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

Diff tool: When detecting encoding of text files, look for an encoding declaration at the top in the format used by vim/emacs etc.

Browse Source
This commit is contained in:
Kovid Goyal 2014-02-01 12:12:03 +05:30
parent fb72c922d6
commit 7cdb090057
1 changed files with 7 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
src/calibre/gui2/tweak_book/diff/main.py
View File

@ -6,7 +6,7 @@ from __future__ import (unicode_literals, division, absolute_import,
__license__ = 'GPL v3' __license__ = 'GPL v3'
__copyright__ = '2014, Kovid Goyal <kovid at kovidgoyal.net>' __copyright__ = '2014, Kovid Goyal <kovid at kovidgoyal.net>'
import sys, os import sys, os, re
from functools import partial from functools import partial
from PyQt4.Qt import ( from PyQt4.Qt import (
@ -93,6 +93,7 @@ def changed_files(list_of_names1, list_of_names2, get_data1, get_data2):
added_names.add(name) added_names.add(name)
return cache, changed_names, renamed_names, removed_names, added_names return cache, changed_names, renamed_names, removed_names, added_names
def get_decoded_raw(name): def get_decoded_raw(name):
from calibre.ebooks.chardet import xml_to_unicode, force_encoding from calibre.ebooks.chardet import xml_to_unicode, force_encoding
with open(name, 'rb') as f: with open(name, 'rb') as f:
@ -107,7 +108,11 @@ def get_decoded_raw(name):
if syntax in {'html', 'xml'}: if syntax in {'html', 'xml'}:
raw = xml_to_unicode(raw, verbose=True)[0] raw = xml_to_unicode(raw, verbose=True)[0]
else: else:
enc = force_encoding(raw, verbose=True) m = re.search(r"coding[:=]\s*([-\w.]+)", raw[:1024], flags=re.I)
if m is not None:
enc = m.group(1)
else:
enc = force_encoding(raw, verbose=True)
try: try:
raw = raw.decode(enc) raw = raw.decode(enc)
except ValueError: except ValueError: