Changelog v1.2.7

• Performance Overhaul: Replaced heavy Regex processing with native XMLSerializer and optimized the internal event loop. Generating EPUBs with thousands of posts is now significantly faster.
• Reduced CPU Usage: The UI no longer freezes during large downloads. The generator now yields to the browser's main thread and throttles progress bar updates to prevent visual lag.
• Image Optimization: Smart detection for image formats. JPEGs are now preserved as-is instead of being forcibly converted to PNGs, reducing the final EPUB file size without losing quality. WebP images are still converted to PNG for maximum e-reader compatibility.
• Rate Limiting Fix: Patched a race condition in the API rate limiter to ensure safer, more reliable communication with the server during bulk fetches.
• Stability: hardened the import logic for single-post downloads to prevent crashes if dependencies aren't immediately available.

Version 1.2.5

• Changed EPUB stylesheet to be more generic and less opinionated. Thanks to user feedback

VERSION 1.2.4

• Improved EPUB generation to better handle edge cases that caused EPUB errors (again), including preventing duplicate images and ensuring valid file structures for better compatibility with e-readers.
• Fixed ToC generation (now inserts a dedicated Table of Contents page right after the cover, with clickable links to chapters).
• Added two more automatic filename generation options in the UI dropdown:
  • For extracting the second number instead of the first (to filter out book or volume numbers, e.g., "Creator_47" from "Book 3 Chapter 47").
  • For extracting both the first and second number into a "Book-Chapter" format (e.g., "Creator_B2C1-B2C4").
• Added a basic word-to-number function that can parse spelled-out numbers like "Four Hundred and Thirty-Nine" into "439" (as a fallback only for the Creator_Numbers pattern; doesn't affect other modes).
• Refactored filenames in the codebase and renamed some internal files for consistency—no user impact.
• Minor reliability boosts: Expanded image server permissions for faster/more stable downloads from Kemono's subdomains; updated extension icon tooltip for clarity.

version 1.2.3

version notes
- re-added bulk chapter downloading
- added tags and custom search filter option
- clicking the extension icon to open the downloader (tabs permission added to manifest for this)
- added a toggle state to turn off the cover image

– HTML is cleaned and normalised to well‑formed XHTML (entity replacement, void‑tag handling) so the output passes EPUB validators and works with strict readers.
– manifest, spine, NCX and EPUB 3 navigation document rebuilt to meet current spec requirements; automatic TOC page added when needed.
– images are now always converted to PNG before being added to the EPUB.

These changes fix validation errors that caused some EPUBs to be rejected by stricter readers.

New icons for version 1.2.1