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

Fix for HTMLZ not converting necessary characters to entities.

Browse Source
This commit is contained in:
John Schember 2011-08-28 09:30:33 -04:00
parent 2222ecca21
commit 472e8d1743
1 changed files with 14 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
src/calibre/ebooks/htmlz/oeb2html.py
View File

@ -126,6 +126,14 @@ class OEB2HTML(object):
css = item.data.cssText css = item.data.cssText
break break
return css return css
def prepare_string_for_html(self, raw):
raw = prepare_string_for_xml(raw)
raw = raw.replace(u'\u00ad', '­')
raw = raw.replace(u'\u2014', '—')
raw = raw.replace(u'\u2013', '–')
raw = raw.replace(u'\u00a0', ' ')
return raw
class OEB2HTMLNoCSSizer(OEB2HTML): class OEB2HTMLNoCSSizer(OEB2HTML):
@ -194,7 +202,7 @@ class OEB2HTMLNoCSSizer(OEB2HTML):
# Process tags that contain text. # Process tags that contain text.
if hasattr(elem, 'text') and elem.text: if hasattr(elem, 'text') and elem.text:
text.append(elem.text) text.append(self.prepare_string_for_html(elem.text))
# Recurse down into tags within the tag we are in. # Recurse down into tags within the tag we are in.
for item in elem: for item in elem:
@ -207,7 +215,7 @@ class OEB2HTMLNoCSSizer(OEB2HTML):
# Add the text that is outside of the tag. # Add the text that is outside of the tag.
if hasattr(elem, 'tail') and elem.tail: if hasattr(elem, 'tail') and elem.tail:
text.append(elem.tail) text.append(self.prepare_string_for_html(elem.tail))
return text return text
@ -267,7 +275,7 @@ class OEB2HTMLInlineCSSizer(OEB2HTML):
# Process tags that contain text. # Process tags that contain text.
if hasattr(elem, 'text') and elem.text: if hasattr(elem, 'text') and elem.text:
text.append(elem.text) text.append(self.prepare_string_for_html(elem.text))
# Recurse down into tags within the tag we are in. # Recurse down into tags within the tag we are in.
for item in elem: for item in elem:
@ -280,7 +288,7 @@ class OEB2HTMLInlineCSSizer(OEB2HTML):
# Add the text that is outside of the tag. # Add the text that is outside of the tag.
if hasattr(elem, 'tail') and elem.tail: if hasattr(elem, 'tail') and elem.tail:
text.append(elem.tail) text.append(self.prepare_string_for_html(elem.tail))
return text return text
@ -347,7 +355,7 @@ class OEB2HTMLClassCSSizer(OEB2HTML):
# Process tags that contain text. # Process tags that contain text.
if hasattr(elem, 'text') and elem.text: if hasattr(elem, 'text') and elem.text:
text.append(elem.text) text.append(self.prepare_string_for_html(elem.text))
# Recurse down into tags within the tag we are in. # Recurse down into tags within the tag we are in.
for item in elem: for item in elem:
@ -360,7 +368,7 @@ class OEB2HTMLClassCSSizer(OEB2HTML):
# Add the text that is outside of the tag. # Add the text that is outside of the tag.
if hasattr(elem, 'tail') and elem.tail: if hasattr(elem, 'tail') and elem.tail:
text.append(elem.tail) text.append(self.prepare_string_for_html(elem.tail))
return text return text