wpivotto/filebox-cli

Filebox

Features

• Easy integration, just write the files in the right folder
• Index documents on demand
  • Asynchronous indexing
  • Automatically handles adding, changing and deleting documents
• Embedded web server with search API
• Aggregates result into a single pdf file
• Serves images with resizing support

Basic Options

• index.folder: Apache Lucene index destination
• docs.folder: Folder where to get the documents to index
• images.folder: Folder where to get the images to serve
• highlight.enabled: Highlight search matches in the output file
• http.port: The port that the embedded web server will listen on when running to handle clients

Tokenization/Indexing Options

• stopwords.txt: list of stop words (used by Lucene StopFilter)
• delimiters.txt: list of custom word delimiters (used in tokenization)

Running Locally

mvn clean exec:java -Dexec.mainClass="br.com.wpivotto.Main"

Build Windows Executable

mvn clean package

Basic Usage

1. Put any text file inside the configured folder. Ex: C:\Filebox\Docs
2. The system will extract the text from the document and index it automatically
3. Open a browser window and access the url http:\\localhost:PORT\search?q=TERM
4. All search results will be grouped into a single PDF document and made available for download

API Usage

Search Query

Returns a single PDF file with all document matches (See Lucene Query Syntax)

• URL

/search?q=[query]

• Method:

  GET

• URL Params

  Required:

  q=[string]

• Success Response:

  • Code: 200
    Content: Complete data stream of the file contents.

• Sample Call:

GET http://localhost:8099/search?q=test

Use Cases

• search in electrical projects
• search in instruction manuals
• search in calibration reports
• search in piping and instrumentation diagrams (P&ID)
• show pictures in embedded browsers (ActiveX) on SCADA Systems

TODO

• Handle text orientation in tokenizer
• Support for pattern replace (CAD files tend to export words without spacing)