lxbfYeaa/BookConverter
Archived
Template
1
0
Fork 0
forked from LiveCarta/BookConverter
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add ability to have several conditions on tags

Browse Source
This commit is contained in:
Kiryl
2022-10-19 18:44:34 +03:00
parent 1363940fa2
commit 658f206f5d
2 changed files with 114 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
src/epub_converter/html_epub_processor.py
View File

@@ -107,12 +107,10 @@ class HtmlEpubProcessor:
len(text_preparing(tag)) != 0 and
re.findall(r"^h[1-5]$", tag.name or chapter_tag.name))
if title_in_text:
self.html_preprocessor._add_span_to_save_ids_for_links(
title_in_text[-1], chapter_tag)
self.html_preprocessor.add_span_to_save_ids_for_links(title_in_text[-1], chapter_tag)
title_in_text[-1].extract()
elif text_in_title:
[self.html_preprocessor._add_span_to_save_ids_for_links(
tag, chapter_tag) for tag in text_in_title]
[self.html_preprocessor.add_span_to_save_ids_for_links(tag, chapter_tag) for tag in text_in_title]
[tag.extract() for tag in text_in_title]
@staticmethod
@@ -135,12 +133,12 @@ class HtmlEpubProcessor:
and (tag.attrs.get("class") not in ["link-anchor", "footnote-element"]):
del tag.attrs["class"]
def prepare_content(self, title_str: str, chapter_tag: BeautifulSoup, remove_title_from_chapter: bool) -> Tag:
def prepare_content(self, title: str, chapter_tag: BeautifulSoup, remove_title_from_chapter: bool) -> Tag:
"""
Function finalise processing/cleaning content
Parameters
----------
title_str: str
title: str
chapter_tag: BeautifulSoup, soup object
@@ -170,7 +168,7 @@ class HtmlEpubProcessor:
self._wrap_strings_with_p(chapter_tag)
# 3.
if remove_title_from_chapter:
self._remove_headings_content(chapter_tag, title_str)
self._remove_headings_content(chapter_tag, title)
# 4.
_process_presets(
html_preprocessor=self.html_preprocessor, html_soup=chapter_tag)