This PDFBox wrapper that can be used for extracting text and text co-ordinates from a printed PDF doc (no OCR)
- JRE 8 or above
- Nodejs 8 or above - only if you want to build from source
- PDFBox v2 or above - only if you want to build from source
Note: Neither the binaries, nor the source files will work without the Java Runtime Environment (JRE)
After cloning the repository, copy the main_java.exe and the BoomPdf.jar files to the desired folder. Make sure you have nodejs installed and then run the following code in the command prompt:
I:\Path> node main_java.exe "Absolute Path of the PDF" FromPage ToPageWhere I:\Path> is the folder path (don't type this in, it will show up by default once you navigate to the folder using cd), FromPage is the first page you want to convert and ToPage is the end page
The zero position/origin (0,0) is on the top left corner of the page:
Out of the box, BoomPdf.java returns each glyph/alphabet and special character with coordinates.
If you're building the code from source, please download the PDFBox jar and use the BoomPdf.java file to customize your solution.
Further customization in the output can be done by altering the main_java.js file. This is the Node.js code that parses the text and returns words with their coordinates (the left-most character's position is taken for reference)
The main code of this project is licensed under the Apache 2.0 License, found at http://www.apache.org/licenses/LICENSE-2.0.html Any code released under a different licenses will be stated in the header.