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

epub converter: add internal links processing

Browse Source 
- fix cleaning headings
- fix when chapter mark is not on the 1 level
This commit is contained in:
shirshasa
2021-05-25 12:58:14 +03:00
parent 0ac20999b5
commit 3eac136e07
2 changed files with 146 additions and 59 deletions
Download Patch File Download Diff File Expand all files Collapse all files

112
src/html_epub_preprocessor.py
View File

@@ -115,8 +115,12 @@ def _process_lists(body_tag):
def clean_headings_content(content: Tag, title: str):
for child in content.contents:
if child.text and re.sub(r'([\n\t\xa0])', '', child.text):
text = re.sub(r'([\n\t\xa0])', ' ', child.text)
if isinstance(child, NavigableString):
text = child
else:
text = child.text
if text and re.sub(r'([\n\t\xa0])', '', text):
text = re.sub(r'([\n\t\xa0])', ' ', text)
text = re.sub(r' +', ' ', text).strip()
if title == text:
child.extract()
@@ -196,10 +200,6 @@ def preprocess_footnotes(source_html_tag: Tag, href2soup_html: dict = None, note
return unicode_string.strip()
def remove_internal_links_with_text(t):
for tag_a in t.find_all('a', {'href': re.compile('(^.+\.(html|xhtml)#.+)|(^#.+)')}):
tag_a.decompose()
for i, noteref_tag in enumerate(noterefs_tags):
href = noteref_tag.attrs['href']
file, element_id = parse_a_tag_href(href)
@@ -208,7 +208,7 @@ def preprocess_footnotes(source_html_tag: Tag, href2soup_html: dict = None, note
else:
target_html_tag = href2soup_html.get(file)
if not target_html_tag:
print(f'Error. for\n{noteref_tag}\ninvalid path: {file} found.')
print(f'Error while footnotes processing. For {noteref_tag} invalid path: {file}.')
continue
possible_footnote = 'note|footnote|endnote|rearenote'
@@ -218,7 +218,6 @@ def preprocess_footnotes(source_html_tag: Tag, href2soup_html: dict = None, note
expected_footnote_tags = verify_footnote_tag(expected_footnote_tags)
footnote_tag = expected_footnote_tags[0]
replace_with_livecarta_anchor_tag(noteref_tag, i)
remove_internal_links_with_text(footnote_tag)
content = get_footnote_tags2str(footnote_tag)
footnote_tag.decompose()
@@ -227,45 +226,19 @@ def preprocess_footnotes(source_html_tag: Tag, href2soup_html: dict = None, note
return footnotes
def add_fonts():
pass
def unwrap_structural_tags(body_tag):
def _add_span_to_save_ids_for_links(tag_to_be_removed):
if tag_to_be_removed.attrs.get('id'):
new_tag = body_tag.new_tag("span")
new_tag.attrs['id'] = tag_to_be_removed.attrs['id']
tag_to_be_removed.insert_before(new_tag)
structural_tags_names = [
'div', 'section', 'article', 'main', 'body', 'html', 'aside', 'canvas', 'data',
'figure', 'footer', 'iframe', 'span', 'p'
]
for div in body_tag.find_all("div"):
if div.contents:
is_not_struct_tag = [child.name not in structural_tags_names for child in div.contents]
if all(is_not_struct_tag):
div.name = 'p'
continue
div.unwrap()
for s in body_tag.find_all("section"):
s.unwrap()
for s in body_tag.find_all("article"):
s.unwrap()
for s in body_tag.find_all("aside"):
s.name = 'blockquote'
for s in body_tag.find_all("main"):
s.unwrap()
for s in body_tag.find_all("body"):
s.unwrap()
for s in body_tag.find_all("html"):
s.unwrap()
for s in body_tag.find_all("header"):
s.name = 'span'
# should be before other tags processing, not to remove converter empty tags with id
# not all cases, if span has <p>s and NavigableString, it won't unwrap
for s in body_tag.find_all("span"):
if s.contents:
@@ -274,6 +247,55 @@ def unwrap_structural_tags(body_tag):
continue
s.unwrap()
for div in body_tag.find_all("div"):
if div.contents:
is_not_struct_tag = [child.name not in structural_tags_names for child in div.contents]
if all(is_not_struct_tag):
div.name = 'p'
continue
_add_span_to_save_ids_for_links(div)
div.unwrap()
for s in body_tag.find_all("section"):
_add_span_to_save_ids_for_links(s)
s.unwrap()
for s in body_tag.find_all("article"):
_add_span_to_save_ids_for_links(s)
s.unwrap()
for s in body_tag.find_all("aside"):
s.name = 'blockquote'
for s in body_tag.find_all("main"):
_add_span_to_save_ids_for_links(s)
s.unwrap()
for s in body_tag.find_all("body"):
_add_span_to_save_ids_for_links(s)
s.unwrap()
for s in body_tag.find_all("html"):
_add_span_to_save_ids_for_links(s)
s.unwrap()
for s in body_tag.find_all("header"):
s.name = 'span'
# check marks for chapter starting are on the same 1 level
marks = body_tag.find_all(attrs={'class': 'converter-chapter-mark'})
parents_marks_are_body = [x.parent == body_tag for x in marks]
# fix marks to be on 1 level
if not all(parents_marks_are_body):
for x in marks:
while x.parent != body_tag:
x.parent.unwrap() # warning! could reflect on formatting/internal links in some cases
parents_marks_are_body = [x.parent == body_tag for x in marks]
assert all(parents_marks_are_body), 'Anchor for chapter is deeper than 2 level.'
_preprocessing_headings(body_tag)
for node in body_tag:
@@ -290,12 +312,12 @@ def unwrap_structural_tags(body_tag):
def get_tags_between_ids(first_id, href, html_soup):
h_marked = html_soup.find(attrs={'id': first_id, 'class': 'internal-mark'})
h_marked = html_soup.find(attrs={'id': first_id, 'class': 'converter-chapter-mark'})
if h_marked:
p = h_marked.next_sibling
tags = []
while p:
if p.name == 'h1' and p.attrs.get('class') == 'internal-mark':
if p.name == 'tmp' and p.attrs.get('class') == 'converter-chapter-mark':
break
tags.append(p)
p = p.next_sibling
@@ -330,7 +352,7 @@ def prepare_title_and_content(title, chapter_tag: BeautifulSoup, remove_title_fr
preprocess_table(chapter_tag)
# 2. class removal
for tag in chapter_tag.find_all(recursive=True):
if hasattr(tag, 'attrs') and tag.attrs.get('class'):
if hasattr(tag, 'attrs') and tag.attrs.get('class') and (tag.attrs.get('class') not in ['link-anchor']):
del tag.attrs['class']
# content_str = re.sub(r'([\n\t\xa0])', ' ', str(content_tag))