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

TXT Input: Retain whitespace at the beginning of lines.

Browse Source
This commit is contained in:
John Schember 2011-01-27 18:30:36 -05:00
parent 8f8c5eaa97
commit 6a5bf738f2
1 changed files with 6 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
src/calibre/ebooks/txt/processor.py
View File

@ -20,9 +20,13 @@ HTML_TEMPLATE = u'<html><head><meta http-equiv="Content-Type" content="text/html
def clean_txt(txt): def clean_txt(txt):
if isbytestring(txt): if isbytestring(txt):
txt = txt.decode('utf-8', 'replace') txt = txt.decode('utf-8', 'replace')
# Strip whitespace from the beginning and end of the line. Also replace # Strip whitespace from the end of the line. Also replace
# all line breaks with \n. # all line breaks with \n.
txt = '\n'.join([line.strip() for line in txt.splitlines()]) txt = '\n'.join([line.rstrip() for line in txt.splitlines()])
# Replace whitespace at the beginning of the list with &nbsp;
txt = re.sub('(?m)(?P<space>[ ]+)', lambda mo: '&nbsp;' * mo.groups('space').count(' '), txt)
txt = re.sub('(?m)(?P<space>[\t]+)', lambda mo: '&nbsp;' * 4 * mo.groups('space').count('\t'), txt)
# Condense redundant spaces # Condense redundant spaces
txt = re.sub('[ ]{2,}', ' ', txt) txt = re.sub('[ ]{2,}', ' ', txt)