mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-07-09 03:04:10 -04:00
Fix #7936 (FB2: Sectionize TOC entries)
This commit is contained in:
Kovid Goyal
commit
c9fe094afa
@ -27,13 +27,10 @@ class FB2MLizer(object):
|
||||
'''
|
||||
Todo: * Include more FB2 specific tags in the conversion.
|
||||
* Handle a tags.
|
||||
* Figure out some way to turn oeb_book.toc items into <section><title>
|
||||
<p> to allow for readers to generate toc from the document.
|
||||
'''
|
||||
|
||||
def __init__(self, log):
|
||||
self.log = log
|
||||
self.image_hrefs = {}
|
||||
self.reset_state()
|
||||
|
||||
def reset_state(self):
|
||||
@ -43,17 +40,25 @@ class FB2MLizer(object):
|
||||
# in different directories. FB2 images are all in a flat layout so we rename all images
|
||||
# into a sequential numbering system to ensure there are no collisions between image names.
|
||||
self.image_hrefs = {}
|
||||
# Mapping of toc items and their
|
||||
self.toc = {}
|
||||
# Used to see whether a new <section> needs to be opened
|
||||
self.section_level = 0
|
||||
|
||||
def extract_content(self, oeb_book, opts):
|
||||
self.log.info('Converting XHTML to FB2 markup...')
|
||||
self.oeb_book = oeb_book
|
||||
self.opts = opts
|
||||
self.reset_state()
|
||||
|
||||
# Used for adding <section>s and <title>s to allow readers
|
||||
# to generate toc from the document.
|
||||
if self.opts.sectionize == 'toc':
|
||||
self.create_flat_toc(self.oeb_book.toc, 1)
|
||||
|
||||
return self.fb2mlize_spine()
|
||||
|
||||
def fb2mlize_spine(self):
|
||||
self.reset_state()
|
||||
|
||||
output = [self.fb2_header()]
|
||||
output.append(self.get_text())
|
||||
output.append(self.fb2mlize_images())
|
||||
@ -66,13 +71,19 @@ class FB2MLizer(object):
|
||||
return u'<?xml version="1.0" encoding="UTF-8"?>' + output
|
||||
|
||||
def clean_text(self, text):
|
||||
text = re.sub(r'(?miu)<section>\s*</section>', '', text)
|
||||
text = re.sub(r'(?miu)\s+</section>', '</section>', text)
|
||||
text = re.sub(r'(?miu)</section><section>', '</section>\n\n<section>', text)
|
||||
|
||||
text = re.sub(r'(?miu)<p>\s*</p>', '', text)
|
||||
text = re.sub(r'(?miu)\s+</p>', '</p>', text)
|
||||
text = re.sub(r'(?miu)</p><p>', '</p>\n\n<p>', text)
|
||||
text = re.sub(r'(?miu)\s*</p>', '</p>', text)
|
||||
text = re.sub(r'(?miu)</p>\s*<p>', '</p>\n\n<p>', text)
|
||||
|
||||
text = re.sub(r'(?miu)<title>\s*</title>', '', text)
|
||||
text = re.sub(r'(?miu)\s+</title>', '</title>', text)
|
||||
|
||||
text = re.sub(r'(?miu)<section>\s*</section>', '', text)
|
||||
text = re.sub(r'(?miu)\s*</section>', '\n</section>', text)
|
||||
text = re.sub(r'(?miu)</section>\s*', '</section>\n\n', text)
|
||||
text = re.sub(r'(?miu)\s*<section>', '\n<section>', text)
|
||||
text = re.sub(r'(?miu)<section>\s*', '<section>\n', text)
|
||||
text = re.sub(r'(?miu)</section><section>', '</section>\n\n<section>', text)
|
||||
|
||||
if self.opts.insert_blank_line:
|
||||
text = re.sub(r'(?miu)</p>', '</p><empty-line />', text)
|
||||
@ -144,12 +155,34 @@ class FB2MLizer(object):
|
||||
|
||||
def get_text(self):
|
||||
text = ['<body>']
|
||||
|
||||
# Create main section if there are no others to create
|
||||
if self.opts.sectionize == 'nothing':
|
||||
text.append('<section>')
|
||||
self.section_level += 1
|
||||
|
||||
for item in self.oeb_book.spine:
|
||||
self.log.debug('Converting %s to FictionBook2 XML' % item.href)
|
||||
stylizer = Stylizer(item.data, item.href, self.oeb_book, self.opts, self.opts.output_profile)
|
||||
|
||||
# Start a <section> if we must sectionize each file or if the TOC references this page
|
||||
page_section_open = False
|
||||
if self.opts.sectionize == 'files' or self.toc.get(item.href) == 'page':
|
||||
text.append('<section>')
|
||||
page_section_open = True
|
||||
self.section_level += 1
|
||||
|
||||
text += self.dump_text(item.data.find(XHTML('body')), stylizer, item)
|
||||
|
||||
if page_section_open:
|
||||
text.append('</section>')
|
||||
self.section_level -= 1
|
||||
|
||||
# Close any open sections
|
||||
while self.section_level > 0:
|
||||
text.append('</section>')
|
||||
self.section_level -= 1
|
||||
|
||||
return ''.join(text) + '</body>'
|
||||
|
||||
def fb2mlize_images(self):
|
||||
@ -184,6 +217,17 @@ class FB2MLizer(object):
|
||||
'%s.' % (item.href, e))
|
||||
return ''.join(images)
|
||||
|
||||
def create_flat_toc(self, nodes, level):
|
||||
for item in nodes:
|
||||
href, mid, id = item.href.partition('#')
|
||||
if not id:
|
||||
self.toc[href] = 'page'
|
||||
else:
|
||||
if not self.toc.get(href, None):
|
||||
self.toc[href] = {}
|
||||
self.toc[href][id] = level
|
||||
self.create_flat_toc(item.nodes, level + 1)
|
||||
|
||||
def ensure_p(self):
|
||||
if self.in_p:
|
||||
return [], []
|
||||
@ -254,10 +298,38 @@ class FB2MLizer(object):
|
||||
# First tag in tree
|
||||
tag = barename(elem_tree.tag)
|
||||
|
||||
# Process the XHTML tag if it needs to be converted to an FB2 tag.
|
||||
if tag == 'h1' and self.opts.h1_to_title or tag == 'h2' and self.opts.h2_to_title or tag == 'h3' and self.opts.h3_to_title:
|
||||
# Convert TOC entries to <title>s and add <section>s
|
||||
if self.opts.sectionize == 'toc':
|
||||
# A section cannot be a child of any other element than another section,
|
||||
# so leave the tag alone if there are parents
|
||||
if not tag_stack:
|
||||
# There are two reasons to start a new section here: the TOC pointed to
|
||||
# this page (then we use the first non-<body> on the page as a <title>), or
|
||||
# the TOC pointed to a specific element
|
||||
newlevel = 0
|
||||
toc_entry = self.toc.get(page.href, None)
|
||||
if toc_entry == 'page':
|
||||
if tag != 'body' and hasattr(elem_tree, 'text') and elem_tree.text:
|
||||
newlevel = 1
|
||||
self.toc[page.href] = None
|
||||
elif toc_entry and elem_tree.attrib.get('id', None):
|
||||
newlevel = toc_entry.get(elem_tree.attrib.get('id', None), None)
|
||||
|
||||
# Start a new section if necessary
|
||||
if newlevel:
|
||||
if not (newlevel > self.section_level):
|
||||
fb2_out.append('</section>')
|
||||
self.section_level -= 1
|
||||
fb2_out.append('<section>')
|
||||
self.section_level += 1
|
||||
fb2_out.append('<title>')
|
||||
tags.append('title')
|
||||
if self.section_level == 0:
|
||||
# If none of the prior processing made a section, make one now to be FB2 spec compliant
|
||||
fb2_out.append('<section>')
|
||||
self.section_level += 1
|
||||
|
||||
# Process the XHTML tag if it needs to be converted to an FB2 tag.
|
||||
if tag == 'img':
|
||||
if elem_tree.attrib.get('src', None):
|
||||
# Only write the image tag if it is in the manifest.
|
||||
|
||||
@ -16,15 +16,15 @@ class FB2Output(OutputFormatPlugin):
|
||||
file_type = 'fb2'
|
||||
|
||||
options = set([
|
||||
OptionRecommendation(name='h1_to_title',
|
||||
recommended_value=False, level=OptionRecommendation.LOW,
|
||||
help=_('Wrap all h1 tags with fb2 title elements.')),
|
||||
OptionRecommendation(name='h2_to_title',
|
||||
recommended_value=False, level=OptionRecommendation.LOW,
|
||||
help=_('Wrap all h2 tags with fb2 title elements.')),
|
||||
OptionRecommendation(name='h3_to_title',
|
||||
recommended_value=False, level=OptionRecommendation.LOW,
|
||||
help=_('Wrap all h3 tags with fb2 title elements.')),
|
||||
OptionRecommendation(name='sectionize',
|
||||
recommended_value='files', level=OptionRecommendation.LOW,
|
||||
choices=['toc', 'files', 'nothing'],
|
||||
help=_('Specify the sectionization of elements. '
|
||||
'A value of "nothing" turns the book into a single section. '
|
||||
'A value of "files" turns each file into a separate section; use this if your device is having trouble. '
|
||||
'A value of "Table of Contents" turns the entries in the Table of Contents into titles and creates sections; '
|
||||
'if it fails, adjust the "Structure Detection" and/or "Table of Contents" settings '
|
||||
'(turn on "Force use of auto-generated Table of Contents).')),
|
||||
])
|
||||
|
||||
def convert(self, oeb_book, output_path, input_plugin, opts, log):
|
||||
|
||||
@ -17,6 +17,8 @@ class PluginWidget(Widget, Ui_Form):
|
||||
ICON = I('mimetypes/fb2.png')
|
||||
|
||||
def __init__(self, parent, get_option, get_help, db=None, book_id=None):
|
||||
Widget.__init__(self, parent, ['h1_to_title', 'h2_to_title', 'h3_to_title'])
|
||||
Widget.__init__(self, parent, ['sectionize'])
|
||||
self.db, self.book_id = db, book_id
|
||||
for x in ('toc', 'files', 'nothing'):
|
||||
self.opt_sectionize.addItem(x)
|
||||
self.initialize_options(get_option, get_help, db, book_id)
|
||||
|
||||
@ -14,7 +14,7 @@
|
||||
<string>Form</string>
|
||||
</property>
|
||||
<layout class="QGridLayout" name="gridLayout">
|
||||
<item row="3" column="0">
|
||||
<item row="1" column="0">
|
||||
<spacer name="verticalSpacer">
|
||||
<property name="orientation">
|
||||
<enum>Qt::Vertical</enum>
|
||||
@ -28,23 +28,19 @@
|
||||
</spacer>
|
||||
</item>
|
||||
<item row="0" column="0">
|
||||
<widget class="QCheckBox" name="opt_h1_to_title">
|
||||
<widget class="QLabel" name="label">
|
||||
<property name="text">
|
||||
<string>Wrap h1 tags with <title> elements</string>
|
||||
<string>Sectionize:</string>
|
||||
</property>
|
||||
<property name="buddy">
|
||||
<cstring>opt_sectionize</cstring>
|
||||
</property>
|
||||
</widget>
|
||||
</item>
|
||||
<item row="1" column="0">
|
||||
<widget class="QCheckBox" name="opt_h2_to_title">
|
||||
<property name="text">
|
||||
<string>Wrap h2 tags with <title> elements</string>
|
||||
</property>
|
||||
</widget>
|
||||
</item>
|
||||
<item row="2" column="0">
|
||||
<widget class="QCheckBox" name="opt_h3_to_title">
|
||||
<property name="text">
|
||||
<string>Wrap h3 tags with <title> elements</string>
|
||||
<item row="0" column="1">
|
||||
<widget class="QComboBox" name="opt_sectionize">
|
||||
<property name="minimumContentsLength">
|
||||
<number>20</number>
|
||||
</property>
|
||||
</widget>
|
||||
</item>
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user