response_file(const char *filename, const char *mime,
        }
    }
    onion_response_set_length(res, length);
+    if (mime != NULL) {
        onion_response_set_header(res, "Content-Type", mime);
+    } else {
+        onion_response_set_header(res, "Content-Type", "application/octet-stream");
+    }
+
    onion_response_write_headers(res);
    if ((onion_request_get_flags(request) & OR_HEAD) == OR_HEAD) {
        length = 0;
@@ -214,21 +219,13 @@ int chunked_response_file(const char *filename, const char *mime,
    return OCS_PROCESSED;
 }

-int search(void *p, onion_request *req, onion_response *res) {
+int search(UNUSED(void *p), onion_request *req, onion_response *res) {

    int flags = onion_request_get_flags(req);
    if ((flags & OR_METHODS) != OR_POST) {
        return OCS_NOT_PROCESSED;
    }

-    char *scroll_param;
-    const char *scroll = onion_request_get_query(req, "scroll");
-    if (scroll != NULL) {
-        scroll_param = "?scroll=3m";
-    } else {
-        scroll_param = "";
-    }
-
    const struct onion_block_t *block = onion_request_get_data(req);

    if (block == NULL) {
@@ -236,7 +233,7 @@ int search(void *p, onion_request *req, onion_response *res) {
    }

    char url[4096];
-    snprintf(url, 4096, "%s/sist2/_search%s", WebCtx.es_url, scroll_param);
+    snprintf(url, 4096, "%s/sist2/_search", WebCtx.es_url);
    response_t *r = web_post(url, onion_block_data(block), "Content-Type: application/json");

    set_default_headers(res);
@@ -254,43 +251,6 @@ int search(void *p, onion_request *req, onion_response *res) {
    return OCS_PROCESSED;
 }

-int scroll(void *p, onion_request *req, onion_response *res) {
-
-    int flags = onion_request_get_flags(req);
-    if ((flags & OR_METHODS) != OR_GET) {
-        return OCS_NOT_PROCESSED;
-    }
-
-    char url[4096];
-    snprintf(url, 4096, "%s/_search/scroll", WebCtx.es_url);
-
-    const char *scroll_id = onion_request_get_query(req, "scroll_id");
-
-    cJSON *json = cJSON_CreateObject();
-    cJSON_AddStringToObject(json, "scroll_id", scroll_id);
-    cJSON_AddStringToObject(json, "scroll", "3m");
-
-    char *json_str = cJSON_PrintUnformatted(json);
-    response_t *r = web_post(url, json_str, "Content-Type: application/json");
-
-    cJSON_Delete(json);
-    cJSON_free(json_str);
-
-    if (r->status_code != 200) {
-        free_response(r);
-        return OCS_NOT_PROCESSED;
-    }
-
-    set_default_headers(res);
-    onion_response_set_header(res, "Content-Type", "application/json");
-    onion_response_set_header(res, "Content-Disposition", "application/json");
-    onion_response_set_length(res, r->size);
-    onion_response_write(res, r->body, r->size);
-    free_response(r);
-
-    return OCS_PROCESSED;
-}
-
 int serve_file_from_url(cJSON *json, index_t *idx, onion_request *req, onion_response *res) {

    const char *path = cJSON_GetObjectItem(json, "path")->valuestring;
@@ -327,7 +287,7 @@ int serve_file_from_disk(cJSON *json, index_t *idx, onion_request *req, onion_re
    return chunked_response_file(full_path, mime, 1, req, res);
 }

-int index_info(void *p, onion_request *req, onion_response *res) {
+int index_info(UNUSED(void *p), onion_request *req, onion_response *res) {
    cJSON *json = cJSON_CreateObject();
    cJSON *arr = cJSON_AddArrayToObject(json, "indices");

@@ -353,14 +313,47 @@ int index_info(void *p, onion_request *req, onion_response *res) {
    return OCS_PROCESSED;
 }

-int file(void *p, onion_request *req, onion_response *res) {
+
+int document_info(UNUSED(void *p), onion_request *req, onion_response *res) {

    const char *arg_uuid = onion_request_get_query(req, "1");
    if (arg_uuid == NULL) {
        return OCS_PROCESSED;
    }

-    char *next = arg_uuid;
+    cJSON *doc = elastic_get_document(arg_uuid);
+    cJSON *source = cJSON_GetObjectItem(doc, "_source");
+
+    cJSON *index_id = cJSON_GetObjectItem(source, "index");
+    if (index_id == NULL) {
+        cJSON_Delete(doc);
+        return OCS_NOT_PROCESSED;
+    }
+
+    index_t *idx = get_index_by_id(index_id->valuestring);
+    if (idx == NULL) {
+        cJSON_Delete(doc);
+        return OCS_NOT_PROCESSED;
+    }
+
+    onion_response_set_header(res, "Content-Type", "application/json");
+
+    char *json_str = cJSON_PrintUnformatted(source);
+    onion_response_write0(res, json_str);
+    free(json_str);
+    cJSON_Delete(doc);
+
+    return OCS_PROCESSED;
+}
+
+int file(UNUSED(void *p), onion_request *req, onion_response *res) {
+
+    const char *arg_uuid = onion_request_get_query(req, "1");
+    if (arg_uuid == NULL) {
+        return OCS_PROCESSED;
+    }
+
+    const char *next = arg_uuid;
    cJSON *doc = NULL;
    cJSON *index_id = NULL;
    cJSON *source = NULL;
@@ -398,6 +391,23 @@ int file(void *p, onion_request *req, onion_response *res) {
    return ret;
 }

+int status(UNUSED(void *p), UNUSED(onion_request *req), onion_response *res) {
+    set_default_headers(res);
+
+    onion_response_set_header(res, "Content-Type", "application/x-empty");
+
+    char *status = elastic_get_status();
+    if (strcmp(status, "open") == 0) {
+        onion_response_set_code(res, 204);
+    } else {
+        onion_response_set_code(res, 500);
+    }
+
+    free(status);
+
+    return OCS_PROCESSED;
+}
+
 void serve(const char *hostname, const char *port) {
    onion *o = onion_new(O_POOL);
    onion_set_timeout(o, 3500);
@@ -416,7 +426,7 @@ void serve(const char *hostname, const char *port) {
    onion_url_add(urls, "img/sprite-skin-flat.png", img_sprite_skin_flag);

    onion_url_add(urls, "es", search);
-    onion_url_add(urls, "scroll", scroll);
+    onion_url_add(urls, "status", status);
    onion_url_add(
            urls,
            "^t/([a-fA-F0-9]{8}-[a-fA-F0-9]{4}-[a-fA-F0-9]{4}-[a-fA-F0-9]{4}-[a-fA-F0-9]{12})/"
@@ -424,6 +434,7 @@ void serve(const char *hostname, const char *port) {
            thumbnail
    );
    onion_url_add(urls, "^f/([a-fA-F0-9]{8}-[a-fA-F0-9]{4}-[a-fA-F0-9]{4}-[a-fA-F0-9]{4}-[a-fA-F0-9]{12})$", file);
+    onion_url_add(urls, "^d/([a-fA-F0-9]{8}-[a-fA-F0-9]{4}-[a-fA-F0-9]{4}-[a-fA-F0-9]{4}-[a-fA-F0-9]{12})$", document_info);
    onion_url_add(urls, "i", index_info);


--- a/src/web/static_generated.c
+++ b/src/web/static_generated.c
--- a/web/css/dark.css
+++ b/web/css/dark.css
@@ -2,6 +2,26 @@
    outline: 0;
 }

+.info-icon {
+    width: 1rem;
+    margin-right: 0.2rem;
+    cursor: pointer;
+    color: #757575;
+    line-height: 1rem;
+    height: 1.1rem;
+}
+
+.info-icon:hover {
+    color: inherit;
+}
+
+.modal-title {
+    max-width: calc(100% - 2rem);
+    overflow: hidden;
+    text-overflow: ellipsis;
+}
+
+
 .path-row {
    display: -ms-flexbox;
    display: flex;
@@ -32,7 +52,7 @@ body {
    margin-top: 1em;
 }

-.card {
+.card, .modal-content {
    margin-top: 1em;
    background: #212121;
    color: #e0e0e0;
@@ -40,6 +60,27 @@ body {
    border: none;
 }

+.table {
+    color: #e0e0e0;
+}
+
+.table td, .table th {
+    border: none;
+}
+
+.table thead th {
+    border-bottom: 1px solid #646464;
+}
+
+.modal-header .close {
+    color: #e0e0e0;
+    text-shadow: none;
+}
+
+.modal-header {
+    border-bottom: 1px solid #646464;
+}
+
 .sub-document {
    background: #37474F !important;
 }
@@ -134,6 +175,8 @@ body {

 .file-title {
    width: 100%;
+    line-height: 1rem;
+    height: 1.1rem;
    font-size: 10pt;
    white-space: nowrap;
    text-overflow: ellipsis;
@@ -380,3 +423,21 @@ option {
    margin-top: -14px;
    font-size: 11px;
 }
+
+@media (min-width: 800px) {
+    .small-btn {
+        display: none;
+    }
+    .large-btn {
+        display: inherit;
+    }
+}
+
+@media (max-width: 801px) {
+    .small-btn {
+        display: inherit;
+    }
+    .large-btn {
+        display: none;
+    }
+}
--- a/web/css/light.css
+++ b/web/css/light.css
@@ -2,6 +2,25 @@
    outline: 0;
 }

+.info-icon {
+    width: 1rem;
+    margin-right: 0.2rem;
+    cursor: pointer;
+    color: #757575;
+    line-height: 1rem;
+    height: 1rem;
+}
+
+.info-icon:hover {
+    color: inherit;
+}
+
+.modal-title {
+    max-width: calc(100% - 2rem);
+    overflow: hidden;
+    text-overflow: ellipsis;
+}
+
 .path-row {
    display: -ms-flexbox;
    display: flex;
@@ -100,6 +119,8 @@ body {

 .file-title {
    width: 100%;
+    line-height: 1rem;
+    height: 1.1rem;
    font-size: 10pt;
    white-space: nowrap;
    text-overflow: ellipsis;
@@ -266,3 +287,21 @@ mark {
    margin-top: -14px;
    font-size: 11px;
 }
+
+@media (min-width: 800px) {
+    .small-btn {
+        display: none;
+    }
+    .large-btn {
+        display: inherit;
+    }
+}
+
+@media (max-width: 801px) {
+    .small-btn {
+        display: inherit;
+    }
+    .large-btn {
+        display: none;
+    }
+}
--- a/web/css/smartphoto.min.css
+++ b/web/css/smartphoto.min.css
--- a/web/js/dom.js
+++ b/web/js/dom.js
@@ -75,6 +75,10 @@ function shouldPlayVideo(hit) {
    return videoc !== "hevc" && videoc !== "mpeg2video" && videoc !== "wmv3";
 }

+function shouldDisplayRawImage(hit) {
+    return hit["_source"]["mime"] && hit["_source"]["mime"].startsWith("image/") && hit["_source"]["videoc"] !== "tiff";
+}
+
 function makePlaceholder(w, h, small) {
    let calc;
    if (small) {
@@ -96,10 +100,14 @@ function makePlaceholder(w, h, small) {
    return el;
 }

+function ext(hit) {
+    return hit["_source"].hasOwnProperty("extension") && hit["_source"]["extension"] !== "" ? "." + hit["_source"]["extension"] : "";
+}
+
 function makeTitle(hit) {
    let title = document.createElement("div");
    title.setAttribute("class", "file-title");
-    let extension = hit["_source"].hasOwnProperty("extension") && hit["_source"]["extension"] !== "" ? "." + hit["_source"]["extension"] : "";
+    let extension = ext(hit);

    applyNameToTitle(hit, title, extension);

@@ -113,7 +121,7 @@ function getTags(hit, mimeCategory) {
    switch (mimeCategory) {
        case "video":
        case "image":
-            if (hit["_source"].hasOwnProperty("videoc")) {
+            if (hit["_source"].hasOwnProperty("videoc") && hit["_source"]["videoc"]) {
                const formatTag = document.createElement("span");
                formatTag.setAttribute("class", "badge badge-pill badge-video");
                formatTag.appendChild(document.createTextNode(hit["_source"]["videoc"].replace(" ", "")));
@@ -121,7 +129,7 @@ function getTags(hit, mimeCategory) {
            }
            break;
        case "audio": {
-            if (hit["_source"].hasOwnProperty("audioc")) {
+            if (hit["_source"].hasOwnProperty("audioc") && hit["_source"]["audioc"]) {
                let formatTag = document.createElement("span");
                formatTag.setAttribute("class", "badge badge-pill badge-audio");
                formatTag.appendChild(document.createTextNode(hit["_source"]["audioc"]));
@@ -153,11 +161,44 @@ function getTags(hit, mimeCategory) {
    return tags
 }

-/**
- *
- * @param hit
- * @returns {Element}
- */
+function infoButtonCb(hit) {
+    return () => {
+        getDocumentInfo(hit["_id"]).then(doc => {
+            $("#modal-title").text(doc["name"] + ext(hit));
+
+            const tbody = $("<tbody>");
+            $("#modal-body").empty()
+                .append($("<table class='table table-sm'>")
+                    .append($("<thead>")
+                        .append($("<tr>")
+                            .append($("<th>").text("Field"))
+                            .append($("<th>").text("Value"))
+                        )
+                    )
+                    .append(tbody)
+                );
+
+            const displayFields = new Set([
+                "mime", "size", "mtime", "path", "title", "width", "height", "duration", "audioc", "videoc",
+                "bitrate", "artist", "album", "album_artist", "genre", "title", "font_name", "tag"
+            ]);
+            Object.keys(doc)
+                .filter(key => key.startsWith("_keyword.") || key.startsWith("_text.") || displayFields.has(key) || key.startsWith("exif_"))
+                .forEach(key => {
+                    tbody.append($("<tr>")
+                        .append($("<td>").text(key))
+                        .append($("<td>").text(doc[key]))
+                    );
+                });
+            if (doc.hasOwnProperty("content") && doc["content"]) {
+                $("#modal-body").append($("<div class='content-div'>").text(doc["content"]))
+            }
+
+            $("#modal").modal();
+        });
+    }
+}
+
 function createDocCard(hit) {
    let docCard = document.createElement("div");
    docCard.setAttribute("class", "card");
@@ -172,6 +213,7 @@ function createDocCard(hit) {
    let link = document.createElement("a");
    link.setAttribute("href", "f/" + hit["_id"]);
    link.setAttribute("target", "_blank");
+    link.style.maxWidth = "calc(100% - 1.2rem)";
    link.appendChild(title);

    if (hit["_source"].hasOwnProperty("parent")) {
@@ -271,7 +313,15 @@ function createDocCard(hit) {
    sizeTag.setAttribute("class", "text-muted");
    tagContainer.appendChild(sizeTag);

-    docCardBody.appendChild(link);
+    const titleWrapper = document.createElement("div");
+    titleWrapper.style.display = "flex";
+
+    const infoButton = makeInfoButton(hit);
+
+    titleWrapper.appendChild(infoButton);
+    titleWrapper.appendChild(link);
+
+    docCardBody.appendChild(titleWrapper);
    docCard.appendChild(docCardBody);

    docCardBody.appendChild(tagContainer);
@@ -337,6 +387,15 @@ function makeThumbnail(mimeCategory, hit, imgWrapper, small) {
        }
        thumbnail.setAttribute("src", `t/${hit["_source"]["index"]}/${hit["_id"]}`);

+        if (!hit["_source"]["parent"] && shouldDisplayRawImage(hit)) {
+            imgWrapper.setAttribute("id", "sp" + hit["_id"]);
+            imgWrapper.setAttribute("data-src", `t/${hit["_source"]["index"]}/${hit["_id"]}`);
+            imgWrapper.setAttribute("href", `f/${hit["_id"]}`);
+            imgWrapper.setAttribute("data-caption", hit["_source"]["path"] + "/" + hit["_source"]["name"] + ext(hit));
+            imgWrapper.setAttribute("data-group", "p" + Math.floor(docCount / SIZE));
+            imgWrapper.classList.add("sp");
+        }
+
        const placeholder = makePlaceholder(hit["_source"]["width"], hit["_source"]["height"], small);
        imgWrapper.appendChild(placeholder);

@@ -352,6 +411,14 @@ function makeThumbnail(mimeCategory, hit, imgWrapper, small) {
    return thumbnail;
 }

+function makeInfoButton(hit) {
+    const infoButton = document.createElement("span");
+    infoButton.appendChild(document.createTextNode("🛈"));
+    infoButton.setAttribute("class", "info-icon");
+    infoButton.addEventListener("click", infoButtonCb(hit));
+    return infoButton;
+}
+
 function createDocLine(hit) {

    const mime = hit["_source"]["mime"];
@@ -372,6 +439,8 @@ function createDocLine(hit) {
        isSubDocument = true;
    }

+    const infoButton = makeInfoButton(hit);
+
    const title = makeTitle(hit);

    let link = document.createElement("a");
@@ -380,8 +449,13 @@ function createDocLine(hit) {
    link.appendChild(title);

    const titleDiv = document.createElement("div");
-    titleDiv.setAttribute("class", "file-title");
-    titleDiv.appendChild(link);
+
+    const titleWrapper = document.createElement("div");
+    titleWrapper.style.display = "flex";
+    titleWrapper.appendChild(infoButton);
+    titleWrapper.appendChild(link);
+
+    titleDiv.appendChild(titleWrapper);

    line.appendChild(media);

@@ -442,8 +516,7 @@ function makePreloader() {
 function makePageIndicator(searchResult) {
    let pageIndicator = document.createElement("div");
    pageIndicator.setAttribute("class", "page-indicator font-weight-light");
-    const totalHits = searchResult["hits"]["total"].hasOwnProperty("value")
-        ? searchResult["hits"]["total"]["value"] : searchResult["hits"]["total"];
+    const totalHits = searchResult["aggregations"]["total_count"]["value"];
    pageIndicator.appendChild(document.createTextNode(docCount + " / " + totalHits));
    return pageIndicator;
 }
@@ -490,8 +563,7 @@ function makeStatsCard(searchResult) {
    });

    let stat = document.createElement("span");
-    const totalHits = searchResult["hits"]["total"].hasOwnProperty("value")
-        ? searchResult["hits"]["total"]["value"] : searchResult["hits"]["total"];
+    const totalHits = searchResult["aggregations"]["total_count"]["value"];
    stat.appendChild(document.createTextNode(totalHits + " results in " + searchResult["took"] + "ms"));

    statsCardBody.appendChild(stat);
--- a/web/js/jquery-smartphoto.min.js
+++ b/web/js/jquery-smartphoto.min.js
--- a/web/js/search.js
+++ b/web/js/search.js
@@ -6,7 +6,8 @@ let tagTree;

 let searchBar = document.getElementById("searchBar");
 let pathBar = document.getElementById("pathBar");
-let scroll_id = null;
+let lastDoc = null;
+let reachedEnd = false;
 let docCount = 0;
 let coolingDown = false;
 let searchBusy = true;
@@ -81,6 +82,13 @@ $.jsonPost("i").then(resp => {
    });
 });

+function getDocumentInfo(id) {
+    return $.getJSON("d/" + id).fail(e => {
+        console.log(e);
+        showEsError();
+    })
+}
+
 function handleTreeClick(tree) {
    return (event, node, handler) => {
        event.preventTreeDefault();
@@ -144,8 +152,8 @@ $.jsonPost("es", {
        target: '#mimeTree'
    });
    mimeTree.on("node.click", handleTreeClick(mimeTree));
-    mimeTree.select();
-    mimeTree.node("any").deselect();
+    mimeTree.deselect();
+    mimeTree.node("any").select();
 });

 function leafTag(tag) {
@@ -252,41 +260,18 @@ function insertHits(resultContainer, hits) {
 }

 window.addEventListener("scroll", function () {
-    if (!coolingDown && !searchBusy) {
+    if (!searchBusy) {
        let threshold = 400;

        if ((window.innerHeight + window.scrollY) >= document.body.offsetHeight - threshold) {
+            if (!reachedEnd) {
                coolingDown = true;
-            doScroll();
+                search(lastDoc);
+            }
        }
    }
 });

-function doScroll() {
-    $.get("scroll", {scroll_id: scroll_id})
-        .then(searchResult => {
-            let searchResults = document.getElementById("searchResults");
-            let hits = searchResult["hits"]["hits"];
-
-            //Page indicator
-            let pageIndicator = makePageIndicator(searchResult);
-            searchResults.appendChild(pageIndicator);
-
-            //Result container
-            let resultContainer = makeResultContainer();
-            searchResults.appendChild(resultContainer);
-
-            insertHits(resultContainer, hits);
-
-            if (hits.length === SIZE) {
-                coolingDown = false;
-            }
-        })
-        .fail(() => {
-            window.location.reload();
-        })
-}
-
 function getSelectedNodes(tree) {
    let selectedNodes = [];

@@ -307,20 +292,24 @@ function getSelectedNodes(tree) {
    return selectedNodes
 }

-function search() {
+function search(after = null) {
+    lastDoc = null;
+
    if (searchBusy) {
        return;
    }
    searchBusy = true;

-    //Clear old search results
    let searchResults = document.getElementById("searchResults");
+    //Clear old search results
+    let preload;
+    if (!after) {
        while (searchResults.firstChild) {
            searchResults.removeChild(searchResults.firstChild);
        }
-
-    const preload = makePreloader();
+        preload = makePreloader();
        searchResults.appendChild(preload);
+    }

    let query = searchBar.value;
    let empty = query === "";
@@ -355,25 +344,25 @@ function search() {
        filters.push([{terms: {"tag": tags}}]);
    }

-    $.jsonPost("es?scroll=1", {
+    let q = {
        "_source": {
-            excludes: ["content"]
+            excludes: ["content", "_tie"]
        },
        query: {
            bool: {
                [condition]: {
-                    multi_match: {
+                    simple_query_string: {
                        query: query,
-                        type: "most_fields",
                        fields: fields,
-                        operator: "and"
+                        default_operator: "and"
                    }
                },
                filter: filters
            }
        },
-        sort: [
-            "_score"
+        "sort": [
+            {"_score": {"order": "desc"}},
+            {"_tie": {"order": "asc"}}
        ],
        highlight: {
            pre_tags: ["<mark>"],
@@ -386,24 +375,45 @@ function search() {
                font_name: {},
            }
        },
-        aggs: {
-            total_size: {"sum": {"field": "size"}}
+        aggs:
+            {
+                total_size: {"sum": {"field": "size"}},
+                total_count: {"value_count": {"field": "size"}}
            },
        size: SIZE,
-    }).then(searchResult => {
-        scroll_id = searchResult["_scroll_id"];
+    };

+    if (after) {
+        q.search_after = [after["_score"], after["_id"]];
+    }
+
+    $.jsonPost("es", q).then(searchResult => {
+        let hits = searchResult["hits"]["hits"];
+        if (hits) {
+            lastDoc = hits[hits.length - 1];
+        }
+
+        if (!after) {
            preload.remove();
-        //Search stats
            searchResults.appendChild(makeStatsCard(searchResult));
+        } else {
+            let pageIndicator = makePageIndicator(searchResult);
+            searchResults.appendChild(pageIndicator);
+        }

        //Setup page
        let resultContainer = makeResultContainer();
        searchResults.appendChild(resultContainer);

-        docCount = 0;
-        insertHits(resultContainer, searchResult["hits"]["hits"]);
+        window.setTimeout(() => {
+            $(".sp").SmartPhoto({animationSpeed: 0, swipeTopToClose: true, showAnimation: false, forceInterval: 50});
+        }, 100);

+        if (!after) {
+            docCount = 0;
+        }
+        reachedEnd = hits.length !== SIZE;
+        insertHits(resultContainer, hits);
        searchBusy = false;
    });
 }
@@ -464,6 +474,15 @@ function updateIndices() {
 document.getElementById("indices").addEventListener("change", updateIndices);
 updateIndices();

+window.onkeyup = function(e) {
+    if (e.key === "/" || e.key === "Escape") {
+        const bar = document.getElementById("searchBar");
+        bar.scrollIntoView();
+        bar.focus();
+    }
+    console.log(e)
+};
+
 //Suggest
 function getPathChoices() {
    return new Promise(getPaths => {
--- a/web/search.html
+++ b/web/search.html
@@ -11,7 +11,7 @@

 <nav class="navbar navbar-expand-lg">
    <a class="navbar-brand" href="/">sist2</a>
-    <span class="badge badge-pill version">v1.1.14</span>
+    <span class="badge badge-pill version">v1.2.12</span>
    <span class="tagline">Lightning-fast file system indexer and search tool </span>
    <a style="margin-left: auto" id="theme" class="btn" title="Toggle theme" href="/">Theme</a>
 </nav>
@@ -31,7 +31,14 @@
                    </div>
                </div>
                <input id="searchBar" type="search" class="form-control" placeholder="Search">
-
+                <div class="input-group-append">
+                    <button class="btn btn-outline-secondary small-btn" type="button" data-toggle="modal"
+                            data-target="#help">?
+                    </button>
+                    <button class="btn btn-outline-secondary large-btn" type="button" data-toggle="modal"
+                            data-target="#help">Help
+                    </button>
+                </div>
            </div>

            <input title="File size" id="sizeSlider" name="size">
@@ -45,10 +52,12 @@
                <div class="col" id="treeTabs">
                    <ul class="nav nav-tabs" role="tablist">
                        <li class="nav-item">
-                            <a class="nav-link active" data-toggle="tab" href="#mime" role="tab" aria-controls="home" aria-selected="true">Mime Types</a>
+                            <a class="nav-link active" data-toggle="tab" href="#mime" role="tab" aria-controls="home"
+                               aria-selected="true">Mime Types</a>
                        </li>
                        <li class="nav-item">
-                            <a class="nav-link" data-toggle="tab" href="#tag" role="tab" aria-controls="profile" aria-selected="false" title="User-defined tags">Tags</a>
+                            <a class="nav-link" data-toggle="tab" href="#tag" role="tab" aria-controls="profile"
+                               aria-selected="false" title="User-defined tags">Tags</a>
                        </li>
                    </ul>
                    <div class="tab-content" id="myTabContent">
@@ -65,6 +74,83 @@
        </div>
    </div>

+    <div class="modal" id="modal" tabindex="-1" role="dialog" aria-labelledby="modal-title" aria-hidden="true">
+        <div class="modal-dialog modal-lg modal-dialog-centered" role="document">
+            <div class="modal-content">
+                <div class="modal-header">
+                    <h5 class="modal-title" id="modal-title"></h5>
+                    <button type="button" class="close" data-dismiss="modal" aria-label="Close">
+                        <span aria-hidden="true">&times;</span>
+                    </button>
+                </div>
+                <div class="modal-body" id="modal-body"></div>
+            </div>
+        </div>
+    </div>
+
+    <div class="modal" id="help" tabindex="-1" role="dialog" aria-labelledby="modal-title" aria-hidden="true">
+        <div class="modal-dialog modal-lg modal-dialog-centered" role="document">
+            <div class="modal-content">
+                <div class="modal-header">
+                    <h5 class="modal-title">Search help</h5>
+                    <button type="button" class="close" data-dismiss="modal" aria-label="Close">
+                        <span aria-hidden="true">&times;</span>
+                    </button>
+                </div>
+                <div class="modal-body">
+
+                    <table class="table">
+                        <tbody>
+                        <tr>
+                            <td><code>+</code></td>
+                            <td>signifies AND operation</td>
+                        </tr>
+                        <tr>
+                            <td><code>|</code></td>
+                            <td>signifies OR operation</td>
+                        </tr>
+                        <tr>
+                            <td><code>-</code></td>
+                            <td>negates a single token</td>
+                        </tr>
+                        <tr>
+                            <td><code>""</code></td>
+                            <td>wraps a number of tokens to signify a phrase for searching</td>
+                        </tr>
+                        <tr>
+                            <td><code>*</code></td>
+                            <td>at the end of a term signifies a prefix query</td>
+                        </tr>
+                        <tr>
+                            <td><code>(</code> and <code>)</code></td>
+                            <td>signify precedence</td>
+                        </tr>
+                        <tr>
+                            <td><code>~N</code></td>
+                            <td>after a word signifies edit distance (fuzziness)</td>
+                        </tr>
+                        <tr>
+                            <td><code>~N</code></td>
+                            <td>after a phrase signifies slop amount</td>
+                        </tr>
+                        </tbody>
+                    </table>
+
+                    <p>For example: <code>"fried eggs" +(eggplant | potato) -frittata</code> will match the phrase
+                        <i>fried eggs</i> and either <i>eggplant</i> or <i>potato</i>, but will ignore results
+                        containing <i>frittata</i>.</p>
+
+                    <p>When neither <code>+</code> or <code>|</code> is specified, the default operator is <code>+</code> (and).</p>
+                    <p>When the <b>Fuzzy</b> option is checked, partial matches are also returned.</p>
+                    <br>
+                    <p>For more information, see <a target="_blank"
+                                                    href="//www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-simple-query-string-query.html">Elasticsearch
+                        documentation</a></p>
+                </div>
+            </div>
+        </div>
+    </div>
+
    <div id="searchResults"></div>
 </div>
Author	SHA1	Message	Date
simon	ae8652d86e	UI tweaks, search syntax (#25 )	2020-02-16 15:24:29 -05:00
simon	849beb09d8	hotfix	2020-02-15 19:33:18 -05:00
simon	e1aaaee617	UI tweak	2020-02-15 09:30:14 -05:00
simon	c02b940945	(I forgot to commit this)	2020-02-14 20:58:10 -05:00
simon	2934ddb07f	Add image viewer (#2 )	2020-02-14 18:28:55 -05:00
simon	7f6f3c02fa	OCR tweaks	2020-02-11 21:13:47 -05:00
simon	7f98d5a682	Fix buffer overflow (whoops)	2020-02-09 18:11:29 -05:00
simon	7eb9c5d7d5	Fix web/index issue with NULL mime types	2020-02-09 17:23:49 -05:00
simon	184439aa38	increase minimum image size for OCR	2020-02-09 14:06:59 -05:00
simon	1ce8b298a1	Display EXIF tags on document info panel, remove march=native on openjp	2020-02-09 13:21:19 -05:00
simon	75f99025d9	add exif dateTime, allow some special characters in text meta	2020-02-09 08:47:13 -05:00
simon	ebe852bd5a	Fix rewrite-url arg	2020-02-09 08:23:17 -05:00
simon	402b103c49	Fix total count for ES 7.5	2020-02-08 09:25:00 -05:00
simon	e9b6e1cdc2	Turn off auto optimisation in libtesseract build	2020-02-08 08:32:04 -05:00
simon	ed1ce8ab5e	Handle XML errors #18	2020-02-07 10:08:01 -05:00
simon	d1fa4febc4	Improve scroll feature, UI fix	2020-02-07 10:08:01 -05:00
simon987	048c55df7b	Update README.md	2020-02-06 19:56:29 -05:00
simon987	f77bc6a025	Update README.md	2020-02-06 19:55:32 -05:00
simon	efdde2734e	version bump	2020-02-06 19:28:05 -05:00
simon	66658fa8f7	Remove trailing/leading white space in text meta fields	2020-02-06 19:27:30 -05:00
simon	df41c251e4	(Breaking!) Add some exif tags	2020-02-06 19:21:50 -05:00
simon	3282ab56ba	Version bump	2020-02-02 09:26:54 -05:00
simon	8300838d30	Suppress XML parsing errors (#18 )	2020-02-02 09:26:03 -05:00
simon	c9870a6d3d	Remove -march=native for release build...	2020-02-02 09:03:06 -05:00
simon	a143cc4fcf	bundle openssl...	2020-02-02 08:39:20 -05:00
simon	9ef1f3781d	fix attempt for #11	2020-02-01 20:04:26 -05:00
simon	bbee8aa721	tesseract ocr path fix	2020-02-01 20:03:59 -05:00
simon	d22f83c797	curl fix	2020-02-01 15:22:43 -05:00
simon	50615486a4	curl fix attempt	2020-02-01 14:42:42 -05:00
simon	ca79e4f797	add /status endpoint	2020-01-28 10:18:37 -05:00
simon987	6a9fd08a80	Merge pull request #21 from simon987/wip-20 Fixes #20	2020-01-27 09:16:00 -05:00
simon987	cab890dc9b	#20 wip	2020-01-27 09:09:42 -05:00
simon987	b3c4faf2df	Update README.md	2020-01-26 12:37:13 -05:00
simon987	353937171a	Update README.md	2020-01-20 15:54:53 -05:00
simon	c80002bea4	Bundle libcurl attempt 2	2020-01-18 11:53:12 -05:00
simon	56adee9d81	Bundle libcurl, libopc bugfix #18	2020-01-18 10:25:02 -05:00
simon	d6493d6d5f	Bundle libpng	2020-01-16 16:21:38 -05:00
simon	0967e9676d	remove static build in CI...	2020-01-16 15:45:18 -05:00
simon	487e998ea0	Display error message on /d/ error	2020-01-16 15:04:50 -05:00
simon	919f45c79c	Document info modal #19	2020-01-16 14:37:19 -05:00
simon987	d42129cfcb	CI fix attempt	2020-01-15 20:11:45 -05:00
simon987	754983e34a	Minor cleanup	2020-01-15 18:16:06 -05:00
simon	7c8a3e2f9d	Support for external json indices	2020-01-14 15:44:31 -05:00
simon	3bb24b4453	Use bundled libtiff	2020-01-14 12:21:26 -05:00
simon	9a56b959d3	Fix build problems...	2020-01-14 10:55:02 -05:00
simon	5e3a2dbcc2	Update README	2020-01-14 10:47:00 -05:00
simon	573f94f24e	OCR support, remove static build	2020-01-14 10:26:40 -05:00
simon	f5db78a69f	Ignore special ascii chars, strip binary in docker build	2020-01-12 10:59:17 -05:00