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

HTMLZ Input: Detect or use input encoding.

Browse Source
This commit is contained in:
John Schember 2011-04-15 08:07:38 -04:00
parent 5b82c42e4b
commit 4f3abc5614
1 changed files with 9 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
src/calibre/ebooks/htmlz/input.py
View File

@ -10,6 +10,7 @@ import os
from calibre import walk from calibre import walk
from calibre.customize.conversion import InputFormatPlugin from calibre.customize.conversion import InputFormatPlugin
from calibre.ebooks.chardet import xml_to_unicode
from calibre.utils.zipfile import ZipFile from calibre.utils.zipfile import ZipFile
class HTMLZInput(InputFormatPlugin): class HTMLZInput(InputFormatPlugin):
@ -34,6 +35,13 @@ class HTMLZInput(InputFormatPlugin):
html = tf.read() html = tf.read()
break break
# Encoding
if options.input_encoding:
ienc = options.input_encoding
else:
ienc = xml_to_unicode(html[:4096])[-1]
html = html.decode(ienc, 'replace')
# Run the HTML through the html processing plugin. # Run the HTML through the html processing plugin.
from calibre.customize.ui import plugin_for_input_format from calibre.customize.ui import plugin_for_input_format
html_input = plugin_for_input_format('html') html_input = plugin_for_input_format('html')
@ -48,7 +56,7 @@ class HTMLZInput(InputFormatPlugin):
fname = 'index%d.html'%c fname = 'index%d.html'%c
htmlfile = open(fname, 'wb') htmlfile = open(fname, 'wb')
with htmlfile: with htmlfile:
htmlfile.write(html.encode('utf-8')) htmlfile.write(html.encode('utf-8', 'replace'))
odi = options.debug_pipeline odi = options.debug_pipeline
options.debug_pipeline = None options.debug_pipeline = None
# Generate oeb from html conversion. # Generate oeb from html conversion.