mirror of
https://github.com/hectorqin/reader.git
synced 2026-11-11 13:31:09 +08:00
hectorqinandClaude Opus 5.5
60d4b41bf1
Source chapters are scraped fragments (`<div> text<br><br><br><br>`) served as text/html, so the TXT cleanup never ran: every paragraph was followed by three empty lines and the indent setting did nothing. Like legado-E, a text-only chapter of a `chapters` book is read back as characters when it is loaded (htmlToText) and typeset by the TXT path. Done at read time, so chapters the server cached earlier are cleaned too; chapters with images or tables keep their markup and EPUBs never take this path. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>