You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
bazarr/libs/markdown/extensions/smarty.py

269 lines
10 KiB

This file contains ambiguous Unicode characters!

This file contains ambiguous Unicode characters that may be confused with others in your current locale. If your use case is intentional and legitimate, you can safely ignore this warning. Use the Escape button to highlight these characters.

# -*- coding: utf-8 -*-
'''
Smarty extension for Python-Markdown
====================================
Adds conversion of ASCII dashes, quotes and ellipses to their HTML
entity equivalents.
See <https://Python-Markdown.github.io/extensions/smarty>
for documentation.
Author: 2013, Dmitry Shachnev <mitya57@gmail.com>
All changes Copyright 2013-2014 The Python Markdown Project
License: [BSD](http://www.opensource.org/licenses/bsd-license.php)
SmartyPants license:
Copyright (c) 2003 John Gruber <http://daringfireball.net/>
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in
the documentation and/or other materials provided with the
distribution.
* Neither the name "SmartyPants" nor the names of its contributors
may be used to endorse or promote products derived from this
software without specific prior written permission.
This software is provided by the copyright holders and contributors "as
is" and any express or implied warranties, including, but not limited
to, the implied warranties of merchantability and fitness for a
particular purpose are disclaimed. In no event shall the copyright
owner or contributors be liable for any direct, indirect, incidental,
special, exemplary, or consequential damages (including, but not
limited to, procurement of substitute goods or services; loss of use,
data, or profits; or business interruption) however caused and on any
theory of liability, whether in contract, strict liability, or tort
(including negligence or otherwise) arising in any way out of the use
of this software, even if advised of the possibility of such damage.
smartypants.py license:
smartypants.py is a derivative work of SmartyPants.
Copyright (c) 2004, 2007 Chad Miller <http://web.chad.org/>
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in
the documentation and/or other materials provided with the
distribution.
This software is provided by the copyright holders and contributors "as
is" and any express or implied warranties, including, but not limited
to, the implied warranties of merchantability and fitness for a
particular purpose are disclaimed. In no event shall the copyright
owner or contributors be liable for any direct, indirect, incidental,
special, exemplary, or consequential damages (including, but not
limited to, procurement of substitute goods or services; loss of use,
data, or profits; or business interruption) however caused and on any
theory of liability, whether in contract, strict liability, or tort
(including negligence or otherwise) arising in any way out of the use
of this software, even if advised of the possibility of such damage.
'''
from __future__ import unicode_literals
from . import Extension
from ..inlinepatterns import HtmlPattern, HTML_RE
from ..odict import OrderedDict
from ..treeprocessors import InlineProcessor
# Constants for quote education.
punctClass = r"""[!"#\$\%'()*+,-.\/:;<=>?\@\[\\\]\^_`{|}~]"""
endOfWordClass = r"[\s.,;:!?)]"
closeClass = r"[^\ \t\r\n\[\{\(\-\u0002\u0003]"
openingQuotesBase = (
r'(\s' # a whitespace char
r'|&nbsp;' # or a non-breaking space entity
r'|--' # or dashes
r'||—' # or unicode
r'|&[mn]dash;' # or named dash entities
r'|&#8211;|&#8212;' # or decimal entities
r')'
)
substitutions = {
'mdash': '&mdash;',
'ndash': '&ndash;',
'ellipsis': '&hellip;',
'left-angle-quote': '&laquo;',
'right-angle-quote': '&raquo;',
'left-single-quote': '&lsquo;',
'right-single-quote': '&rsquo;',
'left-double-quote': '&ldquo;',
'right-double-quote': '&rdquo;',
}
# Special case if the very first character is a quote
# followed by punctuation at a non-word-break. Close the quotes by brute force:
singleQuoteStartRe = r"^'(?=%s\B)" % punctClass
doubleQuoteStartRe = r'^"(?=%s\B)' % punctClass
# Special case for double sets of quotes, e.g.:
# <p>He said, "'Quoted' words in a larger quote."</p>
doubleQuoteSetsRe = r""""'(?=\w)"""
singleQuoteSetsRe = r"""'"(?=\w)"""
# Special case for decade abbreviations (the '80s):
decadeAbbrRe = r"(?<!\w)'(?=\d{2}s)"
# Get most opening double quotes:
openingDoubleQuotesRegex = r'%s"(?=\w)' % openingQuotesBase
# Double closing quotes:
closingDoubleQuotesRegex = r'"(?=\s)'
closingDoubleQuotesRegex2 = '(?<=%s)"' % closeClass
# Get most opening single quotes:
openingSingleQuotesRegex = r"%s'(?=\w)" % openingQuotesBase
# Single closing quotes:
closingSingleQuotesRegex = r"(?<=%s)'(?!\s|s\b|\d)" % closeClass
closingSingleQuotesRegex2 = r"(?<=%s)'(\s|s\b)" % closeClass
# All remaining quotes should be opening ones
remainingSingleQuotesRegex = r"'"
remainingDoubleQuotesRegex = r'"'
HTML_STRICT_RE = HTML_RE + r'(?!\>)'
class SubstituteTextPattern(HtmlPattern):
def __init__(self, pattern, replace, markdown_instance):
""" Replaces matches with some text. """
HtmlPattern.__init__(self, pattern)
self.replace = replace
self.markdown = markdown_instance
def handleMatch(self, m):
result = ''
for part in self.replace:
if isinstance(part, int):
result += m.group(part)
else:
result += self.markdown.htmlStash.store(part, safe=True)
return result
class SmartyExtension(Extension):
def __init__(self, *args, **kwargs):
self.config = {
'smart_quotes': [True, 'Educate quotes'],
'smart_angled_quotes': [False, 'Educate angled quotes'],
'smart_dashes': [True, 'Educate dashes'],
'smart_ellipses': [True, 'Educate ellipses'],
'substitutions': [{}, 'Overwrite default substitutions'],
}
super(SmartyExtension, self).__init__(*args, **kwargs)
self.substitutions = dict(substitutions)
self.substitutions.update(self.getConfig('substitutions', default={}))
def _addPatterns(self, md, patterns, serie):
for ind, pattern in enumerate(patterns):
pattern += (md,)
pattern = SubstituteTextPattern(*pattern)
after = ('>smarty-%s-%d' % (serie, ind - 1) if ind else '_begin')
name = 'smarty-%s-%d' % (serie, ind)
self.inlinePatterns.add(name, pattern, after)
def educateDashes(self, md):
emDashesPattern = SubstituteTextPattern(
r'(?<!-)---(?!-)', (self.substitutions['mdash'],), md
)
enDashesPattern = SubstituteTextPattern(
r'(?<!-)--(?!-)', (self.substitutions['ndash'],), md
)
self.inlinePatterns.add('smarty-em-dashes', emDashesPattern, '_begin')
self.inlinePatterns.add(
'smarty-en-dashes', enDashesPattern, '>smarty-em-dashes'
)
def educateEllipses(self, md):
ellipsesPattern = SubstituteTextPattern(
r'(?<!\.)\.{3}(?!\.)', (self.substitutions['ellipsis'],), md
)
self.inlinePatterns.add('smarty-ellipses', ellipsesPattern, '_begin')
def educateAngledQuotes(self, md):
leftAngledQuotePattern = SubstituteTextPattern(
r'\<\<', (self.substitutions['left-angle-quote'],), md
)
rightAngledQuotePattern = SubstituteTextPattern(
r'\>\>', (self.substitutions['right-angle-quote'],), md
)
self.inlinePatterns.add(
'smarty-left-angle-quotes', leftAngledQuotePattern, '_begin'
)
self.inlinePatterns.add(
'smarty-right-angle-quotes',
rightAngledQuotePattern,
'>smarty-left-angle-quotes'
)
def educateQuotes(self, md):
lsquo = self.substitutions['left-single-quote']
rsquo = self.substitutions['right-single-quote']
ldquo = self.substitutions['left-double-quote']
rdquo = self.substitutions['right-double-quote']
patterns = (
(singleQuoteStartRe, (rsquo,)),
(doubleQuoteStartRe, (rdquo,)),
(doubleQuoteSetsRe, (ldquo + lsquo,)),
(singleQuoteSetsRe, (lsquo + ldquo,)),
(decadeAbbrRe, (rsquo,)),
(openingSingleQuotesRegex, (2, lsquo)),
(closingSingleQuotesRegex, (rsquo,)),
(closingSingleQuotesRegex2, (rsquo, 2)),
(remainingSingleQuotesRegex, (lsquo,)),
(openingDoubleQuotesRegex, (2, ldquo)),
(closingDoubleQuotesRegex, (rdquo,)),
(closingDoubleQuotesRegex2, (rdquo,)),
(remainingDoubleQuotesRegex, (ldquo,))
)
self._addPatterns(md, patterns, 'quotes')
def extendMarkdown(self, md, md_globals):
configs = self.getConfigs()
self.inlinePatterns = OrderedDict()
if configs['smart_ellipses']:
self.educateEllipses(md)
if configs['smart_quotes']:
self.educateQuotes(md)
if configs['smart_angled_quotes']:
self.educateAngledQuotes(md)
# Override HTML_RE from inlinepatterns.py so that it does not
# process tags with duplicate closing quotes.
md.inlinePatterns["html"] = HtmlPattern(HTML_STRICT_RE, md)
if configs['smart_dashes']:
self.educateDashes(md)
inlineProcessor = InlineProcessor(md)
inlineProcessor.inlinePatterns = self.inlinePatterns
md.treeprocessors.add('smarty', inlineProcessor, '_end')
md.ESCAPED_CHARS.extend(['"', "'"])
def makeExtension(*args, **kwargs):
return SmartyExtension(*args, **kwargs)