From c1dc4ad09d15ba60cf009b934541c7cc297fbcfb Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Fri, 29 Oct 2010 08:17:14 -0600 Subject: [PATCH] Fix #7328 (Why are Unicode strings with encoding declaration not supported.) --- src/calibre/ebooks/chardet/__init__.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/calibre/ebooks/chardet/__init__.py b/src/calibre/ebooks/chardet/__init__.py index f094df3cd5..8c101f9de5 100644 --- a/src/calibre/ebooks/chardet/__init__.py +++ b/src/calibre/ebooks/chardet/__init__.py @@ -30,9 +30,9 @@ def detect(aBuf): # Added by Kovid ENCODING_PATS = [ - re.compile(r'<\?[^<>]+encoding=[\'"](.*?)[\'"][^<>]*>', + re.compile(r'<\?[^<>]+encoding\s*=\s*[\'"](.*?)[\'"][^<>]*>', re.IGNORECASE), - re.compile(r''']+?content=['"][^'"]*?charset=([-a-z0-9]+)[^'"]*?['"][^<>]*>''', + re.compile(r''']+?content\s*=\s*['"][^'"]*?charset=([-a-z0-9]+)[^'"]*?['"][^<>]*>''', re.IGNORECASE) ] ENTITY_PATTERN = re.compile(r'&(\S+?);')