Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2026-03-20 08:27:52 -04:00
Code Issues Packages Projects Releases Wiki Activity

DOCX: Generate ToC from headings

Browse Source
This commit is contained in:
Kovid Goyal 2013-05-24 17:04:31 +05:30
parent ae1029f452
commit ea7292f6ed
2 changed files with 48 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
src/calibre/ebooks/docx/styles.py
View File

@ -394,6 +394,9 @@ class Styles(object):
dl.notes dt a { text-decoration: none }
dl.notes dd { page-break-after: always }
dl.notes dd:last-of-type { page-break-after: avoid }
''') % (self.body_font_family, self.body_font_size)
if ef:
prefix = ef + '\n' + prefix

46
src/calibre/ebooks/docx/to_html.py
View File

@ -23,6 +23,8 @@ from calibre.ebooks.docx.fonts import Fonts
from calibre.ebooks.docx.images import Images
from calibre.ebooks.docx.footnotes import Footnotes
from calibre.ebooks.metadata.opf2 import OPFCreator
from calibre.ebooks.metadata.toc import TOC
from calibre.ebooks.oeb.polish.toc import elem_to_toc_text
from calibre.utils.localization import canonicalize_lang, lang_as_iso639_1
class Text:
@ -227,7 +229,48 @@ class Convert(object):
self.styles.resolve_numbering(numbering)
def create_toc(self):
' Create a TOC from headings in the document '
root = self.body
headings = ('h1', 'h2', 'h3')
tocroot = TOC()
xpaths = [XPath('//%s' % x) for x in headings]
level_prev = {i+1:None for i in xrange(len(xpaths))}
level_prev[0] = tocroot
level_item_map = {i+1:frozenset(xp(root)) for i, xp in enumerate(xpaths)}
item_level_map = {e:i for i, elems in level_item_map.iteritems() for e in elems}
self.idcount = 0
def ensure_id(elem):
ans = elem.get('id', None)
if not ans:
self.idcount += 1
ans = 'toc_id_%d' % self.idcount
elem.set('id', ans)
return ans
for item in root.iterdescendants(*headings):
lvl = plvl = item_level_map.get(item, None)
if lvl is None:
continue
parent = None
while parent is None:
plvl -= 1
parent = level_prev[plvl]
lvl = plvl + 1
elem_id = ensure_id(item)
text = elem_to_toc_text(item)
toc = parent.add_item('index.html', elem_id, text)
level_prev[lvl] = toc
for i in xrange(lvl+1, len(xpaths)+1):
level_prev[i] = None
if len(tuple(tocroot.flat())) > 1:
return tocroot
def write(self):
toc = self.create_toc()
raw = html.tostring(self.html, encoding='utf-8', doctype='<!DOCTYPE html>')
with open(os.path.join(self.dest_dir, 'index.html'), 'wb') as f:
f.write(raw)
@ -237,6 +280,7 @@ class Convert(object):
f.write(css.encode('utf-8'))
opf = OPFCreator(self.dest_dir, self.mi)
opf.toc = toc
opf.create_manifest_from_files_in([self.dest_dir])
opf.create_spine(['index.html'])
with open(os.path.join(self.dest_dir, 'metadata.opf'), 'wb') as of, open(os.path.join(self.dest_dir, 'toc.ncx'), 'wb') as ncx:
@ -276,7 +320,7 @@ class Convert(object):
m = re.match(r'heading\s+(\d+)$', style.style_name or '', re.IGNORECASE)
if m is not None:
n = min(1, max(6, int(m.group(1))))
n = min(6, max(1, int(m.group(1))))
dest.tag = 'h%d' % n
if style.direction == 'rtl':