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

HTML Input: Limit emory consumption when converting HTML files that link to lage binary files. Fixes #884821 (python terminated)

Browse Source
This commit is contained in:
Kovid Goyal 2011-11-02 09:14:21 +05:30
parent 1e07787d02
commit 0600e911f1
1 changed files with 4 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/calibre/ebooks/html/input.py
View File

@ -109,14 +109,16 @@ class HTMLFile(object):
try: try:
with open(self.path, 'rb') as f: with open(self.path, 'rb') as f:
src = f.read() src = f.read(4096)
self.is_binary = level > 0 and not bool(self.HTML_PAT.search(src))
if not self.is_binary:
src += f.read()
except IOError as err: except IOError as err:
msg = 'Could not read from file: %s with error: %s'%(self.path, as_unicode(err)) msg = 'Could not read from file: %s with error: %s'%(self.path, as_unicode(err))
if level == 0: if level == 0:
raise IOError(msg) raise IOError(msg)
raise IgnoreFile(msg, err.errno) raise IgnoreFile(msg, err.errno)
self.is_binary = level > 0 and not bool(self.HTML_PAT.search(src[:4096]))
if not self.is_binary: if not self.is_binary:
if not encoding: if not encoding:
encoding = xml_to_unicode(src[:4096], verbose=verbose)[-1] encoding = xml_to_unicode(src[:4096], verbose=verbose)[-1]