Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-07 10:14:46 -04:00
Code Issues Packages Projects Releases Wiki Activity

Fix unicode issues in new preprocess code

Browse Source
This commit is contained in:
Kovid Goyal 2010-09-30 17:03:51 -06:00
parent 47f93f4c9c
commit 54ee947072
2 changed files with 22 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/calibre/ebooks/conversion/preprocess.py
View File

@ -184,14 +184,14 @@ class Dehyphenator(object):
wraptags = match.group('wraptags')
except:
wraptags = ''
hyphenated = str(firsthalf) + "-" + str(secondhalf)
dehyphenated = str(firsthalf) + str(secondhalf)
hyphenated = unicode(firsthalf) + "-" + unicode(secondhalf)
dehyphenated = unicode(firsthalf) + unicode(secondhalf)
lookupword = self.removesuffixes.sub('', dehyphenated)
if self.prefixes.match(firsthalf) is None:
lookupword = self.removeprefix.sub('', lookupword)
#print "lookup word is: "+str(lookupword)+", orig is: " + str(hyphenated)
try:
searchresult = self.html.find(str.lower(lookupword))
searchresult = self.html.find(lookupword.lower())
except:
return hyphenated
if self.format == 'html_cleanup':

32
src/calibre/ebooks/conversion/utils.py
View File

@ -22,18 +22,21 @@ class PreProcessor(object):
title = match.group('title')
if not title:
self.html_preprocess_sections = self.html_preprocess_sections + 1
self.log("found " + str(self.html_preprocess_sections) + " chapters. - " + str(chap))
self.log("found " + unicode(self.html_preprocess_sections) +
" chapters. - " + unicode(chap))
return '<h2>'+chap+'</h2>\n'
else:
self.html_preprocess_sections = self.html_preprocess_sections + 1
self.log("found " + str(self.html_preprocess_sections) + " chapters & titles. - " + str(chap) + ", " + str(title))
self.log("found " + unicode(self.html_preprocess_sections) +
" chapters & titles. - " + unicode(chap) + ", " + unicode(title))
return '<h2>'+chap+'</h2>\n<h3>'+title+'</h3>\n'
def chapter_break(self, match):
chap = match.group('section')
styles = match.group('styles')
self.html_preprocess_sections = self.html_preprocess_sections + 1
self.log("marked " + str(self.html_preprocess_sections) + " section markers based on punctuation. - " + str(chap))
self.log("marked " + unicode(self.html_preprocess_sections) +
" section markers based on punctuation. - " + unicode(chap))
return '<'+styles+' style="page-break-before:always">'+chap
def insert_indent(self, match):
@ -63,7 +66,8 @@ class PreProcessor(object):
line_end = line_end_ere.findall(raw)
tot_htm_ends = len(htm_end)
tot_ln_fds = len(line_end)
self.log("There are " + str(tot_ln_fds) + " total Line feeds, and " + str(tot_htm_ends) + " marked up endings")
self.log("There are " + unicode(tot_ln_fds) + " total Line feeds, and " +
unicode(tot_htm_ends) + " marked up endings")
if percent > 1:
percent = 1
@ -71,7 +75,7 @@ class PreProcessor(object):
percent = 0
min_lns = tot_ln_fds * percent
self.log("There must be fewer than " + str(min_lns) + " unmarked lines to add markup")
self.log("There must be fewer than " + unicode(min_lns) + " unmarked lines to add markup")
if min_lns > tot_htm_ends:
return True
@ -112,7 +116,7 @@ class PreProcessor(object):
txtindent = re.compile(ur'<p(?P<formatting>[^>]*)>\s*(?P<span>(<span[^>]*>\s*)+)?\s*(\u00a0){2,}', re.IGNORECASE)
html = txtindent.sub(self.insert_indent, html)
if self.found_indents > 1:
self.log("replaced "+str(self.found_indents)+ " nbsp indents with inline styles")
self.log("replaced "+unicode(self.found_indents)+ " nbsp indents with inline styles")
# remove remaining non-breaking spaces
html = re.sub(ur'\u00a0', ' ', html)
# Get rid of empty <o:p> tags to simplify other processing
@ -131,7 +135,8 @@ class PreProcessor(object):
lines = linereg.findall(html)
blanks_between_paragraphs = False
if len(lines) > 1:
self.log("There are " + str(len(blanklines)) + " blank lines. " + str(float(len(blanklines)) / float(len(lines))) + " percent blank")
self.log("There are " + unicode(len(blanklines)) + " blank lines. " +
unicode(float(len(blanklines)) / float(len(lines))) + " percent blank")
if float(len(blanklines)) / float(len(lines)) > 0.40 and getattr(self.extra_opts,
'remove_paragraph_spacing', False):
self.log("deleting blank lines")
@ -170,20 +175,20 @@ class PreProcessor(object):
#print chapter_marker
heading = re.compile('<h[1-3][^>]*>', re.IGNORECASE)
self.html_preprocess_sections = len(heading.findall(html))
self.log("found " + str(self.html_preprocess_sections) + " pre-existing headings")
self.log("found " + unicode(self.html_preprocess_sections) + " pre-existing headings")
#
# Start with most typical chapter headings, get more aggressive until one works
if self.html_preprocess_sections < 10:
chapdetect = re.compile(r'%s' % chapter_marker, re.IGNORECASE)
html = chapdetect.sub(self.chapter_head, html)
if self.html_preprocess_sections < 10:
self.log("not enough chapters, only " + str(self.html_preprocess_sections) + ", trying numeric chapters")
self.log("not enough chapters, only " + unicode(self.html_preprocess_sections) + ", trying numeric chapters")
chapter_marker = lookahead+chapter_line_open+chapter_header_open+numeric_chapters+chapter_header_close+chapter_line_close+blank_lines+opt_title_open+title_line_open+title_header_open+default_title+title_header_close+title_line_close+opt_title_close
chapdetect2 = re.compile(r'%s' % chapter_marker, re.IGNORECASE)
html = chapdetect2.sub(self.chapter_head, html)
if self.html_preprocess_sections < 10:
self.log("not enough chapters, only " + str(self.html_preprocess_sections) + ", trying with uppercase words")
self.log("not enough chapters, only " + unicode(self.html_preprocess_sections) + ", trying with uppercase words")
chapter_marker = lookahead+chapter_line_open+chapter_header_open+uppercase_chapters+chapter_header_close+chapter_line_close+blank_lines+opt_title_open+title_line_open+title_header_open+default_title+title_header_close+title_line_close+opt_title_close
chapdetect2 = re.compile(r'%s' % chapter_marker, re.UNICODE)
html = chapdetect2.sub(self.chapter_head, html)
@ -207,11 +212,11 @@ class PreProcessor(object):
# more of the lines break in the same region of the document then unwrapping is required
docanalysis = DocAnalysis(format, html)
hardbreaks = docanalysis.line_histogram(.50)
self.log("Hard line breaks check returned "+str(hardbreaks))
self.log("Hard line breaks check returned "+unicode(hardbreaks))
# Calculate Length
unwrap_factor = getattr(self.extra_opts, 'html_unwrap_factor', 0.4)
length = docanalysis.line_length(unwrap_factor)
self.log("*** Median line length is " + str(length) + ", calculated with " + format + " format ***")
self.log("*** Median line length is " + unicode(length) + ", calculated with " + format + " format ***")
# only go through unwrapping code if the histogram shows unwrapping is required or if the user decreased the default unwrap_factor
if hardbreaks or unwrap_factor < 0.4:
self.log("Unwrapping required, unwrapping Lines")
@ -240,7 +245,8 @@ class PreProcessor(object):
# If still no sections after unwrapping mark split points on lines with no punctuation
if self.html_preprocess_sections < 10:
self.log("Looking for more split points based on punctuation, currently have " + str(self.html_preprocess_sections))
self.log("Looking for more split points based on punctuation,"
" currently have " + unicode(self.html_preprocess_sections))
chapdetect3 = re.compile(r'<(?P<styles>(p|div)[^>]*)>\s*(?P<section>(<span[^>]*>)?\s*(<[ibu][^>]*>){0,2}\s*(<span[^>]*>)?\s*(<[ibu][^>]*>){0,2}\s*(<span[^>]*>)?\s*.?(?=[a-z#\-*\s]+<)([a-z#-*]+\s*){1,5}\s*\s*(</span>)?(</[ibu]>){0,2}\s*(</span>)?\s*(</[ibu]>){0,2}\s*(</span>)?\s*</(p|div)>)', re.IGNORECASE)
html = chapdetect3.sub(self.chapter_break, html)
# search for places where a first or second level heading is immediately followed by another