Changelog

Copy Markdown View Source

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

Unreleased

0.10.0 - 2026-10-17

  • Breaking: extract_tables/3 and extract_tables_from_binary/3 return each table as a map with its rows and the left, right, top and bottom edges of its bounding box, instead of only the rows
  • Add extract_page_info/3 and extract_page_info_from_binary/3 to get the number of pages and the size and rotation of each one
  • Add font: true option to extract_words/3 and extract_words_from_binary/3 to get the font name and size of each word
  • Add render_pages/3 and render_pages_from_binary/3 to render pages or areas to PNG images

0.9.0 - 2026-10-17

  • Add extract_words/3 and extract_words_from_binary/3 to get every word with its bounding box on the page
  • Add layout: true option to extract_text/3 and extract_text_from_binary/3 to keep the position of the text on the page
  • Add extract_tables/3 and extract_tables_from_binary/3 with vertical_strategy and horizontal_strategy options
  • Add password option to every function to open encrypted PDFs

0.8.0 - 2026-10-16

  • Upgrade Python to v3.14 and pin uv to v0.12.24 so recent CPython builds are available
  • Upgrade pdfplumber to v0.11.10 (bundles pdfminer.six 20260107)
  • Upgrade pythonx to v0.4.10
  • Upgrade dev tooling: credo v1.7.19, dialyxir v1.4.8, ex_doc v0.40.4, styler v1.12.2

0.7.0 - 2026-09-27

  • Fix out-of-range page numbers shifting the extracted text onto the wrong pages
  • Return {:error, exception} for invalid arguments instead of crashing the server

0.6.0 - 2026-04-03

  • Add configurable timeout option to all extraction functions (default: 5000ms)
  • Add name option to target a specific server process

0.5.1 - 2026-04-03

  • Upgrade pdfplumber to v0.11.9
  • Upgrade pythonx to v0.4.9

0.5.0 - 2025-08-23

  • Simplify arguments to avoid repetition
  • Upgrade Python to v3.12 and pdfplumber to v0.11.7
  • Lower Elixir requirement to v1.15

0.4.1 - 2025-07-30

0.4.0 - 2025-07-21

Changed

  • Made PdfExtractor a single process to avoid issues with the Python GIL

0.3.0 - 2025-07-20

Added

  • Multiple Areas Support: Extract text from multiple bounding box areas on the same page
  • Metadata Extraction: New extract_metadata/1 and extract_metadata_from_binary/1 functions
  • Binary PDF Processing: Extract text and metadata directly from PDF binary data
  • Enhanced Documentation: Comprehensive doctests and improved API documentation
  • Improved Test Coverage: Added extensive test suite for new functionality

Changed

  • Enhanced area-based extraction to support lists of areas per page
  • Improved error handling and edge case management
  • Updated type specifications for better developer experience

Fixed

  • Better handling of invalid page numbers and area coordinates
  • Improved Python environment initialization

0.2.1 - 2025-06-27

Fixed

  • Added automatic Python dependencies download and installation
  • Improved application startup process

0.2.0 - 2025-06-22

Added

  • Project badges and improved README documentation
  • Enhanced configuration and documentation setup

Changed

  • Improved function naming and API consistency
  • Better documentation structure

0.1.0 - 2025-06-21

Added

  • Initial release of PdfExtractor
  • Support for extracting text from PDF files using Python's pdfplumber
  • Single page text extraction
  • Multi-page text extraction
  • Basic area-based text extraction with bounding boxes
  • Initial test suite
  • Basic documentation and examples

Dependencies

  • pythonx ~> 0.4.0 for Python integration
  • Requires Python with pdfplumber package installed