Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-08 10:44:09 -04:00
Code Issues Packages Projects Releases Wiki Activity

DOCX Input: Fix conversion breaking for files that use heading style

Browse Source 
paragraphs to insert line rules
This commit is contained in:
Kovid Goyal 2013-07-27 08:09:26 +05:30
parent 2a0f6bbeae
commit 6afb55fdfb
1 changed files with 11 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
src/calibre/ebooks/docx/cleanup.py
View File

@ -8,7 +8,7 @@ __copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
import os import os
from calibre.ebooks.docx.names import ancestor from calibre.ebooks.docx.names import XPath
def mergeable(previous, current): def mergeable(previous, current):
if previous.tail or current.tail: if previous.tail or current.tail:
@ -100,14 +100,17 @@ def before_count(root, tag, limit=10):
def cleanup_markup(log, root, styles, dest_dir, detect_cover): def cleanup_markup(log, root, styles, dest_dir, detect_cover):
# Move <hr>s outside paragraphs, if possible. # Move <hr>s outside paragraphs, if possible.
pancestor = XPath('|'.join('ancestor::%s[1]' % x for x in ('p', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6')))
for hr in root.xpath('//span/hr'): for hr in root.xpath('//span/hr'):
p = ancestor(hr, 'p') p = pancestor(hr)
descendants = tuple(p.iterdescendants()) if p:
if descendants[-1] is hr: p = p[0]
parent = p.getparent() descendants = tuple(p.iterdescendants())
idx = parent.index(p) if descendants[-1] is hr:
parent.insert(idx+1, hr) parent = p.getparent()
hr.tail = '\n\t' idx = parent.index(p)
parent.insert(idx+1, hr)
hr.tail = '\n\t'
# Merge consecutive spans that have the same styling # Merge consecutive spans that have the same styling
current_run = [] current_run = []
@ -176,5 +179,3 @@ def cleanup_markup(log, root, styles, dest_dir, detect_cover):
img.getparent().remove(img) img.getparent().remove(img)
return path return path