Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-08 02:34:06 -04:00
Code Issues Packages Projects Releases Wiki Activity

EPUB Output: Make the file size splitting algorithm more intelligent. If a split results ina tree that is very small, choose another split point.

Browse Source
This commit is contained in:
Kovid Goyal 2010-05-02 15:17:09 -06:00
parent 3df90b8926
commit 32f8611a76
2 changed files with 11 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
src/calibre/ebooks/oeb/base.py
View File

@ -295,7 +295,7 @@ def xml2unicode(root, pretty_print=False):
return etree.tostring(root, pretty_print=pretty_print) return etree.tostring(root, pretty_print=pretty_print)
def xml2text(elem): def xml2text(elem):
return etree.tostring(elem, method='text', encoding=unicode) return etree.tostring(elem, method='text', encoding=unicode, with_tail=False)
ASCII_CHARS = set(chr(x) for x in xrange(128)) ASCII_CHARS = set(chr(x) for x in xrange(128))
UNIBYTE_CHARS = set(chr(x) for x in xrange(256)) UNIBYTE_CHARS = set(chr(x) for x in xrange(256))

14
src/calibre/ebooks/oeb/transforms/split.py
View File

@ -5,7 +5,7 @@ __docformat__ = 'restructuredtext en'
''' '''
Splitting of the XHTML flows. Splitting can happen on page boundaries or can be Splitting of the XHTML flows. Splitting can happen on page boundaries or can be
forces at "likely" locations to conform to size limitations. This transform forced at "likely" locations to conform to size limitations. This transform
assumes a prior call to the flatcss transform. assumes a prior call to the flatcss transform.
''' '''
@ -385,12 +385,18 @@ class FlowSplitter(object):
raise SplitError(self.item.href, root) raise SplitError(self.item.href, root)
self.log.debug('\t\t\tSplit point:', split_point.tag, tree.getpath(split_point)) self.log.debug('\t\t\tSplit point:', split_point.tag, tree.getpath(split_point))
for t in self.do_split(tree, split_point, before): trees = self.do_split(tree, split_point, before)
sizes = [len(tostring(t.getroot())) for t in trees]
if min(sizes) < 5*1024:
self.log.debug('\t\t\tSplit tree too small')
self.split_to_size(tree)
return
for t, size in zip(trees, sizes):
r = t.getroot() r = t.getroot()
if self.is_page_empty(r): if self.is_page_empty(r):
continue continue
size = len(tostring(r)) elif size <= self.max_flow_size:
if size <= self.max_flow_size:
self.split_trees.append(t) self.split_trees.append(t)
self.log.debug( self.log.debug(
'\t\t\tCommitted sub-tree #%d (%d KB)'%( '\t\t\tCommitted sub-tree #%d (%d KB)'%(