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

More changes.

Browse Source
This commit is contained in:
John Schember 2011-05-09 21:19:28 -04:00
parent c384188057
commit 842ba75557
2 changed files with 61 additions and 32 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
src/calibre/ebooks/textile/functions.py
View File

@ -12,7 +12,7 @@ A Humane Web Text Generator
#__date__ = '2009/12/04' #__date__ = '2009/12/04'
__copyright__ = """ __copyright__ = """
Copyright (c) 2011, Leigh Parry <leighparry@blueyonder.co.uk> Copyright (c) 2011, Leigh Parry
Copyright (c) 2011, John Schember <john@nachtimwald.com> Copyright (c) 2011, John Schember <john@nachtimwald.com>
Copyright (c) 2009, Jason Samsa, http://jsamsa.com/ Copyright (c) 2009, Jason Samsa, http://jsamsa.com/
Copyright (c) 2004, Roberto A. F. De Almeida, http://dealmeida.net/ Copyright (c) 2004, Roberto A. F. De Almeida, http://dealmeida.net/
@ -219,14 +219,13 @@ class Textile(object):
] ]
glyph_defaults = [ glyph_defaults = [
(re.compile(r'(\d+\'?\"?)( ?)x( ?)(?=\d+)'), r'\1\2&#215;\3'), # dimension sign (re.compile(r'(\d+\'?\"?)( ?)x( ?)(?=\d+)'), r'\1\2&#215;\3'), # dimension sign
(re.compile(r'(\d+)\'', re.I), r'\1&#8242;'), # prime (re.compile(r'(\d+)\'(\s)', re.I), r'\1&#8242;\2'), # prime
(re.compile(r'(\d+)\"', re.I), r'\1&#8243;'), # prime-double (re.compile(r'(\d+)\"(\s)', re.I), r'\1&#8243;\2'), # prime-double
(re.compile(r'\b([A-Z][A-Z0-9]{2,})\b(?:[(]([^)]*)[)])'), r'<acronym title="\2">\1</acronym>'), # 3+ uppercase acronym (re.compile(r'\b([A-Z][A-Z0-9]{2,})\b(?:[(]([^)]*)[)])'), r'<acronym title="\2">\1</acronym>'), # 3+ uppercase acronym
(re.compile(r'\b([A-Z][A-Z\'\-]+[A-Z])(?=[\s.,\)>])'), r'<span class="caps">\1</span>'), # 3+ uppercase (re.compile(r'\b([A-Z][A-Z\'\-]+[A-Z])(?=[\s.,\)>])'), r'<span class="caps">\1</span>'), # 3+ uppercase
(re.compile(r'\b(\s{0,1})?\.{3}'), r'\1&#8230;'), # ellipsis (re.compile(r'\b(\s{0,1})?\.{3}'), r'\1&#8230;'), # ellipsis
(re.compile(r'^[\*_-]{3,}$', re.M), r'<hr />'), # <hr> scene-break (re.compile(r'^[\*_-]{3,}$', re.M), r'<hr />'), # <hr> scene-break
# (re.compile(r'\b--\b'), r'&#8212;'), # em dash (re.compile(r'(^|[^-])--([^-]|$)'), r'\1&#8212;\2'), # em dash
(re.compile(r'([^-])--([^-])'), r'\1&#8212;\2'), # em dash
(re.compile(r'\s-(?:\s|$)'), r' &#8211; '), # en dash (re.compile(r'\s-(?:\s|$)'), r' &#8211; '), # en dash
(re.compile(r'\b( ?)[([]TM[])]', re.I), r'\1&#8482;'), # trademark (re.compile(r'\b( ?)[([]TM[])]', re.I), r'\1&#8482;'), # trademark
(re.compile(r'\b( ?)[([]R[])]', re.I), r'\1&#174;'), # registered (re.compile(r'\b( ?)[([]R[])]', re.I), r'\1&#174;'), # registered
@ -706,6 +705,21 @@ class Textile(object):
result.append(line) result.append(line)
return ''.join(result) return ''.join(result)
def glyphs_only(self, text):
# fix: hackish
text = re.sub(r'"\Z', '\" ', text)
result = []
for line in re.compile(r'(<.*?>)', re.U).split(text):
if not re.search(r'<.*>', line):
rules = []
if re.search(r'{.+?}', line):
rules = self.macro_defaults
for s, r in rules:
line = s.sub(r, line)
result.append(line)
return ''.join(result)
def vAlign(self, input): def vAlign(self, input):
d = {'^':'top', '-':'middle', '~':'bottom'} d = {'^':'top', '-':'middle', '~':'bottom'}
return d.get(input, '') return d.get(input, '')
@ -792,7 +806,6 @@ class Textile(object):
text = self.noTextile(text) text = self.noTextile(text)
text = self.code(text) text = self.code(text)
text = self.glyphs(text)
text = self.links(text) text = self.links(text)
if not self.noimage: if not self.noimage:
@ -804,6 +817,7 @@ class Textile(object):
text = self.span(text) text = self.span(text)
text = self.footnoteRef(text) text = self.footnoteRef(text)
text = self.glyphs(text)
return text.rstrip('\n') return text.rstrip('\n')
@ -814,6 +828,7 @@ class Textile(object):
'fooobar ... and hello world ...' 'fooobar ... and hello world ...'
""" """
text = self.glyphs_only(text)
punct = '!"#$%&\'*+,-./:;=?@\\^_`|~' punct = '!"#$%&\'*+,-./:;=?@\\^_`|~'
pattern = r''' pattern = r'''
@ -868,7 +883,7 @@ class Textile(object):
>>> t.span(r"hello %(bob)span *strong* and **bold**% goodbye") >>> t.span(r"hello %(bob)span *strong* and **bold**% goodbye")
'hello <span class="bob">span <strong>strong</strong> and <b>bold</b></span> goodbye' 'hello <span class="bob">span <strong>strong</strong> and <b>bold</b></span> goodbye'
""" """
qtags = (r'\*\*', r'\*', r'\?\?', r'\-', r'__', r'_', r'%', r'\+', r'~', r'\^', r'&') qtags = (r'\*\*', r'\*', r'\?\?', r'\-', r'__', r'_', r'%', r'\+', r'~', r'\^')
pnct = ".,\"'?!;:" pnct = ".,\"'?!;:"
for qtag in qtags: for qtag in qtags:
@ -900,9 +915,7 @@ class Textile(object):
'%' : 'span', '%' : 'span',
'+' : 'ins', '+' : 'ins',
'~' : 'sub', '~' : 'sub',
'^' : 'sup', '^' : 'sup'
'&' : 'span style="font-variant:small-caps;"'
# '&' : 'span style="font-transform:uppercase;font-size:smaller;"'
} }
tag = qtags[tag] tag = qtags[tag]
atts = self.pba(atts) atts = self.pba(atts)
@ -1046,4 +1059,3 @@ def textile_restricted(text, lite=True, noimage=True, html_type='xhtml'):
return Textile(restricted=True, lite=lite, return Textile(restricted=True, lite=lite,
noimage=noimage).textile(text, rel='nofollow', noimage=noimage).textile(text, rel='nofollow',
html_type=html_type) html_type=html_type)

43
src/calibre/ebooks/txt/textileml.py
View File

@ -69,7 +69,8 @@ class TextileMLizer(OEB2HTML):
txt = '%s' % t txt = '%s' % t
if txt != '%': if txt != '%':
text = re.sub(r'([^'+t+'|^\n])'+t+'\]\['+t+'([^'+t+'])', r'\1\2', text) text = re.sub(r'([^'+t+'|^\n])'+t+'\]\['+t+'([^'+t+'])', r'\1\2', text)
text = re.sub(r'(\s|[*_])\[('+t+'[a-zA-Z0-9 \',.*_]+'+t+')\](\s|[*_])', r'\1\2\3', text) text = re.sub(r'([^'+t+'|^\n])'+t+t+'([^'+t+'])', r'\1\2', text)
text = re.sub(r'(\s|[*_\'"])\[('+t+'[a-zA-Z0-9 \'",.*_]+'+t+')\](\s|[*_\'"?!,.])', r'\1\2\3', text)
return text return text
# Now tidyup links and ids - remove ones that don't have a correponding opposite # Now tidyup links and ids - remove ones that don't have a correponding opposite
@ -77,14 +78,17 @@ class TextileMLizer(OEB2HTML):
for i in self.our_links: for i in self.our_links:
if i[0] == '#': if i[0] == '#':
if i not in self.our_ids: if i not in self.our_ids:
text = re.sub(r'"(.+)":'+i, '\1', text) self.log.debug('Link has no target - %s ...' % i)
text = re.sub(r'"(.+)":'+i+'(\s)', r'\1\2', text)
for i in self.our_ids: for i in self.our_ids:
if i not in self.our_links: if i not in self.our_links:
text = re.sub(r'\('+i+'\)', '', text) self.log.debug('ID has no link - %s ...' % i)
text = re.sub(r'%?\('+i+'\)\xa0?%?', r'', text)
# Note - I'm not checking for escaped '-' as this will also get hypenated words # Remove obvious non-needed escaping, add sub/sup-script ones
text = check_escaping(text, ['\*', '_', '\+', '-']) text = check_escaping(text, ['\*', '_', '\*'])
# text = check_escaping(text, ['\*', '_', '\+', '-']) text = re.sub(r'(\w)([~^]\w+[~^])', r'\1[\2]', text) # escape the super/sub-scripts if needed
text = re.sub(r'([~^]\w+[~^])(\w)', r'[\1]\2', text) # escape the super/sub-scripts if needed
text = re.sub(r'%\xa0+', r'%', text) #remove empty spans text = re.sub(r'%\xa0+', r'%', text) #remove empty spans
text = re.sub(r'%%', r'', text) #remove empty spans - MAY MERGE SOME ? text = re.sub(r'%%', r'', text) #remove empty spans - MAY MERGE SOME ?
@ -96,13 +100,14 @@ class TextileMLizer(OEB2HTML):
# text = re.sub(r'\n{4,}', r'\n\np. \n\n', text) #reduce blank lines + insert blank para # text = re.sub(r'\n{4,}', r'\n\np. \n\n', text) #reduce blank lines + insert blank para
text = re.sub(r'\n{3}', r'\n\n', text) #reduce blank lines text = re.sub(r'\n{3}', r'\n\n', text) #reduce blank lines
# text = re.sub(r' ((\* ?)+) ', r' ==\1== ', text) # text = re.sub(r' ((\* ?)+) ', r' ==\1== ', text)
text = re.sub(u'%\n(p[<>=]{1,2}\.)', r'%\n\n\1', text) text = re.sub(u'%\n(p[<>=]{1,2}\.|p\.)', r'%\n\n\1', text)
text = re.sub(r'\n\n {2,4}%', r'%', text) #Check span following blank para
text = re.sub(u'p[<>=]{1,2}\.\n\n?', r'', text) text = re.sub(u'p[<>=]{1,2}\.\n\n?', r'', text)
text = re.sub(r'(^|\n)p\.\n', r'\1p. \n', text) # blank paragraph text = re.sub(r'(^|\n)p\.\n', r'\1p. \n', text) # blank paragraph
text = re.sub(u'\n\xa0', r'\np. ', text) # blank paragraph text = re.sub(u'\n\xa0', r'\np. ', text) # blank paragraph
text = re.sub(u'\np[<>=]{1,2}?\. \xa0', r'\np. ', text) # blank paragraph text = re.sub(u'\np[<>=]{1,2}?\. \xa0', r'\np. ', text) # blank paragraph
text = re.sub(r'(^|\n)(p.*\. ?\n)(p.*\.)', r'\1\3', text) text = re.sub(r'(^|\n)(p.*\. ?\n)(p.*\.)', r'\1\3', text)
text = re.sub(r'\n(p\. \n)(p.*\.)', r'\n\2', text) text = re.sub(r'\n(p\. \n)(p.*\.|h.*\.)', r'\n\2', text)
text = re.sub(r' {2,}\|', r' |', text) #sort out spaces in tables text = re.sub(r' {2,}\|', r' |', text) #sort out spaces in tables
# Now put back spaces removed earlier as they're needed here # Now put back spaces removed earlier as they're needed here
@ -193,7 +198,8 @@ class TextileMLizer(OEB2HTML):
return txt return txt
def prepare_string_for_textile(self, txt): def prepare_string_for_textile(self, txt):
if re.search(r'(\s([*&_+\-=~@%|]|\?{2}))|(([*&_+\-=~@%|]|\?{2})\s)', txt): # if re.search(r'(\s([*&_+\-~@%|]|\?{2}))|(([*&_+\-~@%|]|\?{2})\s)', txt):
if re.search(r'(\s([*&_+\-~@%|]|\?{2})\S)|(\S([*&_+\-~@%|]|\?{2})\s)', txt):
return ' ==%s== ' % txt return ' ==%s== ' % txt
return txt return txt
@ -240,6 +246,10 @@ class TextileMLizer(OEB2HTML):
if style['font-style'] == 'italic' or tag in ('i', 'em'): if style['font-style'] == 'italic' or tag in ('i', 'em'):
if tag not in ('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'cite'): if tag not in ('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'cite'):
if self.style_italic == False: if self.style_italic == False:
if self.in_a_link:
text.append('_')
tags.append('_')
else:
text.append('[_') text.append('[_')
tags.append('_]') tags.append('_]')
self.style_embed.append('_') self.style_embed.append('_')
@ -247,6 +257,10 @@ class TextileMLizer(OEB2HTML):
if style['font-weight'] in ('bold', 'bolder') or tag in ('b', 'strong'): if style['font-weight'] in ('bold', 'bolder') or tag in ('b', 'strong'):
if tag not in ('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'th'): if tag not in ('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'th'):
if self.style_bold == False: if self.style_bold == False:
if self.in_a_link:
text.append('*')
tags.append('*')
else:
text.append('[*') text.append('[*')
tags.append('*]') tags.append('*]')
self.style_embed.append('*') self.style_embed.append('*')
@ -304,14 +318,17 @@ class TextileMLizer(OEB2HTML):
tags.append('pre\n') tags.append('pre\n')
elif tag == 'a': elif tag == 'a':
if self.opts.keep_links: if self.opts.keep_links:
if attribs.has_key('href'):
text.append('"') text.append('"')
tags.append('a') tags.append('a')
if attribs.has_key('href'):
tags.append('":' + attribs['href']) tags.append('":' + attribs['href'])
self.our_links.append(attribs['href']) self.our_links.append(attribs['href'])
if attribs.has_key('title'): if attribs.has_key('title'):
tags.append('(' + attribs['title'] + ')') tags.append('(' + attribs['title'] + ')')
self.in_a_link = True self.in_a_link = True
else:
text.append('%')
tags.append('%')
elif tag == 'img': elif tag == 'img':
if self.opts.keep_image_references: if self.opts.keep_image_references:
txt = '!' + self.check_halign(style) txt = '!' + self.check_halign(style)
@ -432,9 +449,9 @@ class TextileMLizer(OEB2HTML):
t = '' t = ''
text.append(self.id_no_text) text.append(self.id_no_text)
self.id_no_text = u'' self.id_no_text = u''
if t == '*]': if t in ('*]', '*'):
self.style_bold = False self.style_bold = False
elif t == '_]': elif t in ('_]', '_'):
self.style_italic = False self.style_italic = False
elif t == '+]': elif t == '+]':
self.style_under = False self.style_under = False
@ -442,7 +459,7 @@ class TextileMLizer(OEB2HTML):
self.style_strike = False self.style_strike = False
elif t == '&': elif t == '&':
self.style_smallcap = False self.style_smallcap = False
if t in ('*]', '_]', '+]', '-]'): if t in ('*]', '_]', '+]', '-]', '*', '_'):
txt = self.style_embed.pop() txt = self.style_embed.pop()
text.append('%s' % t) text.append('%s' % t)