v0.1.0

What's Changed

• Fix YouTube transcript scraping intermittent failures by @hadv in #45
• ✅ WORKING: Fix YouTube transcript extraction with YouTube internal API by @hadv in #47
• feat: Add YouTube video chunk management system by @hadv in #48

Full Changelog: v0.1.0-beta...v0.1.0

v0.1.0-beta

What's Changed

• Fix YouTube oversized chunks issue by improving chunking mechanisms by @prpeh in #34
• Fix: Add retry mechanism for empty YouTube caption data by @prpeh in #35
• Implement self-contained job queue for YouTube processing by @prpeh in #36
• Improve YouTube UI to show domains in progress & deletion sections by @prpeh in #37

Full Changelog: v0.1.0-alpha...v0.1.0-beta

v0.1.0-alpha

What's Changed

• Improve error handling for Qdrant and pdf-parse issues by @prpeh in #1
• Add multi-step document upload with domain selection feature by @prpeh in #2
• refactor: remove postfix naming patterns from filenames by @prpeh in #3
• chore: fix naming convention and remove unused imports by @prpeh in #4
• Fix: Use UUID for Qdrant document IDs to prevent special character errors by @prpeh in #5
• Add Gemini embedding support with text chunking by @prpeh in #6
• Rename chunkPdfDocument to chunkDocument for better semantics by @prpeh in #9
• Fix: Use random UUIDs for Qdrant point IDs to meet format requirements by @prpeh in #13
• Implement enhanced document chunking with page-based processing by @hadv in #8
• Add file cleanup after document processing using try-finally pattern by @prpeh in #15
• Improve document chunking to preserve headings, paragraphs and sentences by @prpeh in #17
• Enhance Vietnamese text handling and improve document processing by @prpeh in #18
• Add generative AI content enhancement capabilities by @prpeh in #19
• Fix: Improve Gemini API reliability and prevent accidental translation of Vietnamese content by @prpeh in #22
• feat: process DOCX files by sections similar to PDF pages to prevent content truncation by @prpeh in #23
• Fix YouTube transcript access with web scraping fallback by @prpeh in #25
• Improve YouTube transcript processing with duplicate validation, enhanced chunks, AI titles/summaries by @prpeh in #26
• Add YouTube video deletion functionality with UI component by @prpeh in #28
• Add simplified document management by @prpeh in #29
• Fix document chunks infinite loading issue and improve UI feedback by @prpeh in #30
• Fix YouTube upload timeout handling for web scraping operations by @prpeh in #31
• Implement real-time YouTube progress tracking via WebSockets by @prpeh in #32
• Fix YouTube transcript chunk detection and display count before deletion by @prpeh in #33

New Contributors

• @prpeh made their first contribution in #1
• @hadv made their first contribution in #8

Full Changelog: https://github.com/hadv/yitam-admin/commits/v0.1.0-alpha