Dymas/abogen
1
0
Fork 0
mirror of https://github.com/denizsafak/abogen.git synced 2026-07-18 13:40:27 +02:00
Code Issues Packages Projects Releases Wiki Activity

feat: Enhance heading processing by stripping numeric prefixes from titles

Browse Source
This commit is contained in:
JB
2025-11-27 19:41:40 -08:00
parent 550fdbf537
commit 775e05f94c
1 changed files with 14 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
abogen/web/conversion_runner.py
+14
View File
@@ -1599,6 +1599,13 @@ def run_conversion_job(job: Job) -> None:
mutated_entry = False mutated_entry = False
if pending_heading_strip and heading_text: if pending_heading_strip and heading_text:
chunk_text, removed_heading = _strip_duplicate_heading_line(chunk_text, heading_text) chunk_text, removed_heading = _strip_duplicate_heading_line(chunk_text, heading_text)
if not removed_heading and raw_title:
match = _HEADING_NUMBER_PREFIX_RE.match(raw_title)
if match:
number = match.group("number")
if number:
chunk_text, removed_heading = _strip_duplicate_heading_line(chunk_text, number)
if removed_heading: if removed_heading:
pending_heading_strip = False pending_heading_strip = False
chunk_entry = dict(chunk_entry) chunk_entry = dict(chunk_entry)
@@ -1677,6 +1684,13 @@ def run_conversion_job(job: Job) -> None:
chapter_text = str(chapter.text or "") chapter_text = str(chapter.text or "")
if pending_heading_strip and heading_text: if pending_heading_strip and heading_text:
chapter_text, removed_heading = _strip_duplicate_heading_line(chapter_text, heading_text) chapter_text, removed_heading = _strip_duplicate_heading_line(chapter_text, heading_text)
if not removed_heading and raw_title:
match = _HEADING_NUMBER_PREFIX_RE.match(raw_title)
if match:
number = match.group("number")
if number:
chapter_text, removed_heading = _strip_duplicate_heading_line(chapter_text, number)
if removed_heading: if removed_heading:
pending_heading_strip = False pending_heading_strip = False
if opening_caps_pending and chapter_text: if opening_caps_pending and chapter_text: