Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Fix block ordering when marking sentences

Browse Source
This commit is contained in:
Kovid Goyal 2024-10-16 10:07:30 +05:30
parent 567404a1ef
commit 4e621117e3
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
2 changed files with 14 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/calibre/ebooks/oeb/polish/tests/structure.py
View File

@ -239,6 +239,12 @@ class Structure(BaseTest):
'<p>Hello, <span data-calibre-tts="moose">world!': '<p>Hello, <span data-calibre-tts="moose">world!':
'<body><p><span id="1">Hello, </span><span data-calibre-tts="moose"><span id="2">world!</span></span></p>', '<body><p><span id="1">Hello, </span><span data-calibre-tts="moose"><span id="2">world!</span></span></p>',
'<p>One<p>Two':
'<body><p><span id="1">One</span></p><p><span id="2">Two</span></p>',
'<div><p>something':
'<body><div><p><span id="1">something</span></p></div>',
}.items()): }.items()):
root = parse(text, namespace_elements=True) root = parse(text, namespace_elements=True)
orig = normalize_markup(root) orig = normalize_markup(root)

22
src/calibre/ebooks/oeb/polish/tts.py
View File

@ -96,7 +96,6 @@ def mark_sentences_in_html(root, lang: str = '', voice: str = '') -> list[Senten
self.texts.append(Chunk(None, elem.text, self.pos)) self.texts.append(Chunk(None, elem.text, self.pos))
self.pos += len(elem.text) self.pos += len(elem.text)
self.children = tuple(elem.iterchildren()) self.children = tuple(elem.iterchildren())
self.child_pos = 0
def add_simple_child(self, elem): def add_simple_child(self, elem):
if text := elem.text: if text := elem.text:
@ -325,27 +324,22 @@ def mark_sentences_in_html(root, lang: str = '', voice: str = '') -> list[Senten
stack_of_parents = [Parent(elem, 'body', root_lang, root_voice) for elem in root.iterchildren('*') if barename(elem.tag).lower() == 'body'] stack_of_parents = [Parent(elem, 'body', root_lang, root_voice) for elem in root.iterchildren('*') if barename(elem.tag).lower() == 'body']
while stack_of_parents: while stack_of_parents:
p = stack_of_parents.pop() p = stack_of_parents.pop()
if len(p.elem) == 1 and not has_text(p.elem): # wrapper simple_allowed = True
c = p.elem[0] children_to_process = []
if isinstance(c.tag, str): for child in p.children:
stack_of_parents.append(Parent(c, barename(c.tag).lower(), p.lang, p.voice))
continue
for i in range(p.child_pos, len(p.children)):
child = p.children[i]
child_voice = child.get('data-calibre-tts', '') child_voice = child.get('data-calibre-tts', '')
child_lang = lang_for_elem(child, p.lang) child_lang = lang_for_elem(child, p.lang)
child_tag_name = barename(child.tag).lower() if isinstance(child.tag, str) else '' child_tag_name = barename(child.tag).lower() if isinstance(child.tag, str) else ''
if child_lang == p.lang and child_voice == p.voice and child_tag_name in continued_tag_names and len(child) == 0: if simple_allowed and child_lang == p.lang and child_voice == p.voice and child_tag_name in continued_tag_names and len(child) == 0:
p.add_simple_child(child) p.add_simple_child(child)
elif child_tag_name not in ignored_tag_names: elif child_tag_name not in ignored_tag_names:
stack_of_parents.append(Parent(child, child_tag_name, p.lang, p.voice, child_lang=child_lang)) simple_allowed = False
children_to_process.append(Parent(child, child_tag_name, p.lang, p.voice, child_lang=child_lang))
p.commit() p.commit()
p.child_pos = i + 1 if simple_allowed and (text := child.tail):
stack_of_parents.append(p)
continue
if text := child.tail:
p.add_tail(child, text) p.add_tail(child, text)
p.commit() p.commit()
stack_of_parents.extend(reversed(children_to_process))
for src_elem, clones in clones_map.items(): for src_elem, clones in clones_map.items():
for clone in clones + [src_elem]: for clone in clones + [src_elem]:
if not clone.text and not clone.tail and not clone.get('id') and not clone.get('name'): if not clone.text and not clone.tail and not clone.get('id') and not clone.get('name'):