|
0
|
1 #!/usr/bin/env python
|
|
|
2 # Copyright (c) 2012 Trent Mick.
|
|
|
3 # Copyright (c) 2007-2008 ActiveState Corp.
|
|
|
4 # License: MIT (http://www.opensource.org/licenses/mit-license.php)
|
|
|
5
|
|
|
6 from __future__ import generators
|
|
|
7
|
|
|
8 r"""A fast and complete Python implementation of Markdown.
|
|
|
9
|
|
|
10 [from http://daringfireball.net/projects/markdown/]
|
|
|
11 > Markdown is a text-to-HTML filter; it translates an easy-to-read /
|
|
|
12 > easy-to-write structured text format into HTML. Markdown's text
|
|
|
13 > format is most similar to that of plain text email, and supports
|
|
|
14 > features such as headers, *emphasis*, code blocks, blockquotes, and
|
|
|
15 > links.
|
|
|
16 >
|
|
|
17 > Markdown's syntax is designed not as a generic markup language, but
|
|
|
18 > specifically to serve as a front-end to (X)HTML. You can use span-level
|
|
|
19 > HTML tags anywhere in a Markdown document, and you can use block level
|
|
|
20 > HTML tags (like <div> and <table> as well).
|
|
|
21
|
|
|
22 Module usage:
|
|
|
23
|
|
|
24 >>> import markdown2
|
|
|
25 >>> markdown2.markdown("*boo!*") # or use `html = markdown_path(PATH)`
|
|
|
26 u'<p><em>boo!</em></p>\n'
|
|
|
27
|
|
|
28 >>> markdowner = Markdown()
|
|
|
29 >>> markdowner.convert("*boo!*")
|
|
|
30 u'<p><em>boo!</em></p>\n'
|
|
|
31 >>> markdowner.convert("**boom!**")
|
|
|
32 u'<p><strong>boom!</strong></p>\n'
|
|
|
33
|
|
|
34 This implementation of Markdown implements the full "core" syntax plus a
|
|
|
35 number of extras (e.g., code syntax coloring, footnotes) as described on
|
|
|
36 <https://github.com/trentm/python-markdown2/wiki/Extras>.
|
|
|
37 """
|
|
|
38
|
|
|
39 cmdln_desc = """A fast and complete Python implementation of Markdown, a
|
|
|
40 text-to-HTML conversion tool for web writers.
|
|
|
41
|
|
|
42 Supported extra syntax options (see -x|--extras option below and
|
|
|
43 see <https://github.com/trentm/python-markdown2/wiki/Extras> for details):
|
|
|
44
|
|
|
45 * code-friendly: Disable _ and __ for em and strong.
|
|
|
46 * cuddled-lists: Allow lists to be cuddled to the preceding paragraph.
|
|
|
47 * fenced-code-blocks: Allows a code block to not have to be indented
|
|
|
48 by fencing it with '```' on a line before and after. Based on
|
|
|
49 <http://github.github.com/github-flavored-markdown/> with support for
|
|
|
50 syntax highlighting.
|
|
|
51 * footnotes: Support footnotes as in use on daringfireball.net and
|
|
|
52 implemented in other Markdown processors (tho not in Markdown.pl v1.0.1).
|
|
|
53 * header-ids: Adds "id" attributes to headers. The id value is a slug of
|
|
|
54 the header text.
|
|
|
55 * html-classes: Takes a dict mapping html tag names (lowercase) to a
|
|
|
56 string to use for a "class" tag attribute. Currently only supports
|
|
|
57 "pre" and "code" tags. Add an issue if you require this for other tags.
|
|
|
58 * markdown-in-html: Allow the use of `markdown="1"` in a block HTML tag to
|
|
|
59 have markdown processing be done on its contents. Similar to
|
|
|
60 <http://michelf.com/projects/php-markdown/extra/#markdown-attr> but with
|
|
|
61 some limitations.
|
|
|
62 * metadata: Extract metadata from a leading '---'-fenced block.
|
|
|
63 See <https://github.com/trentm/python-markdown2/issues/77> for details.
|
|
|
64 * nofollow: Add `rel="nofollow"` to add `<a>` tags with an href. See
|
|
|
65 <http://en.wikipedia.org/wiki/Nofollow>.
|
|
|
66 * pyshell: Treats unindented Python interactive shell sessions as <code>
|
|
|
67 blocks.
|
|
|
68 * link-patterns: Auto-link given regex patterns in text (e.g. bug number
|
|
|
69 references, revision number references).
|
|
|
70 * smarty-pants: Replaces ' and " with curly quotation marks or curly
|
|
|
71 apostrophes. Replaces --, ---, ..., and . . . with en dashes, em dashes,
|
|
|
72 and ellipses.
|
|
|
73 * toc: The returned HTML string gets a new "toc_html" attribute which is
|
|
|
74 a Table of Contents for the document. (experimental)
|
|
|
75 * xml: Passes one-liner processing instructions and namespaced XML tags.
|
|
|
76 * tables: Tables using the same format as GFM
|
|
|
77 <https://help.github.com/articles/github-flavored-markdown#tables> and
|
|
|
78 PHP-Markdown Extra <https://michelf.ca/projects/php-markdown/extra/#table>.
|
|
|
79 * wiki-tables: Google Code Wiki-style tables. See
|
|
|
80 <http://code.google.com/p/support/wiki/WikiSyntax#Tables>.
|
|
|
81 """
|
|
|
82
|
|
|
83 # Dev Notes:
|
|
|
84 # - Python's regex syntax doesn't have '\z', so I'm using '\Z'. I'm
|
|
|
85 # not yet sure if there implications with this. Compare 'pydoc sre'
|
|
|
86 # and 'perldoc perlre'.
|
|
|
87
|
|
|
88 __version_info__ = (2, 3, 1)
|
|
|
89 __version__ = '.'.join(map(str, __version_info__))
|
|
|
90 __author__ = "Trent Mick"
|
|
|
91
|
|
|
92 import os
|
|
|
93 import sys
|
|
|
94 from pprint import pprint, pformat
|
|
|
95 import re
|
|
|
96 import logging
|
|
|
97 try:
|
|
|
98 from hashlib import md5
|
|
|
99 except ImportError:
|
|
|
100 from md5 import md5
|
|
|
101 import optparse
|
|
|
102 from random import random, randint
|
|
|
103 import codecs
|
|
|
104
|
|
|
105
|
|
|
106 #---- Python version compat
|
|
|
107
|
|
|
108 try:
|
|
|
109 from urllib.parse import quote # python3
|
|
|
110 except ImportError:
|
|
|
111 from urllib import quote # python2
|
|
|
112
|
|
|
113 if sys.version_info[:2] < (2,4):
|
|
|
114 from sets import Set as set
|
|
|
115 def reversed(sequence):
|
|
|
116 for i in sequence[::-1]:
|
|
|
117 yield i
|
|
|
118
|
|
|
119 # Use `bytes` for byte strings and `unicode` for unicode strings (str in Py3).
|
|
|
120 if sys.version_info[0] <= 2:
|
|
|
121 py3 = False
|
|
|
122 try:
|
|
|
123 bytes
|
|
|
124 except NameError:
|
|
|
125 bytes = str
|
|
|
126 base_string_type = basestring
|
|
|
127 elif sys.version_info[0] >= 3:
|
|
|
128 py3 = True
|
|
|
129 unicode = str
|
|
|
130 base_string_type = str
|
|
|
131
|
|
|
132
|
|
|
133
|
|
|
134 #---- globals
|
|
|
135
|
|
|
136 DEBUG = False
|
|
|
137 log = logging.getLogger("markdown")
|
|
|
138
|
|
|
139 DEFAULT_TAB_WIDTH = 4
|
|
|
140
|
|
|
141
|
|
|
142 SECRET_SALT = bytes(randint(0, 1000000))
|
|
|
143 def _hash_text(s):
|
|
|
144 return 'md5-' + md5(SECRET_SALT + s.encode("utf-8")).hexdigest()
|
|
|
145
|
|
|
146 # Table of hash values for escaped characters:
|
|
|
147 g_escape_table = dict([(ch, _hash_text(ch))
|
|
|
148 for ch in '\\`*_{}[]()>#+-.!'])
|
|
|
149
|
|
|
150
|
|
|
151
|
|
|
152 #---- exceptions
|
|
|
153
|
|
|
154 class MarkdownError(Exception):
|
|
|
155 pass
|
|
|
156
|
|
|
157
|
|
|
158
|
|
|
159 #---- public api
|
|
|
160
|
|
|
161 def markdown_path(path, encoding="utf-8",
|
|
|
162 html4tags=False, tab_width=DEFAULT_TAB_WIDTH,
|
|
|
163 safe_mode=None, extras=None, link_patterns=None,
|
|
|
164 use_file_vars=False):
|
|
|
165 fp = codecs.open(path, 'r', encoding)
|
|
|
166 text = fp.read()
|
|
|
167 fp.close()
|
|
|
168 return Markdown(html4tags=html4tags, tab_width=tab_width,
|
|
|
169 safe_mode=safe_mode, extras=extras,
|
|
|
170 link_patterns=link_patterns,
|
|
|
171 use_file_vars=use_file_vars).convert(text)
|
|
|
172
|
|
|
173 def markdown(text, html4tags=False, tab_width=DEFAULT_TAB_WIDTH,
|
|
|
174 safe_mode=None, extras=None, link_patterns=None,
|
|
|
175 use_file_vars=False):
|
|
|
176 return Markdown(html4tags=html4tags, tab_width=tab_width,
|
|
|
177 safe_mode=safe_mode, extras=extras,
|
|
|
178 link_patterns=link_patterns,
|
|
|
179 use_file_vars=use_file_vars).convert(text)
|
|
|
180
|
|
|
181 class Markdown(object):
|
|
|
182 # The dict of "extras" to enable in processing -- a mapping of
|
|
|
183 # extra name to argument for the extra. Most extras do not have an
|
|
|
184 # argument, in which case the value is None.
|
|
|
185 #
|
|
|
186 # This can be set via (a) subclassing and (b) the constructor
|
|
|
187 # "extras" argument.
|
|
|
188 extras = None
|
|
|
189
|
|
|
190 urls = None
|
|
|
191 titles = None
|
|
|
192 html_blocks = None
|
|
|
193 html_spans = None
|
|
|
194 html_removed_text = "[HTML_REMOVED]" # for compat with markdown.py
|
|
|
195
|
|
|
196 # Used to track when we're inside an ordered or unordered list
|
|
|
197 # (see _ProcessListItems() for details):
|
|
|
198 list_level = 0
|
|
|
199
|
|
|
200 _ws_only_line_re = re.compile(r"^[ \t]+$", re.M)
|
|
|
201
|
|
|
202 def __init__(self, html4tags=False, tab_width=4, safe_mode=None,
|
|
|
203 extras=None, link_patterns=None, use_file_vars=False):
|
|
|
204 if html4tags:
|
|
|
205 self.empty_element_suffix = ">"
|
|
|
206 else:
|
|
|
207 self.empty_element_suffix = " />"
|
|
|
208 self.tab_width = tab_width
|
|
|
209
|
|
|
210 # For compatibility with earlier markdown2.py and with
|
|
|
211 # markdown.py's safe_mode being a boolean,
|
|
|
212 # safe_mode == True -> "replace"
|
|
|
213 if safe_mode is True:
|
|
|
214 self.safe_mode = "replace"
|
|
|
215 else:
|
|
|
216 self.safe_mode = safe_mode
|
|
|
217
|
|
|
218 # Massaging and building the "extras" info.
|
|
|
219 if self.extras is None:
|
|
|
220 self.extras = {}
|
|
|
221 elif not isinstance(self.extras, dict):
|
|
|
222 self.extras = dict([(e, None) for e in self.extras])
|
|
|
223 if extras:
|
|
|
224 if not isinstance(extras, dict):
|
|
|
225 extras = dict([(e, None) for e in extras])
|
|
|
226 self.extras.update(extras)
|
|
|
227 assert isinstance(self.extras, dict)
|
|
|
228 if "toc" in self.extras and not "header-ids" in self.extras:
|
|
|
229 self.extras["header-ids"] = None # "toc" implies "header-ids"
|
|
|
230 self._instance_extras = self.extras.copy()
|
|
|
231
|
|
|
232 self.link_patterns = link_patterns
|
|
|
233 self.use_file_vars = use_file_vars
|
|
|
234 self._outdent_re = re.compile(r'^(\t|[ ]{1,%d})' % tab_width, re.M)
|
|
|
235
|
|
|
236 self._escape_table = g_escape_table.copy()
|
|
|
237 if "smarty-pants" in self.extras:
|
|
|
238 self._escape_table['"'] = _hash_text('"')
|
|
|
239 self._escape_table["'"] = _hash_text("'")
|
|
|
240
|
|
|
241 def reset(self):
|
|
|
242 self.urls = {}
|
|
|
243 self.titles = {}
|
|
|
244 self.html_blocks = {}
|
|
|
245 self.html_spans = {}
|
|
|
246 self.list_level = 0
|
|
|
247 self.extras = self._instance_extras.copy()
|
|
|
248 if "footnotes" in self.extras:
|
|
|
249 self.footnotes = {}
|
|
|
250 self.footnote_ids = []
|
|
|
251 if "header-ids" in self.extras:
|
|
|
252 self._count_from_header_id = {} # no `defaultdict` in Python 2.4
|
|
|
253 if "metadata" in self.extras:
|
|
|
254 self.metadata = {}
|
|
|
255
|
|
|
256 # Per <https://developer.mozilla.org/en-US/docs/HTML/Element/a> "rel"
|
|
|
257 # should only be used in <a> tags with an "href" attribute.
|
|
|
258 _a_nofollow = re.compile(r"<(a)([^>]*href=)", re.IGNORECASE)
|
|
|
259
|
|
|
260 def convert(self, text):
|
|
|
261 """Convert the given text."""
|
|
|
262 # Main function. The order in which other subs are called here is
|
|
|
263 # essential. Link and image substitutions need to happen before
|
|
|
264 # _EscapeSpecialChars(), so that any *'s or _'s in the <a>
|
|
|
265 # and <img> tags get encoded.
|
|
|
266
|
|
|
267 # Clear the global hashes. If we don't clear these, you get conflicts
|
|
|
268 # from other articles when generating a page which contains more than
|
|
|
269 # one article (e.g. an index page that shows the N most recent
|
|
|
270 # articles):
|
|
|
271 self.reset()
|
|
|
272
|
|
|
273 if not isinstance(text, unicode):
|
|
|
274 #TODO: perhaps shouldn't presume UTF-8 for string input?
|
|
|
275 text = unicode(text, 'utf-8')
|
|
|
276
|
|
|
277 if self.use_file_vars:
|
|
|
278 # Look for emacs-style file variable hints.
|
|
|
279 emacs_vars = self._get_emacs_vars(text)
|
|
|
280 if "markdown-extras" in emacs_vars:
|
|
|
281 splitter = re.compile("[ ,]+")
|
|
|
282 for e in splitter.split(emacs_vars["markdown-extras"]):
|
|
|
283 if '=' in e:
|
|
|
284 ename, earg = e.split('=', 1)
|
|
|
285 try:
|
|
|
286 earg = int(earg)
|
|
|
287 except ValueError:
|
|
|
288 pass
|
|
|
289 else:
|
|
|
290 ename, earg = e, None
|
|
|
291 self.extras[ename] = earg
|
|
|
292
|
|
|
293 # Standardize line endings:
|
|
|
294 text = re.sub("\r\n|\r", "\n", text)
|
|
|
295
|
|
|
296 # Make sure $text ends with a couple of newlines:
|
|
|
297 text += "\n\n"
|
|
|
298
|
|
|
299 # Convert all tabs to spaces.
|
|
|
300 text = self._detab(text)
|
|
|
301
|
|
|
302 # Strip any lines consisting only of spaces and tabs.
|
|
|
303 # This makes subsequent regexen easier to write, because we can
|
|
|
304 # match consecutive blank lines with /\n+/ instead of something
|
|
|
305 # contorted like /[ \t]*\n+/ .
|
|
|
306 text = self._ws_only_line_re.sub("", text)
|
|
|
307
|
|
|
308 # strip metadata from head and extract
|
|
|
309 if "metadata" in self.extras:
|
|
|
310 text = self._extract_metadata(text)
|
|
|
311
|
|
|
312 text = self.preprocess(text)
|
|
|
313
|
|
|
314 if "fenced-code-blocks" in self.extras and not self.safe_mode:
|
|
|
315 text = self._do_fenced_code_blocks(text)
|
|
|
316
|
|
|
317 if self.safe_mode:
|
|
|
318 text = self._hash_html_spans(text)
|
|
|
319
|
|
|
320 # Turn block-level HTML blocks into hash entries
|
|
|
321 text = self._hash_html_blocks(text, raw=True)
|
|
|
322
|
|
|
323 if "fenced-code-blocks" in self.extras and self.safe_mode:
|
|
|
324 text = self._do_fenced_code_blocks(text)
|
|
|
325
|
|
|
326 # Strip link definitions, store in hashes.
|
|
|
327 if "footnotes" in self.extras:
|
|
|
328 # Must do footnotes first because an unlucky footnote defn
|
|
|
329 # looks like a link defn:
|
|
|
330 # [^4]: this "looks like a link defn"
|
|
|
331 text = self._strip_footnote_definitions(text)
|
|
|
332 text = self._strip_link_definitions(text)
|
|
|
333
|
|
|
334 text = self._run_block_gamut(text)
|
|
|
335
|
|
|
336 if "footnotes" in self.extras:
|
|
|
337 text = self._add_footnotes(text)
|
|
|
338
|
|
|
339 text = self.postprocess(text)
|
|
|
340
|
|
|
341 text = self._unescape_special_chars(text)
|
|
|
342
|
|
|
343 if self.safe_mode:
|
|
|
344 text = self._unhash_html_spans(text)
|
|
|
345
|
|
|
346 if "nofollow" in self.extras:
|
|
|
347 text = self._a_nofollow.sub(r'<\1 rel="nofollow"\2', text)
|
|
|
348
|
|
|
349 text += "\n"
|
|
|
350
|
|
|
351 rv = UnicodeWithAttrs(text)
|
|
|
352 if "toc" in self.extras:
|
|
|
353 rv._toc = self._toc
|
|
|
354 if "metadata" in self.extras:
|
|
|
355 rv.metadata = self.metadata
|
|
|
356 return rv
|
|
|
357
|
|
|
358 def postprocess(self, text):
|
|
|
359 """A hook for subclasses to do some postprocessing of the html, if
|
|
|
360 desired. This is called before unescaping of special chars and
|
|
|
361 unhashing of raw HTML spans.
|
|
|
362 """
|
|
|
363 return text
|
|
|
364
|
|
|
365 def preprocess(self, text):
|
|
|
366 """A hook for subclasses to do some preprocessing of the Markdown, if
|
|
|
367 desired. This is called after basic formatting of the text, but prior
|
|
|
368 to any extras, safe mode, etc. processing.
|
|
|
369 """
|
|
|
370 return text
|
|
|
371
|
|
|
372 # Is metadata if the content starts with '---'-fenced `key: value`
|
|
|
373 # pairs. E.g. (indented for presentation):
|
|
|
374 # ---
|
|
|
375 # foo: bar
|
|
|
376 # another-var: blah blah
|
|
|
377 # ---
|
|
|
378 _metadata_pat = re.compile("""^---[ \t]*\n((?:[ \t]*[^ \t:]+[ \t]*:[^\n]*\n)+)---[ \t]*\n""")
|
|
|
379
|
|
|
380 def _extract_metadata(self, text):
|
|
|
381 # fast test
|
|
|
382 if not text.startswith("---"):
|
|
|
383 return text
|
|
|
384 match = self._metadata_pat.match(text)
|
|
|
385 if not match:
|
|
|
386 return text
|
|
|
387
|
|
|
388 tail = text[len(match.group(0)):]
|
|
|
389 metadata_str = match.group(1).strip()
|
|
|
390 for line in metadata_str.split('\n'):
|
|
|
391 key, value = line.split(':', 1)
|
|
|
392 self.metadata[key.strip()] = value.strip()
|
|
|
393
|
|
|
394 return tail
|
|
|
395
|
|
|
396
|
|
|
397 _emacs_oneliner_vars_pat = re.compile(r"-\*-\s*([^\r\n]*?)\s*-\*-", re.UNICODE)
|
|
|
398 # This regular expression is intended to match blocks like this:
|
|
|
399 # PREFIX Local Variables: SUFFIX
|
|
|
400 # PREFIX mode: Tcl SUFFIX
|
|
|
401 # PREFIX End: SUFFIX
|
|
|
402 # Some notes:
|
|
|
403 # - "[ \t]" is used instead of "\s" to specifically exclude newlines
|
|
|
404 # - "(\r\n|\n|\r)" is used instead of "$" because the sre engine does
|
|
|
405 # not like anything other than Unix-style line terminators.
|
|
|
406 _emacs_local_vars_pat = re.compile(r"""^
|
|
|
407 (?P<prefix>(?:[^\r\n|\n|\r])*?)
|
|
|
408 [\ \t]*Local\ Variables:[\ \t]*
|
|
|
409 (?P<suffix>.*?)(?:\r\n|\n|\r)
|
|
|
410 (?P<content>.*?\1End:)
|
|
|
411 """, re.IGNORECASE | re.MULTILINE | re.DOTALL | re.VERBOSE)
|
|
|
412
|
|
|
413 def _get_emacs_vars(self, text):
|
|
|
414 """Return a dictionary of emacs-style local variables.
|
|
|
415
|
|
|
416 Parsing is done loosely according to this spec (and according to
|
|
|
417 some in-practice deviations from this):
|
|
|
418 http://www.gnu.org/software/emacs/manual/html_node/emacs/Specifying-File-Variables.html#Specifying-File-Variables
|
|
|
419 """
|
|
|
420 emacs_vars = {}
|
|
|
421 SIZE = pow(2, 13) # 8kB
|
|
|
422
|
|
|
423 # Search near the start for a '-*-'-style one-liner of variables.
|
|
|
424 head = text[:SIZE]
|
|
|
425 if "-*-" in head:
|
|
|
426 match = self._emacs_oneliner_vars_pat.search(head)
|
|
|
427 if match:
|
|
|
428 emacs_vars_str = match.group(1)
|
|
|
429 assert '\n' not in emacs_vars_str
|
|
|
430 emacs_var_strs = [s.strip() for s in emacs_vars_str.split(';')
|
|
|
431 if s.strip()]
|
|
|
432 if len(emacs_var_strs) == 1 and ':' not in emacs_var_strs[0]:
|
|
|
433 # While not in the spec, this form is allowed by emacs:
|
|
|
434 # -*- Tcl -*-
|
|
|
435 # where the implied "variable" is "mode". This form
|
|
|
436 # is only allowed if there are no other variables.
|
|
|
437 emacs_vars["mode"] = emacs_var_strs[0].strip()
|
|
|
438 else:
|
|
|
439 for emacs_var_str in emacs_var_strs:
|
|
|
440 try:
|
|
|
441 variable, value = emacs_var_str.strip().split(':', 1)
|
|
|
442 except ValueError:
|
|
|
443 log.debug("emacs variables error: malformed -*- "
|
|
|
444 "line: %r", emacs_var_str)
|
|
|
445 continue
|
|
|
446 # Lowercase the variable name because Emacs allows "Mode"
|
|
|
447 # or "mode" or "MoDe", etc.
|
|
|
448 emacs_vars[variable.lower()] = value.strip()
|
|
|
449
|
|
|
450 tail = text[-SIZE:]
|
|
|
451 if "Local Variables" in tail:
|
|
|
452 match = self._emacs_local_vars_pat.search(tail)
|
|
|
453 if match:
|
|
|
454 prefix = match.group("prefix")
|
|
|
455 suffix = match.group("suffix")
|
|
|
456 lines = match.group("content").splitlines(0)
|
|
|
457 #print "prefix=%r, suffix=%r, content=%r, lines: %s"\
|
|
|
458 # % (prefix, suffix, match.group("content"), lines)
|
|
|
459
|
|
|
460 # Validate the Local Variables block: proper prefix and suffix
|
|
|
461 # usage.
|
|
|
462 for i, line in enumerate(lines):
|
|
|
463 if not line.startswith(prefix):
|
|
|
464 log.debug("emacs variables error: line '%s' "
|
|
|
465 "does not use proper prefix '%s'"
|
|
|
466 % (line, prefix))
|
|
|
467 return {}
|
|
|
468 # Don't validate suffix on last line. Emacs doesn't care,
|
|
|
469 # neither should we.
|
|
|
470 if i != len(lines)-1 and not line.endswith(suffix):
|
|
|
471 log.debug("emacs variables error: line '%s' "
|
|
|
472 "does not use proper suffix '%s'"
|
|
|
473 % (line, suffix))
|
|
|
474 return {}
|
|
|
475
|
|
|
476 # Parse out one emacs var per line.
|
|
|
477 continued_for = None
|
|
|
478 for line in lines[:-1]: # no var on the last line ("PREFIX End:")
|
|
|
479 if prefix: line = line[len(prefix):] # strip prefix
|
|
|
480 if suffix: line = line[:-len(suffix)] # strip suffix
|
|
|
481 line = line.strip()
|
|
|
482 if continued_for:
|
|
|
483 variable = continued_for
|
|
|
484 if line.endswith('\\'):
|
|
|
485 line = line[:-1].rstrip()
|
|
|
486 else:
|
|
|
487 continued_for = None
|
|
|
488 emacs_vars[variable] += ' ' + line
|
|
|
489 else:
|
|
|
490 try:
|
|
|
491 variable, value = line.split(':', 1)
|
|
|
492 except ValueError:
|
|
|
493 log.debug("local variables error: missing colon "
|
|
|
494 "in local variables entry: '%s'" % line)
|
|
|
495 continue
|
|
|
496 # Do NOT lowercase the variable name, because Emacs only
|
|
|
497 # allows "mode" (and not "Mode", "MoDe", etc.) in this block.
|
|
|
498 value = value.strip()
|
|
|
499 if value.endswith('\\'):
|
|
|
500 value = value[:-1].rstrip()
|
|
|
501 continued_for = variable
|
|
|
502 else:
|
|
|
503 continued_for = None
|
|
|
504 emacs_vars[variable] = value
|
|
|
505
|
|
|
506 # Unquote values.
|
|
|
507 for var, val in list(emacs_vars.items()):
|
|
|
508 if len(val) > 1 and (val.startswith('"') and val.endswith('"')
|
|
|
509 or val.startswith('"') and val.endswith('"')):
|
|
|
510 emacs_vars[var] = val[1:-1]
|
|
|
511
|
|
|
512 return emacs_vars
|
|
|
513
|
|
|
514 # Cribbed from a post by Bart Lateur:
|
|
|
515 # <http://www.nntp.perl.org/group/perl.macperl.anyperl/154>
|
|
|
516 _detab_re = re.compile(r'(.*?)\t', re.M)
|
|
|
517 def _detab_sub(self, match):
|
|
|
518 g1 = match.group(1)
|
|
|
519 return g1 + (' ' * (self.tab_width - len(g1) % self.tab_width))
|
|
|
520 def _detab(self, text):
|
|
|
521 r"""Remove (leading?) tabs from a file.
|
|
|
522
|
|
|
523 >>> m = Markdown()
|
|
|
524 >>> m._detab("\tfoo")
|
|
|
525 ' foo'
|
|
|
526 >>> m._detab(" \tfoo")
|
|
|
527 ' foo'
|
|
|
528 >>> m._detab("\t foo")
|
|
|
529 ' foo'
|
|
|
530 >>> m._detab(" foo")
|
|
|
531 ' foo'
|
|
|
532 >>> m._detab(" foo\n\tbar\tblam")
|
|
|
533 ' foo\n bar blam'
|
|
|
534 """
|
|
|
535 if '\t' not in text:
|
|
|
536 return text
|
|
|
537 return self._detab_re.subn(self._detab_sub, text)[0]
|
|
|
538
|
|
|
539 # I broke out the html5 tags here and add them to _block_tags_a and
|
|
|
540 # _block_tags_b. This way html5 tags are easy to keep track of.
|
|
|
541 _html5tags = '|article|aside|header|hgroup|footer|nav|section|figure|figcaption'
|
|
|
542
|
|
|
543 _block_tags_a = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|script|noscript|form|fieldset|iframe|math|ins|del'
|
|
|
544 _block_tags_a += _html5tags
|
|
|
545
|
|
|
546 _strict_tag_block_re = re.compile(r"""
|
|
|
547 ( # save in \1
|
|
|
548 ^ # start of line (with re.M)
|
|
|
549 <(%s) # start tag = \2
|
|
|
550 \b # word break
|
|
|
551 (.*\n)*? # any number of lines, minimally matching
|
|
|
552 </\2> # the matching end tag
|
|
|
553 [ \t]* # trailing spaces/tabs
|
|
|
554 (?=\n+|\Z) # followed by a newline or end of document
|
|
|
555 )
|
|
|
556 """ % _block_tags_a,
|
|
|
557 re.X | re.M)
|
|
|
558
|
|
|
559 _block_tags_b = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|script|noscript|form|fieldset|iframe|math'
|
|
|
560 _block_tags_b += _html5tags
|
|
|
561
|
|
|
562 _liberal_tag_block_re = re.compile(r"""
|
|
|
563 ( # save in \1
|
|
|
564 ^ # start of line (with re.M)
|
|
|
565 <(%s) # start tag = \2
|
|
|
566 \b # word break
|
|
|
567 (.*\n)*? # any number of lines, minimally matching
|
|
|
568 .*</\2> # the matching end tag
|
|
|
569 [ \t]* # trailing spaces/tabs
|
|
|
570 (?=\n+|\Z) # followed by a newline or end of document
|
|
|
571 )
|
|
|
572 """ % _block_tags_b,
|
|
|
573 re.X | re.M)
|
|
|
574
|
|
|
575 _html_markdown_attr_re = re.compile(
|
|
|
576 r'''\s+markdown=("1"|'1')''')
|
|
|
577 def _hash_html_block_sub(self, match, raw=False):
|
|
|
578 html = match.group(1)
|
|
|
579 if raw and self.safe_mode:
|
|
|
580 html = self._sanitize_html(html)
|
|
|
581 elif 'markdown-in-html' in self.extras and 'markdown=' in html:
|
|
|
582 first_line = html.split('\n', 1)[0]
|
|
|
583 m = self._html_markdown_attr_re.search(first_line)
|
|
|
584 if m:
|
|
|
585 lines = html.split('\n')
|
|
|
586 middle = '\n'.join(lines[1:-1])
|
|
|
587 last_line = lines[-1]
|
|
|
588 first_line = first_line[:m.start()] + first_line[m.end():]
|
|
|
589 f_key = _hash_text(first_line)
|
|
|
590 self.html_blocks[f_key] = first_line
|
|
|
591 l_key = _hash_text(last_line)
|
|
|
592 self.html_blocks[l_key] = last_line
|
|
|
593 return ''.join(["\n\n", f_key,
|
|
|
594 "\n\n", middle, "\n\n",
|
|
|
595 l_key, "\n\n"])
|
|
|
596 key = _hash_text(html)
|
|
|
597 self.html_blocks[key] = html
|
|
|
598 return "\n\n" + key + "\n\n"
|
|
|
599
|
|
|
600 def _hash_html_blocks(self, text, raw=False):
|
|
|
601 """Hashify HTML blocks
|
|
|
602
|
|
|
603 We only want to do this for block-level HTML tags, such as headers,
|
|
|
604 lists, and tables. That's because we still want to wrap <p>s around
|
|
|
605 "paragraphs" that are wrapped in non-block-level tags, such as anchors,
|
|
|
606 phrase emphasis, and spans. The list of tags we're looking for is
|
|
|
607 hard-coded.
|
|
|
608
|
|
|
609 @param raw {boolean} indicates if these are raw HTML blocks in
|
|
|
610 the original source. It makes a difference in "safe" mode.
|
|
|
611 """
|
|
|
612 if '<' not in text:
|
|
|
613 return text
|
|
|
614
|
|
|
615 # Pass `raw` value into our calls to self._hash_html_block_sub.
|
|
|
616 hash_html_block_sub = _curry(self._hash_html_block_sub, raw=raw)
|
|
|
617
|
|
|
618 # First, look for nested blocks, e.g.:
|
|
|
619 # <div>
|
|
|
620 # <div>
|
|
|
621 # tags for inner block must be indented.
|
|
|
622 # </div>
|
|
|
623 # </div>
|
|
|
624 #
|
|
|
625 # The outermost tags must start at the left margin for this to match, and
|
|
|
626 # the inner nested divs must be indented.
|
|
|
627 # We need to do this before the next, more liberal match, because the next
|
|
|
628 # match will start at the first `<div>` and stop at the first `</div>`.
|
|
|
629 text = self._strict_tag_block_re.sub(hash_html_block_sub, text)
|
|
|
630
|
|
|
631 # Now match more liberally, simply from `\n<tag>` to `</tag>\n`
|
|
|
632 text = self._liberal_tag_block_re.sub(hash_html_block_sub, text)
|
|
|
633
|
|
|
634 # Special case just for <hr />. It was easier to make a special
|
|
|
635 # case than to make the other regex more complicated.
|
|
|
636 if "<hr" in text:
|
|
|
637 _hr_tag_re = _hr_tag_re_from_tab_width(self.tab_width)
|
|
|
638 text = _hr_tag_re.sub(hash_html_block_sub, text)
|
|
|
639
|
|
|
640 # Special case for standalone HTML comments:
|
|
|
641 if "<!--" in text:
|
|
|
642 start = 0
|
|
|
643 while True:
|
|
|
644 # Delimiters for next comment block.
|
|
|
645 try:
|
|
|
646 start_idx = text.index("<!--", start)
|
|
|
647 except ValueError:
|
|
|
648 break
|
|
|
649 try:
|
|
|
650 end_idx = text.index("-->", start_idx) + 3
|
|
|
651 except ValueError:
|
|
|
652 break
|
|
|
653
|
|
|
654 # Start position for next comment block search.
|
|
|
655 start = end_idx
|
|
|
656
|
|
|
657 # Validate whitespace before comment.
|
|
|
658 if start_idx:
|
|
|
659 # - Up to `tab_width - 1` spaces before start_idx.
|
|
|
660 for i in range(self.tab_width - 1):
|
|
|
661 if text[start_idx - 1] != ' ':
|
|
|
662 break
|
|
|
663 start_idx -= 1
|
|
|
664 if start_idx == 0:
|
|
|
665 break
|
|
|
666 # - Must be preceded by 2 newlines or hit the start of
|
|
|
667 # the document.
|
|
|
668 if start_idx == 0:
|
|
|
669 pass
|
|
|
670 elif start_idx == 1 and text[0] == '\n':
|
|
|
671 start_idx = 0 # to match minute detail of Markdown.pl regex
|
|
|
672 elif text[start_idx-2:start_idx] == '\n\n':
|
|
|
673 pass
|
|
|
674 else:
|
|
|
675 break
|
|
|
676
|
|
|
677 # Validate whitespace after comment.
|
|
|
678 # - Any number of spaces and tabs.
|
|
|
679 while end_idx < len(text):
|
|
|
680 if text[end_idx] not in ' \t':
|
|
|
681 break
|
|
|
682 end_idx += 1
|
|
|
683 # - Must be following by 2 newlines or hit end of text.
|
|
|
684 if text[end_idx:end_idx+2] not in ('', '\n', '\n\n'):
|
|
|
685 continue
|
|
|
686
|
|
|
687 # Escape and hash (must match `_hash_html_block_sub`).
|
|
|
688 html = text[start_idx:end_idx]
|
|
|
689 if raw and self.safe_mode:
|
|
|
690 html = self._sanitize_html(html)
|
|
|
691 key = _hash_text(html)
|
|
|
692 self.html_blocks[key] = html
|
|
|
693 text = text[:start_idx] + "\n\n" + key + "\n\n" + text[end_idx:]
|
|
|
694
|
|
|
695 if "xml" in self.extras:
|
|
|
696 # Treat XML processing instructions and namespaced one-liner
|
|
|
697 # tags as if they were block HTML tags. E.g., if standalone
|
|
|
698 # (i.e. are their own paragraph), the following do not get
|
|
|
699 # wrapped in a <p> tag:
|
|
|
700 # <?foo bar?>
|
|
|
701 #
|
|
|
702 # <xi:include xmlns:xi="http://www.w3.org/2001/XInclude" href="chapter_1.md"/>
|
|
|
703 _xml_oneliner_re = _xml_oneliner_re_from_tab_width(self.tab_width)
|
|
|
704 text = _xml_oneliner_re.sub(hash_html_block_sub, text)
|
|
|
705
|
|
|
706 return text
|
|
|
707
|
|
|
708 def _strip_link_definitions(self, text):
|
|
|
709 # Strips link definitions from text, stores the URLs and titles in
|
|
|
710 # hash references.
|
|
|
711 less_than_tab = self.tab_width - 1
|
|
|
712
|
|
|
713 # Link defs are in the form:
|
|
|
714 # [id]: url "optional title"
|
|
|
715 _link_def_re = re.compile(r"""
|
|
|
716 ^[ ]{0,%d}\[(.+)\]: # id = \1
|
|
|
717 [ \t]*
|
|
|
718 \n? # maybe *one* newline
|
|
|
719 [ \t]*
|
|
|
720 <?(.+?)>? # url = \2
|
|
|
721 [ \t]*
|
|
|
722 (?:
|
|
|
723 \n? # maybe one newline
|
|
|
724 [ \t]*
|
|
|
725 (?<=\s) # lookbehind for whitespace
|
|
|
726 ['"(]
|
|
|
727 ([^\n]*) # title = \3
|
|
|
728 ['")]
|
|
|
729 [ \t]*
|
|
|
730 )? # title is optional
|
|
|
731 (?:\n+|\Z)
|
|
|
732 """ % less_than_tab, re.X | re.M | re.U)
|
|
|
733 return _link_def_re.sub(self._extract_link_def_sub, text)
|
|
|
734
|
|
|
735 def _extract_link_def_sub(self, match):
|
|
|
736 id, url, title = match.groups()
|
|
|
737 key = id.lower() # Link IDs are case-insensitive
|
|
|
738 self.urls[key] = self._encode_amps_and_angles(url)
|
|
|
739 if title:
|
|
|
740 self.titles[key] = title
|
|
|
741 return ""
|
|
|
742
|
|
|
743 def _extract_footnote_def_sub(self, match):
|
|
|
744 id, text = match.groups()
|
|
|
745 text = _dedent(text, skip_first_line=not text.startswith('\n')).strip()
|
|
|
746 normed_id = re.sub(r'\W', '-', id)
|
|
|
747 # Ensure footnote text ends with a couple newlines (for some
|
|
|
748 # block gamut matches).
|
|
|
749 self.footnotes[normed_id] = text + "\n\n"
|
|
|
750 return ""
|
|
|
751
|
|
|
752 def _strip_footnote_definitions(self, text):
|
|
|
753 """A footnote definition looks like this:
|
|
|
754
|
|
|
755 [^note-id]: Text of the note.
|
|
|
756
|
|
|
757 May include one or more indented paragraphs.
|
|
|
758
|
|
|
759 Where,
|
|
|
760 - The 'note-id' can be pretty much anything, though typically it
|
|
|
761 is the number of the footnote.
|
|
|
762 - The first paragraph may start on the next line, like so:
|
|
|
763
|
|
|
764 [^note-id]:
|
|
|
765 Text of the note.
|
|
|
766 """
|
|
|
767 less_than_tab = self.tab_width - 1
|
|
|
768 footnote_def_re = re.compile(r'''
|
|
|
769 ^[ ]{0,%d}\[\^(.+)\]: # id = \1
|
|
|
770 [ \t]*
|
|
|
771 ( # footnote text = \2
|
|
|
772 # First line need not start with the spaces.
|
|
|
773 (?:\s*.*\n+)
|
|
|
774 (?:
|
|
|
775 (?:[ ]{%d} | \t) # Subsequent lines must be indented.
|
|
|
776 .*\n+
|
|
|
777 )*
|
|
|
778 )
|
|
|
779 # Lookahead for non-space at line-start, or end of doc.
|
|
|
780 (?:(?=^[ ]{0,%d}\S)|\Z)
|
|
|
781 ''' % (less_than_tab, self.tab_width, self.tab_width),
|
|
|
782 re.X | re.M)
|
|
|
783 return footnote_def_re.sub(self._extract_footnote_def_sub, text)
|
|
|
784
|
|
|
785 _hr_re = re.compile(r'^[ ]{0,3}([-_*][ ]{0,2}){3,}$', re.M)
|
|
|
786
|
|
|
787 def _run_block_gamut(self, text):
|
|
|
788 # These are all the transformations that form block-level
|
|
|
789 # tags like paragraphs, headers, and list items.
|
|
|
790
|
|
|
791 if "fenced-code-blocks" in self.extras:
|
|
|
792 text = self._do_fenced_code_blocks(text)
|
|
|
793
|
|
|
794 text = self._do_headers(text)
|
|
|
795
|
|
|
796 # Do Horizontal Rules:
|
|
|
797 # On the number of spaces in horizontal rules: The spec is fuzzy: "If
|
|
|
798 # you wish, you may use spaces between the hyphens or asterisks."
|
|
|
799 # Markdown.pl 1.0.1's hr regexes limit the number of spaces between the
|
|
|
800 # hr chars to one or two. We'll reproduce that limit here.
|
|
|
801 hr = "\n<hr"+self.empty_element_suffix+"\n"
|
|
|
802 text = re.sub(self._hr_re, hr, text)
|
|
|
803
|
|
|
804 text = self._do_lists(text)
|
|
|
805
|
|
|
806 if "pyshell" in self.extras:
|
|
|
807 text = self._prepare_pyshell_blocks(text)
|
|
|
808 if "wiki-tables" in self.extras:
|
|
|
809 text = self._do_wiki_tables(text)
|
|
|
810 if "tables" in self.extras:
|
|
|
811 text = self._do_tables(text)
|
|
|
812
|
|
|
813 text = self._do_code_blocks(text)
|
|
|
814
|
|
|
815 text = self._do_block_quotes(text)
|
|
|
816
|
|
|
817 # We already ran _HashHTMLBlocks() before, in Markdown(), but that
|
|
|
818 # was to escape raw HTML in the original Markdown source. This time,
|
|
|
819 # we're escaping the markup we've just created, so that we don't wrap
|
|
|
820 # <p> tags around block-level tags.
|
|
|
821 text = self._hash_html_blocks(text)
|
|
|
822
|
|
|
823 text = self._form_paragraphs(text)
|
|
|
824
|
|
|
825 return text
|
|
|
826
|
|
|
827 def _pyshell_block_sub(self, match):
|
|
|
828 lines = match.group(0).splitlines(0)
|
|
|
829 _dedentlines(lines)
|
|
|
830 indent = ' ' * self.tab_width
|
|
|
831 s = ('\n' # separate from possible cuddled paragraph
|
|
|
832 + indent + ('\n'+indent).join(lines)
|
|
|
833 + '\n\n')
|
|
|
834 return s
|
|
|
835
|
|
|
836 def _prepare_pyshell_blocks(self, text):
|
|
|
837 """Ensure that Python interactive shell sessions are put in
|
|
|
838 code blocks -- even if not properly indented.
|
|
|
839 """
|
|
|
840 if ">>>" not in text:
|
|
|
841 return text
|
|
|
842
|
|
|
843 less_than_tab = self.tab_width - 1
|
|
|
844 _pyshell_block_re = re.compile(r"""
|
|
|
845 ^([ ]{0,%d})>>>[ ].*\n # first line
|
|
|
846 ^(\1.*\S+.*\n)* # any number of subsequent lines
|
|
|
847 ^\n # ends with a blank line
|
|
|
848 """ % less_than_tab, re.M | re.X)
|
|
|
849
|
|
|
850 return _pyshell_block_re.sub(self._pyshell_block_sub, text)
|
|
|
851
|
|
|
852 def _table_sub(self, match):
|
|
|
853 head, underline, body = match.groups()
|
|
|
854
|
|
|
855 # Determine aligns for columns.
|
|
|
856 cols = [cell.strip() for cell in underline.strip('| \t\n').split('|')]
|
|
|
857 align_from_col_idx = {}
|
|
|
858 for col_idx, col in enumerate(cols):
|
|
|
859 if col[0] == ':' and col[-1] == ':':
|
|
|
860 align_from_col_idx[col_idx] = ' align="center"'
|
|
|
861 elif col[0] == ':':
|
|
|
862 align_from_col_idx[col_idx] = ' align="left"'
|
|
|
863 elif col[-1] == ':':
|
|
|
864 align_from_col_idx[col_idx] = ' align="right"'
|
|
|
865
|
|
|
866 # thead
|
|
|
867 hlines = ['<table>', '<thead>', '<tr>']
|
|
|
868 cols = [cell.strip() for cell in head.strip('| \t\n').split('|')]
|
|
|
869 for col_idx, col in enumerate(cols):
|
|
|
870 hlines.append(' <th%s>%s</th>' % (
|
|
|
871 align_from_col_idx.get(col_idx, ''),
|
|
|
872 self._run_span_gamut(col)
|
|
|
873 ))
|
|
|
874 hlines.append('</tr>')
|
|
|
875 hlines.append('</thead>')
|
|
|
876
|
|
|
877 # tbody
|
|
|
878 hlines.append('<tbody>')
|
|
|
879 for line in body.strip('\n').split('\n'):
|
|
|
880 hlines.append('<tr>')
|
|
|
881 cols = [cell.strip() for cell in line.strip('| \t\n').split('|')]
|
|
|
882 for col_idx, col in enumerate(cols):
|
|
|
883 hlines.append(' <td%s>%s</td>' % (
|
|
|
884 align_from_col_idx.get(col_idx, ''),
|
|
|
885 self._run_span_gamut(col)
|
|
|
886 ))
|
|
|
887 hlines.append('</tr>')
|
|
|
888 hlines.append('</tbody>')
|
|
|
889 hlines.append('</table>')
|
|
|
890
|
|
|
891 return '\n'.join(hlines) + '\n'
|
|
|
892
|
|
|
893 def _do_tables(self, text):
|
|
|
894 """Copying PHP-Markdown and GFM table syntax. Some regex borrowed from
|
|
|
895 https://github.com/michelf/php-markdown/blob/lib/Michelf/Markdown.php#L2538
|
|
|
896 """
|
|
|
897 less_than_tab = self.tab_width - 1
|
|
|
898 table_re = re.compile(r'''
|
|
|
899 (?:(?<=\n\n)|\A\n?) # leading blank line
|
|
|
900
|
|
|
901 ^[ ]{0,%d} # allowed whitespace
|
|
|
902 (.*[|].*) \n # $1: header row (at least one pipe)
|
|
|
903
|
|
|
904 ^[ ]{0,%d} # allowed whitespace
|
|
|
905 ( # $2: underline row
|
|
|
906 # underline row with leading bar
|
|
|
907 (?: \|\ *:?-+:?\ * )+ \|? \n
|
|
|
908 |
|
|
|
909 # or, underline row without leading bar
|
|
|
910 (?: \ *:?-+:?\ *\| )+ (?: \ *:?-+:?\ * )? \n
|
|
|
911 )
|
|
|
912
|
|
|
913 ( # $3: data rows
|
|
|
914 (?:
|
|
|
915 ^[ ]{0,%d}(?!\ ) # ensure line begins with 0 to less_than_tab spaces
|
|
|
916 .*\|.* \n
|
|
|
917 )+
|
|
|
918 )
|
|
|
919 ''' % (less_than_tab, less_than_tab, less_than_tab), re.M | re.X)
|
|
|
920 return table_re.sub(self._table_sub, text)
|
|
|
921
|
|
|
922 def _wiki_table_sub(self, match):
|
|
|
923 ttext = match.group(0).strip()
|
|
|
924 #print 'wiki table: %r' % match.group(0)
|
|
|
925 rows = []
|
|
|
926 for line in ttext.splitlines(0):
|
|
|
927 line = line.strip()[2:-2].strip()
|
|
|
928 row = [c.strip() for c in re.split(r'(?<!\\)\|\|', line)]
|
|
|
929 rows.append(row)
|
|
|
930 #pprint(rows)
|
|
|
931 hlines = ['<table>', '<tbody>']
|
|
|
932 for row in rows:
|
|
|
933 hrow = ['<tr>']
|
|
|
934 for cell in row:
|
|
|
935 hrow.append('<td>')
|
|
|
936 hrow.append(self._run_span_gamut(cell))
|
|
|
937 hrow.append('</td>')
|
|
|
938 hrow.append('</tr>')
|
|
|
939 hlines.append(''.join(hrow))
|
|
|
940 hlines += ['</tbody>', '</table>']
|
|
|
941 return '\n'.join(hlines) + '\n'
|
|
|
942
|
|
|
943 def _do_wiki_tables(self, text):
|
|
|
944 # Optimization.
|
|
|
945 if "||" not in text:
|
|
|
946 return text
|
|
|
947
|
|
|
948 less_than_tab = self.tab_width - 1
|
|
|
949 wiki_table_re = re.compile(r'''
|
|
|
950 (?:(?<=\n\n)|\A\n?) # leading blank line
|
|
|
951 ^([ ]{0,%d})\|\|.+?\|\|[ ]*\n # first line
|
|
|
952 (^\1\|\|.+?\|\|\n)* # any number of subsequent lines
|
|
|
953 ''' % less_than_tab, re.M | re.X)
|
|
|
954 return wiki_table_re.sub(self._wiki_table_sub, text)
|
|
|
955
|
|
|
956 def _run_span_gamut(self, text):
|
|
|
957 # These are all the transformations that occur *within* block-level
|
|
|
958 # tags like paragraphs, headers, and list items.
|
|
|
959
|
|
|
960 text = self._do_code_spans(text)
|
|
|
961
|
|
|
962 text = self._escape_special_chars(text)
|
|
|
963
|
|
|
964 # Process anchor and image tags.
|
|
|
965 text = self._do_links(text)
|
|
|
966
|
|
|
967 # Make links out of things like `<http://example.com/>`
|
|
|
968 # Must come after _do_links(), because you can use < and >
|
|
|
969 # delimiters in inline links like [this](<url>).
|
|
|
970 text = self._do_auto_links(text)
|
|
|
971
|
|
|
972 if "link-patterns" in self.extras:
|
|
|
973 text = self._do_link_patterns(text)
|
|
|
974
|
|
|
975 text = self._encode_amps_and_angles(text)
|
|
|
976
|
|
|
977 text = self._do_italics_and_bold(text)
|
|
|
978
|
|
|
979 if "smarty-pants" in self.extras:
|
|
|
980 text = self._do_smart_punctuation(text)
|
|
|
981
|
|
|
982 # Do hard breaks:
|
|
|
983 if "break-on-newline" in self.extras:
|
|
|
984 text = re.sub(r" *\n", "<br%s\n" % self.empty_element_suffix, text)
|
|
|
985 else:
|
|
|
986 text = re.sub(r" {2,}\n", " <br%s\n" % self.empty_element_suffix, text)
|
|
|
987
|
|
|
988 return text
|
|
|
989
|
|
|
990 # "Sorta" because auto-links are identified as "tag" tokens.
|
|
|
991 _sorta_html_tokenize_re = re.compile(r"""
|
|
|
992 (
|
|
|
993 # tag
|
|
|
994 </?
|
|
|
995 (?:\w+) # tag name
|
|
|
996 (?:\s+(?:[\w-]+:)?[\w-]+=(?:".*?"|'.*?'))* # attributes
|
|
|
997 \s*/?>
|
|
|
998 |
|
|
|
999 # auto-link (e.g., <http://www.activestate.com/>)
|
|
|
1000 <\w+[^>]*>
|
|
|
1001 |
|
|
|
1002 <!--.*?--> # comment
|
|
|
1003 |
|
|
|
1004 <\?.*?\?> # processing instruction
|
|
|
1005 )
|
|
|
1006 """, re.X)
|
|
|
1007
|
|
|
1008 def _escape_special_chars(self, text):
|
|
|
1009 # Python markdown note: the HTML tokenization here differs from
|
|
|
1010 # that in Markdown.pl, hence the behaviour for subtle cases can
|
|
|
1011 # differ (I believe the tokenizer here does a better job because
|
|
|
1012 # it isn't susceptible to unmatched '<' and '>' in HTML tags).
|
|
|
1013 # Note, however, that '>' is not allowed in an auto-link URL
|
|
|
1014 # here.
|
|
|
1015 escaped = []
|
|
|
1016 is_html_markup = False
|
|
|
1017 for token in self._sorta_html_tokenize_re.split(text):
|
|
|
1018 if is_html_markup:
|
|
|
1019 # Within tags/HTML-comments/auto-links, encode * and _
|
|
|
1020 # so they don't conflict with their use in Markdown for
|
|
|
1021 # italics and strong. We're replacing each such
|
|
|
1022 # character with its corresponding MD5 checksum value;
|
|
|
1023 # this is likely overkill, but it should prevent us from
|
|
|
1024 # colliding with the escape values by accident.
|
|
|
1025 escaped.append(token.replace('*', self._escape_table['*'])
|
|
|
1026 .replace('_', self._escape_table['_']))
|
|
|
1027 else:
|
|
|
1028 escaped.append(self._encode_backslash_escapes(token))
|
|
|
1029 is_html_markup = not is_html_markup
|
|
|
1030 return ''.join(escaped)
|
|
|
1031
|
|
|
1032 def _hash_html_spans(self, text):
|
|
|
1033 # Used for safe_mode.
|
|
|
1034
|
|
|
1035 def _is_auto_link(s):
|
|
|
1036 if ':' in s and self._auto_link_re.match(s):
|
|
|
1037 return True
|
|
|
1038 elif '@' in s and self._auto_email_link_re.match(s):
|
|
|
1039 return True
|
|
|
1040 return False
|
|
|
1041
|
|
|
1042 tokens = []
|
|
|
1043 is_html_markup = False
|
|
|
1044 for token in self._sorta_html_tokenize_re.split(text):
|
|
|
1045 if is_html_markup and not _is_auto_link(token):
|
|
|
1046 sanitized = self._sanitize_html(token)
|
|
|
1047 key = _hash_text(sanitized)
|
|
|
1048 self.html_spans[key] = sanitized
|
|
|
1049 tokens.append(key)
|
|
|
1050 else:
|
|
|
1051 tokens.append(token)
|
|
|
1052 is_html_markup = not is_html_markup
|
|
|
1053 return ''.join(tokens)
|
|
|
1054
|
|
|
1055 def _unhash_html_spans(self, text):
|
|
|
1056 for key, sanitized in list(self.html_spans.items()):
|
|
|
1057 text = text.replace(key, sanitized)
|
|
|
1058 return text
|
|
|
1059
|
|
|
1060 def _sanitize_html(self, s):
|
|
|
1061 if self.safe_mode == "replace":
|
|
|
1062 return self.html_removed_text
|
|
|
1063 elif self.safe_mode == "escape":
|
|
|
1064 replacements = [
|
|
|
1065 ('&', '&'),
|
|
|
1066 ('<', '<'),
|
|
|
1067 ('>', '>'),
|
|
|
1068 ]
|
|
|
1069 for before, after in replacements:
|
|
|
1070 s = s.replace(before, after)
|
|
|
1071 return s
|
|
|
1072 else:
|
|
|
1073 raise MarkdownError("invalid value for 'safe_mode': %r (must be "
|
|
|
1074 "'escape' or 'replace')" % self.safe_mode)
|
|
|
1075
|
|
|
1076 _inline_link_title = re.compile(r'''
|
|
|
1077 ( # \1
|
|
|
1078 [ \t]+
|
|
|
1079 (['"]) # quote char = \2
|
|
|
1080 (?P<title>.*?)
|
|
|
1081 \2
|
|
|
1082 )? # title is optional
|
|
|
1083 \)$
|
|
|
1084 ''', re.X | re.S)
|
|
|
1085 _tail_of_reference_link_re = re.compile(r'''
|
|
|
1086 # Match tail of: [text][id]
|
|
|
1087 [ ]? # one optional space
|
|
|
1088 (?:\n[ ]*)? # one optional newline followed by spaces
|
|
|
1089 \[
|
|
|
1090 (?P<id>.*?)
|
|
|
1091 \]
|
|
|
1092 ''', re.X | re.S)
|
|
|
1093
|
|
|
1094 _whitespace = re.compile(r'\s*')
|
|
|
1095
|
|
|
1096 _strip_anglebrackets = re.compile(r'<(.*)>.*')
|
|
|
1097
|
|
|
1098 def _find_non_whitespace(self, text, start):
|
|
|
1099 """Returns the index of the first non-whitespace character in text
|
|
|
1100 after (and including) start
|
|
|
1101 """
|
|
|
1102 match = self._whitespace.match(text, start)
|
|
|
1103 return match.end()
|
|
|
1104
|
|
|
1105 def _find_balanced(self, text, start, open_c, close_c):
|
|
|
1106 """Returns the index where the open_c and close_c characters balance
|
|
|
1107 out - the same number of open_c and close_c are encountered - or the
|
|
|
1108 end of string if it's reached before the balance point is found.
|
|
|
1109 """
|
|
|
1110 i = start
|
|
|
1111 l = len(text)
|
|
|
1112 count = 1
|
|
|
1113 while count > 0 and i < l:
|
|
|
1114 if text[i] == open_c:
|
|
|
1115 count += 1
|
|
|
1116 elif text[i] == close_c:
|
|
|
1117 count -= 1
|
|
|
1118 i += 1
|
|
|
1119 return i
|
|
|
1120
|
|
|
1121 def _extract_url_and_title(self, text, start):
|
|
|
1122 """Extracts the url and (optional) title from the tail of a link"""
|
|
|
1123 # text[start] equals the opening parenthesis
|
|
|
1124 idx = self._find_non_whitespace(text, start+1)
|
|
|
1125 if idx == len(text):
|
|
|
1126 return None, None, None
|
|
|
1127 end_idx = idx
|
|
|
1128 has_anglebrackets = text[idx] == "<"
|
|
|
1129 if has_anglebrackets:
|
|
|
1130 end_idx = self._find_balanced(text, end_idx+1, "<", ">")
|
|
|
1131 end_idx = self._find_balanced(text, end_idx, "(", ")")
|
|
|
1132 match = self._inline_link_title.search(text, idx, end_idx)
|
|
|
1133 if not match:
|
|
|
1134 return None, None, None
|
|
|
1135 url, title = text[idx:match.start()], match.group("title")
|
|
|
1136 if has_anglebrackets:
|
|
|
1137 url = self._strip_anglebrackets.sub(r'\1', url)
|
|
|
1138 return url, title, end_idx
|
|
|
1139
|
|
|
1140 def _do_links(self, text):
|
|
|
1141 """Turn Markdown link shortcuts into XHTML <a> and <img> tags.
|
|
|
1142
|
|
|
1143 This is a combination of Markdown.pl's _DoAnchors() and
|
|
|
1144 _DoImages(). They are done together because that simplified the
|
|
|
1145 approach. It was necessary to use a different approach than
|
|
|
1146 Markdown.pl because of the lack of atomic matching support in
|
|
|
1147 Python's regex engine used in $g_nested_brackets.
|
|
|
1148 """
|
|
|
1149 MAX_LINK_TEXT_SENTINEL = 3000 # markdown2 issue 24
|
|
|
1150
|
|
|
1151 # `anchor_allowed_pos` is used to support img links inside
|
|
|
1152 # anchors, but not anchors inside anchors. An anchor's start
|
|
|
1153 # pos must be `>= anchor_allowed_pos`.
|
|
|
1154 anchor_allowed_pos = 0
|
|
|
1155
|
|
|
1156 curr_pos = 0
|
|
|
1157 while True: # Handle the next link.
|
|
|
1158 # The next '[' is the start of:
|
|
|
1159 # - an inline anchor: [text](url "title")
|
|
|
1160 # - a reference anchor: [text][id]
|
|
|
1161 # - an inline img: 
|
|
|
1162 # - a reference img: ![text][id]
|
|
|
1163 # - a footnote ref: [^id]
|
|
|
1164 # (Only if 'footnotes' extra enabled)
|
|
|
1165 # - a footnote defn: [^id]: ...
|
|
|
1166 # (Only if 'footnotes' extra enabled) These have already
|
|
|
1167 # been stripped in _strip_footnote_definitions() so no
|
|
|
1168 # need to watch for them.
|
|
|
1169 # - a link definition: [id]: url "title"
|
|
|
1170 # These have already been stripped in
|
|
|
1171 # _strip_link_definitions() so no need to watch for them.
|
|
|
1172 # - not markup: [...anything else...
|
|
|
1173 try:
|
|
|
1174 start_idx = text.index('[', curr_pos)
|
|
|
1175 except ValueError:
|
|
|
1176 break
|
|
|
1177 text_length = len(text)
|
|
|
1178
|
|
|
1179 # Find the matching closing ']'.
|
|
|
1180 # Markdown.pl allows *matching* brackets in link text so we
|
|
|
1181 # will here too. Markdown.pl *doesn't* currently allow
|
|
|
1182 # matching brackets in img alt text -- we'll differ in that
|
|
|
1183 # regard.
|
|
|
1184 bracket_depth = 0
|
|
|
1185 for p in range(start_idx+1, min(start_idx+MAX_LINK_TEXT_SENTINEL,
|
|
|
1186 text_length)):
|
|
|
1187 ch = text[p]
|
|
|
1188 if ch == ']':
|
|
|
1189 bracket_depth -= 1
|
|
|
1190 if bracket_depth < 0:
|
|
|
1191 break
|
|
|
1192 elif ch == '[':
|
|
|
1193 bracket_depth += 1
|
|
|
1194 else:
|
|
|
1195 # Closing bracket not found within sentinel length.
|
|
|
1196 # This isn't markup.
|
|
|
1197 curr_pos = start_idx + 1
|
|
|
1198 continue
|
|
|
1199 link_text = text[start_idx+1:p]
|
|
|
1200
|
|
|
1201 # Possibly a footnote ref?
|
|
|
1202 if "footnotes" in self.extras and link_text.startswith("^"):
|
|
|
1203 normed_id = re.sub(r'\W', '-', link_text[1:])
|
|
|
1204 if normed_id in self.footnotes:
|
|
|
1205 self.footnote_ids.append(normed_id)
|
|
|
1206 result = '<sup class="footnote-ref" id="fnref-%s">' \
|
|
|
1207 '<a href="#fn-%s">%s</a></sup>' \
|
|
|
1208 % (normed_id, normed_id, len(self.footnote_ids))
|
|
|
1209 text = text[:start_idx] + result + text[p+1:]
|
|
|
1210 else:
|
|
|
1211 # This id isn't defined, leave the markup alone.
|
|
|
1212 curr_pos = p+1
|
|
|
1213 continue
|
|
|
1214
|
|
|
1215 # Now determine what this is by the remainder.
|
|
|
1216 p += 1
|
|
|
1217 if p == text_length:
|
|
|
1218 return text
|
|
|
1219
|
|
|
1220 # Inline anchor or img?
|
|
|
1221 if text[p] == '(': # attempt at perf improvement
|
|
|
1222 url, title, url_end_idx = self._extract_url_and_title(text, p)
|
|
|
1223 if url is not None:
|
|
|
1224 # Handle an inline anchor or img.
|
|
|
1225 is_img = start_idx > 0 and text[start_idx-1] == "!"
|
|
|
1226 if is_img:
|
|
|
1227 start_idx -= 1
|
|
|
1228
|
|
|
1229 # We've got to encode these to avoid conflicting
|
|
|
1230 # with italics/bold.
|
|
|
1231 url = url.replace('*', self._escape_table['*']) \
|
|
|
1232 .replace('_', self._escape_table['_'])
|
|
|
1233 if title:
|
|
|
1234 title_str = ' title="%s"' % (
|
|
|
1235 _xml_escape_attr(title)
|
|
|
1236 .replace('*', self._escape_table['*'])
|
|
|
1237 .replace('_', self._escape_table['_']))
|
|
|
1238 else:
|
|
|
1239 title_str = ''
|
|
|
1240 if is_img:
|
|
|
1241 img_class_str = self._html_class_str_from_tag("img")
|
|
|
1242 result = '<img src="%s" alt="%s"%s%s%s' \
|
|
|
1243 % (url.replace('"', '"'),
|
|
|
1244 _xml_escape_attr(link_text),
|
|
|
1245 title_str, img_class_str, self.empty_element_suffix)
|
|
|
1246 if "smarty-pants" in self.extras:
|
|
|
1247 result = result.replace('"', self._escape_table['"'])
|
|
|
1248 curr_pos = start_idx + len(result)
|
|
|
1249 text = text[:start_idx] + result + text[url_end_idx:]
|
|
|
1250 elif start_idx >= anchor_allowed_pos:
|
|
|
1251 result_head = '<a href="%s"%s>' % (url, title_str)
|
|
|
1252 result = '%s%s</a>' % (result_head, link_text)
|
|
|
1253 if "smarty-pants" in self.extras:
|
|
|
1254 result = result.replace('"', self._escape_table['"'])
|
|
|
1255 # <img> allowed from curr_pos on, <a> from
|
|
|
1256 # anchor_allowed_pos on.
|
|
|
1257 curr_pos = start_idx + len(result_head)
|
|
|
1258 anchor_allowed_pos = start_idx + len(result)
|
|
|
1259 text = text[:start_idx] + result + text[url_end_idx:]
|
|
|
1260 else:
|
|
|
1261 # Anchor not allowed here.
|
|
|
1262 curr_pos = start_idx + 1
|
|
|
1263 continue
|
|
|
1264
|
|
|
1265 # Reference anchor or img?
|
|
|
1266 else:
|
|
|
1267 match = self._tail_of_reference_link_re.match(text, p)
|
|
|
1268 if match:
|
|
|
1269 # Handle a reference-style anchor or img.
|
|
|
1270 is_img = start_idx > 0 and text[start_idx-1] == "!"
|
|
|
1271 if is_img:
|
|
|
1272 start_idx -= 1
|
|
|
1273 link_id = match.group("id").lower()
|
|
|
1274 if not link_id:
|
|
|
1275 link_id = link_text.lower() # for links like [this][]
|
|
|
1276 if link_id in self.urls:
|
|
|
1277 url = self.urls[link_id]
|
|
|
1278 # We've got to encode these to avoid conflicting
|
|
|
1279 # with italics/bold.
|
|
|
1280 url = url.replace('*', self._escape_table['*']) \
|
|
|
1281 .replace('_', self._escape_table['_'])
|
|
|
1282 title = self.titles.get(link_id)
|
|
|
1283 if title:
|
|
|
1284 before = title
|
|
|
1285 title = _xml_escape_attr(title) \
|
|
|
1286 .replace('*', self._escape_table['*']) \
|
|
|
1287 .replace('_', self._escape_table['_'])
|
|
|
1288 title_str = ' title="%s"' % title
|
|
|
1289 else:
|
|
|
1290 title_str = ''
|
|
|
1291 if is_img:
|
|
|
1292 img_class_str = self._html_class_str_from_tag("img")
|
|
|
1293 result = '<img src="%s" alt="%s"%s%s%s' \
|
|
|
1294 % (url.replace('"', '"'),
|
|
|
1295 link_text.replace('"', '"'),
|
|
|
1296 title_str, img_class_str, self.empty_element_suffix)
|
|
|
1297 if "smarty-pants" in self.extras:
|
|
|
1298 result = result.replace('"', self._escape_table['"'])
|
|
|
1299 curr_pos = start_idx + len(result)
|
|
|
1300 text = text[:start_idx] + result + text[match.end():]
|
|
|
1301 elif start_idx >= anchor_allowed_pos:
|
|
|
1302 result = '<a href="%s"%s>%s</a>' \
|
|
|
1303 % (url, title_str, link_text)
|
|
|
1304 result_head = '<a href="%s"%s>' % (url, title_str)
|
|
|
1305 result = '%s%s</a>' % (result_head, link_text)
|
|
|
1306 if "smarty-pants" in self.extras:
|
|
|
1307 result = result.replace('"', self._escape_table['"'])
|
|
|
1308 # <img> allowed from curr_pos on, <a> from
|
|
|
1309 # anchor_allowed_pos on.
|
|
|
1310 curr_pos = start_idx + len(result_head)
|
|
|
1311 anchor_allowed_pos = start_idx + len(result)
|
|
|
1312 text = text[:start_idx] + result + text[match.end():]
|
|
|
1313 else:
|
|
|
1314 # Anchor not allowed here.
|
|
|
1315 curr_pos = start_idx + 1
|
|
|
1316 else:
|
|
|
1317 # This id isn't defined, leave the markup alone.
|
|
|
1318 curr_pos = match.end()
|
|
|
1319 continue
|
|
|
1320
|
|
|
1321 # Otherwise, it isn't markup.
|
|
|
1322 curr_pos = start_idx + 1
|
|
|
1323
|
|
|
1324 return text
|
|
|
1325
|
|
|
1326 def header_id_from_text(self, text, prefix, n):
|
|
|
1327 """Generate a header id attribute value from the given header
|
|
|
1328 HTML content.
|
|
|
1329
|
|
|
1330 This is only called if the "header-ids" extra is enabled.
|
|
|
1331 Subclasses may override this for different header ids.
|
|
|
1332
|
|
|
1333 @param text {str} The text of the header tag
|
|
|
1334 @param prefix {str} The requested prefix for header ids. This is the
|
|
|
1335 value of the "header-ids" extra key, if any. Otherwise, None.
|
|
|
1336 @param n {int} The <hN> tag number, i.e. `1` for an <h1> tag.
|
|
|
1337 @returns {str} The value for the header tag's "id" attribute. Return
|
|
|
1338 None to not have an id attribute and to exclude this header from
|
|
|
1339 the TOC (if the "toc" extra is specified).
|
|
|
1340 """
|
|
|
1341 header_id = _slugify(text)
|
|
|
1342 if prefix and isinstance(prefix, base_string_type):
|
|
|
1343 header_id = prefix + '-' + header_id
|
|
|
1344 if header_id in self._count_from_header_id:
|
|
|
1345 self._count_from_header_id[header_id] += 1
|
|
|
1346 header_id += '-%s' % self._count_from_header_id[header_id]
|
|
|
1347 else:
|
|
|
1348 self._count_from_header_id[header_id] = 1
|
|
|
1349 return header_id
|
|
|
1350
|
|
|
1351 _toc = None
|
|
|
1352 def _toc_add_entry(self, level, id, name):
|
|
|
1353 if self._toc is None:
|
|
|
1354 self._toc = []
|
|
|
1355 self._toc.append((level, id, self._unescape_special_chars(name)))
|
|
|
1356
|
|
|
1357 _h_re_base = r'''
|
|
|
1358 (^(.+)[ \t]*\n(=+|-+)[ \t]*\n+)
|
|
|
1359 |
|
|
|
1360 (^(\#{1,6}) # \1 = string of #'s
|
|
|
1361 [ \t]%s
|
|
|
1362 (.+?) # \2 = Header text
|
|
|
1363 [ \t]*
|
|
|
1364 (?<!\\) # ensure not an escaped trailing '#'
|
|
|
1365 \#* # optional closing #'s (not counted)
|
|
|
1366 \n+
|
|
|
1367 )
|
|
|
1368 '''
|
|
|
1369
|
|
|
1370 _h_re = re.compile(_h_re_base % '*', re.X | re.M)
|
|
|
1371 _h_re_tag_friendly = re.compile(_h_re_base % '+', re.X | re.M)
|
|
|
1372
|
|
|
1373 def _h_sub(self, match):
|
|
|
1374 if match.group(1) is not None:
|
|
|
1375 # Setext header
|
|
|
1376 n = {"=": 1, "-": 2}[match.group(3)[0]]
|
|
|
1377 header_group = match.group(2)
|
|
|
1378 else:
|
|
|
1379 # atx header
|
|
|
1380 n = len(match.group(5))
|
|
|
1381 header_group = match.group(6)
|
|
|
1382
|
|
|
1383 demote_headers = self.extras.get("demote-headers")
|
|
|
1384 if demote_headers:
|
|
|
1385 n = min(n + demote_headers, 6)
|
|
|
1386 header_id_attr = ""
|
|
|
1387 if "header-ids" in self.extras:
|
|
|
1388 header_id = self.header_id_from_text(header_group,
|
|
|
1389 self.extras["header-ids"], n)
|
|
|
1390 if header_id:
|
|
|
1391 header_id_attr = ' id="%s"' % header_id
|
|
|
1392 html = self._run_span_gamut(header_group)
|
|
|
1393 if "toc" in self.extras and header_id:
|
|
|
1394 self._toc_add_entry(n, header_id, html)
|
|
|
1395 return "<h%d%s>%s</h%d>\n\n" % (n, header_id_attr, html, n)
|
|
|
1396
|
|
|
1397 def _do_headers(self, text):
|
|
|
1398 # Setext-style headers:
|
|
|
1399 # Header 1
|
|
|
1400 # ========
|
|
|
1401 #
|
|
|
1402 # Header 2
|
|
|
1403 # --------
|
|
|
1404
|
|
|
1405 # atx-style headers:
|
|
|
1406 # # Header 1
|
|
|
1407 # ## Header 2
|
|
|
1408 # ## Header 2 with closing hashes ##
|
|
|
1409 # ...
|
|
|
1410 # ###### Header 6
|
|
|
1411
|
|
|
1412 if 'tag-friendly' in self.extras:
|
|
|
1413 return self._h_re_tag_friendly.sub(self._h_sub, text)
|
|
|
1414 return self._h_re.sub(self._h_sub, text)
|
|
|
1415
|
|
|
1416 _marker_ul_chars = '*+-'
|
|
|
1417 _marker_any = r'(?:[%s]|\d+\.)' % _marker_ul_chars
|
|
|
1418 _marker_ul = '(?:[%s])' % _marker_ul_chars
|
|
|
1419 _marker_ol = r'(?:\d+\.)'
|
|
|
1420
|
|
|
1421 def _list_sub(self, match):
|
|
|
1422 lst = match.group(1)
|
|
|
1423 lst_type = match.group(3) in self._marker_ul_chars and "ul" or "ol"
|
|
|
1424 result = self._process_list_items(lst)
|
|
|
1425 if self.list_level:
|
|
|
1426 return "<%s>\n%s</%s>\n" % (lst_type, result, lst_type)
|
|
|
1427 else:
|
|
|
1428 return "<%s>\n%s</%s>\n\n" % (lst_type, result, lst_type)
|
|
|
1429
|
|
|
1430 def _do_lists(self, text):
|
|
|
1431 # Form HTML ordered (numbered) and unordered (bulleted) lists.
|
|
|
1432
|
|
|
1433 # Iterate over each *non-overlapping* list match.
|
|
|
1434 pos = 0
|
|
|
1435 while True:
|
|
|
1436 # Find the *first* hit for either list style (ul or ol). We
|
|
|
1437 # match ul and ol separately to avoid adjacent lists of different
|
|
|
1438 # types running into each other (see issue #16).
|
|
|
1439 hits = []
|
|
|
1440 for marker_pat in (self._marker_ul, self._marker_ol):
|
|
|
1441 less_than_tab = self.tab_width - 1
|
|
|
1442 whole_list = r'''
|
|
|
1443 ( # \1 = whole list
|
|
|
1444 ( # \2
|
|
|
1445 [ ]{0,%d}
|
|
|
1446 (%s) # \3 = first list item marker
|
|
|
1447 [ \t]+
|
|
|
1448 (?!\ *\3\ ) # '- - - ...' isn't a list. See 'not_quite_a_list' test case.
|
|
|
1449 )
|
|
|
1450 (?:.+?)
|
|
|
1451 ( # \4
|
|
|
1452 \Z
|
|
|
1453 |
|
|
|
1454 \n{2,}
|
|
|
1455 (?=\S)
|
|
|
1456 (?! # Negative lookahead for another list item marker
|
|
|
1457 [ \t]*
|
|
|
1458 %s[ \t]+
|
|
|
1459 )
|
|
|
1460 )
|
|
|
1461 )
|
|
|
1462 ''' % (less_than_tab, marker_pat, marker_pat)
|
|
|
1463 if self.list_level: # sub-list
|
|
|
1464 list_re = re.compile("^"+whole_list, re.X | re.M | re.S)
|
|
|
1465 else:
|
|
|
1466 list_re = re.compile(r"(?:(?<=\n\n)|\A\n?)"+whole_list,
|
|
|
1467 re.X | re.M | re.S)
|
|
|
1468 match = list_re.search(text, pos)
|
|
|
1469 if match:
|
|
|
1470 hits.append((match.start(), match))
|
|
|
1471 if not hits:
|
|
|
1472 break
|
|
|
1473 hits.sort()
|
|
|
1474 match = hits[0][1]
|
|
|
1475 start, end = match.span()
|
|
|
1476 middle = self._list_sub(match)
|
|
|
1477 text = text[:start] + middle + text[end:]
|
|
|
1478 pos = start + len(middle) # start pos for next attempted match
|
|
|
1479
|
|
|
1480 return text
|
|
|
1481
|
|
|
1482 _list_item_re = re.compile(r'''
|
|
|
1483 (\n)? # leading line = \1
|
|
|
1484 (^[ \t]*) # leading whitespace = \2
|
|
|
1485 (?P<marker>%s) [ \t]+ # list marker = \3
|
|
|
1486 ((?:.+?) # list item text = \4
|
|
|
1487 (\n{1,2})) # eols = \5
|
|
|
1488 (?= \n* (\Z | \2 (?P<next_marker>%s) [ \t]+))
|
|
|
1489 ''' % (_marker_any, _marker_any),
|
|
|
1490 re.M | re.X | re.S)
|
|
|
1491
|
|
|
1492 _last_li_endswith_two_eols = False
|
|
|
1493 def _list_item_sub(self, match):
|
|
|
1494 item = match.group(4)
|
|
|
1495 leading_line = match.group(1)
|
|
|
1496 leading_space = match.group(2)
|
|
|
1497 if leading_line or "\n\n" in item or self._last_li_endswith_two_eols:
|
|
|
1498 item = self._run_block_gamut(self._outdent(item))
|
|
|
1499 else:
|
|
|
1500 # Recursion for sub-lists:
|
|
|
1501 item = self._do_lists(self._outdent(item))
|
|
|
1502 if item.endswith('\n'):
|
|
|
1503 item = item[:-1]
|
|
|
1504 item = self._run_span_gamut(item)
|
|
|
1505 self._last_li_endswith_two_eols = (len(match.group(5)) == 2)
|
|
|
1506 return "<li>%s</li>\n" % item
|
|
|
1507
|
|
|
1508 def _process_list_items(self, list_str):
|
|
|
1509 # Process the contents of a single ordered or unordered list,
|
|
|
1510 # splitting it into individual list items.
|
|
|
1511
|
|
|
1512 # The $g_list_level global keeps track of when we're inside a list.
|
|
|
1513 # Each time we enter a list, we increment it; when we leave a list,
|
|
|
1514 # we decrement. If it's zero, we're not in a list anymore.
|
|
|
1515 #
|
|
|
1516 # We do this because when we're not inside a list, we want to treat
|
|
|
1517 # something like this:
|
|
|
1518 #
|
|
|
1519 # I recommend upgrading to version
|
|
|
1520 # 8. Oops, now this line is treated
|
|
|
1521 # as a sub-list.
|
|
|
1522 #
|
|
|
1523 # As a single paragraph, despite the fact that the second line starts
|
|
|
1524 # with a digit-period-space sequence.
|
|
|
1525 #
|
|
|
1526 # Whereas when we're inside a list (or sub-list), that line will be
|
|
|
1527 # treated as the start of a sub-list. What a kludge, huh? This is
|
|
|
1528 # an aspect of Markdown's syntax that's hard to parse perfectly
|
|
|
1529 # without resorting to mind-reading. Perhaps the solution is to
|
|
|
1530 # change the syntax rules such that sub-lists must start with a
|
|
|
1531 # starting cardinal number; e.g. "1." or "a.".
|
|
|
1532 self.list_level += 1
|
|
|
1533 self._last_li_endswith_two_eols = False
|
|
|
1534 list_str = list_str.rstrip('\n') + '\n'
|
|
|
1535 list_str = self._list_item_re.sub(self._list_item_sub, list_str)
|
|
|
1536 self.list_level -= 1
|
|
|
1537 return list_str
|
|
|
1538
|
|
|
1539 def _get_pygments_lexer(self, lexer_name):
|
|
|
1540 try:
|
|
|
1541 from pygments import lexers, util
|
|
|
1542 except ImportError:
|
|
|
1543 return None
|
|
|
1544 try:
|
|
|
1545 return lexers.get_lexer_by_name(lexer_name)
|
|
|
1546 except util.ClassNotFound:
|
|
|
1547 return None
|
|
|
1548
|
|
|
1549 def _color_with_pygments(self, codeblock, lexer, **formatter_opts):
|
|
|
1550 import pygments
|
|
|
1551 import pygments.formatters
|
|
|
1552
|
|
|
1553 class HtmlCodeFormatter(pygments.formatters.HtmlFormatter):
|
|
|
1554 def _wrap_code(self, inner):
|
|
|
1555 """A function for use in a Pygments Formatter which
|
|
|
1556 wraps in <code> tags.
|
|
|
1557 """
|
|
|
1558 yield 0, "<code>"
|
|
|
1559 for tup in inner:
|
|
|
1560 yield tup
|
|
|
1561 yield 0, "</code>"
|
|
|
1562
|
|
|
1563 def wrap(self, source, outfile):
|
|
|
1564 """Return the source with a code, pre, and div."""
|
|
|
1565 return self._wrap_div(self._wrap_pre(self._wrap_code(source)))
|
|
|
1566
|
|
|
1567 formatter_opts.setdefault("cssclass", "codehilite")
|
|
|
1568 formatter = HtmlCodeFormatter(**formatter_opts)
|
|
|
1569 return pygments.highlight(codeblock, lexer, formatter)
|
|
|
1570
|
|
|
1571 def _code_block_sub(self, match, is_fenced_code_block=False):
|
|
|
1572 lexer_name = None
|
|
|
1573 if is_fenced_code_block:
|
|
|
1574 lexer_name = match.group(1)
|
|
|
1575 if lexer_name:
|
|
|
1576 formatter_opts = self.extras['fenced-code-blocks'] or {}
|
|
|
1577 codeblock = match.group(2)
|
|
|
1578 codeblock = codeblock[:-1] # drop one trailing newline
|
|
|
1579 else:
|
|
|
1580 codeblock = match.group(1)
|
|
|
1581 codeblock = self._outdent(codeblock)
|
|
|
1582 codeblock = self._detab(codeblock)
|
|
|
1583 codeblock = codeblock.lstrip('\n') # trim leading newlines
|
|
|
1584 codeblock = codeblock.rstrip() # trim trailing whitespace
|
|
|
1585
|
|
|
1586 # Note: "code-color" extra is DEPRECATED.
|
|
|
1587 if "code-color" in self.extras and codeblock.startswith(":::"):
|
|
|
1588 lexer_name, rest = codeblock.split('\n', 1)
|
|
|
1589 lexer_name = lexer_name[3:].strip()
|
|
|
1590 codeblock = rest.lstrip("\n") # Remove lexer declaration line.
|
|
|
1591 formatter_opts = self.extras['code-color'] or {}
|
|
|
1592
|
|
|
1593 if lexer_name:
|
|
|
1594 def unhash_code( codeblock ):
|
|
|
1595 for key, sanitized in list(self.html_spans.items()):
|
|
|
1596 codeblock = codeblock.replace(key, sanitized)
|
|
|
1597 replacements = [
|
|
|
1598 ("&", "&"),
|
|
|
1599 ("<", "<"),
|
|
|
1600 (">", ">")
|
|
|
1601 ]
|
|
|
1602 for old, new in replacements:
|
|
|
1603 codeblock = codeblock.replace(old, new)
|
|
|
1604 return codeblock
|
|
|
1605 lexer = self._get_pygments_lexer(lexer_name)
|
|
|
1606 if lexer:
|
|
|
1607 codeblock = unhash_code( codeblock )
|
|
|
1608 colored = self._color_with_pygments(codeblock, lexer,
|
|
|
1609 **formatter_opts)
|
|
|
1610 return "\n\n%s\n\n" % colored
|
|
|
1611
|
|
|
1612 codeblock = self._encode_code(codeblock)
|
|
|
1613 pre_class_str = self._html_class_str_from_tag("pre")
|
|
|
1614 code_class_str = self._html_class_str_from_tag("code")
|
|
|
1615 return "\n\n<pre%s><code%s>%s\n</code></pre>\n\n" % (
|
|
|
1616 pre_class_str, code_class_str, codeblock)
|
|
|
1617
|
|
|
1618 def _html_class_str_from_tag(self, tag):
|
|
|
1619 """Get the appropriate ' class="..."' string (note the leading
|
|
|
1620 space), if any, for the given tag.
|
|
|
1621 """
|
|
|
1622 if "html-classes" not in self.extras:
|
|
|
1623 return ""
|
|
|
1624 try:
|
|
|
1625 html_classes_from_tag = self.extras["html-classes"]
|
|
|
1626 except TypeError:
|
|
|
1627 return ""
|
|
|
1628 else:
|
|
|
1629 if tag in html_classes_from_tag:
|
|
|
1630 return ' class="%s"' % html_classes_from_tag[tag]
|
|
|
1631 return ""
|
|
|
1632
|
|
|
1633 def _do_code_blocks(self, text):
|
|
|
1634 """Process Markdown `<pre><code>` blocks."""
|
|
|
1635 code_block_re = re.compile(r'''
|
|
|
1636 (?:\n\n|\A\n?)
|
|
|
1637 ( # $1 = the code block -- one or more lines, starting with a space/tab
|
|
|
1638 (?:
|
|
|
1639 (?:[ ]{%d} | \t) # Lines must start with a tab or a tab-width of spaces
|
|
|
1640 .*\n+
|
|
|
1641 )+
|
|
|
1642 )
|
|
|
1643 ((?=^[ ]{0,%d}\S)|\Z) # Lookahead for non-space at line-start, or end of doc
|
|
|
1644 # Lookahead to make sure this block isn't already in a code block.
|
|
|
1645 # Needed when syntax highlighting is being used.
|
|
|
1646 (?![^<]*\</code\>)
|
|
|
1647 ''' % (self.tab_width, self.tab_width),
|
|
|
1648 re.M | re.X)
|
|
|
1649 return code_block_re.sub(self._code_block_sub, text)
|
|
|
1650
|
|
|
1651 _fenced_code_block_re = re.compile(r'''
|
|
|
1652 (?:\n\n|\A\n?)
|
|
|
1653 ^```([\w+-]+)?[ \t]*\n # opening fence, $1 = optional lang
|
|
|
1654 (.*?) # $2 = code block content
|
|
|
1655 ^```[ \t]*\n # closing fence
|
|
|
1656 ''', re.M | re.X | re.S)
|
|
|
1657
|
|
|
1658 def _fenced_code_block_sub(self, match):
|
|
|
1659 return self._code_block_sub(match, is_fenced_code_block=True);
|
|
|
1660
|
|
|
1661 def _do_fenced_code_blocks(self, text):
|
|
|
1662 """Process ```-fenced unindented code blocks ('fenced-code-blocks' extra)."""
|
|
|
1663 return self._fenced_code_block_re.sub(self._fenced_code_block_sub, text)
|
|
|
1664
|
|
|
1665 # Rules for a code span:
|
|
|
1666 # - backslash escapes are not interpreted in a code span
|
|
|
1667 # - to include one or or a run of more backticks the delimiters must
|
|
|
1668 # be a longer run of backticks
|
|
|
1669 # - cannot start or end a code span with a backtick; pad with a
|
|
|
1670 # space and that space will be removed in the emitted HTML
|
|
|
1671 # See `test/tm-cases/escapes.text` for a number of edge-case
|
|
|
1672 # examples.
|
|
|
1673 _code_span_re = re.compile(r'''
|
|
|
1674 (?<!\\)
|
|
|
1675 (`+) # \1 = Opening run of `
|
|
|
1676 (?!`) # See Note A test/tm-cases/escapes.text
|
|
|
1677 (.+?) # \2 = The code block
|
|
|
1678 (?<!`)
|
|
|
1679 \1 # Matching closer
|
|
|
1680 (?!`)
|
|
|
1681 ''', re.X | re.S)
|
|
|
1682
|
|
|
1683 def _code_span_sub(self, match):
|
|
|
1684 c = match.group(2).strip(" \t")
|
|
|
1685 c = self._encode_code(c)
|
|
|
1686 return "<code>%s</code>" % c
|
|
|
1687
|
|
|
1688 def _do_code_spans(self, text):
|
|
|
1689 # * Backtick quotes are used for <code></code> spans.
|
|
|
1690 #
|
|
|
1691 # * You can use multiple backticks as the delimiters if you want to
|
|
|
1692 # include literal backticks in the code span. So, this input:
|
|
|
1693 #
|
|
|
1694 # Just type ``foo `bar` baz`` at the prompt.
|
|
|
1695 #
|
|
|
1696 # Will translate to:
|
|
|
1697 #
|
|
|
1698 # <p>Just type <code>foo `bar` baz</code> at the prompt.</p>
|
|
|
1699 #
|
|
|
1700 # There's no arbitrary limit to the number of backticks you
|
|
|
1701 # can use as delimters. If you need three consecutive backticks
|
|
|
1702 # in your code, use four for delimiters, etc.
|
|
|
1703 #
|
|
|
1704 # * You can use spaces to get literal backticks at the edges:
|
|
|
1705 #
|
|
|
1706 # ... type `` `bar` `` ...
|
|
|
1707 #
|
|
|
1708 # Turns to:
|
|
|
1709 #
|
|
|
1710 # ... type <code>`bar`</code> ...
|
|
|
1711 return self._code_span_re.sub(self._code_span_sub, text)
|
|
|
1712
|
|
|
1713 def _encode_code(self, text):
|
|
|
1714 """Encode/escape certain characters inside Markdown code runs.
|
|
|
1715 The point is that in code, these characters are literals,
|
|
|
1716 and lose their special Markdown meanings.
|
|
|
1717 """
|
|
|
1718 replacements = [
|
|
|
1719 # Encode all ampersands; HTML entities are not
|
|
|
1720 # entities within a Markdown code span.
|
|
|
1721 ('&', '&'),
|
|
|
1722 # Do the angle bracket song and dance:
|
|
|
1723 ('<', '<'),
|
|
|
1724 ('>', '>'),
|
|
|
1725 ]
|
|
|
1726 for before, after in replacements:
|
|
|
1727 text = text.replace(before, after)
|
|
|
1728 hashed = _hash_text(text)
|
|
|
1729 self._escape_table[text] = hashed
|
|
|
1730 return hashed
|
|
|
1731
|
|
|
1732 _strong_re = re.compile(r"(\*\*|__)(?=\S)(.+?[*_]*)(?<=\S)\1", re.S)
|
|
|
1733 _em_re = re.compile(r"(\*|_)(?=\S)(.+?)(?<=\S)\1", re.S)
|
|
|
1734 _code_friendly_strong_re = re.compile(r"\*\*(?=\S)(.+?[*_]*)(?<=\S)\*\*", re.S)
|
|
|
1735 _code_friendly_em_re = re.compile(r"\*(?=\S)(.+?)(?<=\S)\*", re.S)
|
|
|
1736 def _do_italics_and_bold(self, text):
|
|
|
1737 # <strong> must go first:
|
|
|
1738 if "code-friendly" in self.extras:
|
|
|
1739 text = self._code_friendly_strong_re.sub(r"<strong>\1</strong>", text)
|
|
|
1740 text = self._code_friendly_em_re.sub(r"<em>\1</em>", text)
|
|
|
1741 else:
|
|
|
1742 text = self._strong_re.sub(r"<strong>\2</strong>", text)
|
|
|
1743 text = self._em_re.sub(r"<em>\2</em>", text)
|
|
|
1744 return text
|
|
|
1745
|
|
|
1746 # "smarty-pants" extra: Very liberal in interpreting a single prime as an
|
|
|
1747 # apostrophe; e.g. ignores the fact that "round", "bout", "twer", and
|
|
|
1748 # "twixt" can be written without an initial apostrophe. This is fine because
|
|
|
1749 # using scare quotes (single quotation marks) is rare.
|
|
|
1750 _apostrophe_year_re = re.compile(r"'(\d\d)(?=(\s|,|;|\.|\?|!|$))")
|
|
|
1751 _contractions = ["tis", "twas", "twer", "neath", "o", "n",
|
|
|
1752 "round", "bout", "twixt", "nuff", "fraid", "sup"]
|
|
|
1753 def _do_smart_contractions(self, text):
|
|
|
1754 text = self._apostrophe_year_re.sub(r"’\1", text)
|
|
|
1755 for c in self._contractions:
|
|
|
1756 text = text.replace("'%s" % c, "’%s" % c)
|
|
|
1757 text = text.replace("'%s" % c.capitalize(),
|
|
|
1758 "’%s" % c.capitalize())
|
|
|
1759 return text
|
|
|
1760
|
|
|
1761 # Substitute double-quotes before single-quotes.
|
|
|
1762 _opening_single_quote_re = re.compile(r"(?<!\S)'(?=\S)")
|
|
|
1763 _opening_double_quote_re = re.compile(r'(?<!\S)"(?=\S)')
|
|
|
1764 _closing_single_quote_re = re.compile(r"(?<=\S)'")
|
|
|
1765 _closing_double_quote_re = re.compile(r'(?<=\S)"(?=(\s|,|;|\.|\?|!|$))')
|
|
|
1766 def _do_smart_punctuation(self, text):
|
|
|
1767 """Fancifies 'single quotes', "double quotes", and apostrophes.
|
|
|
1768 Converts --, ---, and ... into en dashes, em dashes, and ellipses.
|
|
|
1769
|
|
|
1770 Inspiration is: <http://daringfireball.net/projects/smartypants/>
|
|
|
1771 See "test/tm-cases/smarty_pants.text" for a full discussion of the
|
|
|
1772 support here and
|
|
|
1773 <http://code.google.com/p/python-markdown2/issues/detail?id=42> for a
|
|
|
1774 discussion of some diversion from the original SmartyPants.
|
|
|
1775 """
|
|
|
1776 if "'" in text: # guard for perf
|
|
|
1777 text = self._do_smart_contractions(text)
|
|
|
1778 text = self._opening_single_quote_re.sub("‘", text)
|
|
|
1779 text = self._closing_single_quote_re.sub("’", text)
|
|
|
1780
|
|
|
1781 if '"' in text: # guard for perf
|
|
|
1782 text = self._opening_double_quote_re.sub("“", text)
|
|
|
1783 text = self._closing_double_quote_re.sub("”", text)
|
|
|
1784
|
|
|
1785 text = text.replace("---", "—")
|
|
|
1786 text = text.replace("--", "–")
|
|
|
1787 text = text.replace("...", "…")
|
|
|
1788 text = text.replace(" . . . ", "…")
|
|
|
1789 text = text.replace(". . .", "…")
|
|
|
1790 return text
|
|
|
1791
|
|
|
1792 _block_quote_re = re.compile(r'''
|
|
|
1793 ( # Wrap whole match in \1
|
|
|
1794 (
|
|
|
1795 ^[ \t]*>[ \t]? # '>' at the start of a line
|
|
|
1796 .+\n # rest of the first line
|
|
|
1797 (.+\n)* # subsequent consecutive lines
|
|
|
1798 \n* # blanks
|
|
|
1799 )+
|
|
|
1800 )
|
|
|
1801 ''', re.M | re.X)
|
|
|
1802 _bq_one_level_re = re.compile('^[ \t]*>[ \t]?', re.M);
|
|
|
1803
|
|
|
1804 _html_pre_block_re = re.compile(r'(\s*<pre>.+?</pre>)', re.S)
|
|
|
1805 def _dedent_two_spaces_sub(self, match):
|
|
|
1806 return re.sub(r'(?m)^ ', '', match.group(1))
|
|
|
1807
|
|
|
1808 def _block_quote_sub(self, match):
|
|
|
1809 bq = match.group(1)
|
|
|
1810 bq = self._bq_one_level_re.sub('', bq) # trim one level of quoting
|
|
|
1811 bq = self._ws_only_line_re.sub('', bq) # trim whitespace-only lines
|
|
|
1812 bq = self._run_block_gamut(bq) # recurse
|
|
|
1813
|
|
|
1814 bq = re.sub('(?m)^', ' ', bq)
|
|
|
1815 # These leading spaces screw with <pre> content, so we need to fix that:
|
|
|
1816 bq = self._html_pre_block_re.sub(self._dedent_two_spaces_sub, bq)
|
|
|
1817
|
|
|
1818 return "<blockquote>\n%s\n</blockquote>\n\n" % bq
|
|
|
1819
|
|
|
1820 def _do_block_quotes(self, text):
|
|
|
1821 if '>' not in text:
|
|
|
1822 return text
|
|
|
1823 return self._block_quote_re.sub(self._block_quote_sub, text)
|
|
|
1824
|
|
|
1825 def _form_paragraphs(self, text):
|
|
|
1826 # Strip leading and trailing lines:
|
|
|
1827 text = text.strip('\n')
|
|
|
1828
|
|
|
1829 # Wrap <p> tags.
|
|
|
1830 grafs = []
|
|
|
1831 for i, graf in enumerate(re.split(r"\n{2,}", text)):
|
|
|
1832 if graf in self.html_blocks:
|
|
|
1833 # Unhashify HTML blocks
|
|
|
1834 grafs.append(self.html_blocks[graf])
|
|
|
1835 else:
|
|
|
1836 cuddled_list = None
|
|
|
1837 if "cuddled-lists" in self.extras:
|
|
|
1838 # Need to put back trailing '\n' for `_list_item_re`
|
|
|
1839 # match at the end of the paragraph.
|
|
|
1840 li = self._list_item_re.search(graf + '\n')
|
|
|
1841 # Two of the same list marker in this paragraph: a likely
|
|
|
1842 # candidate for a list cuddled to preceding paragraph
|
|
|
1843 # text (issue 33). Note the `[-1]` is a quick way to
|
|
|
1844 # consider numeric bullets (e.g. "1." and "2.") to be
|
|
|
1845 # equal.
|
|
|
1846 if (li and len(li.group(2)) <= 3 and li.group("next_marker")
|
|
|
1847 and li.group("marker")[-1] == li.group("next_marker")[-1]):
|
|
|
1848 start = li.start()
|
|
|
1849 cuddled_list = self._do_lists(graf[start:]).rstrip("\n")
|
|
|
1850 assert cuddled_list.startswith("<ul>") or cuddled_list.startswith("<ol>")
|
|
|
1851 graf = graf[:start]
|
|
|
1852
|
|
|
1853 # Wrap <p> tags.
|
|
|
1854 graf = self._run_span_gamut(graf)
|
|
|
1855 grafs.append("<p>" + graf.lstrip(" \t") + "</p>")
|
|
|
1856
|
|
|
1857 if cuddled_list:
|
|
|
1858 grafs.append(cuddled_list)
|
|
|
1859
|
|
|
1860 return "\n\n".join(grafs)
|
|
|
1861
|
|
|
1862 def _add_footnotes(self, text):
|
|
|
1863 if self.footnotes:
|
|
|
1864 footer = [
|
|
|
1865 '<div class="footnotes">',
|
|
|
1866 '<hr' + self.empty_element_suffix,
|
|
|
1867 '<ol>',
|
|
|
1868 ]
|
|
|
1869 for i, id in enumerate(self.footnote_ids):
|
|
|
1870 if i != 0:
|
|
|
1871 footer.append('')
|
|
|
1872 footer.append('<li id="fn-%s">' % id)
|
|
|
1873 footer.append(self._run_block_gamut(self.footnotes[id]))
|
|
|
1874 backlink = ('<a href="#fnref-%s" '
|
|
|
1875 'class="footnoteBackLink" '
|
|
|
1876 'title="Jump back to footnote %d in the text.">'
|
|
|
1877 '↩</a>' % (id, i+1))
|
|
|
1878 if footer[-1].endswith("</p>"):
|
|
|
1879 footer[-1] = footer[-1][:-len("</p>")] \
|
|
|
1880 + ' ' + backlink + "</p>"
|
|
|
1881 else:
|
|
|
1882 footer.append("\n<p>%s</p>" % backlink)
|
|
|
1883 footer.append('</li>')
|
|
|
1884 footer.append('</ol>')
|
|
|
1885 footer.append('</div>')
|
|
|
1886 return text + '\n\n' + '\n'.join(footer)
|
|
|
1887 else:
|
|
|
1888 return text
|
|
|
1889
|
|
|
1890 # Ampersand-encoding based entirely on Nat Irons's Amputator MT plugin:
|
|
|
1891 # http://bumppo.net/projects/amputator/
|
|
|
1892 _ampersand_re = re.compile(r'&(?!#?[xX]?(?:[0-9a-fA-F]+|\w+);)')
|
|
|
1893 _naked_lt_re = re.compile(r'<(?![a-z/?\$!])', re.I)
|
|
|
1894 _naked_gt_re = re.compile(r'''(?<![a-z0-9?!/'"-])>''', re.I)
|
|
|
1895
|
|
|
1896 def _encode_amps_and_angles(self, text):
|
|
|
1897 # Smart processing for ampersands and angle brackets that need
|
|
|
1898 # to be encoded.
|
|
|
1899 text = self._ampersand_re.sub('&', text)
|
|
|
1900
|
|
|
1901 # Encode naked <'s
|
|
|
1902 text = self._naked_lt_re.sub('<', text)
|
|
|
1903
|
|
|
1904 # Encode naked >'s
|
|
|
1905 # Note: Other markdown implementations (e.g. Markdown.pl, PHP
|
|
|
1906 # Markdown) don't do this.
|
|
|
1907 text = self._naked_gt_re.sub('>', text)
|
|
|
1908 return text
|
|
|
1909
|
|
|
1910 def _encode_backslash_escapes(self, text):
|
|
|
1911 for ch, escape in list(self._escape_table.items()):
|
|
|
1912 text = text.replace("\\"+ch, escape)
|
|
|
1913 return text
|
|
|
1914
|
|
|
1915 _auto_link_re = re.compile(r'<((https?|ftp):[^\'">\s]+)>', re.I)
|
|
|
1916 def _auto_link_sub(self, match):
|
|
|
1917 g1 = match.group(1)
|
|
|
1918 return '<a href="%s">%s</a>' % (g1, g1)
|
|
|
1919
|
|
|
1920 _auto_email_link_re = re.compile(r"""
|
|
|
1921 <
|
|
|
1922 (?:mailto:)?
|
|
|
1923 (
|
|
|
1924 [-.\w]+
|
|
|
1925 \@
|
|
|
1926 [-\w]+(\.[-\w]+)*\.[a-z]+
|
|
|
1927 )
|
|
|
1928 >
|
|
|
1929 """, re.I | re.X | re.U)
|
|
|
1930 def _auto_email_link_sub(self, match):
|
|
|
1931 return self._encode_email_address(
|
|
|
1932 self._unescape_special_chars(match.group(1)))
|
|
|
1933
|
|
|
1934 def _do_auto_links(self, text):
|
|
|
1935 text = self._auto_link_re.sub(self._auto_link_sub, text)
|
|
|
1936 text = self._auto_email_link_re.sub(self._auto_email_link_sub, text)
|
|
|
1937 return text
|
|
|
1938
|
|
|
1939 def _encode_email_address(self, addr):
|
|
|
1940 # Input: an email address, e.g. "foo@example.com"
|
|
|
1941 #
|
|
|
1942 # Output: the email address as a mailto link, with each character
|
|
|
1943 # of the address encoded as either a decimal or hex entity, in
|
|
|
1944 # the hopes of foiling most address harvesting spam bots. E.g.:
|
|
|
1945 #
|
|
|
1946 # <a href="mailto:foo@e
|
|
|
1947 # xample.com">foo
|
|
|
1948 # @example.com</a>
|
|
|
1949 #
|
|
|
1950 # Based on a filter by Matthew Wickline, posted to the BBEdit-Talk
|
|
|
1951 # mailing list: <http://tinyurl.com/yu7ue>
|
|
|
1952 chars = [_xml_encode_email_char_at_random(ch)
|
|
|
1953 for ch in "mailto:" + addr]
|
|
|
1954 # Strip the mailto: from the visible part.
|
|
|
1955 addr = '<a href="%s">%s</a>' \
|
|
|
1956 % (''.join(chars), ''.join(chars[7:]))
|
|
|
1957 return addr
|
|
|
1958
|
|
|
1959 def _do_link_patterns(self, text):
|
|
|
1960 """Caveat emptor: there isn't much guarding against link
|
|
|
1961 patterns being formed inside other standard Markdown links, e.g.
|
|
|
1962 inside a [link def][like this].
|
|
|
1963
|
|
|
1964 Dev Notes: *Could* consider prefixing regexes with a negative
|
|
|
1965 lookbehind assertion to attempt to guard against this.
|
|
|
1966 """
|
|
|
1967 link_from_hash = {}
|
|
|
1968 for regex, repl in self.link_patterns:
|
|
|
1969 replacements = []
|
|
|
1970 for match in regex.finditer(text):
|
|
|
1971 if hasattr(repl, "__call__"):
|
|
|
1972 href = repl(match)
|
|
|
1973 else:
|
|
|
1974 href = match.expand(repl)
|
|
|
1975 replacements.append((match.span(), href))
|
|
|
1976 for (start, end), href in reversed(replacements):
|
|
|
1977 escaped_href = (
|
|
|
1978 href.replace('"', '"') # b/c of attr quote
|
|
|
1979 # To avoid markdown <em> and <strong>:
|
|
|
1980 .replace('*', self._escape_table['*'])
|
|
|
1981 .replace('_', self._escape_table['_']))
|
|
|
1982 link = '<a href="%s">%s</a>' % (escaped_href, text[start:end])
|
|
|
1983 hash = _hash_text(link)
|
|
|
1984 link_from_hash[hash] = link
|
|
|
1985 text = text[:start] + hash + text[end:]
|
|
|
1986 for hash, link in list(link_from_hash.items()):
|
|
|
1987 text = text.replace(hash, link)
|
|
|
1988 return text
|
|
|
1989
|
|
|
1990 def _unescape_special_chars(self, text):
|
|
|
1991 # Swap back in all the special characters we've hidden.
|
|
|
1992 for ch, hash in list(self._escape_table.items()):
|
|
|
1993 text = text.replace(hash, ch)
|
|
|
1994 return text
|
|
|
1995
|
|
|
1996 def _outdent(self, text):
|
|
|
1997 # Remove one level of line-leading tabs or spaces
|
|
|
1998 return self._outdent_re.sub('', text)
|
|
|
1999
|
|
|
2000
|
|
|
2001 class MarkdownWithExtras(Markdown):
|
|
|
2002 """A markdowner class that enables most extras:
|
|
|
2003
|
|
|
2004 - footnotes
|
|
|
2005 - code-color (only has effect if 'pygments' Python module on path)
|
|
|
2006
|
|
|
2007 These are not included:
|
|
|
2008 - pyshell (specific to Python-related documenting)
|
|
|
2009 - code-friendly (because it *disables* part of the syntax)
|
|
|
2010 - link-patterns (because you need to specify some actual
|
|
|
2011 link-patterns anyway)
|
|
|
2012 """
|
|
|
2013 extras = ["footnotes", "code-color"]
|
|
|
2014
|
|
|
2015
|
|
|
2016 #---- internal support functions
|
|
|
2017
|
|
|
2018 class UnicodeWithAttrs(unicode):
|
|
|
2019 """A subclass of unicode used for the return value of conversion to
|
|
|
2020 possibly attach some attributes. E.g. the "toc_html" attribute when
|
|
|
2021 the "toc" extra is used.
|
|
|
2022 """
|
|
|
2023 metadata = None
|
|
|
2024 _toc = None
|
|
|
2025 def toc_html(self):
|
|
|
2026 """Return the HTML for the current TOC.
|
|
|
2027
|
|
|
2028 This expects the `_toc` attribute to have been set on this instance.
|
|
|
2029 """
|
|
|
2030 if self._toc is None:
|
|
|
2031 return None
|
|
|
2032
|
|
|
2033 def indent():
|
|
|
2034 return ' ' * (len(h_stack) - 1)
|
|
|
2035 lines = []
|
|
|
2036 h_stack = [0] # stack of header-level numbers
|
|
|
2037 for level, id, name in self._toc:
|
|
|
2038 if level > h_stack[-1]:
|
|
|
2039 lines.append("%s<ul>" % indent())
|
|
|
2040 h_stack.append(level)
|
|
|
2041 elif level == h_stack[-1]:
|
|
|
2042 lines[-1] += "</li>"
|
|
|
2043 else:
|
|
|
2044 while level < h_stack[-1]:
|
|
|
2045 h_stack.pop()
|
|
|
2046 if not lines[-1].endswith("</li>"):
|
|
|
2047 lines[-1] += "</li>"
|
|
|
2048 lines.append("%s</ul></li>" % indent())
|
|
|
2049 lines.append('%s<li><a href="#%s">%s</a>' % (
|
|
|
2050 indent(), id, name))
|
|
|
2051 while len(h_stack) > 1:
|
|
|
2052 h_stack.pop()
|
|
|
2053 if not lines[-1].endswith("</li>"):
|
|
|
2054 lines[-1] += "</li>"
|
|
|
2055 lines.append("%s</ul>" % indent())
|
|
|
2056 return '\n'.join(lines) + '\n'
|
|
|
2057 toc_html = property(toc_html)
|
|
|
2058
|
|
|
2059 ## {{{ http://code.activestate.com/recipes/577257/ (r1)
|
|
|
2060 _slugify_strip_re = re.compile(r'[^\w\s-]')
|
|
|
2061 _slugify_hyphenate_re = re.compile(r'[-\s]+')
|
|
|
2062 def _slugify(value):
|
|
|
2063 """
|
|
|
2064 Normalizes string, converts to lowercase, removes non-alpha characters,
|
|
|
2065 and converts spaces to hyphens.
|
|
|
2066
|
|
|
2067 From Django's "django/template/defaultfilters.py".
|
|
|
2068 """
|
|
|
2069 import unicodedata
|
|
|
2070 value = unicodedata.normalize('NFKD', value).encode('ascii', 'ignore').decode()
|
|
|
2071 value = _slugify_strip_re.sub('', value).strip().lower()
|
|
|
2072 return _slugify_hyphenate_re.sub('-', value)
|
|
|
2073 ## end of http://code.activestate.com/recipes/577257/ }}}
|
|
|
2074
|
|
|
2075
|
|
|
2076 # From http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/52549
|
|
|
2077 def _curry(*args, **kwargs):
|
|
|
2078 function, args = args[0], args[1:]
|
|
|
2079 def result(*rest, **kwrest):
|
|
|
2080 combined = kwargs.copy()
|
|
|
2081 combined.update(kwrest)
|
|
|
2082 return function(*args + rest, **combined)
|
|
|
2083 return result
|
|
|
2084
|
|
|
2085 # Recipe: regex_from_encoded_pattern (1.0)
|
|
|
2086 def _regex_from_encoded_pattern(s):
|
|
|
2087 """'foo' -> re.compile(re.escape('foo'))
|
|
|
2088 '/foo/' -> re.compile('foo')
|
|
|
2089 '/foo/i' -> re.compile('foo', re.I)
|
|
|
2090 """
|
|
|
2091 if s.startswith('/') and s.rfind('/') != 0:
|
|
|
2092 # Parse it: /PATTERN/FLAGS
|
|
|
2093 idx = s.rfind('/')
|
|
|
2094 pattern, flags_str = s[1:idx], s[idx+1:]
|
|
|
2095 flag_from_char = {
|
|
|
2096 "i": re.IGNORECASE,
|
|
|
2097 "l": re.LOCALE,
|
|
|
2098 "s": re.DOTALL,
|
|
|
2099 "m": re.MULTILINE,
|
|
|
2100 "u": re.UNICODE,
|
|
|
2101 }
|
|
|
2102 flags = 0
|
|
|
2103 for char in flags_str:
|
|
|
2104 try:
|
|
|
2105 flags |= flag_from_char[char]
|
|
|
2106 except KeyError:
|
|
|
2107 raise ValueError("unsupported regex flag: '%s' in '%s' "
|
|
|
2108 "(must be one of '%s')"
|
|
|
2109 % (char, s, ''.join(list(flag_from_char.keys()))))
|
|
|
2110 return re.compile(s[1:idx], flags)
|
|
|
2111 else: # not an encoded regex
|
|
|
2112 return re.compile(re.escape(s))
|
|
|
2113
|
|
|
2114 # Recipe: dedent (0.1.2)
|
|
|
2115 def _dedentlines(lines, tabsize=8, skip_first_line=False):
|
|
|
2116 """_dedentlines(lines, tabsize=8, skip_first_line=False) -> dedented lines
|
|
|
2117
|
|
|
2118 "lines" is a list of lines to dedent.
|
|
|
2119 "tabsize" is the tab width to use for indent width calculations.
|
|
|
2120 "skip_first_line" is a boolean indicating if the first line should
|
|
|
2121 be skipped for calculating the indent width and for dedenting.
|
|
|
2122 This is sometimes useful for docstrings and similar.
|
|
|
2123
|
|
|
2124 Same as dedent() except operates on a sequence of lines. Note: the
|
|
|
2125 lines list is modified **in-place**.
|
|
|
2126 """
|
|
|
2127 DEBUG = False
|
|
|
2128 if DEBUG:
|
|
|
2129 print("dedent: dedent(..., tabsize=%d, skip_first_line=%r)"\
|
|
|
2130 % (tabsize, skip_first_line))
|
|
|
2131 indents = []
|
|
|
2132 margin = None
|
|
|
2133 for i, line in enumerate(lines):
|
|
|
2134 if i == 0 and skip_first_line: continue
|
|
|
2135 indent = 0
|
|
|
2136 for ch in line:
|
|
|
2137 if ch == ' ':
|
|
|
2138 indent += 1
|
|
|
2139 elif ch == '\t':
|
|
|
2140 indent += tabsize - (indent % tabsize)
|
|
|
2141 elif ch in '\r\n':
|
|
|
2142 continue # skip all-whitespace lines
|
|
|
2143 else:
|
|
|
2144 break
|
|
|
2145 else:
|
|
|
2146 continue # skip all-whitespace lines
|
|
|
2147 if DEBUG: print("dedent: indent=%d: %r" % (indent, line))
|
|
|
2148 if margin is None:
|
|
|
2149 margin = indent
|
|
|
2150 else:
|
|
|
2151 margin = min(margin, indent)
|
|
|
2152 if DEBUG: print("dedent: margin=%r" % margin)
|
|
|
2153
|
|
|
2154 if margin is not None and margin > 0:
|
|
|
2155 for i, line in enumerate(lines):
|
|
|
2156 if i == 0 and skip_first_line: continue
|
|
|
2157 removed = 0
|
|
|
2158 for j, ch in enumerate(line):
|
|
|
2159 if ch == ' ':
|
|
|
2160 removed += 1
|
|
|
2161 elif ch == '\t':
|
|
|
2162 removed += tabsize - (removed % tabsize)
|
|
|
2163 elif ch in '\r\n':
|
|
|
2164 if DEBUG: print("dedent: %r: EOL -> strip up to EOL" % line)
|
|
|
2165 lines[i] = lines[i][j:]
|
|
|
2166 break
|
|
|
2167 else:
|
|
|
2168 raise ValueError("unexpected non-whitespace char %r in "
|
|
|
2169 "line %r while removing %d-space margin"
|
|
|
2170 % (ch, line, margin))
|
|
|
2171 if DEBUG:
|
|
|
2172 print("dedent: %r: %r -> removed %d/%d"\
|
|
|
2173 % (line, ch, removed, margin))
|
|
|
2174 if removed == margin:
|
|
|
2175 lines[i] = lines[i][j+1:]
|
|
|
2176 break
|
|
|
2177 elif removed > margin:
|
|
|
2178 lines[i] = ' '*(removed-margin) + lines[i][j+1:]
|
|
|
2179 break
|
|
|
2180 else:
|
|
|
2181 if removed:
|
|
|
2182 lines[i] = lines[i][removed:]
|
|
|
2183 return lines
|
|
|
2184
|
|
|
2185 def _dedent(text, tabsize=8, skip_first_line=False):
|
|
|
2186 """_dedent(text, tabsize=8, skip_first_line=False) -> dedented text
|
|
|
2187
|
|
|
2188 "text" is the text to dedent.
|
|
|
2189 "tabsize" is the tab width to use for indent width calculations.
|
|
|
2190 "skip_first_line" is a boolean indicating if the first line should
|
|
|
2191 be skipped for calculating the indent width and for dedenting.
|
|
|
2192 This is sometimes useful for docstrings and similar.
|
|
|
2193
|
|
|
2194 textwrap.dedent(s), but don't expand tabs to spaces
|
|
|
2195 """
|
|
|
2196 lines = text.splitlines(1)
|
|
|
2197 _dedentlines(lines, tabsize=tabsize, skip_first_line=skip_first_line)
|
|
|
2198 return ''.join(lines)
|
|
|
2199
|
|
|
2200
|
|
|
2201 class _memoized(object):
|
|
|
2202 """Decorator that caches a function's return value each time it is called.
|
|
|
2203 If called later with the same arguments, the cached value is returned, and
|
|
|
2204 not re-evaluated.
|
|
|
2205
|
|
|
2206 http://wiki.python.org/moin/PythonDecoratorLibrary
|
|
|
2207 """
|
|
|
2208 def __init__(self, func):
|
|
|
2209 self.func = func
|
|
|
2210 self.cache = {}
|
|
|
2211 def __call__(self, *args):
|
|
|
2212 try:
|
|
|
2213 return self.cache[args]
|
|
|
2214 except KeyError:
|
|
|
2215 self.cache[args] = value = self.func(*args)
|
|
|
2216 return value
|
|
|
2217 except TypeError:
|
|
|
2218 # uncachable -- for instance, passing a list as an argument.
|
|
|
2219 # Better to not cache than to blow up entirely.
|
|
|
2220 return self.func(*args)
|
|
|
2221 def __repr__(self):
|
|
|
2222 """Return the function's docstring."""
|
|
|
2223 return self.func.__doc__
|
|
|
2224
|
|
|
2225
|
|
|
2226 def _xml_oneliner_re_from_tab_width(tab_width):
|
|
|
2227 """Standalone XML processing instruction regex."""
|
|
|
2228 return re.compile(r"""
|
|
|
2229 (?:
|
|
|
2230 (?<=\n\n) # Starting after a blank line
|
|
|
2231 | # or
|
|
|
2232 \A\n? # the beginning of the doc
|
|
|
2233 )
|
|
|
2234 ( # save in $1
|
|
|
2235 [ ]{0,%d}
|
|
|
2236 (?:
|
|
|
2237 <\?\w+\b\s+.*?\?> # XML processing instruction
|
|
|
2238 |
|
|
|
2239 <\w+:\w+\b\s+.*?/> # namespaced single tag
|
|
|
2240 )
|
|
|
2241 [ \t]*
|
|
|
2242 (?=\n{2,}|\Z) # followed by a blank line or end of document
|
|
|
2243 )
|
|
|
2244 """ % (tab_width - 1), re.X)
|
|
|
2245 _xml_oneliner_re_from_tab_width = _memoized(_xml_oneliner_re_from_tab_width)
|
|
|
2246
|
|
|
2247 def _hr_tag_re_from_tab_width(tab_width):
|
|
|
2248 return re.compile(r"""
|
|
|
2249 (?:
|
|
|
2250 (?<=\n\n) # Starting after a blank line
|
|
|
2251 | # or
|
|
|
2252 \A\n? # the beginning of the doc
|
|
|
2253 )
|
|
|
2254 ( # save in \1
|
|
|
2255 [ ]{0,%d}
|
|
|
2256 <(hr) # start tag = \2
|
|
|
2257 \b # word break
|
|
|
2258 ([^<>])*? #
|
|
|
2259 /?> # the matching end tag
|
|
|
2260 [ \t]*
|
|
|
2261 (?=\n{2,}|\Z) # followed by a blank line or end of document
|
|
|
2262 )
|
|
|
2263 """ % (tab_width - 1), re.X)
|
|
|
2264 _hr_tag_re_from_tab_width = _memoized(_hr_tag_re_from_tab_width)
|
|
|
2265
|
|
|
2266
|
|
|
2267 def _xml_escape_attr(attr, skip_single_quote=True):
|
|
|
2268 """Escape the given string for use in an HTML/XML tag attribute.
|
|
|
2269
|
|
|
2270 By default this doesn't bother with escaping `'` to `'`, presuming that
|
|
|
2271 the tag attribute is surrounded by double quotes.
|
|
|
2272 """
|
|
|
2273 escaped = (attr
|
|
|
2274 .replace('&', '&')
|
|
|
2275 .replace('"', '"')
|
|
|
2276 .replace('<', '<')
|
|
|
2277 .replace('>', '>'))
|
|
|
2278 if not skip_single_quote:
|
|
|
2279 escaped = escaped.replace("'", "'")
|
|
|
2280 return escaped
|
|
|
2281
|
|
|
2282
|
|
|
2283 def _xml_encode_email_char_at_random(ch):
|
|
|
2284 r = random()
|
|
|
2285 # Roughly 10% raw, 45% hex, 45% dec.
|
|
|
2286 # '@' *must* be encoded. I [John Gruber] insist.
|
|
|
2287 # Issue 26: '_' must be encoded.
|
|
|
2288 if r > 0.9 and ch not in "@_":
|
|
|
2289 return ch
|
|
|
2290 elif r < 0.45:
|
|
|
2291 # The [1:] is to drop leading '0': 0x63 -> x63
|
|
|
2292 return '&#%s;' % hex(ord(ch))[1:]
|
|
|
2293 else:
|
|
|
2294 return '&#%s;' % ord(ch)
|
|
|
2295
|
|
|
2296
|
|
|
2297
|
|
|
2298 #---- mainline
|
|
|
2299
|
|
|
2300 class _NoReflowFormatter(optparse.IndentedHelpFormatter):
|
|
|
2301 """An optparse formatter that does NOT reflow the description."""
|
|
|
2302 def format_description(self, description):
|
|
|
2303 return description or ""
|
|
|
2304
|
|
|
2305 def _test():
|
|
|
2306 import doctest
|
|
|
2307 doctest.testmod()
|
|
|
2308
|
|
|
2309 def main(argv=None):
|
|
|
2310 if argv is None:
|
|
|
2311 argv = sys.argv
|
|
|
2312 if not logging.root.handlers:
|
|
|
2313 logging.basicConfig()
|
|
|
2314
|
|
|
2315 usage = "usage: %prog [PATHS...]"
|
|
|
2316 version = "%prog "+__version__
|
|
|
2317 parser = optparse.OptionParser(prog="markdown2", usage=usage,
|
|
|
2318 version=version, description=cmdln_desc,
|
|
|
2319 formatter=_NoReflowFormatter())
|
|
|
2320 parser.add_option("-v", "--verbose", dest="log_level",
|
|
|
2321 action="store_const", const=logging.DEBUG,
|
|
|
2322 help="more verbose output")
|
|
|
2323 parser.add_option("--encoding",
|
|
|
2324 help="specify encoding of text content")
|
|
|
2325 parser.add_option("--html4tags", action="store_true", default=False,
|
|
|
2326 help="use HTML 4 style for empty element tags")
|
|
|
2327 parser.add_option("-s", "--safe", metavar="MODE", dest="safe_mode",
|
|
|
2328 help="sanitize literal HTML: 'escape' escapes "
|
|
|
2329 "HTML meta chars, 'replace' replaces with an "
|
|
|
2330 "[HTML_REMOVED] note")
|
|
|
2331 parser.add_option("-x", "--extras", action="append",
|
|
|
2332 help="Turn on specific extra features (not part of "
|
|
|
2333 "the core Markdown spec). See above.")
|
|
|
2334 parser.add_option("--use-file-vars",
|
|
|
2335 help="Look for and use Emacs-style 'markdown-extras' "
|
|
|
2336 "file var to turn on extras. See "
|
|
|
2337 "<https://github.com/trentm/python-markdown2/wiki/Extras>")
|
|
|
2338 parser.add_option("--link-patterns-file",
|
|
|
2339 help="path to a link pattern file")
|
|
|
2340 parser.add_option("--self-test", action="store_true",
|
|
|
2341 help="run internal self-tests (some doctests)")
|
|
|
2342 parser.add_option("--compare", action="store_true",
|
|
|
2343 help="run against Markdown.pl as well (for testing)")
|
|
|
2344 parser.set_defaults(log_level=logging.INFO, compare=False,
|
|
|
2345 encoding="utf-8", safe_mode=None, use_file_vars=False)
|
|
|
2346 opts, paths = parser.parse_args()
|
|
|
2347 log.setLevel(opts.log_level)
|
|
|
2348
|
|
|
2349 if opts.self_test:
|
|
|
2350 return _test()
|
|
|
2351
|
|
|
2352 if opts.extras:
|
|
|
2353 extras = {}
|
|
|
2354 for s in opts.extras:
|
|
|
2355 splitter = re.compile("[,;: ]+")
|
|
|
2356 for e in splitter.split(s):
|
|
|
2357 if '=' in e:
|
|
|
2358 ename, earg = e.split('=', 1)
|
|
|
2359 try:
|
|
|
2360 earg = int(earg)
|
|
|
2361 except ValueError:
|
|
|
2362 pass
|
|
|
2363 else:
|
|
|
2364 ename, earg = e, None
|
|
|
2365 extras[ename] = earg
|
|
|
2366 else:
|
|
|
2367 extras = None
|
|
|
2368
|
|
|
2369 if opts.link_patterns_file:
|
|
|
2370 link_patterns = []
|
|
|
2371 f = open(opts.link_patterns_file)
|
|
|
2372 try:
|
|
|
2373 for i, line in enumerate(f.readlines()):
|
|
|
2374 if not line.strip(): continue
|
|
|
2375 if line.lstrip().startswith("#"): continue
|
|
|
2376 try:
|
|
|
2377 pat, href = line.rstrip().rsplit(None, 1)
|
|
|
2378 except ValueError:
|
|
|
2379 raise MarkdownError("%s:%d: invalid link pattern line: %r"
|
|
|
2380 % (opts.link_patterns_file, i+1, line))
|
|
|
2381 link_patterns.append(
|
|
|
2382 (_regex_from_encoded_pattern(pat), href))
|
|
|
2383 finally:
|
|
|
2384 f.close()
|
|
|
2385 else:
|
|
|
2386 link_patterns = None
|
|
|
2387
|
|
|
2388 from os.path import join, dirname, abspath, exists
|
|
|
2389 markdown_pl = join(dirname(dirname(abspath(__file__))), "test",
|
|
|
2390 "Markdown.pl")
|
|
|
2391 if not paths:
|
|
|
2392 paths = ['-']
|
|
|
2393 for path in paths:
|
|
|
2394 if path == '-':
|
|
|
2395 text = sys.stdin.read()
|
|
|
2396 else:
|
|
|
2397 fp = codecs.open(path, 'r', opts.encoding)
|
|
|
2398 text = fp.read()
|
|
|
2399 fp.close()
|
|
|
2400 if opts.compare:
|
|
|
2401 from subprocess import Popen, PIPE
|
|
|
2402 print("==== Markdown.pl ====")
|
|
|
2403 p = Popen('perl %s' % markdown_pl, shell=True, stdin=PIPE, stdout=PIPE, close_fds=True)
|
|
|
2404 p.stdin.write(text.encode('utf-8'))
|
|
|
2405 p.stdin.close()
|
|
|
2406 perl_html = p.stdout.read().decode('utf-8')
|
|
|
2407 if py3:
|
|
|
2408 sys.stdout.write(perl_html)
|
|
|
2409 else:
|
|
|
2410 sys.stdout.write(perl_html.encode(
|
|
|
2411 sys.stdout.encoding or "utf-8", 'xmlcharrefreplace'))
|
|
|
2412 print("==== markdown2.py ====")
|
|
|
2413 html = markdown(text,
|
|
|
2414 html4tags=opts.html4tags,
|
|
|
2415 safe_mode=opts.safe_mode,
|
|
|
2416 extras=extras, link_patterns=link_patterns,
|
|
|
2417 use_file_vars=opts.use_file_vars)
|
|
|
2418 if py3:
|
|
|
2419 sys.stdout.write(html)
|
|
|
2420 else:
|
|
|
2421 sys.stdout.write(html.encode(
|
|
|
2422 sys.stdout.encoding or "utf-8", 'xmlcharrefreplace'))
|
|
|
2423 if extras and "toc" in extras:
|
|
|
2424 log.debug("toc_html: " +
|
|
|
2425 html.toc_html.encode(sys.stdout.encoding or "utf-8", 'xmlcharrefreplace'))
|
|
|
2426 if opts.compare:
|
|
|
2427 test_dir = join(dirname(dirname(abspath(__file__))), "test")
|
|
|
2428 if exists(join(test_dir, "test_markdown2.py")):
|
|
|
2429 sys.path.insert(0, test_dir)
|
|
|
2430 from test_markdown2 import norm_html_from_html
|
|
|
2431 norm_html = norm_html_from_html(html)
|
|
|
2432 norm_perl_html = norm_html_from_html(perl_html)
|
|
|
2433 else:
|
|
|
2434 norm_html = html
|
|
|
2435 norm_perl_html = perl_html
|
|
|
2436 print("==== match? %r ====" % (norm_perl_html == norm_html))
|
|
|
2437
|
|
|
2438
|
|
|
2439 if __name__ == "__main__":
|
|
|
2440 sys.exit( main(sys.argv) )
|
