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

Change line length amount to work with duplicates removed lengths. Enhance the unwrapping regex to account for more cases.

Browse Source
This commit is contained in:
John Schember 2009-06-22 21:12:17 -04:00
parent e426c9e60d
commit 3a87f0d065
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/calibre/ebooks/conversion/preprocess.py
View File

@ -183,12 +183,12 @@ class HTMLPreProcessor(object):
elif self.is_book_designer(html): elif self.is_book_designer(html):
rules = self.BOOK_DESIGNER rules = self.BOOK_DESIGNER
elif self.is_pdftohtml(html): elif self.is_pdftohtml(html):
length = line_length(html, .3) length = line_length(html, .5)
line_length_rules = [] line_length_rules = []
if length: if length:
line_length_rules = [ line_length_rules = [
# Un wrap using punctuation # Un wrap using punctuation
(re.compile(r'(?<=.{%i}[a-z,;:-IA])\s*(?P<ital></(i|b|u)>)?\s*(<p.*?>)\s*(?=(<(i|b|u)>)?[\w\d])' % length, re.UNICODE), wrap_lines), (re.compile(r'(?<=.{%i}[a-z\.,;:)-IA])\s*(?P<ital></(i|b|u)>)?\s*(<p.*?>)\s*(?=(<(i|b|u)>)?\s*[\w\d(])' % length, re.UNICODE), wrap_lines),
] ]
rules = self.PDFTOHTML + line_length_rules rules = self.PDFTOHTML + line_length_rules