Settings menu (#30) and UI tweaks

Fixes #20

.gitmodules

@@ -37,3 +37,9 @@
 [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

CMakeLists.txt

@@ -26,6 +26,7 @@ add_executable(
         src/parsing/arc.c src/parsing/arc.h
         src/parsing/doc.c src/parsing/doc.h
         src/log.c src/log.h
+        src/parsing/cbr.h src/parsing/cbr.c
 
         # argparse
         argparse/argparse.h argparse/argparse.c
@@ -45,7 +46,6 @@ add_executable(
 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)
@@ -84,7 +84,8 @@ target_link_directories(
 target_compile_options(sist2
         PRIVATE
         -Ofast
-        # -march=native
+#        -march=native
+        -fPIC
         -fno-stack-protector
         -fomit-frame-pointer
         )
@@ -111,7 +112,7 @@ TARGET_LINK_LIBRARIES(
         ${PROJECT_SOURCE_DIR}/lib/libonion_static.a
 
         pthread
-        curl
+
         m
         bz2
         ${PROJECT_SOURCE_DIR}/lib/libmagic.a
@@ -127,9 +128,17 @@ TARGET_LINK_LIBRARIES(
 
         ${PROJECT_SOURCE_DIR}/lib/libtesseract.a
         ${PROJECT_SOURCE_DIR}/lib/liblept.a
-        png
-        tiff
+        ${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
+
+        pcre
 )
 
 add_custom_target(

Docker/Dockerfile

@@ -3,7 +3,7 @@ MAINTAINER simon987 <me@simon987.net>
 
 RUN apt update
 RUN apt install -y libglib2.0-0 libcurl4 libmagic1 libharfbuzz-bin libopenjp2-7 libarchive13 liblzma5 libzstd1 liblz4-1 \
- curl libtiff5 libpng16-16
+ curl libtiff5 libpng16-16 libpcre3
 
 RUN mkdir -p /usr/share/tessdata && \
     cd /usr/share/tessdata/ && \
@@ -16,4 +16,7 @@ RUN mkdir -p /usr/share/tessdata && \
 
 ADD sist2 /root/sist2
 
+ENV LANG C.UTF-8
+ENV LC_ALL C.UTF-8
+
 ENTRYPOINT ["/root/sist2"]

Docker/build.sh

@@ -5,6 +5,11 @@ 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

README.md

@@ -8,9 +8,12 @@ sist2 (Simple incremental search tool)
 
 *Warning: sist2 is in early development*
 
+![sist2.png](sist2.png)
+
 ## Features
 
 * Fast, low memory usage, multi-threaded
+* Mobile-friendly Web interface
 * Portable (all its features are packaged in a single executable)
 * Extracts text from common file types \*
 * Generates thumbnails \*
@@ -26,73 +29,49 @@ sist2 (Simple incremental search tool)
 
 ## Getting Started
 
-1. Have an [Elasticsearch](https://www.elastic.co/downloads/elasticsearch) instance running
-1. 
+1. Have an Elasticsearch (>= 6.X.X) instance running
+    1. Download [from official website](https://www.elastic.co/downloads/elasticsearch)
+    1. *(or)* Run using docker:
+        ```bash
+       docker run -d --name es1 --net sist2_net -p 9200:9200 \
+            -e "discovery.type=single-node" elasticsearch:7.5.2
+        ```
+    1. *(or)* Run using docker-compose:
+        ```yaml
+          elasticsearch:
+            image: docker.elastic.co/elasticsearch/elasticsearch:7.5.2
+            environment:
+              - discovery.type=single-node
+              - "ES_JAVA_OPTS=-Xms1G -Xmx2G"
+        ```
+1. Download sist2 executable
     1. Download the [latest sist2 release](https://github.com/simon987/sist2/releases) *
-    1. *(or)* Download an [development snapshot](https://files.simon987.net/artifacts/Sist2/Build/) *(Not recommended!)*
+    1. *(or)* Download a [development snapshot](https://files.simon987.net/artifacts/Sist2/Build/) *(Not recommended!)*
     1. *(or)* `docker pull simon987/sist2:latest`
+
+1. See [Usage guide](USAGE.md)
    
 
 \* *Windows users*: **sist2** runs under [WSL](https://en.wikipedia.org/wiki/Windows_Subsystem_for_Linux)    
-\* *Mac users*: See [#1](https://github.com/simon987/sist2/issues/1)
 
 
 ## Example usage
 
-See help page `sist2 --help` for more details.
+See [Usage guide](USAGE.md) for more details
 
-**Scan a directory**
-```bash
-sist2 scan ~/Documents -o ./orig_idx/
-sist2 scan --threads 4 --content-size 16384 /mnt/Pictures
-sist2 scan --incremental ./orig_idx/ -o ./updated_idx/ ~/Documents
-```
-
-**Push index to Elasticsearch or file**
-```bash
-sist2 index --force-reset ./my_idx
-sist2 index --print ./my_idx > raw_documents.ndjson
-```
-
-**Start web interface**
-```bash
-sist2 web --bind 0.0.0.0 --port 4321 ./my_idx1 ./my_idx2 ./my_idx3
-```
-
-### Use sist2 with docker
-
-**scan**
-```bash
-docker run -it \
-    -v /path/to/files/:/files \
-    -v $PWD/out/:/out \
-    simon987/sist2 scan -t 4 /files -o /out/my_idx1
-```
-**index**
-```bash
-docker run -it --network host\
-    -v $PWD/out/:/out \
-    simon987/sist2 index /out/my_idx1
-```
-
-**web**
-```bash
-docker run --rm --network host -d --name sist2\
-    -v $PWD/out/my_idx:/idx \
-    -v $PWD/my/files:/files
-    simon987/sist2 web --bind 0.0.0.0 /idx
-docker stop sist2
-```
+1. Scan a directory: `sist2 scan ~/Documents -o ./docs_idx`
+1. Push index to Elasticsearch: `sist2 index ./docs_idx`
+1. Start web interface: `sist2 web ./docs_idx`
 
 
 ## Format support
 
 File type | Library | Content | Thumbnail | Metadata
 :---|:---|:---|:---|:---
-pdf,xps,cbz,fb2,epub | MuPDF | text+ocr | yes, `png` | title |
+pdf,xps,cbz,cbr,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 |
@@ -117,7 +96,7 @@ To check if a media file can be parsed without *seek*, execute `cat file.mp4 | f
  
 ### OCR
 
-You can enable OCR support for pdf,xps,cbz,fb2,epub file types with the
+You can enable OCR support for pdf,xps,cbz,cbr,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).
 
@@ -142,8 +121,9 @@ binaries.
     ```bash
     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 libtiff5 libpng16-16
+        libcurl4-openssl-dev libbz2-dev yasm libharfbuzz-dev ragel \
+        libarchive-dev libtiff5 libpng16-16 libpango1.0-dev \
+        libxml2-dev
    ```
 
 2. Build

USAGE.md

@@ -0,0 +1,275 @@
+# Usage
+
+*More examples (specifically with docker/compose) are in progress*
+
+* [scan](#scan)
+    * [options](#scan-options)
+    * [examples](#scan-examples)
+    * [index format](#index-format)
+* [index](#index)
+    * [options](#index-options)
+    * [examples](#index-examples)
+* [web](#web)
+    * [options](#web-options)
+    * [examples](#web-examples)
+    * [rewrite_url](#rewrite_url)
+    * [link to specific indices](#link-to-specific-indices)
+
+```
+Usage: sist2 scan [OPTION]... PATH
+   or: sist2 index [OPTION]... INDEX
+   or: sist2 web [OPTION]... INDEX...
+Lightning-fast file system indexer and search tool.
+
+    -h, --help            show this help message and exit
+    -v, --version         Show version and exit
+    --verbose             Turn on logging
+    --very-verbose        Turn on debug messages
+
+Scan options
+    -t, --threads=<int>   Number of threads. DEFAULT=1
+    -q, --quality=<flt>   Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best. DEFAULT=5
+    --size=<int>          Thumbnail size, in pixels. Use negative value to disable. DEFAULT=500
+    --content-size=<int>  Number of bytes to be extracted from text documents. Use negative value to disable. DEFAULT=32768
+    --incremental=<str>   Reuse an existing index and only scan modified files.
+    -o, --output=<str>    Output directory. DEFAULT=index.sist2/
+    --rewrite-url=<str>   Serve files from this url instead of from disk.
+    --name=<str>          Index display name. DEFAULT: (name of the directory)
+    --depth=<int>         Scan up to DEPTH subdirectories deep. Use 0 to only scan files in PATH. DEFAULT: -1
+    --archive=<str>       Archive file mode (skip|list|shallow|recurse). skip: Don't parse, list: only get file names as text, shallow: Don't parse archives inside archives. DEFAULT: recurse
+    --ocr=<str>           Tesseract language (use tesseract --list-langs to see which are installed on your machine)
+    -e, --exclude=<str>   Files that match this regex will not be scanned
+    --fast                Only index file names & mime type
+
+Index options
+    --es-url=<str>        Elasticsearch url with port. DEFAULT=http://localhost:9200
+    -p, --print           Just print JSON documents to stdout.
+    --script-file=<str>   Path to user script.
+    --batch-size=<int>    Index batch size. DEFAULT: 100
+    -f, --force-reset     Reset Elasticsearch mappings and settings. (You must use this option the first time you use the index command)
+
+Web options
+    --es-url=<str>        Elasticsearch url. DEFAULT=http://localhost:9200
+    --bind=<str>          Listen on this address. DEFAULT=localhost
+    --port=<str>          Listen on this port. DEFAULT=4090
+    --auth=<str>          Basic auth in user:password format
+Made by simon987 <me@simon987.net>. Released under GPL-3.0
+
+```
+
+## Scan
+
+### Scan options
+
+* `-t, --threads` 
+      Number of threads for file parsing. **Do not set a number higher than `$(nproc)`!**.
+* `-q, --quality` 
+    Thumbnail quality, on a scale of 1.0 to 31.0, 1.0 being the best. *Does not affect PDF thumbnails quality*
+* `--size` 
+    Thumbnail size in pixels.
+* `--content-size` 
+    Number of bytes of text to be extracted from the content of files (plain text and PDFs).
+    Repeated whitespace and special characters do not count toward this limit.
+* `--incremental`
+    Specify an existing index. Information about files in this index that were not modified (based on *mtime* attribute)
+    will be copied to the new index and will not be parsed again.
+* `-o, --output` Output directory. 
+* `--rewrite-url` Set the `rewrite_url` option for the web module (See [rewrite_url](#rewrite_url)) 
+* `--name` Set the `name` option for the web module
+* `--depth` Maximum scan dept. Set to 0 only scan files directly in the root directory, set to -1 for infinite depth
+* `--archive` Archive file mode.
+    * skip: Don't parse
+    * list: Only get file names as text
+    * shallow: Don't parse archives inside archives.
+    * recurse: Scan archives recursively (default)
+* `--ocr` See [OCR](README.md#OCR)
+* `-e, --exclude` Regex pattern to exclude files. A file is excluded if the pattern matches any 
+    part of the full absolute path.
+    
+    Examples: 
+    * `-e ".*\.ttf"`: Ignore ttf files
+    * `-e ".*\.(ttf|rar)"`: Ignore ttf and rar files
+    * `-e "^/mnt/backups/"`: Ignore all files in the `/mnt/backups/` directory
+    * `-e "^/mnt/Data[12]/"`: Ignore all files in the `/mnt/Data1/` and `/mnt/Data2/` directory
+    * `-e "(^/usr/)|(^/var/)|(^/media/DRIVE-A/tmp/)|(^/media/DRIVE-B/Trash/)"` Exclude the
+     `/usr`, `/var`, `/media/DRIVE-A/tmp`, `/media/DRIVE-B/Trash` directories
+* `--fast` Only index file names and mime type
+
+### Scan examples
+
+Simple scan
+```bash
+sist2 scan ~/Documents
+
+sist2 scan \
+    --threads 4 --content-size 16000000 --quality 1.0 --archive shallow \
+    --name "My Documents" --rewrite-url "http://nas.domain.local/My Documents/" \
+    ~/Documents -o ./documents.idx/
+```
+
+Incremental scan
+```
+sist2 scan --incremental ./orig_idx/ -o ./updated_idx/ ~/Documents
+```
+
+### Index format
+
+A typical `binary` type index structure looks like this:
+```
+documents.idx/
+├── descriptor.json
+├── _index_139965416830720
+├── _index_139965425223424
+├── _index_139965433616128
+├── _index_139965442008832
+└── thumbs
+    ├── data.mdb
+    └── lock.mdb
+```
+
+The `_index_*` files contain the raw binary index data and are not meant to be
+read by other applications. The format is generally compatible across different 
+sist2 versions.
+
+The `thumbs/` folder is a [LMDB](https://en.wikipedia.org/wiki/Lightning_Memory-Mapped_Database)
+database containing the thumbnails.
+
+The `descriptor.json` file contains general information about the index. The 
+following fields are safe to modify manually: `root`, `name`, [rewrite_url](#rewrite_url) and `timestamp`.
+
+
+*Advanced usage*
+
+Instead of using the `scan` module, you can also import an index generated
+by a third party application. The 'external' index must have the following format:
+
+```
+my_index/
+├── descriptor.json
+├── _index_0
+└── thumbs
+    ├── data.mdb
+    └── lock.mdb
+```
+
+*descriptor.json*:
+```json
+{
+    "uuid": "<valid UUID4>",
+    "version": "_external_v1",
+    "root": "(optional)",
+    "name": "<name>",
+    "rewrite_url": "(optional)",
+    "type": "json",
+    "timestamp": 1578971024
+}
+```
+
+*_index_0*: NDJSON format (One json object per line)
+
+```json
+{
+  "_id": "unique uuid for the file",
+  "index": "index uuid4 (same one as descriptor.json!)",
+  "mime": "application/x-cbz",
+  "size": 14341204,
+  "mtime": 1578882996,
+  "extension": "cbz",
+  "name": "my_book",
+  "path": "path/to/books",
+  "content": "text contents of the book",
+  "title": "Title of the book",
+  "tag": ["genre.fiction", "author.someguy", "etc..."],
+  "_keyword": [
+    {"k": "ISBN", "v": "ABCD34789231"}
+  ],
+  "_text": [
+    {"k": "other", "v": "This will be indexed as text"}
+  ]
+}
+```
+
+You can find the full list of supported fields [here](src/io/serialize.c#L90)
+
+The `_keyword.*` items will be indexed and searchable as **keyword** fields (only full matches allowed).
+The `_text.*` items will be indexed and searchable as **text** fields (fuzzy searching allowed)
+
+
+*thumbs/*:
+
+LMDB key-value store. Keys are **binary** 128-bit UUID4s (`_id` field)
+and values are raw image bytes.
+
+Importing an external `binary` type index is technically possible but
+it is currently unsupported and has no guaranties of back/forward compatibility.
+
+
+## Index
+### Index options
+ * `--es-url` 
+ Elasticsearch url and port. If you are using docker, make sure that both containers are on the
+ same network.
+ * `-p, --print` 
+    Print index in JSON format to stdout.
+ * `--script-file` 
+    Path to user script. See [Scripting](scripting/README.md).
+ * `--batch-size=<int>` 
+    Index batch size. Indexing is generally faster with larger batches, but payloads that
+    are too large will fail and additional overhead for retrying with smaller sizes may slow
+    down the process.
+ * `-f, --force-reset` 
+    Reset Elasticsearch mappings and settings.
+    **(You must use this option the first time you use the index command)**.
+    
+### Index examples
+
+**Push to elasticsearch**
+```bash
+sist2 index --force-reset --batch-size 1000 --es-url http://localhost:9200 ./my_index/
+sist2 index ./my_index/
+```
+
+**Save index in JSON format**
+```bash
+sist2 index --print ./my_index/ > my_index.ndjson
+```
+
+**Inspect contents of an index**
+```bash
+sist2 index --print ./my_index/ | jq | less
+```
+
+## Web
+
+### Web options
+ * `--es-url=<str>` Elasticsearch url.
+ * `--bind=<str>` Listen on this address.
+ * `--port=<str>` Listen on this port.
+ * `--auth=<str>` Basic auth in user:password format
+ 
+### Web examples
+
+**Single index**
+```bash
+sist2 web --auth admin:hunter2 --bind 0.0.0.0 --port 8888 my_index
+```
+
+**Multiple indices**
+```bash
+# Indices will be displayed in this order in the web interface
+sist2 web index1 index2 index3 index4
+```
+
+### rewrite_url
+
+When the `rewrite_url` field is not empty, the web module ignores the `root`
+field and will return a HTTP redirect to `<rewrite_url><path>/<name><extension>`
+instead of serving the file from disk. 
+Both the `root` and `rewrite_url` fields are safe to manually modify from the 
+`descriptor.json` file.
+
+### Link to specific indices
+
+To link to specific indices, you can add a list of comma-separated index name to 
+the URL: `?i=<name>,<name>`. By default, indices with `"(nsfw)"` in their name are
+not displayed.

ci/build.sh

@@ -5,4 +5,3 @@
 cmake .
 make
 strip sist2
-strip sist2_scan

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

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"
     }
   }
 }

schema/pipeline.json

@@ -0,0 +1,10 @@
+{
+  "description": "Copy _id to _tie",
+  "processors": [
+    {
+      "script": {
+        "source": "ctx._tie = ctx._id;"
+      }
+    }
+  ]
+}

schema/settings.json

@@ -21,6 +21,12 @@
           "lowercase"
         ]
       },
+      "case_insensitive_kw_analyzer": {
+        "tokenizer": "keyword",
+        "filter": [
+          "lowercase"
+        ]
+      },
       "my_nGram": {
         "tokenizer": "my_nGram_tokenizer",
         "filter": [

scripting/README.md

@@ -54,6 +54,11 @@ script.painless.regex.enabled: true
 ```
 Or, if you're using docker add `-e "script.painless.regex.enabled=true"`
 
+**Tag color**
+
+You can specify the color for an individual tag by appending an 
+hexadecimal color code (`#RRGGBBAA`) to the tag name.
+
 ### Examples
 
 If `(20XX)` is in the file name, add the `year.<year>` tag:
@@ -115,3 +120,33 @@ if (ctx._source.path != "") {
     tags.add("studio." + names[names.length-1]);
 }
 ```
+
+Parse `EXIF:F Number` tag
+```Java
+if (ctx._source?.exif_fnumber != null) {
+    String[] values = ctx._source.exif_fnumber.splitOnToken(' ');
+    String aperture = String.valueOf(Float.parseFloat(values[0]) / Float.parseFloat(values[1]));
+    if (aperture == "NaN") {
+        aperture = "0,0";
+    }
+    tags.add("Aperture.f/" + aperture.replace(".", ","));
+}
+```
+
+Display year and months from `EXIF:DateTime` tag
+```Java
+if (ctx._source?.exif_datetime != null) {
+    SimpleDateFormat parser = new SimpleDateFormat("yyyy:MM:dd HH:mm:ss");
+    Date date = parser.parse(ctx._source.exif_datetime);
+
+    SimpleDateFormat yp = new SimpleDateFormat("yyyy");
+    SimpleDateFormat mp = new SimpleDateFormat("MMMMMMMMM");
+
+    String year = yp.format(date);
+    String month = mp.format(date);
+
+    tags.add("Month." + month);
+    tags.add("Year." + year);
+}
+
+```

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 ..
 
@@ -74,7 +74,8 @@ mv libmagic/src/.libs/libmagic.a .
 cd tesseract
 mkdir build
 cd build
-cmake -DSTATIC=on -DBUILD_TRAINING_TOOLS=off ..
+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 .
@@ -82,9 +83,46 @@ mv tesseract/build/libtesseract.a .
 # leptonica
 cd leptonica
 ./autogen.sh
-./configure --without-zlib --without-jpeg --without-giflib \
+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 leptonica/src/.libs/liblept.a .
+
+# tiff
+cd libtiff
+./autogen.sh
+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 libtiff/libtiff/.libs/libtiff.a .
+
+# png
+cd libpng
+CFLAGS="-fPIC" ./configure --enable-static --disable-shared
+make -j $THREADS
+cd ..
+mv libpng/.libs/libpng16.a .
+
+# openssl...
+git clone --depth 1 -b OpenSSL_1_1_0-stable https://github.com/openssl/openssl
+cd openssl
+./config --prefix=$(pwd)/../ssl
+make depend
+make -j $THREADS
+make install
+cd ..
+mv ./openssl/libcrypto.a ./openssl/libssl.a .
+
+# curl
+wget -nc https://curl.haxx.se/download/curl-7.68.0.tar.gz
+tar -xzf curl-7.68.0.tar.gz
+cd curl-7.68.0
+./configure --disable-ldap --disable-ldaps --without-librtmp --disable-rtsp --disable-crypto-auth \
+  --disable-smtp --without-libidn2 --without-nghttp2 --without-brotli --enable-static --disable-shared \
+  --without-libpsl --with-ssl=$(pwd)/../ssl
+make -j $THREADS
+cd ..
+mv curl-7.68.0/lib/.libs/libcurl.a .

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 ..

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)))

src/cli.c

@@ -15,6 +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);
@@ -136,13 +143,43 @@ int scan_args_validate(scan_args_t *args, int argc, const char **argv) {
 
     if (args->tesseract_lang != NULL) {
         TessBaseAPI *api = TessBaseAPICreate();
-        ret = TessBaseAPIInit3(api, TESS_DATAPATH, args->tesseract_lang);
+
+        char filename[128];
+        sprintf(filename, "%s.traineddata", args->tesseract_lang);
+        const char * path = find_file_in_paths(TESS_DATAPATHS, filename);
+        if (path == NULL) {
+            LOG_FATAL("cli.c", "Could not find tesseract language file!");
+        }
+
+        ret = TessBaseAPIInit3(api, path, args->tesseract_lang);
         if (ret != 0) {
             fprintf(stderr, "Could not initialize tesseract with lang '%s'\n", args->tesseract_lang);
             return 1;
         }
         TessBaseAPIEnd(api);
         TessBaseAPIDelete(api);
+
+        args->tesseract_path = path;
+    }
+
+    if (args->exclude_regex != NULL) {
+        const char *error;
+        int error_offset;
+
+        pcre *re = pcre_compile(args->exclude_regex, 0, &error, &error_offset, 0);
+        if (error != NULL) {
+            LOG_FATALF("cli.c", "pcre_compile returned error: %s (offset:%d)", error, error_offset)
+        }
+
+        pcre_extra *re_extra = pcre_study(re, 0, &error);
+        if (error != NULL) {
+            LOG_FATALF("cli.c", "pcre_study returned error: %s", error)
+        }
+
+        ScanCtx.exclude = re;
+        ScanCtx.exclude_extra = re_extra;
+    } else {
+        ScanCtx.exclude = NULL;
     }
 
     LOG_DEBUGF("cli.c", "arg quality=%f", args->quality)
@@ -156,13 +193,18 @@ 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 ocr=%s", args->tesseract_lang)
+    LOG_DEBUGF("cli.c", "arg tesseract_lang=%s", args->tesseract_lang)
+    LOG_DEBUGF("cli.c", "arg tesseract_path=%s", args->tesseract_path)
+    LOG_DEBUGF("cli.c", "arg exclude=%s", args->exclude_regex)
+    LOG_DEBUGF("cli.c", "arg fast=%d", args->fast)
 
     return 0;
 }
 
 int index_args_validate(index_args_t *args, int argc, const char **argv) {
 
+    LogCtx.verbose = 1;
+
     if (argc < 2) {
         fprintf(stderr, "Required positional argument: PATH.\n");
         return 1;
@@ -224,6 +266,8 @@ 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;

src/cli.h

@@ -17,10 +17,15 @@ typedef struct scan_args {
     char *archive;
     archive_mode_t archive_mode;
     char *tesseract_lang;
+    const char *tesseract_path;
+    char *exclude_regex;
+    int fast;
 } scan_args_t;
 
 scan_args_t *scan_args_create();
+
 void scan_args_destroy(scan_args_t *args);
+
 int scan_args_validate(scan_args_t *args, int argc, const char **argv);
 
 typedef struct index_args {
@@ -44,12 +49,15 @@ typedef struct web_args {
 } web_args_t;
 
 index_args_t *index_args_create();
+
 void index_args_destroy(index_args_t *args);
 
 web_args_t *web_args_create();
+
 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

src/ctx.h

@@ -28,6 +28,10 @@ struct {
 
     pthread_mutex_t mupdf_mu;
     char * tesseract_lang;
+    const char * tesseract_path;
+    pcre *exclude;
+    pcre_extra *exclude_extra;
+    int fast;
 } ScanCtx;
 
 struct {

src/index/elastic.c

@@ -20,6 +20,8 @@ typedef struct es_indexer {
 
 static es_indexer_t *Indexer;
 
+void delete_queue(int max);
+
 void print_json(cJSON *document, const char uuid_str[UUID_STR_LEN]) {
 
     cJSON *line = cJSON_CreateObject();
@@ -27,13 +29,14 @@ 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]) {
@@ -63,12 +66,12 @@ void execute_update_script(const char *script, const char index_id[UUID_STR_LEN]
     cJSON *term_obj = cJSON_AddObjectToObject(query, "term");
     cJSON_AddStringToObject(term_obj, "index", index_id);
 
-    char * str = cJSON_Print(body);
+    char *str = cJSON_Print(body);
 
     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);
@@ -79,31 +82,25 @@ 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);
     }
 
     cJSON_Delete(resp);
 }
 
-void elastic_flush() {
-
-    if (Indexer == NULL) {
-        Indexer = create_indexer(IndexCtx.es_url);
-    }
-
+void *create_bulk_buffer(int max, int *count, size_t *buf_len) {
     es_bulk_line_t *line = Indexer->line_head;
-
-    int count = 0;
+    *count = 0;
 
     size_t buf_size = 0;
     size_t buf_cur = 0;
     char *buf = malloc(1);
 
-    while (line != NULL) {
+    while (line != NULL && *count < max) {
         char action_str[512];
         snprintf(action_str, 512,
-                "{\"index\":{\"_id\":\"%s\", \"_type\":\"_doc\", \"_index\":\"sist2\"}}\n", line->uuid_str);
+                 "{\"index\":{\"_id\":\"%s\", \"_type\":\"_doc\", \"_index\":\"sist2\"}}\n", line->uuid_str);
         size_t action_str_len = strlen(action_str);
 
         size_t line_len = strlen(line->line);
@@ -115,47 +112,101 @@ void elastic_flush() {
         memcpy(buf + buf_cur, line->line, line_len);
         buf_cur += line_len;
 
-        es_bulk_line_t *tmp = line;
         line = line->next;
-        free(tmp);
-        count++;
+        (*count)++;
     }
     buf = realloc(buf, buf_size + 1);
-    *(buf+buf_cur) = '\0';
+    *(buf + buf_cur) = '\0';
 
-    Indexer->line_head = NULL;
-    Indexer->line_tail = NULL;
-    Indexer->queued = 0;
+    *buf_len = buf_cur;
+    return buf;
+}
+
+void _elastic_flush(int max) {
+    size_t buf_len;
+    int count;
+    void *buf = create_bulk_buffer(max, &count, &buf_len);
 
     char bulk_url[4096];
-    snprintf(bulk_url, 4096, "%s/sist2/_bulk", 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);
+    if (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);
-                cJSON_free(str);
+        if (max <= 1) {
+            LOG_ERRORF("elastic.c", "Single document too large, giving up: {%s}", Indexer->line_head->uuid_str)
+            free_response(r);
+            free(buf);
+            delete_queue(1);
+            if (Indexer->queued != 0) {
+                elastic_flush();
             }
+            return;
+        }
+
+        LOG_WARNINGF("elastic.c", "Payload too large, retrying (%d documents)", count);
+
+        free_response(r);
+        free(buf);
+        _elastic_flush(max / 2);
+        return;
+
+    } else if (r->status_code != 200) {
+        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);
+                    LOG_ERRORF("elastic.c", "%s\n", str);
+                    cJSON_free(str);
+                }
+            }
+        }
+
+        cJSON_Delete(ret_json);
+        delete_queue(Indexer->queued);
+
+    } else {
+        LOG_INFOF("elastic.c", "Indexed %d documents (%zukB) <%d>", count, buf_len / 1024, r->status_code);
+
+        delete_queue(max);
+
+        if (Indexer->queued != 0) {
+            elastic_flush();
         }
     }
 
-    cJSON_Delete(ret_json);
-
     free_response(r);
     free(buf);
 }
 
+void delete_queue(int max) {
+    for (int i = 0; i < max; i++) {
+        es_bulk_line_t *tmp = Indexer->line_head;
+        Indexer->line_head = tmp->next;
+        if (Indexer->line_head == NULL) {
+            Indexer->line_tail = NULL;
+        } else {
+            free(tmp);
+        }
+        Indexer->queued -= 1;
+    }
+}
+
+void elastic_flush() {
+
+    if (Indexer == NULL) {
+        Indexer = create_indexer(IndexCtx.es_url);
+    }
+
+    _elastic_flush(Indexer->queued);
+}
+
 void elastic_index_line(es_bulk_line_t *line) {
 
     if (Indexer == NULL) {
@@ -192,13 +243,13 @@ es_indexer_t *create_indexer(const char *url) {
     return indexer;
 }
 
-void destroy_indexer(char * script, char index_id[UUID_STR_LEN]) {
+void destroy_indexer(char *script, char index_id[UUID_STR_LEN]) {
 
     char url[4096];
 
     snprintf(url, sizeof(url), "%s/sist2/_refresh", IndexCtx.es_url);
     response_t *r = web_post(url, "", NULL);
-    printf("Refresh index <%d>\n", r->status_code);
+    LOG_INFOF("elastic.c", "Refresh index <%d>", r->status_code);
     free_response(r);
 
     if (script != NULL) {
@@ -207,12 +258,12 @@ 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 +283,37 @@ 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 +330,28 @@ 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;
+}

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

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);
 
-        cJSON_AddStringToObject(document, "mime", mime_get_mime_text(line.mime));
+        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) {
@@ -262,7 +299,7 @@ void read_index(const char *path, const char index_id[UUID_STR_LEN], index_func
                     break;
                 }
                 default:
-                    LOG_FATALF("serialize.c", "Invalid meta key (corrupt index): %x", key)
+                LOG_FATALF("serialize.c", "Invalid meta key (corrupt index): %x", key)
             }
 
             key = getc(file);
@@ -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;

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);
 

src/io/walk.c

@@ -28,8 +28,18 @@ parse_job_t *create_fs_parse_job(const char *filepath, const struct stat *info,
     return job;
 }
 
+int sub_strings[30];
+#define EXCLUDED(str) (pcre_exec(ScanCtx.exclude, ScanCtx.exclude_extra, filepath, strlen(filepath), 0, 0, sub_strings, sizeof(sub_strings)) >= 0)
+
 int handle_entry(const char *filepath, const struct stat *info, int typeflag, struct FTW *ftw) {
-    if (ftw->level <= ScanCtx.depth && typeflag == FTW_F && S_ISREG(info->st_mode)) {
+
+    if (typeflag == FTW_F && S_ISREG(info->st_mode) && ftw->level <= ScanCtx.depth) {
+
+        if (ScanCtx.exclude != NULL && EXCLUDED(filepath)) {
+            LOG_DEBUGF("walk.c", "Excluded: %s", filepath)
+            return 0;
+        }
+
         parse_job_t *job = create_fs_parse_job(filepath, info, ftw->base);
         tpool_add_work(ScanCtx.pool, parse, job);
     }

src/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) {

src/main.c

@@ -6,7 +6,7 @@
 #define EPILOG "Made by simon987 <me@simon987.net>. Released under GPL-3.0"
 
 
-static const char *const Version = "1.2.0";
+static const char *const Version = "1.2.17";
 static const char *const usage[] = {
         "sist2 scan [OPTION]... PATH",
         "sist2 index [OPTION]... INDEX",
@@ -29,6 +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);
 }
@@ -48,14 +49,19 @@ 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;
+    ScanCtx.fast = args->fast;
 
     init_dir(ScanCtx.index.path);
 
     ScanCtx.mime_table = mime_get_mime_table();
     ScanCtx.ext_table = mime_get_ext_table();
 
+    cbr_init();
+
     char store_path[PATH_MAX];
     snprintf(store_path, PATH_MAX, "%sthumbs", ScanCtx.index.path);
     mkdir(store_path, S_IWUSR | S_IRUSR | S_IXUSR);
@@ -130,8 +136,12 @@ 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;
     }
 
@@ -153,7 +163,7 @@ 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);
@@ -231,6 +241,8 @@ int main(int argc, const char *argv[]) {
                                                           "shallow: Don't parse archives inside archives. DEFAULT: recurse"),
             OPT_STRING(0, "ocr", &scan_args->tesseract_lang, "Tesseract language (use tesseract --list-langs to see "
                                                              "which are installed on your machine)"),
+            OPT_STRING('e', "exclude", &scan_args->exclude_regex, "Files that match this regex will not be scanned"),
+            OPT_BOOLEAN(0, "fast", &scan_args->fast, "Only index file names & mime type"),
 
             OPT_GROUP("Index options"),
             OPT_STRING(0, "es-url", &common_es_url, "Elasticsearch url with port. DEFAULT=http://localhost:9200"),
@@ -277,9 +289,7 @@ int main(int argc, const char *argv[]) {
         }
         sist2_scan(scan_args);
 
-    }
-
-    else if (strcmp(argv[0], "index") == 0) {
+    } else if (strcmp(argv[0], "index") == 0) {
 
         int err = index_args_validate(index_args, argc, argv);
         if (err != 0) {
@@ -295,8 +305,7 @@ int main(int argc, const char *argv[]) {
         }
         sist2_web(web_args);
 
-    }
-    else {
+    } else {
         fprintf(stderr, "Invalid command: '%s'\n", argv[0]);
         argparse_usage(&argparse);
         return 1;

src/parsing/arc.c

@@ -1,8 +1,6 @@
 #include "arc.h"
 #include "src/ctx.h"
 
-#define ARC_BUF_SIZE 8192
-
 int should_parse_filtered_file(const char *filepath, int ext) {
     char tmp[PATH_MAX * 2];
 

src/parsing/arc.h

@@ -2,6 +2,7 @@
 #define SIST2_ARC_H
 
 #include "src/sist.h"
+#define ARC_BUF_SIZE 8192
 
 int should_parse_filtered_file(const char *filepath, int ext);
 

src/parsing/cbr.c

@@ -0,0 +1,52 @@
+#import "cbr.h"
+#import "src/ctx.h"
+
+unsigned int cbr_mime;
+unsigned int cbz_mime;
+
+void cbr_init() {
+    cbr_mime = mime_get_mime_by_string(ScanCtx.mime_table, "application/x-cbr");
+    cbz_mime = mime_get_mime_by_string(ScanCtx.mime_table, "application/x-cbz");
+}
+
+int is_cbr(unsigned int mime) {
+    return mime == cbr_mime;
+}
+
+void parse_cbr(void *buf, size_t buf_len, document_t *doc) {
+    char *out_buf = malloc(buf_len * 2);
+    size_t out_buf_used = 0;
+
+    struct archive *rar_in = archive_read_new();
+    archive_read_support_filter_none(rar_in);
+    archive_read_support_format_rar(rar_in);
+
+    archive_read_open_memory(rar_in, buf, buf_len);
+
+    struct archive *zip_out = archive_write_new();
+    archive_write_set_format_zip(zip_out);
+    archive_write_open_memory(zip_out, out_buf, buf_len * 2, &out_buf_used);
+
+    struct archive_entry *entry;
+    while (archive_read_next_header(rar_in, &entry) == ARCHIVE_OK) {
+        archive_write_header(zip_out, entry);
+
+        char arc_buf[ARC_BUF_SIZE];
+        int len = archive_read_data(rar_in, arc_buf, ARC_BUF_SIZE);
+        while (len > 0) {
+            archive_write_data(zip_out, arc_buf, len);
+            len = archive_read_data(rar_in, arc_buf, ARC_BUF_SIZE);
+        }
+    }
+
+    archive_write_close(zip_out);
+    archive_write_free(zip_out);
+
+    archive_read_close(rar_in);
+    archive_read_free(rar_in);
+
+    doc->mime = cbz_mime;
+    parse_pdf(out_buf, out_buf_used, doc);
+    doc->mime = cbr_mime;
+    free(out_buf);
+}

src/parsing/cbr.h

@@ -0,0 +1,12 @@
+#ifndef SIST2_CBR_H
+#define SIST2_CBR_H
+
+#include "src/sist.h"
+
+void cbr_init();
+
+int is_cbr(unsigned int mime);
+
+void parse_cbr(void *buf, size_t buf_len, document_t *doc);
+
+#endif

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)));
 

src/parsing/media.c

@@ -193,6 +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)
             }
         }
     }

src/parsing/mime.h

@@ -8,7 +8,7 @@
 #define MIME_EMPTY 1
 
 #define DONT_PARSE 0x80000000
-#define SHOULD_PARSE(mime_id) (mime_id & DONT_PARSE) != DONT_PARSE && mime_id != 0
+#define SHOULD_PARSE(mime_id) (ScanCtx.fast == 0 && (mime_id & DONT_PARSE) != DONT_PARSE && mime_id != 0)
 
 #define PDF_MASK 0x40000000
 #define IS_PDF(mime_id) (mime_id & PDF_MASK) == PDF_MASK

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);

src/parsing/parse.c

@@ -149,6 +149,13 @@ void parse(void *arg) {
         if (doc_buf != buf && doc_buf != NULL) {
             free(doc_buf);
         }
+    } else if (is_cbr(doc.mime)) {
+        void *cbr_buf = read_all(job, (char *) buf, bytes_read);
+        parse_cbr(cbr_buf, doc.size, &doc);
+
+        if (cbr_buf != buf && cbr_buf != NULL) {
+            free(cbr_buf);
+        }
     }
 
     //Parent meta

src/parsing/pdf.c

@@ -1,7 +1,8 @@
 #include "pdf.h"
 #include "src/ctx.h"
 
-#define MIN_OCR_SIZE 128
+#define MIN_OCR_SIZE 350
+#define MIN_OCR_LEN 10
 __thread text_buffer_t thread_buffer;
 
 
@@ -82,7 +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;
     }
@@ -127,37 +129,40 @@ 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)) {
+                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) {
+    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, TESS_DATAPATH, ScanCtx.tesseract_lang);
+            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);
-            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
-            )
+            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);
         }
+        fz_drop_pixmap(ctx, pix);
     }
 }
 

src/parsing/pdf.h

@@ -2,7 +2,6 @@
 #define SIST2_PDF_H
 
 #include "src/sist.h"
-#include <tesseract/capi.h>
 
 
 void parse_pdf(void *buf, size_t buf_len, document_t *doc);

src/sist.h

@@ -3,7 +3,6 @@
 
 #define UUID_STR_LEN 37
 #define UNUSED(x) __attribute__((__unused__))  x
-#define TESS_DATAPATH "/usr/share/tessdata/"
 
 #include <glib-2.0/glib.h>
 #include <unistd.h>
@@ -34,6 +33,9 @@
 #include <archive_entry.h>
 #include <opc/opc.h>
 #include <libxml/xmlstring.h>
+#define BOOL int
+#include <tesseract/capi.h>
+#include <pcre.h>
 
 #include <onion/onion.h>
 #include <onion/handler.h>
@@ -58,6 +60,7 @@
 #include "parsing/font.h"
 #include "parsing/arc.h"
 #include "parsing/doc.h"
+#include "parsing/cbr.h"
 #include "cli.h"
 #include "log.h"
 #include "utf8.h/utf8.h"

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 {
@@ -80,7 +95,6 @@ typedef int (*read_func_t)(struct vfile *, void *buf, size_t size);
 typedef void (*close_func_t)(struct vfile *);
 
 typedef struct vfile {
-
     union {
         int fd;
         struct archive *arc;

src/util.c

@@ -1,4 +1,5 @@
 #include "util.h"
+#include "src/ctx.h"
 
 dyn_buffer_t dyn_buffer_create() {
     dyn_buffer_t buf;
@@ -90,7 +91,11 @@ text_buffer_t text_buffer_create(int max_size) {
 }
 
 void text_buffer_terminate_string(text_buffer_t *buf) {
-    dyn_buffer_write_char(&buf->dyn_buffer, '\0');
+    if (buf->dyn_buffer.cur > 0 && *(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
@@ -171,8 +176,8 @@ int text_buffer_append_string0(text_buffer_t *buf, char *str) {
 
 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;
 
@@ -254,8 +259,10 @@ char *abspath(const char *path) {
     if (abs == NULL) {
         return NULL;
     }
-    abs = realloc(abs, strlen(abs) + 2);
-    strcat(abs, "/");
+    if (strlen(abs) > 1) {
+        abs = realloc(abs, strlen(abs) + 2);
+        strcat(abs, "/");
+    }
 
     wordfree(&w);
     return abs;
@@ -317,4 +324,29 @@ 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;
+}
+
 

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 >= '0' && c <= '9') || (c >= 'A' && c <= 'z') || (c > 127))
+#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

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);
-    onion_response_set_header(res, "Content-Type", mime);
+    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);
 
 

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;
 }
@@ -61,6 +102,7 @@ body {
     border-bottom: none;
     border-left: none;
     border-right: none;
+    padding: .25rem 0.5rem;
 }
 
 .list-group-item:first-child {
@@ -134,6 +176,8 @@ body {
 
 .file-title {
     width: 100%;
+    line-height: 1rem;
+    height: 1.1rem;
     font-size: 10pt;
     white-space: nowrap;
     text-overflow: ellipsis;
@@ -156,7 +200,7 @@ body {
     max-width: 100%;
     max-height: 175px;
     margin: 0 auto 0;
-    padding: 3px 3px 0 3px;
+    padding: 3px 3px 0;
     width: auto;
     height: auto;
 }
@@ -165,7 +209,7 @@ body {
     display: block;
     max-width: 64px;
     max-height: 64px;
-    margin: 0 auto 0;
+    margin: 0 auto;
     width: auto;
     height: auto;
 }
@@ -200,11 +244,18 @@ body {
 }
 
 mark {
-    background: #fff217;
+    background: rgba(251, 191, 41, 0.25);
     border-radius: 0;
     padding: 1px 0;
+    color: inherit;
 }
 
+.content-div mark {
+    background: rgba(251, 191, 41, 0.40);
+    color: white;
+}
+
+
 .content-div {
     font-family: SFMono-Regular, Menlo, Monaco, Consolas, "Liberation Mono", "Courier New", monospace;
     font-size: 13px;
@@ -348,10 +399,6 @@ option {
     margin-top: 1em;
 }
 
-.list-group-item {
-    padding: .25rem 0.5rem;
-}
-
 .wrapper-sm {
     min-width: 64px;
 }
@@ -380,3 +427,25 @@ 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;
+    }
+}
+
+.input-group > .custom-select:not(:first-child), .input-group > .form-control:not(:first-child) {
+    border-right: none;
+}

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;
@@ -163,6 +184,7 @@ mark {
     background: #fff217;
     border-radius: 0;
     padding: 1px 0;
+    color: inherit;
 }
 
 .content-div {
@@ -266,3 +288,25 @@ 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;
+    }
+}
+
+.input-group > .custom-select:not(:first-child), .input-group > .form-control:not(:first-child) {
+    border-right: none;
+}

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);
@@ -327,6 +377,7 @@ function makeThumbnail(mimeCategory, hit, imgWrapper, small) {
         || hit["_source"]["mime"] === "application/pdf"
         || hit["_source"]["mime"] === "application/epub+zip"
         || hit["_source"]["mime"] === "application/x-cbz"
+        || hit["_source"]["mime"] === "application/x-cbr"
         || hit["_source"].hasOwnProperty("font_name")
     ) {
         thumbnail = document.createElement("img");
@@ -337,6 +388,15 @@ 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 +412,14 @@ 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 +440,8 @@ function createDocLine(hit) {
         isSubDocument = true;
     }
 
+    const infoButton = makeInfoButton(hit);
+
     const title = makeTitle(hit);
 
     let link = document.createElement("a");
@@ -380,8 +450,13 @@ 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 +517,7 @@ 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;
 }
@@ -472,26 +546,29 @@ function makeStatsCard(searchResult) {
     resultMode.appendChild(gridMode);
     resultMode.appendChild(listMode);
 
-    if (mode === "grid") {
+    if (CONF.options.display === "grid") {
         gridMode.classList.add("active")
     } else {
         listMode.classList.add("active")
     }
 
     gridMode.addEventListener("click", () => {
-        mode = "grid";
-        localStorage.setItem("mode", mode);
+        console.log("what");
+        console.log(CONF.options);
+        CONF.options.display = "grid";
+        console.log(CONF.options);
+        CONF.save();
+        console.log(CONF.options);
         searchDebounced();
     });
     listMode.addEventListener("click", () => {
-        mode = "list";
-        localStorage.setItem("mode", mode);
+        CONF.options.display = "list";
+        CONF.save();
         searchDebounced();
     });
 
     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);
@@ -511,7 +588,7 @@ function makeStatsCard(searchResult) {
 function makeResultContainer() {
     let resultContainer = document.createElement("div");
 
-    if (mode === "grid") {
+    if (CONF.options.display === "grid") {
         resultContainer.setAttribute("class", "card-columns");
     } else {
         resultContainer.setAttribute("class", "list-group");

web/js/search.js

@@ -6,19 +6,46 @@ 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;
 let selectedIndices = [];
 
-let mode;
-if (localStorage.getItem("mode") === null) {
-    mode = "grid";
-} else {
-    mode = localStorage.getItem("mode")
+const CONF = new Settings();
+
+const _defaults = {
+    display: "grid",
+    fuzzy: true,
+    highlight: true
+};
+
+function Settings() {
+    this.options = {};
+
+    this._onUpdate = function() {
+        $("#fuzzyToggle").prop("checked", this.options.fuzzy);
+    }
+
+    this.load = function() {
+        const raw = window.localStorage.getItem("options");
+        if (raw === null) {
+            this.options = _defaults;
+        } else {
+            this.options = JSON.parse(raw);
+        }
+
+        this._onUpdate();
+    }
+
+    this.save = function() {
+        window.localStorage.setItem("options", JSON.stringify(this.options));
+        this._onUpdate();
+    }
 }
 
+
 function showEsError() {
     $.toast({
         heading: "Elasticsearch connection error",
@@ -53,6 +80,7 @@ window.onload = () => {
         }
         window.location.reload();
     })
+    CONF.load();
 };
 
 function toggleFuzzy() {
@@ -81,6 +109,13 @@ $.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 +179,8 @@ $.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) {
@@ -242,7 +277,7 @@ new autoComplete({
 function insertHits(resultContainer, hits) {
     for (let i = 0; i < hits.length; i++) {
 
-        if (mode === "grid") {
+        if (CONF.options.display === "grid") {
             resultContainer.appendChild(createDocCard(hits[i]));
         } else {
             resultContainer.appendChild(createDocLine(hits[i]));
@@ -252,41 +287,18 @@ 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) {
-            coolingDown = true;
-            doScroll();
+            if (!reachedEnd) {
+                coolingDown = true;
+                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 = [];
 
@@ -307,21 +319,25 @@ 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");
-    while (searchResults.firstChild) {
-        searchResults.removeChild(searchResults.firstChild);
+    //Clear old search results
+    let preload;
+    if (!after) {
+        while (searchResults.firstChild) {
+            searchResults.removeChild(searchResults.firstChild);
+        }
+        preload = makePreloader();
+        searchResults.appendChild(preload);
     }
 
-    const preload = makePreloader();
-    searchResults.appendChild(preload);
-
     let query = searchBar.value;
     let empty = query === "";
     let condition = empty ? "should" : "must";
@@ -355,55 +371,79 @@ 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: {
+        aggs:
+            {
+                total_size: {"sum": {"field": "size"}},
+                total_count: {"value_count": {"field": "size"}}
+            },
+        size: SIZE,
+    };
+
+    if (after) {
+        q.search_after = [after["_score"], after["_id"]];
+    }
+
+    if (CONF.options.highlight) {
+        q.highlight = {
             pre_tags: ["<mark>"],
-            post_tags: ["</mark>"],
-            fields: {
+                post_tags: ["</mark>"],
+                fields: {
                 content: {},
                 // "content.nGram": {},
                 name: {},
                 "name.nGram": {},
                 font_name: {},
             }
-        },
-        aggs: {
-            total_size: {"sum": {"field": "size"}}
-        },
-        size: SIZE,
-    }).then(searchResult => {
-        scroll_id = searchResult["_scroll_id"];
+        };
+    }
 
-        preload.remove();
-        //Search stats
-        searchResults.appendChild(makeStatsCard(searchResult));
+    $.jsonPost("es", q).then(searchResult => {
+        let hits = searchResult["hits"]["hits"];
+        if (hits) {
+            lastDoc = hits[hits.length - 1];
+        }
+
+        if (!after) {
+            preload.remove();
+            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 +504,14 @@ 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();
+    }
+};
+
 //Suggest
 function getPathChoices() {
     return new Promise(getPaths => {
@@ -482,3 +530,30 @@ function getPathChoices() {
     })
 }
 
+function updateSettings() {
+    CONF.options.display = $("#settingDisplay").val();
+    CONF.options.fuzzy = $("#settingFuzzy").prop("checked");
+    CONF.options.highlight = $("#settingHighlight").prop("checked");
+    CONF.save();
+
+    searchDebounced();
+
+    $.toast({
+        heading: "Settings updated",
+        text: "Settings saved to browser storage",
+        stack: 3,
+        bgColor: "#00a4bc",
+        textColor: "#fff",
+        position: 'bottom-right',
+        hideAfter: 3000,
+        loaderBg: "#08c7e8",
+    });
+}
+
+function loadSettings() {
+    CONF.load();
+
+    $("#settingDisplay").val(CONF.options.display);
+    $("#settingFuzzy").prop("checked", CONF.options.fuzzy);
+    $("#settingHighlight").prop("checked", CONF.options.highlight);
+}

web/search.html

@@ -11,9 +11,10 @@
 
 <nav class="navbar navbar-expand-lg">
     <a class="navbar-brand" href="/">sist2</a>
-    <span class="badge badge-pill version">v1.2.0</span>
+    <span class="badge badge-pill version">v1.2.17</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>
+    <button style="margin-left: auto" class="btn" type="button" data-toggle="modal" data-target="#settings" onclick="loadSettings()">Settings</button>
+    <a id="theme" class="btn" title="Toggle theme" href="/">Theme</a>
 </nav>
 
 <div class="container">
@@ -25,13 +26,20 @@
             <div class="input-group">
                 <div class="input-group-prepend">
                     <div class="input-group-text">
-                        <span title="Toggle fuzzy searching"  onclick="document.getElementById('fuzzyToggle').click()">Fuzzy&nbsp</span>
+                        <span title="Toggle fuzzy searching" onclick="document.getElementById('fuzzyToggle').click()">Fuzzy&nbsp</span>
                         <input title="Toggle fuzzy searching" type="checkbox" id="fuzzyToggle"
                                onclick="toggleFuzzy()" checked>
                     </div>
                 </div>
                 <input id="searchBar" type="search" class="form-control" placeholder="Search">
-
+                <div class="input-group-append">
+                    <button class="btn btn-outline-secondary small-btn" type="button" data-toggle="modal"
+                            data-target="#help">?
+                    </button>
+                    <button class="btn btn-outline-secondary large-btn" type="button" data-toggle="modal"
+                            data-target="#help">Help
+                    </button>
+                </div>
             </div>
 
             <input title="File size" id="sizeSlider" name="size">
@@ -45,10 +53,12 @@
                 <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 +75,116 @@
         </div>
     </div>
 
+    <div class="modal" id="modal" tabindex="-1" role="dialog" aria-labelledby="modal-title" aria-hidden="true">
+        <div class="modal-dialog modal-lg modal-dialog-centered" role="document">
+            <div class="modal-content">
+                <div class="modal-header">
+                    <h5 class="modal-title" id="modal-title"></h5>
+                    <button type="button" class="close" data-dismiss="modal" aria-label="Close">
+                        <span aria-hidden="true">&times;</span>
+                    </button>
+                </div>
+                <div class="modal-body" id="modal-body"></div>
+            </div>
+        </div>
+    </div>
+
+    <div class="modal" id="help" tabindex="-1" role="dialog" aria-labelledby="modal-title" aria-hidden="true">
+        <div class="modal-dialog modal-lg modal-dialog-centered" role="document">
+            <div class="modal-content">
+                <div class="modal-header">
+                    <h5 class="modal-title">Search help</h5>
+                    <button type="button" class="close" data-dismiss="modal" aria-label="Close">
+                        <span aria-hidden="true">&times;</span>
+                    </button>
+                </div>
+                <div class="modal-body">
+
+                    <table class="table">
+                        <tbody>
+                        <tr>
+                            <td><code>+</code></td>
+                            <td>signifies AND operation</td>
+                        </tr>
+                        <tr>
+                            <td><code>|</code></td>
+                            <td>signifies OR operation</td>
+                        </tr>
+                        <tr>
+                            <td><code>-</code></td>
+                            <td>negates a single token</td>
+                        </tr>
+                        <tr>
+                            <td><code>""</code></td>
+                            <td>wraps a number of tokens to signify a phrase for searching</td>
+                        </tr>
+                        <tr>
+                            <td><code>*</code></td>
+                            <td>at the end of a term signifies a prefix query</td>
+                        </tr>
+                        <tr>
+                            <td><code>(</code> and <code>)</code></td>
+                            <td>signify precedence</td>
+                        </tr>
+                        <tr>
+                            <td><code>~N</code></td>
+                            <td>after a word signifies edit distance (fuzziness)</td>
+                        </tr>
+                        <tr>
+                            <td><code>~N</code></td>
+                            <td>after a phrase signifies slop amount</td>
+                        </tr>
+                        </tbody>
+                    </table>
+
+                    <p>For example: <code>"fried eggs" +(eggplant | potato) -frittata</code> will match the phrase
+                        <i>fried eggs</i> and either <i>eggplant</i> or <i>potato</i>, but will ignore results
+                        containing <i>frittata</i>.</p>
+
+                    <p>When neither <code>+</code> or <code>|</code> is specified, the default operator is <code>+</code> (and).</p>
+                    <p>When the <b>Fuzzy</b> option is checked, partial matches are also returned.</p>
+                    <br>
+                    <p>For more information, see <a target="_blank"
+                                                    href="//www.elastic.co/guide/en/elasticsearch/reference/current/query-dsl-simple-query-string-query.html">Elasticsearch
+                        documentation</a></p>
+                </div>
+            </div>
+        </div>
+    </div>
+
+    <div class="modal" id="settings" tabindex="-1" role="dialog" aria-labelledby="modal-title" aria-hidden="true">
+        <div class="modal-dialog modal-dialog-centered" role="document">
+            <div class="modal-content">
+                <div class="modal-header">
+                    <h5 class="modal-title">Settings</h5>
+                    <button type="button" class="close" data-dismiss="modal" aria-label="Close">
+                        <span aria-hidden="true">&times;</span>
+                    </button>
+                </div>
+                <div class="modal-body">
+                    <div class="custom-control custom-checkbox">
+                        <input type="checkbox" class="custom-control-input" id="settingHighlight">
+                        <label class="custom-control-label" for="settingHighlight">Enable highlighting</label>
+                    </div>
+
+                    <div class="custom-control custom-checkbox">
+                        <input type="checkbox" class="custom-control-input" id="settingFuzzy">
+                        <label class="custom-control-label" for="settingFuzzy">Set fuzzy search by default</label>
+                    </div>
+
+                    <label for="settingDisplay">Display</label>
+                    <select id="settingDisplay" class="form-control form-control-sm">
+                        <option value="grid">Grid</option>
+                        <option value="list">List</option>
+                    </select>
+
+                    <br>
+                    <button style="float: right" class="btn btn-primary" onclick="updateSettings()">Update settings</button>
+                </div>
+            </div>
+        </div>
+    </div>
+
     <div id="searchResults"></div>
 </div>