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

pep8

Browse Source
This commit is contained in:
Kovid Goyal 2013-05-12 10:18:35 +05:30
parent 70a6852ab6
commit abad7da850
1 changed files with 11 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
src/calibre/ebooks/conversion/preprocess.py
View File

@ -14,7 +14,7 @@ SVG_NS = 'http://www.w3.org/2000/svg'
XLINK_NS = 'http://www.w3.org/1999/xlink' XLINK_NS = 'http://www.w3.org/1999/xlink'
convert_entities = functools.partial(entity_to_unicode, convert_entities = functools.partial(entity_to_unicode,
result_exceptions = { result_exceptions={
u'<' : '&lt;', u'<' : '&lt;',
u'>' : '&gt;', u'>' : '&gt;',
u"'" : '&apos;', u"'" : '&apos;',
@ -144,9 +144,9 @@ class DocAnalysis(object):
percent is the percentage of lines that should be in a single bucket to return true percent is the percentage of lines that should be in a single bucket to return true
The majority of the lines will exist in 1-2 buckets in typical docs with hard line breaks The majority of the lines will exist in 1-2 buckets in typical docs with hard line breaks
''' '''
minLineLength=20 # Ignore lines under 20 chars (typical of spaces) minLineLength=20 # Ignore lines under 20 chars (typical of spaces)
maxLineLength=1900 # Discard larger than this to stay in range maxLineLength=1900 # Discard larger than this to stay in range
buckets=20 # Each line is divided into a bucket based on length buckets=20 # Each line is divided into a bucket based on length
#print "there are "+str(len(lines))+" lines" #print "there are "+str(len(lines))+" lines"
#max = 0 #max = 0
@ -156,7 +156,7 @@ class DocAnalysis(object):
# max = l # max = l
#print "max line found is "+str(max) #print "max line found is "+str(max)
# Build the line length histogram # Build the line length histogram
hRaw = [ 0 for i in range(0,buckets) ] hRaw = [0 for i in range(0,buckets)]
for line in self.lines: for line in self.lines:
l = len(line) l = len(line)
if l > minLineLength and l < maxLineLength: if l > minLineLength and l < maxLineLength:
@ -167,7 +167,7 @@ class DocAnalysis(object):
# Normalize the histogram into percents # Normalize the histogram into percents
totalLines = len(self.lines) totalLines = len(self.lines)
if totalLines > 0: if totalLines > 0:
h = [ float(count)/totalLines for count in hRaw ] h = [float(count)/totalLines for count in hRaw]
else: else:
h = [] h = []
#print "\nhRaw histogram lengths are: "+str(hRaw) #print "\nhRaw histogram lengths are: "+str(hRaw)
@ -200,7 +200,7 @@ class Dehyphenator(object):
# Add common suffixes to the regex below to increase the likelihood of a match - # Add common suffixes to the regex below to increase the likelihood of a match -
# don't add suffixes which are also complete words, such as 'able' or 'sex' # don't add suffixes which are also complete words, such as 'able' or 'sex'
# only remove if it's not already the point of hyphenation # only remove if it's not already the point of hyphenation
self.suffix_string = "((ed)?ly|'?e?s||a?(t|s)?ion(s|al(ly)?)?|ings?|er|(i)?ous|(i|a)ty|(it)?ies|ive|gence|istic(ally)?|(e|a)nce|m?ents?|ism|ated|(e|u)ct(ed)?|ed|(i|ed)?ness|(e|a)ncy|ble|ier|al|ex|ian)$" self.suffix_string = "((ed)?ly|'?e?s||a?(t|s)?ion(s|al(ly)?)?|ings?|er|(i)?ous|(i|a)ty|(it)?ies|ive|gence|istic(ally)?|(e|a)nce|m?ents?|ism|ated|(e|u)ct(ed)?|ed|(i|ed)?ness|(e|a)ncy|ble|ier|al|ex|ian)$" # noqa
self.suffixes = re.compile(r"^%s" % self.suffix_string, re.IGNORECASE) self.suffixes = re.compile(r"^%s" % self.suffix_string, re.IGNORECASE)
self.removesuffixes = re.compile(r"%s" % self.suffix_string, re.IGNORECASE) self.removesuffixes = re.compile(r"%s" % self.suffix_string, re.IGNORECASE)
# remove prefixes if the prefix was not already the point of hyphenation # remove prefixes if the prefix was not already the point of hyphenation
@ -265,19 +265,18 @@ class Dehyphenator(object):
self.html = html self.html = html
self.format = format self.format = format
if format == 'html': if format == 'html':
intextmatch = re.compile(u'(?<=.{%i})(?P<firstpart>[^\W\-]+)(-|‐)\s*(?=<)(?P<wraptags>(</span>)?\s*(</[iubp]>\s*){1,2}(?P<up2threeblanks><(p|div)[^>]*>\s*(<p[^>]*>\s*</p>\s*)?</(p|div)>\s+){0,3}\s*(<[iubp][^>]*>\s*){1,2}(<span[^>]*>)?)\s*(?P<secondpart>[\w\d]+)' % length) intextmatch = re.compile(u'(?<=.{%i})(?P<firstpart>[^\W\-]+)(-|‐)\s*(?=<)(?P<wraptags>(</span>)?\s*(</[iubp]>\s*){1,2}(?P<up2threeblanks><(p|div)[^>]*>\s*(<p[^>]*>\s*</p>\s*)?</(p|div)>\s+){0,3}\s*(<[iubp][^>]*>\s*){1,2}(<span[^>]*>)?)\s*(?P<secondpart>[\w\d]+)' % length) # noqa
elif format == 'pdf': elif format == 'pdf':
intextmatch = re.compile(u'(?<=.{%i})(?P<firstpart>[^\W\-]+)(-|‐)\s*(?P<wraptags><p>|</[iub]>\s*<p>\s*<[iub]>)\s*(?P<secondpart>[\w\d]+)'% length) intextmatch = re.compile(u'(?<=.{%i})(?P<firstpart>[^\W\-]+)(-|‐)\s*(?P<wraptags><p>|</[iub]>\s*<p>\s*<[iub]>)\s*(?P<secondpart>[\w\d]+)'% length)
elif format == 'txt': elif format == 'txt':
intextmatch = re.compile(u'(?<=.{%i})(?P<firstpart>[^\W\-]+)(-|‐)(\u0020|\u0009)*(?P<wraptags>(\n(\u0020|\u0009)*)+)(?P<secondpart>[\w\d]+)'% length) intextmatch = re.compile(u'(?<=.{%i})(?P<firstpart>[^\W\-]+)(-|‐)(\u0020|\u0009)*(?P<wraptags>(\n(\u0020|\u0009)*)+)(?P<secondpart>[\w\d]+)'% length) # noqa
elif format == 'individual_words': elif format == 'individual_words':
intextmatch = re.compile(u'(?!<)(?P<firstpart>[^\W\-]+)(-|‐)\s*(?P<secondpart>\w+)(?![^<]*?>)') intextmatch = re.compile(u'(?!<)(?P<firstpart>[^\W\-]+)(-|‐)\s*(?P<secondpart>\w+)(?![^<]*?>)')
elif format == 'html_cleanup': elif format == 'html_cleanup':
intextmatch = re.compile(u'(?P<firstpart>[^\W\-]+)(-|‐)\s*(?=<)(?P<wraptags></span>\s*(</[iubp]>\s*<[iubp][^>]*>\s*)?<span[^>]*>|</[iubp]>\s*<[iubp][^>]*>)?\s*(?P<secondpart>[\w\d]+)') intextmatch = re.compile(u'(?P<firstpart>[^\W\-]+)(-|‐)\s*(?=<)(?P<wraptags></span>\s*(</[iubp]>\s*<[iubp][^>]*>\s*)?<span[^>]*>|</[iubp]>\s*<[iubp][^>]*>)?\s*(?P<secondpart>[\w\d]+)') # noqa
elif format == 'txt_cleanup': elif format == 'txt_cleanup':
intextmatch = re.compile(u'(?P<firstpart>[^\W\-]+)(-|‐)(?P<wraptags>\s+)(?P<secondpart>[\w\d]+)') intextmatch = re.compile(u'(?P<firstpart>[^\W\-]+)(-|‐)(?P<wraptags>\s+)(?P<secondpart>[\w\d]+)')
html = intextmatch.sub(self.dehyphenate, html) html = intextmatch.sub(self.dehyphenate, html)
return html return html
@ -581,7 +580,7 @@ class HTMLPreProcessor(object):
end_rules.append((re.compile(u'(?<=.{%i}[–—])\s*<p>\s*(?=[[a-z\d])' % length), lambda match: '')) end_rules.append((re.compile(u'(?<=.{%i}[–—])\s*<p>\s*(?=[[a-z\d])' % length), lambda match: ''))
end_rules.append( end_rules.append(
# Un wrap using punctuation # Un wrap using punctuation
(re.compile(u'(?<=.{%i}([a-zäëïöüàèìòùáćéíĺóŕńśúýâêîôûçąężıãõñæøþðßěľščťžňďřů,:)\IA\u00DF]|(?<!\&\w{4});))\s*(?P<ital></(i|b|u)>)?\s*(</p>\s*<p>\s*)+\s*(?=(<(i|b|u)>)?\s*[\w\d$(])' % length, re.UNICODE), wrap_lines), (re.compile(u'(?<=.{%i}([a-zäëïöüàèìòùáćéíĺóŕńśúýâêîôûçąężıãõñæøþðßěľščťžňďřů,:)\IA\u00DF]|(?<!\&\w{4});))\s*(?P<ital></(i|b|u)>)?\s*(</p>\s*<p>\s*)+\s*(?=(<(i|b|u)>)?\s*[\w\d$(])' % length, re.UNICODE), wrap_lines), # noqa
) )
for rule in self.PREPROCESS + start_rules: for rule in self.PREPROCESS + start_rules: