tied all the new heuristics options to preprocess.utils

This commit is contained in:
ldolse 2011-01-15 13:34:35 +08:00
parent 01b12c5e8a
commit 6f252bb105
2 changed files with 97 additions and 82 deletions

View File

@ -523,7 +523,7 @@ OptionRecommendation(name='delete_blank_paragraphs',
OptionRecommendation(name='format_scene_breaks',
recommended_value=False, level=OptionRecommendation.LOW,
help=_('Detects left aligned scene break markers and center aligns them. '
help=_('left aligned scene break markers are center aligned. '
'Replace soft scene breaks that use multiple blank lines with'
'horizontal rules.')),

View File

@ -18,6 +18,11 @@ class PreProcessor(object):
self.html_preprocess_sections = 0
self.found_indents = 0
self.extra_opts = extra_opts
self.deleted_nbsps = False
self.min_chapters = 1
self.linereg = re.compile('(?<=<p).*?(?=</p>)', re.IGNORECASE|re.DOTALL)
self.blankreg = re.compile(r'\s*(?P<openline><p(?!\sid=\"softbreak\")[^>]*>)\s*(?P<closeline></p>)', re.IGNORECASE)
self.multi_blank = re.compile(r'(\s*<p[^>]*>\s*</p>){2,}', re.IGNORECASE)
def is_pdftohtml(self, src):
return '<!-- created by calibre\'s pdftohtml -->' in src[:1000]
@ -120,7 +125,6 @@ class PreProcessor(object):
'''
# Typical chapters are between 2000 and 7000 words, use the larger number to decide the
# minimum of chapters to search for
self.min_chapters = 1
if wordcount > 7000:
self.min_chapters = int(ceil(wordcount / 7000.))
#print "minimum chapters required are: "+str(self.min_chapters)
@ -192,21 +196,28 @@ class PreProcessor(object):
def punctuation_unwrap(self, length, content, format):
'''
Unwraps lines based on line length and punctuation
supports range of potential html markup and text files
supports a range of html markup and text files
'''
# define the pieces of the regex
lookahead = "(?<=.{"+str(length)+"}([a-zäëïöüàèìòùáćéíóńśúâêîôûçąężıãõñæøþðß,:)\IA\u00DF]|(?<!\&\w{4});))" # (?<!\&\w{4});) is a semicolon not part of an entity
line_ending = "\s*</(span|p|div)>\s*(</(p|span|div)>)?"
em_en_lookahead = "(?<=.{"+str(length)+"}[\u2013\u2014])"
line_ending = "\s*</(span|[iubp]|div)>\s*(</(span|[iubp]|div)>)?"
blanklines = "\s*(?P<up2threeblanks><(p|span|div)[^>]*>\s*(<(p|span|div)[^>]*>\s*</(span|p|div)>\s*)</(span|p|div)>\s*){0,3}\s*"
line_opening = "<(span|div|p)[^>]*>\s*(<(span|div|p)[^>]*>)?\s*"
line_opening = "<(span|[iubp]|div)[^>]*>\s*(<(span|[iubp]|div)[^>]*>)?\s*"
txt_line_wrap = u"((\u0020|\u0009)*\n){1,4}"
unwrap_regex = lookahead+line_ending+blanklines+line_opening
em_en_unwrap_regex = em_en_lookahead+line_ending+blanklines+line_opening
if format == 'txt':
unwrap_regex = lookahead+txt_line_wrap
em_en_unwrap_regex = em_en_lookahead+txt_line_wrap
unwrap = re.compile(u"%s" % unwrap_regex, re.UNICODE)
em_en_unwrap = re.compile(u"%s" % em_en_unwrap_regex, re.UNICODE)
content = unwrap.sub(' ', content)
content = em_en_unwrap.sub('', content)
return content
@ -253,8 +264,38 @@ class PreProcessor(object):
html = re.sub(r"\s*<span[^>]*>\s*(<span[^>]*>\s*</span>){0,2}\s*</span>\s*", " ", html)
html = re.sub(r"\s*<[ibu][^>]*>\s*(<[ibu][^>]*>\s*</[ibu]>\s*){0,2}\s*</[ibu]>", " ", html)
html = re.sub(r"\s*<span[^>]*>\s*(<span[^>]>\s*</span>){0,2}\s*</span>\s*", " ", html)
self.deleted_nbsps = True
return html
def analyze_line_endings(self, html):
'''
determines the type of html line ending used most commonly in a document
use before calling docanalysis functions
'''
paras_reg = re.compile('<p[^>]*>', re.IGNORECASE)
spans_reg = re.compile('<span[^>]*>', re.IGNORECASE)
paras = len(paras_reg.findall(html))
spans = len(spans_reg.findall(html))
if spans > 1:
if float(paras) / float(spans) < 0.75:
return 'spanned_html'
else:
return 'html'
else:
return 'html'
def analyze_blanks(self, html):
blanklines = self.blankreg.findall(html)
lines = self.linereg.findall(html)
if len(lines) > 1:
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:
return True
else:
return False
def __call__(self, html):
self.log("********* Preprocessing HTML *********")
@ -281,9 +322,8 @@ class PreProcessor(object):
# check if content is in pre tags, use txt processor to mark up if so
html = self.text_process_pre(html)
###### Mark Indents/Cleanup ######
#
# Replace series of non-breaking spaces with text-indent
if getattr(self.extra_opts, 'fix_indents', True):
html = self.fix_nbsp_indents(html)
html = self.cleanup_markup(html)
@ -291,85 +331,58 @@ class PreProcessor(object):
# ADE doesn't render <br />, change to empty paragraphs
#html = re.sub('<br[^>]*>', u'<p>\u00a0</p>', html)
# If more than 40% of the lines are empty paragraphs and the user has enabled delete
# blank paragraphs then delete blank lines to clean up spacing
linereg = re.compile('(?<=<p).*?(?=</p>)', re.IGNORECASE|re.DOTALL)
blankreg = re.compile(r'\s*(?P<openline><p(?!\sid=\"softbreak\")[^>]*>)\s*(?P<closeline></p>)', re.IGNORECASE)
multi_blank = re.compile(r'(\s*<p[^>]*>\s*</p>){2,}', re.IGNORECASE)
blanklines = blankreg.findall(html)
lines = linereg.findall(html)
blanks_between_paragraphs = False
print "delete blank paragraphs is "+str(getattr(self.extra_opts, 'delete_blank_paragraphs', False))
if len(lines) > 1:
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:
blanks_between_paragraphs = True
print "blanks between paragraphs is marked True"
else:
blanks_between_paragraphs = False
# Determine whether the document uses interleaved blank lines
blanks_between_paragraphs = self.analyze_blanks(html)
#self.dump(html, 'before_chapter_markup')
# detect chapters/sections to match xpath or splitting logic
#
if getattr(self.extra_opts, 'markup_chapter_headings', True):
html = self.markup_chapters(html, totalwords, blanks_between_paragraphs)
# If more than 40% of the lines are empty paragraphs and the user has enabled delete
# blank paragraphs then delete blank lines to clean up spacing
if blanks_between_paragraphs and getattr(self.extra_opts,
'delete_blank_paragraphs', False):
self.log("deleting blank lines")
html = multi_blank.sub('\n<p id="softbreak" style="margin-top:1.5em; margin-bottom:1.5em"> </p>', html)
html = blankreg.sub('', html)
html = self.multi_blank.sub('\n<p id="softbreak" style="margin-top:1.5em; margin-bottom:1.5em"> </p>', html)
html = self.blankreg.sub('', html)
###### Unwrap lines ######
#
if getattr(self.extra_opts, 'unwrap_lines', True):
# Determine line ending type
# Some OCR sourced files have line breaks in the html using a combination of span & p tags
# span are used for hard line breaks, p for new paragraphs. Determine which is used so
# that lines can be un-wrapped across page boundaries
paras_reg = re.compile('<p[^>]*>', re.IGNORECASE)
spans_reg = re.compile('<span[^>]*>', re.IGNORECASE)
paras = len(paras_reg.findall(html))
spans = len(spans_reg.findall(html))
if spans > 1:
if float(paras) / float(spans) < 0.75:
format = 'spanned_html'
else:
format = 'html'
else:
format = 'html'
format = self.analyze_line_endings(html)
# Check Line histogram to determine if the document uses hard line breaks, If 50% or
# 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 "+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 " + 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")
# Unwrap em/en dashes
html = re.sub(u'(?<=.{%i}[\u2013\u2014])\s*(?=<)(</span>\s*(</[iubp]>\s*<[iubp][^>]*>\s*)?<span[^>]*>|</[iubp]>\s*<[iubp][^>]*>)?\s*(?=[[a-z\d])' % length, '', html)
# Dehyphenate
self.log("Unwrapping/Removing hyphens")
# Dehyphenate with line length limiters
dehyphenator = Dehyphenator()
html = dehyphenator(html,'html', length)
self.log("Done dehyphenating")
# Unwrap lines using punctation and line length
#unwrap_quotes = re.compile(u"(?<=.{%i}\"')\s*</(span|p|div)>\s*(</(p|span|div)>)?\s*(?P<up2threeblanks><(p|span|div)[^>]*>\s*(<(p|span|div)[^>]*>\s*</(span|p|div)>\s*)</(span|p|div)>\s*){0,3}\s*<(span|div|p)[^>]*>\s*(<(span|div|p)[^>]*>)?\s*(?=[a-z])" % length, re.UNICODE)
html = self.punctuation_unwrap(length, html, 'html')
#check any remaining hyphens, but only unwrap if there is a match
dehyphenator = Dehyphenator()
html = dehyphenator(html,'html_cleanup', length)
else:
if getattr(self.extra_opts, 'dehyphenate', True):
# dehyphenate in cleanup mode to fix anything previous conversions/editing missed
self.log("Cleaning up hyphenation")
self.log("Fixing hyphenated content")
dehyphenator = Dehyphenator()
html = dehyphenator(html,'html_cleanup', length)
self.log("Done dehyphenating")
# delete soft hyphens
html = re.sub(u'\xad\s*(</span>\s*(</[iubp]>\s*<[iubp][^>]*>\s*)?<span[^>]*>|</[iubp]>\s*<[iubp][^>]*>)?\s*', '', html)
@ -385,10 +398,12 @@ class PreProcessor(object):
doubleheading = re.compile(r'(?P<firsthead><h(1|2)[^>]*>.+?</h(1|2)>\s*(<(?!h\d)[^>]*>\s*)*)<h(1|2)(?P<secondhead>[^>]*>.+?)</h(1|2)>', re.IGNORECASE)
html = doubleheading.sub('\g<firsthead>'+'\n<h3'+'\g<secondhead>'+'</h3>', html)
# put back non-breaking spaces in empty paragraphs to preserve original formatting
html = blankreg.sub('\n'+r'\g<openline>'+u'\u00a0'+r'\g<closeline>', html)
if getattr(self.extra_opts, 'dehyphenate', True):
# Center separator lines
html = re.sub(u'<(?P<outer>p|div)[^>]*>\s*(<(?P<inner1>font|span|[ibu])[^>]*>)?\s*(<(?P<inner2>font|span|[ibu])[^>]*>)?\s*(<(?P<inner3>font|span|[ibu])[^>]*>)?\s*(?P<break>([*#•]+\s*)+)\s*(</(?P=inner3)>)?\s*(</(?P=inner2)>)?\s*(</(?P=inner1)>)?\s*</(?P=outer)>', '<p style="text-align:center">' + '\g<break>' + '</p>', html)
html = re.sub(u'<(?P<outer>p|div)[^>]*>\s*(<(?P<inner1>font|span|[ibu])[^>]*>)?\s*(<(?P<inner2>font|span|[ibu])[^>]*>)?\s*(<(?P<inner3>font|span|[ibu])[^>]*>)?\s*(?P<break>([*#•]+\s*)+)\s*(</(?P=inner3)>)?\s*(</(?P=inner2)>)?\s*(</(?P=inner1)>)?\s*</(?P=outer)>', '<p style="text-align:center; margin-top:1.25em; margin-bottom:1.25em">' + '\g<break>' + '</p>', html)
if self.deleted_nbsps:
# put back non-breaking spaces in empty paragraphs to preserve original formatting
html = self.blankreg.sub('\n'+r'\g<openline>'+u'\u00a0'+r'\g<closeline>', html)
return html