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

text processing tweaks

Browse Source
This commit is contained in:
ldolse 2011-01-09 02:12:09 +08:00
parent f88045c162
commit 9bbff15c27
2 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

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

@ -190,7 +190,7 @@ class PreProcessor(object):
line_ending = "\s*</(span|p|div)>\s*(</(p|span|div)>)?" line_ending = "\s*</(span|p|div)>\s*(</(p|span|div)>)?"
blanklines = "\s*(?P<up2threeblanks><(p|span|div)[^>]*>\s*(<(p|span|div)[^>]*>\s*</(span|p|div)>\s*)</(span|p|div)>\s*){0,3}\s*" blanklines = "\s*(?P<up2threeblanks><(p|span|div)[^>]*>\s*(<(p|span|div)[^>]*>\s*</(span|p|div)>\s*)</(span|p|div)>\s*){0,3}\s*"
line_opening = "<(span|div|p)[^>]*>\s*(<(span|div|p)[^>]*>)?\s*" line_opening = "<(span|div|p)[^>]*>\s*(<(span|div|p)[^>]*>)?\s*"
txt_line_wrap = u"(\u0020|\u0009)*\n" txt_line_wrap = u"((\u0020|\u0009)*\n){1,4}"
unwrap_regex = lookahead+line_ending+blanklines+line_opening unwrap_regex = lookahead+line_ending+blanklines+line_opening
if format == 'txt': if format == 'txt':
@ -357,6 +357,6 @@ class PreProcessor(object):
html = blankreg.sub('\n'+r'\g<openline>'+u'\u00a0'+r'\g<closeline>', html) html = blankreg.sub('\n'+r'\g<openline>'+u'\u00a0'+r'\g<closeline>', html)
# Center separator lines # Center separator lines
html = re.sub(u'<p>\s*(?P<break>([*#•]+\s*)+)\s*</p>', '<p style="text-align:center">' + '\g<break>' + '</p>', html) html = re.sub(u'<(?P<outer>p|div)[^>]*>\s*(<(?P<inner1>font|span|[ibu])[^>]*>)?\s*(<(?P<inner2>font|span|[ibu])[^>]*>)?\s*(<(?P<inner3>font|span|[ibu])[^>]*>)?\s*(?P<break>([*#•]+\s*)+)\s*(</(?P=inner3)>)?\s*(</(?P=inner2)>)?\s*(</(?P=inner1)>)?\s*</(?P=outer)>', '<p style="text-align:center">' + '\g<break>' + '</p>', html)
return html return html

2
src/calibre/ebooks/txt/input.py
View File

@ -90,7 +90,7 @@ class TXTInput(InputFormatPlugin):
# We don't check for block because the processor assumes block. # We don't check for block because the processor assumes block.
# single and print at transformed to block for processing. # single and print at transformed to block for processing.
if options.paragraph_type == 'single' or 'unformatted': if options.paragraph_type in ('single', 'unformatted'):
txt = separate_paragraphs_single_line(txt) txt = separate_paragraphs_single_line(txt)
elif options.paragraph_type == 'print': elif options.paragraph_type == 'print':
txt = separate_paragraphs_print_formatted(txt) txt = separate_paragraphs_print_formatted(txt)