Demo: sist2.simon987.net

Community URL: Discord

sist2

sist2 (Simple incremental search tool)

Warning: sist2 is in early development

Features

• Fast, low memory usage, multi-threaded
• Manage & schedule scan jobs with simple web interface (Docker only)
• Mobile-friendly Web interface
• Extracts text and metadata from common file types *
• Generates thumbnails *
• Incremental scanning
• Manual tagging from the UI and automatic tagging based on file attributes via user scripts
• Recursive scan inside archive files **
• OCR support with tesseract ***
• Stats page & disk utilisation visualization
• Named-entity recognition (client-side) ****

* See format support
** See Archive files
*** See OCR
**** See Named-Entity Recognition

Getting Started

Using Docker Compose (Windows/Linux/Mac)

version: "3"

services:
  elasticsearch:
    image: elasticsearch:7.17.9
    restart: unless-stopped
    volumes:
      # This directory must have 1000:1000 permissions (or update PUID & PGID below)
      - /data/sist2-es-data/:/usr/share/elasticsearch/data
    environment:
      - "discovery.type=single-node"
      - "ES_JAVA_OPTS=-Xms2g -Xmx2g"
      - "PUID=1000"
      - "PGID=1000"
  sist2-admin:
    image: simon987/sist2:3.4.2-x64-linux
    restart: unless-stopped
    volumes:
      - /data/sist2-admin-data/:/sist2-admin/
      - /:/host
    ports:
      - 4090:4090
      # NOTE: Don't expose this port publicly!
      - 8080:8080
    working_dir: /root/sist2-admin/
    entrypoint: python3
    command:
      - /root/sist2-admin/sist2_admin/app.py

Navigate to http://localhost:8080/ to configure sist2-admin.

Using the executable file (Linux/WSL only)

1. Choose search backend (See comparison):

   • Elasticsearch: have an Elasticsearch (version >= 6.8.X, ideally >=7.14.0) instance running
     1. Download from official website
     2. (or) Run using docker:

        docker run -d -p 9200:9200 -e "discovery.type=single-node" elasticsearch:7.17.9

   • SQLite: No installation required

2. Download the latest sist2 release. Select the file corresponding to your CPU architecture and mark the binary as executable with chmod +x.

3. See usage guide for command line usage.

Example usage:

1. Scan a directory: sist2 scan ~/Documents --output ./documents.sist2
2. Prepare search index:
   • Elasticsearch: sist2 index --es-url http://localhost:9200 ./documents.sist2
   • SQLite: sist2 sqlite-index --search-index ./search.sist2 ./documents.sist2
3. Start web interface:
   • Elasticsearch: sist2 web ./documents.sist2
   • SQLite: sist2 web --search-index ./search.sist2 ./documents.sist2

Format support

File type	Library	Content	Thumbnail	Metadata
pdf,xps,fb2,epub	MuPDF	text+ocr	yes	author, title
cbz,cbr	libscan	-	yes	-
audio/*	ffmpeg	-	yes	ID3 tags
video/*	ffmpeg	-	yes	title, comment, artist
image/*	ffmpeg	ocr	yes	Common EXIF tags, GPS tags
raw, rw2, dng, cr2, crw, dcr, k25, kdc, mrw, pef, xf3, arw, sr2, srf, erf	LibRaw	no	yes	Common EXIF tags, GPS tags
ttf,ttc,cff,woff,fnt,otf	Freetype2	-	yes, bmp	Name & style
text/plain	libscan	yes	no	-
html, xml	libscan	yes	no	-
tar, zip, rar, 7z, ar ...	Libarchive	yes*	-	no
docx, xlsx, pptx	libscan	yes	if embedded	creator, modified_by, title
doc (MS Word 97-2003)	antiword	yes	no	author, title
mobi, azw, azw3	libmobi	yes	yes	author, title
wpd (WordPerfect)	libwpd	yes	no	planned
json, jsonl, ndjson	libscan	yes	-	-

* See Archive files

Archive files

sist2 will scan files stored into archive files (zip, tar, 7z...) as if they were directly in the file system. Recursive (archives inside archives) scan is also supported.

Limitations:

• Support for parsing media files with formats that require seek (e.g. .gif, .mp4 w/ fragmented metadata etc.) is limitted (see --mem-buffer option)
• Archive files are scanned sequentially, by a single thread. On systems where sist2 is not I/O bound, scans might be faster when larger archives are split into smaller parts.

OCR

You can enable OCR support for ebook (pdf,xps,fb2,epub) or image file types with the --ocr-lang <lang> option in combination with --ocr-images and/or --ocr-ebooks. Download the language data files with your package manager (apt install tesseract-ocr-eng) or directly from Github.

The simon987/sist2 image comes with common languages (hin, jpn, eng, fra, rus, spa, chi_sim, deu, pol) pre-installed.

You can use the + separator to specify multiple languages. The language name must be identical to the *.traineddata file installed on your system (use chi_sim rather than chi-sim).

Examples:

sist2 scan --ocr-ebooks --ocr-lang jpn ~/Books/Manga/
sist2 scan --ocr-images --ocr-lang eng ~/Images/Screenshots/
sist2 scan --ocr-ebooks --ocr-images --ocr-lang eng+chi_sim ~/Chinese-Bilingual/

Search backends

sist2 v3.0.7+ supports SQLite search backend. The SQLite search backend has fewer features and generally comparable query performance for medium-size indices, but it uses much less memory and is easier to set up.

	SQLite	Elasticsearch
Requires separate search engine installation		✓
Memory footprint	~20MB	>500MB
Query syntax	fts5	query_string
Fuzzy search		✓
Media Types tree real-time updating		✓
Manual tagging	✓	✓
User scripts	✓	✓
Media Type breakdown for search results		✓
Embeddings search	✓ O(n)	✓ O(logn)

NER

sist2 v3.0.4+ supports named-entity recognition (NER). Simply add a supported repository URL to Configuration > Machine learning options > Model repositories to enable it.

The text processing is done in your browser, no data is sent to any third-party services. See simon987/sist2-ner-models for more details.

List of available repositories:

URL	Maintainer	Purpose
simon987/sist2-ner-models	simon987	General

Screenshot

Build from source

You can compile sist2 by yourself if you don't want to use the pre-compiled binaries

Using docker

git clone --recursive https://github.com/simon987/sist2/
cd sist2
docker build . -t my-sist2-image
# Copy sist2 executable from docker image
docker run --rm --entrypoint cat my-sist2-image /root/sist2 > sist2-x64-linux

Using a linux computer

1. Install compile-time dependencies

   apt install gcc g++ python3 yasm ragel automake autotools-dev wget libtool libssl-dev curl zip unzip tar xorg-dev libglu1-mesa-dev libxcursor-dev libxml2-dev libxinerama-dev gettext nasm git nodejs

2. Install vcpkg using my fork: https://github.com/simon987/vcpkg

3. Install vcpkg dependencies

   vcpkg install openblas curl[core,openssl] sqlite3[core,fts5,json1] cpp-jwt pcre cjson brotli libarchive[core,bzip2,libxml2,lz4,lzma,lzo] pthread tesseract libxml2 libmupdf[ocr] gtest mongoose libmagic libraw gumbo ffmpeg[core,avcodec,avformat,swscale,swresample,webp,opus,mp3lame,vpx,zlib]

4. Build

   git clone --recursive https://github.com/simon987/sist2/
   (cd sist2-vue; npm install; npm run build)
   (cd sist2-admin/frontend; npm install; npm run build)
   cmake -DSIST_DEBUG=off -DCMAKE_TOOLCHAIN_FILE=<VCPKG_ROOT>/scripts/buildsystems/vcpkg.cmake .
   make