|
1
|
|
|
# coding: utf-8 |
|
2
|
|
|
""" |
|
3
|
|
|
mistune |
|
4
|
|
|
~~~~~~~ |
|
5
|
|
|
|
|
6
|
|
|
The fastest markdown parser in pure Python with renderer feature. |
|
7
|
|
|
|
|
8
|
|
|
:copyright: (c) 2014 - 2017 by Hsiaoming Yang. |
|
9
|
|
|
""" |
|
10
|
|
|
|
|
11
|
|
|
import re |
|
12
|
|
|
import inspect |
|
13
|
|
|
|
|
14
|
|
|
__version__ = '0.8.3' |
|
15
|
|
|
__author__ = 'Hsiaoming Yang <[email protected]>' |
|
16
|
|
|
__all__ = [ |
|
17
|
|
|
'BlockGrammar', 'BlockLexer', |
|
18
|
|
|
'InlineGrammar', 'InlineLexer', |
|
19
|
|
|
'Renderer', 'Markdown', |
|
20
|
|
|
'markdown', 'escape', |
|
21
|
|
|
] |
|
22
|
|
|
|
|
23
|
|
|
|
|
24
|
|
|
_key_pattern = re.compile(r'\s+') |
|
25
|
|
|
_nonalpha_pattern = re.compile(r'\W') |
|
26
|
|
|
_escape_pattern = re.compile(r'&(?!#?\w+;)') |
|
27
|
|
|
_newline_pattern = re.compile(r'\r\n|\r') |
|
28
|
|
|
_block_quote_leading_pattern = re.compile(r'^ *> ?', flags=re.M) |
|
29
|
|
|
_block_code_leading_pattern = re.compile(r'^ {4}', re.M) |
|
30
|
|
|
_inline_tags = [ |
|
31
|
|
|
'a', 'em', 'strong', 'small', 's', 'cite', 'q', 'dfn', 'abbr', 'data', |
|
32
|
|
|
'time', 'code', 'var', 'samp', 'kbd', 'sub', 'sup', 'i', 'b', 'u', 'mark', |
|
33
|
|
|
'ruby', 'rt', 'rp', 'bdi', 'bdo', 'span', 'br', 'wbr', 'ins', 'del', |
|
34
|
|
|
'img', 'font', |
|
35
|
|
|
] |
|
36
|
|
|
_pre_tags = ['pre', 'script', 'style'] |
|
37
|
|
|
_valid_end = r'(?!:/|[^\w\s@]*@)\b' |
|
38
|
|
|
_valid_attr = r'''\s*[a-zA-Z\-](?:\=(?:"[^"]*"|'[^']*'|[^\s'">]+))?''' |
|
39
|
|
|
_block_tag = r'(?!(?:%s)\b)\w+%s' % ('|'.join(_inline_tags), _valid_end) |
|
40
|
|
|
_scheme_blacklist = ('javascript:', 'vbscript:') |
|
41
|
|
|
|
|
42
|
|
|
|
|
43
|
|
|
def _pure_pattern(regex): |
|
44
|
|
|
pattern = regex.pattern |
|
45
|
|
|
if pattern.startswith('^'): |
|
46
|
|
|
pattern = pattern[1:] |
|
47
|
|
|
return pattern |
|
48
|
|
|
|
|
49
|
|
|
|
|
50
|
|
|
def _keyify(key): |
|
51
|
|
|
key = escape(key.lower(), quote=True) |
|
52
|
|
|
return _key_pattern.sub(' ', key) |
|
53
|
|
|
|
|
54
|
|
|
|
|
55
|
|
|
def escape(text, quote=False, smart_amp=True): |
|
56
|
|
|
"""Replace special characters "&", "<" and ">" to HTML-safe sequences. |
|
57
|
|
|
|
|
58
|
|
|
The original cgi.escape will always escape "&", but you can control |
|
59
|
|
|
this one for a smart escape amp. |
|
60
|
|
|
|
|
61
|
|
|
:param quote: if set to True, " and ' will be escaped. |
|
62
|
|
|
:param smart_amp: if set to False, & will always be escaped. |
|
63
|
|
|
""" |
|
64
|
|
|
if smart_amp: |
|
65
|
|
|
text = _escape_pattern.sub('&', text) |
|
66
|
|
|
else: |
|
67
|
|
|
text = text.replace('&', '&') |
|
68
|
|
|
text = text.replace('<', '<') |
|
69
|
|
|
text = text.replace('>', '>') |
|
70
|
|
|
if quote: |
|
71
|
|
|
text = text.replace('"', '"') |
|
72
|
|
|
text = text.replace("'", ''') |
|
73
|
|
|
return text |
|
74
|
|
|
|
|
75
|
|
|
|
|
76
|
|
|
def escape_link(url): |
|
77
|
|
|
"""Remove dangerous URL schemes like javascript: and escape afterwards.""" |
|
78
|
|
|
lower_url = url.lower().strip('\x00\x1a \n\r\t') |
|
79
|
|
|
|
|
80
|
|
|
for scheme in _scheme_blacklist: |
|
81
|
|
|
if re.sub(r'[^A-Za-z0-9\/:]+', '', lower_url).startswith(scheme): |
|
82
|
|
|
return '' |
|
83
|
|
|
return escape(url, quote=True, smart_amp=False) |
|
84
|
|
|
|
|
85
|
|
|
|
|
86
|
|
|
def preprocessing(text, tab=4): |
|
87
|
|
|
text = _newline_pattern.sub('\n', text) |
|
88
|
|
|
text = text.expandtabs(tab) |
|
89
|
|
|
text = text.replace('\u2424', '\n') |
|
90
|
|
|
pattern = re.compile(r'^ +$', re.M) |
|
91
|
|
|
return pattern.sub('', text) |
|
92
|
|
|
|
|
93
|
|
|
|
|
94
|
|
|
class BlockGrammar(object): |
|
95
|
|
|
"""Grammars for block level tokens.""" |
|
96
|
|
|
|
|
97
|
|
|
def_links = re.compile( |
|
98
|
|
|
r'^ *\[([^^\]]+)\]: *' # [key]: |
|
99
|
|
|
r'<?([^\s>]+)>?' # <link> or link |
|
100
|
|
|
r'(?: +["(]([^\n]+)[")])? *(?:\n+|$)' |
|
101
|
|
|
) |
|
102
|
|
|
def_footnotes = re.compile( |
|
103
|
|
|
r'^\[\^([^\]]+)\]: *(' |
|
104
|
|
|
r'[^\n]*(?:\n+|$)' # [^key]: |
|
105
|
|
|
r'(?: {1,}[^\n]*(?:\n+|$))*' |
|
106
|
|
|
r')' |
|
107
|
|
|
) |
|
108
|
|
|
|
|
109
|
|
|
newline = re.compile(r'^\n+') |
|
110
|
|
|
block_code = re.compile(r'^( {4}[^\n]+\n*)+') |
|
111
|
|
|
fences = re.compile( |
|
112
|
|
|
r'^ *(`{3,}|~{3,}) *(\S+)? *\n' # ```lang |
|
113
|
|
|
r'([\s\S]+?)\s*' |
|
114
|
|
|
r'\1 *(?:\n+|$)' # ``` |
|
115
|
|
|
) |
|
116
|
|
|
hrule = re.compile(r'^ {0,3}[-*_](?: *[-*_]){2,} *(?:\n+|$)') |
|
117
|
|
|
heading = re.compile(r'^ *(#{1,6}) *([^\n]+?) *#* *(?:\n+|$)') |
|
118
|
|
|
lheading = re.compile(r'^([^\n]+)\n *(=|-)+ *(?:\n+|$)') |
|
119
|
|
|
block_quote = re.compile(r'^( *>[^\n]+(\n[^\n]+)*\n*)+') |
|
120
|
|
|
list_block = re.compile( |
|
121
|
|
|
r'^( *)(?=[*+-]|\d+\.)(([*+-])?(?:\d+\.)?) [\s\S]+?' |
|
122
|
|
|
r'(?:' |
|
123
|
|
|
r'\n+(?=\1?(?:[-*_] *){3,}(?:\n+|$))' # hrule |
|
124
|
|
|
r'|\n+(?=%s)' # def links |
|
125
|
|
|
r'|\n+(?=%s)' # def footnotes\ |
|
126
|
|
|
r'|\n+(?=\1(?(3)\d+\.|[*+-]) )' # heterogeneous bullet |
|
127
|
|
|
r'|\n{2,}' |
|
128
|
|
|
r'(?! )' |
|
129
|
|
|
r'(?!\1(?:[*+-]|\d+\.) )\n*' |
|
130
|
|
|
r'|' |
|
131
|
|
|
r'\s*$)' % ( |
|
132
|
|
|
_pure_pattern(def_links), |
|
133
|
|
|
_pure_pattern(def_footnotes), |
|
134
|
|
|
) |
|
135
|
|
|
) |
|
136
|
|
|
list_item = re.compile( |
|
137
|
|
|
r'^(( *)(?:[*+-]|\d+\.) [^\n]*' |
|
138
|
|
|
r'(?:\n(?!\2(?:[*+-]|\d+\.) )[^\n]*)*)', |
|
139
|
|
|
flags=re.M |
|
140
|
|
|
) |
|
141
|
|
|
list_bullet = re.compile(r'^ *(?:[*+-]|\d+\.) +') |
|
142
|
|
|
paragraph = re.compile( |
|
143
|
|
|
r'^((?:[^\n]+\n?(?!' |
|
144
|
|
|
r'%s|%s|%s|%s|%s|%s|%s|%s|%s' |
|
145
|
|
|
r'))+)\n*' % ( |
|
146
|
|
|
_pure_pattern(fences).replace(r'\1', r'\2'), |
|
147
|
|
|
_pure_pattern(list_block).replace(r'\1', r'\3'), |
|
148
|
|
|
_pure_pattern(hrule), |
|
149
|
|
|
_pure_pattern(heading), |
|
150
|
|
|
_pure_pattern(lheading), |
|
151
|
|
|
_pure_pattern(block_quote), |
|
152
|
|
|
_pure_pattern(def_links), |
|
153
|
|
|
_pure_pattern(def_footnotes), |
|
154
|
|
|
'<' + _block_tag, |
|
155
|
|
|
) |
|
156
|
|
|
) |
|
157
|
|
|
block_html = re.compile( |
|
158
|
|
|
r'^ *(?:%s|%s|%s) *(?:\n{2,}|\s*$)' % ( |
|
159
|
|
|
r'<!--[\s\S]*?-->', |
|
160
|
|
|
r'<(%s)((?:%s)*?)>([\s\S]*?)<\/\1>' % (_block_tag, _valid_attr), |
|
161
|
|
|
r'<%s(?:%s)*?\s*\/?>' % (_block_tag, _valid_attr), |
|
162
|
|
|
) |
|
163
|
|
|
) |
|
164
|
|
|
table = re.compile( |
|
165
|
|
|
r'^ *\|(.+)\n *\|( *[-:]+[-| :]*)\n((?: *\|.*(?:\n|$))*)\n*' |
|
166
|
|
|
) |
|
167
|
|
|
nptable = re.compile( |
|
168
|
|
|
r'^ *(\S.*\|.*)\n *([-:]+ *\|[-| :]*)\n((?:.*\|.*(?:\n|$))*)\n*' |
|
169
|
|
|
) |
|
170
|
|
|
text = re.compile(r'^[^\n]+') |
|
171
|
|
|
|
|
172
|
|
|
|
|
173
|
|
|
class BlockLexer(object): |
|
174
|
|
|
"""Block level lexer for block grammars.""" |
|
175
|
|
|
grammar_class = BlockGrammar |
|
176
|
|
|
|
|
177
|
|
|
default_rules = [ |
|
178
|
|
|
'newline', 'hrule', 'block_code', 'fences', 'heading', |
|
179
|
|
|
'nptable', 'lheading', 'block_quote', |
|
180
|
|
|
'list_block', 'block_html', 'def_links', |
|
181
|
|
|
'def_footnotes', 'table', 'paragraph', 'text' |
|
182
|
|
|
] |
|
183
|
|
|
|
|
184
|
|
|
list_rules = ( |
|
185
|
|
|
'newline', 'block_code', 'fences', 'lheading', 'hrule', |
|
186
|
|
|
'block_quote', 'list_block', 'block_html', 'text', |
|
187
|
|
|
) |
|
188
|
|
|
|
|
189
|
|
|
footnote_rules = ( |
|
190
|
|
|
'newline', 'block_code', 'fences', 'heading', |
|
191
|
|
|
'nptable', 'lheading', 'hrule', 'block_quote', |
|
192
|
|
|
'list_block', 'block_html', 'table', 'paragraph', 'text' |
|
193
|
|
|
) |
|
194
|
|
|
|
|
195
|
|
|
def __init__(self, rules=None, **kwargs): |
|
196
|
|
|
self.tokens = [] |
|
197
|
|
|
self.def_links = {} |
|
198
|
|
|
self.def_footnotes = {} |
|
199
|
|
|
|
|
200
|
|
|
if not rules: |
|
201
|
|
|
rules = self.grammar_class() |
|
202
|
|
|
|
|
203
|
|
|
self.rules = rules |
|
204
|
|
|
|
|
205
|
|
|
def __call__(self, text, rules=None): |
|
206
|
|
|
return self.parse(text, rules) |
|
207
|
|
|
|
|
208
|
|
|
def parse(self, text, rules=None): |
|
209
|
|
|
text = text.rstrip('\n') |
|
210
|
|
|
|
|
211
|
|
|
if not rules: |
|
212
|
|
|
rules = self.default_rules |
|
213
|
|
|
|
|
214
|
|
|
def manipulate(text): |
|
215
|
|
|
for key in rules: |
|
216
|
|
|
rule = getattr(self.rules, key) |
|
217
|
|
|
m = rule.match(text) |
|
218
|
|
|
if not m: |
|
219
|
|
|
continue |
|
220
|
|
|
getattr(self, 'parse_%s' % key)(m) |
|
221
|
|
|
return m |
|
222
|
|
|
return False # pragma: no cover |
|
223
|
|
|
|
|
224
|
|
|
while text: |
|
225
|
|
|
m = manipulate(text) |
|
226
|
|
|
if m is not False: |
|
227
|
|
|
text = text[len(m.group(0)):] |
|
228
|
|
|
continue |
|
229
|
|
|
if text: # pragma: no cover |
|
230
|
|
|
raise RuntimeError('Infinite loop at: %s' % text) |
|
231
|
|
|
return self.tokens |
|
232
|
|
|
|
|
233
|
|
|
def parse_newline(self, m): |
|
234
|
|
|
length = len(m.group(0)) |
|
235
|
|
|
if length > 1: |
|
236
|
|
|
self.tokens.append({'type': 'newline'}) |
|
237
|
|
|
|
|
238
|
|
|
def parse_block_code(self, m): |
|
239
|
|
|
# clean leading whitespace |
|
240
|
|
|
code = _block_code_leading_pattern.sub('', m.group(0)) |
|
241
|
|
|
self.tokens.append({ |
|
242
|
|
|
'type': 'code', |
|
243
|
|
|
'lang': None, |
|
244
|
|
|
'text': code, |
|
245
|
|
|
}) |
|
246
|
|
|
|
|
247
|
|
|
def parse_fences(self, m): |
|
248
|
|
|
self.tokens.append({ |
|
249
|
|
|
'type': 'code', |
|
250
|
|
|
'lang': m.group(2), |
|
251
|
|
|
'text': m.group(3), |
|
252
|
|
|
}) |
|
253
|
|
|
|
|
254
|
|
|
def parse_heading(self, m): |
|
255
|
|
|
self.tokens.append({ |
|
256
|
|
|
'type': 'heading', |
|
257
|
|
|
'level': len(m.group(1)), |
|
258
|
|
|
'text': m.group(2), |
|
259
|
|
|
}) |
|
260
|
|
|
|
|
261
|
|
|
def parse_lheading(self, m): |
|
262
|
|
|
"""Parse setext heading.""" |
|
263
|
|
|
self.tokens.append({ |
|
264
|
|
|
'type': 'heading', |
|
265
|
|
|
'level': 1 if m.group(2) == '=' else 2, |
|
266
|
|
|
'text': m.group(1), |
|
267
|
|
|
}) |
|
268
|
|
|
|
|
269
|
|
|
def parse_hrule(self, m): |
|
270
|
|
|
self.tokens.append({'type': 'hrule'}) |
|
271
|
|
|
|
|
272
|
|
|
def parse_list_block(self, m): |
|
273
|
|
|
bull = m.group(2) |
|
274
|
|
|
self.tokens.append({ |
|
275
|
|
|
'type': 'list_start', |
|
276
|
|
|
'ordered': '.' in bull, |
|
277
|
|
|
}) |
|
278
|
|
|
cap = m.group(0) |
|
279
|
|
|
self._process_list_item(cap, bull) |
|
280
|
|
|
self.tokens.append({'type': 'list_end'}) |
|
281
|
|
|
|
|
282
|
|
|
def _process_list_item(self, cap, bull): |
|
283
|
|
|
cap = self.rules.list_item.findall(cap) |
|
284
|
|
|
|
|
285
|
|
|
_next = False |
|
286
|
|
|
length = len(cap) |
|
287
|
|
|
|
|
288
|
|
|
for i in range(length): |
|
289
|
|
|
item = cap[i][0] |
|
290
|
|
|
|
|
291
|
|
|
# remove the bullet |
|
292
|
|
|
space = len(item) |
|
293
|
|
|
item = self.rules.list_bullet.sub('', item) |
|
294
|
|
|
|
|
295
|
|
|
# outdent |
|
296
|
|
|
if '\n ' in item: |
|
297
|
|
|
space = space - len(item) |
|
298
|
|
|
pattern = re.compile(r'^ {1,%d}' % space, flags=re.M) |
|
299
|
|
|
item = pattern.sub('', item) |
|
300
|
|
|
|
|
301
|
|
|
# determine whether item is loose or not |
|
302
|
|
|
loose = _next |
|
303
|
|
|
if not loose and re.search(r'\n\n(?!\s*$)', item): |
|
304
|
|
|
loose = True |
|
305
|
|
|
|
|
306
|
|
|
rest = len(item) |
|
307
|
|
|
if i != length - 1 and rest: |
|
308
|
|
|
_next = item[rest-1] == '\n' |
|
309
|
|
|
if not loose: |
|
310
|
|
|
loose = _next |
|
311
|
|
|
|
|
312
|
|
|
if loose: |
|
313
|
|
|
t = 'loose_item_start' |
|
314
|
|
|
else: |
|
315
|
|
|
t = 'list_item_start' |
|
316
|
|
|
|
|
317
|
|
|
self.tokens.append({'type': t}) |
|
318
|
|
|
# recurse |
|
319
|
|
|
self.parse(item, self.list_rules) |
|
320
|
|
|
self.tokens.append({'type': 'list_item_end'}) |
|
321
|
|
|
|
|
322
|
|
|
def parse_block_quote(self, m): |
|
323
|
|
|
self.tokens.append({'type': 'block_quote_start'}) |
|
324
|
|
|
# clean leading > |
|
325
|
|
|
cap = _block_quote_leading_pattern.sub('', m.group(0)) |
|
326
|
|
|
self.parse(cap) |
|
327
|
|
|
self.tokens.append({'type': 'block_quote_end'}) |
|
328
|
|
|
|
|
329
|
|
|
def parse_def_links(self, m): |
|
330
|
|
|
key = _keyify(m.group(1)) |
|
331
|
|
|
self.def_links[key] = { |
|
332
|
|
|
'link': m.group(2), |
|
333
|
|
|
'title': m.group(3), |
|
334
|
|
|
} |
|
335
|
|
|
|
|
336
|
|
|
def parse_def_footnotes(self, m): |
|
337
|
|
|
key = _keyify(m.group(1)) |
|
338
|
|
|
if key in self.def_footnotes: |
|
339
|
|
|
# footnote is already defined |
|
340
|
|
|
return |
|
341
|
|
|
|
|
342
|
|
|
self.def_footnotes[key] = 0 |
|
343
|
|
|
|
|
344
|
|
|
self.tokens.append({ |
|
345
|
|
|
'type': 'footnote_start', |
|
346
|
|
|
'key': key, |
|
347
|
|
|
}) |
|
348
|
|
|
|
|
349
|
|
|
text = m.group(2) |
|
350
|
|
|
|
|
351
|
|
|
if '\n' in text: |
|
352
|
|
|
lines = text.split('\n') |
|
353
|
|
|
whitespace = None |
|
354
|
|
|
for line in lines[1:]: |
|
355
|
|
|
space = len(line) - len(line.lstrip()) |
|
356
|
|
|
if space and (not whitespace or space < whitespace): |
|
357
|
|
|
whitespace = space |
|
358
|
|
|
newlines = [lines[0]] |
|
359
|
|
|
for line in lines[1:]: |
|
360
|
|
|
newlines.append(line[whitespace:]) |
|
361
|
|
|
text = '\n'.join(newlines) |
|
362
|
|
|
|
|
363
|
|
|
self.parse(text, self.footnote_rules) |
|
364
|
|
|
|
|
365
|
|
|
self.tokens.append({ |
|
366
|
|
|
'type': 'footnote_end', |
|
367
|
|
|
'key': key, |
|
368
|
|
|
}) |
|
369
|
|
|
|
|
370
|
|
|
def parse_table(self, m): |
|
371
|
|
|
item = self._process_table(m) |
|
372
|
|
|
|
|
373
|
|
|
cells = re.sub(r'(?: *\| *)?\n$', '', m.group(3)) |
|
374
|
|
|
cells = cells.split('\n') |
|
375
|
|
|
for i, v in enumerate(cells): |
|
376
|
|
|
v = re.sub(r'^ *\| *| *\| *$', '', v) |
|
377
|
|
|
cells[i] = re.split(r' *(?<!\\)\| *', v) |
|
378
|
|
|
|
|
379
|
|
|
item['cells'] = self._process_cells(cells) |
|
380
|
|
|
self.tokens.append(item) |
|
381
|
|
|
|
|
382
|
|
|
def parse_nptable(self, m): |
|
383
|
|
|
item = self._process_table(m) |
|
384
|
|
|
|
|
385
|
|
|
cells = re.sub(r'\n$', '', m.group(3)) |
|
386
|
|
|
cells = cells.split('\n') |
|
387
|
|
|
for i, v in enumerate(cells): |
|
388
|
|
|
cells[i] = re.split(r' *(?<!\\)\| *', v) |
|
389
|
|
|
|
|
390
|
|
|
item['cells'] = self._process_cells(cells) |
|
391
|
|
|
self.tokens.append(item) |
|
392
|
|
|
|
|
393
|
|
|
def _process_table(self, m): |
|
394
|
|
|
header = re.sub(r'^ *| *\| *$', '', m.group(1)) |
|
395
|
|
|
header = re.split(r' *\| *', header) |
|
396
|
|
|
align = re.sub(r' *|\| *$', '', m.group(2)) |
|
397
|
|
|
align = re.split(r' *\| *', align) |
|
398
|
|
|
|
|
399
|
|
|
for i, v in enumerate(align): |
|
400
|
|
|
if re.search(r'^ *-+: *$', v): |
|
401
|
|
|
align[i] = 'right' |
|
402
|
|
|
elif re.search(r'^ *:-+: *$', v): |
|
403
|
|
|
align[i] = 'center' |
|
404
|
|
|
elif re.search(r'^ *:-+ *$', v): |
|
405
|
|
|
align[i] = 'left' |
|
406
|
|
|
else: |
|
407
|
|
|
align[i] = None |
|
408
|
|
|
|
|
409
|
|
|
item = { |
|
410
|
|
|
'type': 'table', |
|
411
|
|
|
'header': header, |
|
412
|
|
|
'align': align, |
|
413
|
|
|
} |
|
414
|
|
|
return item |
|
415
|
|
|
|
|
416
|
|
|
def _process_cells(self, cells): |
|
417
|
|
|
for i, line in enumerate(cells): |
|
418
|
|
|
for c, cell in enumerate(line): |
|
419
|
|
|
# de-escape any pipe inside the cell here |
|
420
|
|
|
cells[i][c] = re.sub('\\\\\|', '|', cell) |
|
421
|
|
|
|
|
422
|
|
|
return cells |
|
423
|
|
|
|
|
424
|
|
|
def parse_block_html(self, m): |
|
425
|
|
|
tag = m.group(1) |
|
426
|
|
|
if not tag: |
|
427
|
|
|
text = m.group(0) |
|
428
|
|
|
self.tokens.append({ |
|
429
|
|
|
'type': 'close_html', |
|
430
|
|
|
'text': text |
|
431
|
|
|
}) |
|
432
|
|
|
else: |
|
433
|
|
|
attr = m.group(2) |
|
434
|
|
|
text = m.group(3) |
|
435
|
|
|
self.tokens.append({ |
|
436
|
|
|
'type': 'open_html', |
|
437
|
|
|
'tag': tag, |
|
438
|
|
|
'extra': attr, |
|
439
|
|
|
'text': text |
|
440
|
|
|
}) |
|
441
|
|
|
|
|
442
|
|
|
def parse_paragraph(self, m): |
|
443
|
|
|
text = m.group(1).rstrip('\n') |
|
444
|
|
|
self.tokens.append({'type': 'paragraph', 'text': text}) |
|
445
|
|
|
|
|
446
|
|
|
def parse_text(self, m): |
|
447
|
|
|
text = m.group(0) |
|
448
|
|
|
self.tokens.append({'type': 'text', 'text': text}) |
|
449
|
|
|
|
|
450
|
|
|
|
|
451
|
|
|
class InlineGrammar(object): |
|
452
|
|
|
"""Grammars for inline level tokens.""" |
|
453
|
|
|
|
|
454
|
|
|
escape = re.compile(r'^\\([\\`*{}\[\]()#+\-.!_>~|])') # \* \+ \! .... |
|
455
|
|
|
inline_html = re.compile( |
|
456
|
|
|
r'^(?:%s|%s|%s)' % ( |
|
457
|
|
|
r'<!--[\s\S]*?-->', |
|
458
|
|
|
r'<(\w+%s)((?:%s)*?)\s*>([\s\S]*?)<\/\1>' % ( |
|
459
|
|
|
_valid_end, _valid_attr), |
|
460
|
|
|
r'<\w+%s(?:%s)*?\s*\/?>' % (_valid_end, _valid_attr), |
|
461
|
|
|
) |
|
462
|
|
|
) |
|
463
|
|
|
autolink = re.compile(r'^<([^ >]+(@|:)[^ >]+)>') |
|
464
|
|
|
link = re.compile( |
|
465
|
|
|
r'^!?\[(' |
|
466
|
|
|
r'(?:\[[^^\]]*\]|[^\[\]]|\](?=[^\[]*\]))*' |
|
467
|
|
|
r')\]\(' |
|
468
|
|
|
r'''\s*(<)?([\s\S]*?)(?(2)>)(?:\s+['"]([\s\S]*?)['"])?\s*''' |
|
469
|
|
|
r'\)' |
|
470
|
|
|
) |
|
471
|
|
|
reflink = re.compile( |
|
472
|
|
|
r'^!?\[(' |
|
473
|
|
|
r'(?:\[[^^\]]*\]|[^\[\]]|\](?=[^\[]*\]))*' |
|
474
|
|
|
r')\]\s*\[([^^\]]*)\]' |
|
475
|
|
|
) |
|
476
|
|
|
nolink = re.compile(r'^!?\[((?:\[[^\]]*\]|[^\[\]])*)\]') |
|
477
|
|
|
url = re.compile(r'''^(https?:\/\/[^\s<]+[^<.,:;"')\]\s])''') |
|
478
|
|
|
double_emphasis = re.compile( |
|
479
|
|
|
r'^_{2}([\s\S]+?)_{2}(?!_)' # __word__ |
|
480
|
|
|
r'|' |
|
481
|
|
|
r'^\*{2}([\s\S]+?)\*{2}(?!\*)' # **word** |
|
482
|
|
|
) |
|
483
|
|
|
emphasis = re.compile( |
|
484
|
|
|
r'^\b_((?:__|[^_])+?)_\b' # _word_ |
|
485
|
|
|
r'|' |
|
486
|
|
|
r'^\*((?:\*\*|[^\*])+?)\*(?!\*)' # *word* |
|
487
|
|
|
) |
|
488
|
|
|
code = re.compile(r'^(`+)\s*([\s\S]*?[^`])\s*\1(?!`)') # `code` |
|
489
|
|
|
linebreak = re.compile(r'^ {2,}\n(?!\s*$)') |
|
490
|
|
|
strikethrough = re.compile(r'^~~(?=\S)([\s\S]*?\S)~~') # ~~word~~ |
|
491
|
|
|
footnote = re.compile(r'^\[\^([^\]]+)\]') |
|
492
|
|
|
text = re.compile(r'^[\s\S]+?(?=[\\<!\[_*`~]|https?://| {2,}\n|$)') |
|
493
|
|
|
|
|
494
|
|
|
def hard_wrap(self): |
|
495
|
|
|
"""Grammar for hard wrap linebreak. You don't need to add two |
|
496
|
|
|
spaces at the end of a line. |
|
497
|
|
|
""" |
|
498
|
|
|
self.linebreak = re.compile(r'^ *\n(?!\s*$)') |
|
499
|
|
|
self.text = re.compile( |
|
500
|
|
|
r'^[\s\S]+?(?=[\\<!\[_*`~]|https?://| *\n|$)' |
|
501
|
|
|
) |
|
502
|
|
|
|
|
503
|
|
|
|
|
504
|
|
|
class InlineLexer(object): |
|
505
|
|
|
"""Inline level lexer for inline grammars.""" |
|
506
|
|
|
grammar_class = InlineGrammar |
|
507
|
|
|
|
|
508
|
|
|
default_rules = [ |
|
509
|
|
|
'escape', 'inline_html', 'autolink', 'url', |
|
510
|
|
|
'footnote', 'link', 'reflink', 'nolink', |
|
511
|
|
|
'double_emphasis', 'emphasis', 'code', |
|
512
|
|
|
'linebreak', 'strikethrough', 'text', |
|
513
|
|
|
] |
|
514
|
|
|
inline_html_rules = [ |
|
515
|
|
|
'escape', 'inline_html', 'autolink', 'url', 'link', 'reflink', |
|
516
|
|
|
'nolink', 'double_emphasis', 'emphasis', 'code', |
|
517
|
|
|
'linebreak', 'strikethrough', 'text', |
|
518
|
|
|
] |
|
519
|
|
|
|
|
520
|
|
|
def __init__(self, renderer, rules=None, **kwargs): |
|
521
|
|
|
self.renderer = renderer |
|
522
|
|
|
self.links = {} |
|
523
|
|
|
self.footnotes = {} |
|
524
|
|
|
self.footnote_index = 0 |
|
525
|
|
|
|
|
526
|
|
|
if not rules: |
|
527
|
|
|
rules = self.grammar_class() |
|
528
|
|
|
|
|
529
|
|
|
kwargs.update(self.renderer.options) |
|
530
|
|
|
if kwargs.get('hard_wrap'): |
|
531
|
|
|
rules.hard_wrap() |
|
532
|
|
|
|
|
533
|
|
|
self.rules = rules |
|
534
|
|
|
|
|
535
|
|
|
self._in_link = False |
|
536
|
|
|
self._in_footnote = False |
|
537
|
|
|
self._parse_inline_html = kwargs.get('parse_inline_html') |
|
538
|
|
|
|
|
539
|
|
|
def __call__(self, text, rules=None): |
|
540
|
|
|
return self.output(text, rules) |
|
541
|
|
|
|
|
542
|
|
|
def setup(self, links, footnotes): |
|
543
|
|
|
self.footnote_index = 0 |
|
544
|
|
|
self.links = links or {} |
|
545
|
|
|
self.footnotes = footnotes or {} |
|
546
|
|
|
|
|
547
|
|
|
def output(self, text, rules=None): |
|
548
|
|
|
text = text.rstrip('\n') |
|
549
|
|
|
if not rules: |
|
550
|
|
|
rules = list(self.default_rules) |
|
551
|
|
|
|
|
552
|
|
|
if self._in_footnote and 'footnote' in rules: |
|
553
|
|
|
rules.remove('footnote') |
|
554
|
|
|
|
|
555
|
|
|
output = self.renderer.placeholder() |
|
556
|
|
|
|
|
557
|
|
|
def manipulate(text): |
|
558
|
|
|
for key in rules: |
|
559
|
|
|
pattern = getattr(self.rules, key) |
|
560
|
|
|
m = pattern.match(text) |
|
561
|
|
|
if not m: |
|
562
|
|
|
continue |
|
563
|
|
|
self.line_match = m |
|
564
|
|
|
out = getattr(self, 'output_%s' % key)(m) |
|
565
|
|
|
if out is not None: |
|
566
|
|
|
return m, out |
|
567
|
|
|
return False # pragma: no cover |
|
568
|
|
|
|
|
569
|
|
|
while text: |
|
570
|
|
|
ret = manipulate(text) |
|
571
|
|
|
if ret is not False: |
|
572
|
|
|
m, out = ret |
|
573
|
|
|
output += out |
|
574
|
|
|
text = text[len(m.group(0)):] |
|
575
|
|
|
continue |
|
576
|
|
|
if text: # pragma: no cover |
|
577
|
|
|
raise RuntimeError('Infinite loop at: %s' % text) |
|
578
|
|
|
|
|
579
|
|
|
return output |
|
580
|
|
|
|
|
581
|
|
|
def output_escape(self, m): |
|
582
|
|
|
text = m.group(1) |
|
583
|
|
|
return self.renderer.escape(text) |
|
584
|
|
|
|
|
585
|
|
|
def output_autolink(self, m): |
|
586
|
|
|
link = m.group(1) |
|
587
|
|
|
if m.group(2) == '@': |
|
588
|
|
|
is_email = True |
|
589
|
|
|
else: |
|
590
|
|
|
is_email = False |
|
591
|
|
|
return self.renderer.autolink(link, is_email) |
|
592
|
|
|
|
|
593
|
|
|
def output_url(self, m): |
|
594
|
|
|
link = m.group(1) |
|
595
|
|
|
if self._in_link: |
|
596
|
|
|
return self.renderer.text(link) |
|
597
|
|
|
return self.renderer.autolink(link, False) |
|
598
|
|
|
|
|
599
|
|
|
def output_inline_html(self, m): |
|
600
|
|
|
tag = m.group(1) |
|
601
|
|
|
if self._parse_inline_html and tag in _inline_tags: |
|
602
|
|
|
text = m.group(3) |
|
603
|
|
|
if tag == 'a': |
|
604
|
|
|
self._in_link = True |
|
605
|
|
|
text = self.output(text, rules=self.inline_html_rules) |
|
606
|
|
|
self._in_link = False |
|
607
|
|
|
else: |
|
608
|
|
|
text = self.output(text, rules=self.inline_html_rules) |
|
609
|
|
|
extra = m.group(2) or '' |
|
610
|
|
|
html = '<%s%s>%s</%s>' % (tag, extra, text, tag) |
|
611
|
|
|
else: |
|
612
|
|
|
html = m.group(0) |
|
613
|
|
|
return self.renderer.inline_html(html) |
|
614
|
|
|
|
|
615
|
|
|
def output_footnote(self, m): |
|
616
|
|
|
key = _keyify(m.group(1)) |
|
617
|
|
|
if key not in self.footnotes: |
|
618
|
|
|
return None |
|
619
|
|
|
if self.footnotes[key]: |
|
620
|
|
|
return None |
|
621
|
|
|
self.footnote_index += 1 |
|
622
|
|
|
self.footnotes[key] = self.footnote_index |
|
623
|
|
|
return self.renderer.footnote_ref(key, self.footnote_index) |
|
624
|
|
|
|
|
625
|
|
|
def output_link(self, m): |
|
626
|
|
|
return self._process_link(m, m.group(3), m.group(4)) |
|
627
|
|
|
|
|
628
|
|
|
def output_reflink(self, m): |
|
629
|
|
|
key = _keyify(m.group(2) or m.group(1)) |
|
630
|
|
|
if key not in self.links: |
|
631
|
|
|
return None |
|
632
|
|
|
ret = self.links[key] |
|
633
|
|
|
return self._process_link(m, ret['link'], ret['title']) |
|
634
|
|
|
|
|
635
|
|
|
def output_nolink(self, m): |
|
636
|
|
|
key = _keyify(m.group(1)) |
|
637
|
|
|
if key not in self.links: |
|
638
|
|
|
return None |
|
639
|
|
|
ret = self.links[key] |
|
640
|
|
|
return self._process_link(m, ret['link'], ret['title']) |
|
641
|
|
|
|
|
642
|
|
|
def _process_link(self, m, link, title=None): |
|
643
|
|
|
line = m.group(0) |
|
644
|
|
|
text = m.group(1) |
|
645
|
|
|
if line[0] == '!': |
|
646
|
|
|
return self.renderer.image(link, title, text) |
|
647
|
|
|
|
|
648
|
|
|
self._in_link = True |
|
649
|
|
|
text = self.output(text) |
|
650
|
|
|
self._in_link = False |
|
651
|
|
|
return self.renderer.link(link, title, text) |
|
652
|
|
|
|
|
653
|
|
|
def output_double_emphasis(self, m): |
|
654
|
|
|
text = m.group(2) or m.group(1) |
|
655
|
|
|
text = self.output(text) |
|
656
|
|
|
return self.renderer.double_emphasis(text) |
|
657
|
|
|
|
|
658
|
|
|
def output_emphasis(self, m): |
|
659
|
|
|
text = m.group(2) or m.group(1) |
|
660
|
|
|
text = self.output(text) |
|
661
|
|
|
return self.renderer.emphasis(text) |
|
662
|
|
|
|
|
663
|
|
|
def output_code(self, m): |
|
664
|
|
|
text = m.group(2) |
|
665
|
|
|
return self.renderer.codespan(text) |
|
666
|
|
|
|
|
667
|
|
|
def output_linebreak(self, m): |
|
668
|
|
|
return self.renderer.linebreak() |
|
669
|
|
|
|
|
670
|
|
|
def output_strikethrough(self, m): |
|
671
|
|
|
text = self.output(m.group(1)) |
|
672
|
|
|
return self.renderer.strikethrough(text) |
|
673
|
|
|
|
|
674
|
|
|
def output_text(self, m): |
|
675
|
|
|
text = m.group(0) |
|
676
|
|
|
return self.renderer.text(text) |
|
677
|
|
|
|
|
678
|
|
|
|
|
679
|
|
|
class Renderer(object): |
|
680
|
|
|
"""The default HTML renderer for rendering Markdown. |
|
681
|
|
|
""" |
|
682
|
|
|
|
|
683
|
|
|
def __init__(self, **kwargs): |
|
684
|
|
|
self.options = kwargs |
|
685
|
|
|
|
|
686
|
|
|
def placeholder(self): |
|
687
|
|
|
"""Returns the default, empty output value for the renderer. |
|
688
|
|
|
|
|
689
|
|
|
All renderer methods use the '+=' operator to append to this value. |
|
690
|
|
|
Default is a string so rendering HTML can build up a result string with |
|
691
|
|
|
the rendered Markdown. |
|
692
|
|
|
|
|
693
|
|
|
Can be overridden by Renderer subclasses to be types like an empty |
|
694
|
|
|
list, allowing the renderer to create a tree-like structure to |
|
695
|
|
|
represent the document (which can then be reprocessed later into a |
|
696
|
|
|
separate format like docx or pdf). |
|
697
|
|
|
""" |
|
698
|
|
|
return '' |
|
699
|
|
|
|
|
700
|
|
|
def block_code(self, code, lang=None): |
|
701
|
|
|
"""Rendering block level code. ``pre > code``. |
|
702
|
|
|
|
|
703
|
|
|
:param code: text content of the code block. |
|
704
|
|
|
:param lang: language of the given code. |
|
705
|
|
|
""" |
|
706
|
|
|
code = code.rstrip('\n') |
|
707
|
|
|
if not lang: |
|
708
|
|
|
code = escape(code, smart_amp=False) |
|
709
|
|
|
return '<pre><code>%s\n</code></pre>\n' % code |
|
710
|
|
|
code = escape(code, quote=True, smart_amp=False) |
|
711
|
|
|
return '<pre><code class="lang-%s">%s\n</code></pre>\n' % (lang, code) |
|
712
|
|
|
|
|
713
|
|
|
def block_quote(self, text): |
|
714
|
|
|
"""Rendering <blockquote> with the given text. |
|
715
|
|
|
|
|
716
|
|
|
:param text: text content of the blockquote. |
|
717
|
|
|
""" |
|
718
|
|
|
return '<blockquote>%s\n</blockquote>\n' % text.rstrip('\n') |
|
719
|
|
|
|
|
720
|
|
|
def block_html(self, html): |
|
721
|
|
|
"""Rendering block level pure html content. |
|
722
|
|
|
|
|
723
|
|
|
:param html: text content of the html snippet. |
|
724
|
|
|
""" |
|
725
|
|
|
if self.options.get('skip_style') and \ |
|
726
|
|
|
html.lower().startswith('<style'): |
|
727
|
|
|
return '' |
|
728
|
|
|
if self.options.get('escape'): |
|
729
|
|
|
return escape(html) |
|
730
|
|
|
return html |
|
731
|
|
|
|
|
732
|
|
|
def header(self, text, level, raw=None): |
|
733
|
|
|
"""Rendering header/heading tags like ``<h1>`` ``<h2>``. |
|
734
|
|
|
|
|
735
|
|
|
:param text: rendered text content for the header. |
|
736
|
|
|
:param level: a number for the header level, for example: 1. |
|
737
|
|
|
:param raw: raw text content of the header. |
|
738
|
|
|
""" |
|
739
|
|
|
return '<h%d>%s</h%d>\n' % (level, text, level) |
|
740
|
|
|
|
|
741
|
|
|
def hrule(self): |
|
742
|
|
|
"""Rendering method for ``<hr>`` tag.""" |
|
743
|
|
|
if self.options.get('use_xhtml'): |
|
744
|
|
|
return '<hr />\n' |
|
745
|
|
|
return '<hr>\n' |
|
746
|
|
|
|
|
747
|
|
|
def list(self, body, ordered=True): |
|
748
|
|
|
"""Rendering list tags like ``<ul>`` and ``<ol>``. |
|
749
|
|
|
|
|
750
|
|
|
:param body: body contents of the list. |
|
751
|
|
|
:param ordered: whether this list is ordered or not. |
|
752
|
|
|
""" |
|
753
|
|
|
tag = 'ul' |
|
754
|
|
|
if ordered: |
|
755
|
|
|
tag = 'ol' |
|
756
|
|
|
return '<%s>\n%s</%s>\n' % (tag, body, tag) |
|
757
|
|
|
|
|
758
|
|
|
def list_item(self, text): |
|
759
|
|
|
"""Rendering list item snippet. Like ``<li>``.""" |
|
760
|
|
|
return '<li>%s</li>\n' % text |
|
761
|
|
|
|
|
762
|
|
|
def paragraph(self, text): |
|
763
|
|
|
"""Rendering paragraph tags. Like ``<p>``.""" |
|
764
|
|
|
return '<p>%s</p>\n' % text.strip(' ') |
|
765
|
|
|
|
|
766
|
|
|
def table(self, header, body): |
|
767
|
|
|
"""Rendering table element. Wrap header and body in it. |
|
768
|
|
|
|
|
769
|
|
|
:param header: header part of the table. |
|
770
|
|
|
:param body: body part of the table. |
|
771
|
|
|
""" |
|
772
|
|
|
return ( |
|
773
|
|
|
'<table>\n<thead>%s</thead>\n' |
|
774
|
|
|
'<tbody>\n%s</tbody>\n</table>\n' |
|
775
|
|
|
) % (header, body) |
|
776
|
|
|
|
|
777
|
|
|
def table_row(self, content): |
|
778
|
|
|
"""Rendering a table row. Like ``<tr>``. |
|
779
|
|
|
|
|
780
|
|
|
:param content: content of current table row. |
|
781
|
|
|
""" |
|
782
|
|
|
return '<tr>\n%s</tr>\n' % content |
|
783
|
|
|
|
|
784
|
|
|
def table_cell(self, content, **flags): |
|
785
|
|
|
"""Rendering a table cell. Like ``<th>`` ``<td>``. |
|
786
|
|
|
|
|
787
|
|
|
:param content: content of current table cell. |
|
788
|
|
|
:param header: whether this is header or not. |
|
789
|
|
|
:param align: align of current table cell. |
|
790
|
|
|
""" |
|
791
|
|
|
if flags['header']: |
|
792
|
|
|
tag = 'th' |
|
793
|
|
|
else: |
|
794
|
|
|
tag = 'td' |
|
795
|
|
|
align = flags['align'] |
|
796
|
|
|
if not align: |
|
797
|
|
|
return '<%s>%s</%s>\n' % (tag, content, tag) |
|
798
|
|
|
return '<%s style="text-align:%s">%s</%s>\n' % ( |
|
799
|
|
|
tag, align, content, tag |
|
800
|
|
|
) |
|
801
|
|
|
|
|
802
|
|
|
def double_emphasis(self, text): |
|
803
|
|
|
"""Rendering **strong** text. |
|
804
|
|
|
|
|
805
|
|
|
:param text: text content for emphasis. |
|
806
|
|
|
""" |
|
807
|
|
|
return '<strong>%s</strong>' % text |
|
808
|
|
|
|
|
809
|
|
|
def emphasis(self, text): |
|
810
|
|
|
"""Rendering *emphasis* text. |
|
811
|
|
|
|
|
812
|
|
|
:param text: text content for emphasis. |
|
813
|
|
|
""" |
|
814
|
|
|
return '<em>%s</em>' % text |
|
815
|
|
|
|
|
816
|
|
|
def codespan(self, text): |
|
817
|
|
|
"""Rendering inline `code` text. |
|
818
|
|
|
|
|
819
|
|
|
:param text: text content for inline code. |
|
820
|
|
|
""" |
|
821
|
|
|
text = escape(text.rstrip(), smart_amp=False) |
|
822
|
|
|
return '<code>%s</code>' % text |
|
823
|
|
|
|
|
824
|
|
|
def linebreak(self): |
|
825
|
|
|
"""Rendering line break like ``<br>``.""" |
|
826
|
|
|
if self.options.get('use_xhtml'): |
|
827
|
|
|
return '<br />\n' |
|
828
|
|
|
return '<br>\n' |
|
829
|
|
|
|
|
830
|
|
|
def strikethrough(self, text): |
|
831
|
|
|
"""Rendering ~~strikethrough~~ text. |
|
832
|
|
|
|
|
833
|
|
|
:param text: text content for strikethrough. |
|
834
|
|
|
""" |
|
835
|
|
|
return '<del>%s</del>' % text |
|
836
|
|
|
|
|
837
|
|
|
def text(self, text): |
|
838
|
|
|
"""Rendering unformatted text. |
|
839
|
|
|
|
|
840
|
|
|
:param text: text content. |
|
841
|
|
|
""" |
|
842
|
|
|
if self.options.get('parse_block_html'): |
|
843
|
|
|
return text |
|
844
|
|
|
return escape(text) |
|
845
|
|
|
|
|
846
|
|
|
def escape(self, text): |
|
847
|
|
|
"""Rendering escape sequence. |
|
848
|
|
|
|
|
849
|
|
|
:param text: text content. |
|
850
|
|
|
""" |
|
851
|
|
|
return escape(text) |
|
852
|
|
|
|
|
853
|
|
|
def autolink(self, link, is_email=False): |
|
854
|
|
|
"""Rendering a given link or email address. |
|
855
|
|
|
|
|
856
|
|
|
:param link: link content or email address. |
|
857
|
|
|
:param is_email: whether this is an email or not. |
|
858
|
|
|
""" |
|
859
|
|
|
text = link = escape_link(link) |
|
860
|
|
|
if is_email: |
|
861
|
|
|
link = 'mailto:%s' % link |
|
862
|
|
|
return '<a href="%s">%s</a>' % (link, text) |
|
863
|
|
|
|
|
864
|
|
|
def link(self, link, title, text): |
|
865
|
|
|
"""Rendering a given link with content and title. |
|
866
|
|
|
|
|
867
|
|
|
:param link: href link for ``<a>`` tag. |
|
868
|
|
|
:param title: title content for `title` attribute. |
|
869
|
|
|
:param text: text content for description. |
|
870
|
|
|
""" |
|
871
|
|
|
link = escape_link(link) |
|
872
|
|
|
if not title: |
|
873
|
|
|
return '<a href="%s">%s</a>' % (link, text) |
|
874
|
|
|
title = escape(title, quote=True) |
|
875
|
|
|
return '<a href="%s" title="%s">%s</a>' % (link, title, text) |
|
876
|
|
|
|
|
877
|
|
|
def image(self, src, title, text): |
|
878
|
|
|
"""Rendering a image with title and text. |
|
879
|
|
|
|
|
880
|
|
|
:param src: source link of the image. |
|
881
|
|
|
:param title: title text of the image. |
|
882
|
|
|
:param text: alt text of the image. |
|
883
|
|
|
""" |
|
884
|
|
|
src = escape_link(src) |
|
885
|
|
|
text = escape(text, quote=True) |
|
886
|
|
|
if title: |
|
887
|
|
|
title = escape(title, quote=True) |
|
888
|
|
|
html = '<img src="%s" alt="%s" title="%s"' % (src, text, title) |
|
889
|
|
|
else: |
|
890
|
|
|
html = '<img src="%s" alt="%s"' % (src, text) |
|
891
|
|
|
if self.options.get('use_xhtml'): |
|
892
|
|
|
return '%s />' % html |
|
893
|
|
|
return '%s>' % html |
|
894
|
|
|
|
|
895
|
|
|
def inline_html(self, html): |
|
896
|
|
|
"""Rendering span level pure html content. |
|
897
|
|
|
|
|
898
|
|
|
:param html: text content of the html snippet. |
|
899
|
|
|
""" |
|
900
|
|
|
if self.options.get('escape'): |
|
901
|
|
|
return escape(html) |
|
902
|
|
|
return html |
|
903
|
|
|
|
|
904
|
|
|
def newline(self): |
|
905
|
|
|
"""Rendering newline element.""" |
|
906
|
|
|
return '' |
|
907
|
|
|
|
|
908
|
|
|
def footnote_ref(self, key, index): |
|
909
|
|
|
"""Rendering the ref anchor of a footnote. |
|
910
|
|
|
|
|
911
|
|
|
:param key: identity key for the footnote. |
|
912
|
|
|
:param index: the index count of current footnote. |
|
913
|
|
|
""" |
|
914
|
|
|
html = ( |
|
915
|
|
|
'<sup class="footnote-ref" id="fnref-%s">' |
|
916
|
|
|
'<a href="#fn-%s">%d</a></sup>' |
|
917
|
|
|
) % (escape(key), escape(key), index) |
|
918
|
|
|
return html |
|
919
|
|
|
|
|
920
|
|
|
def footnote_item(self, key, text): |
|
921
|
|
|
"""Rendering a footnote item. |
|
922
|
|
|
|
|
923
|
|
|
:param key: identity key for the footnote. |
|
924
|
|
|
:param text: text content of the footnote. |
|
925
|
|
|
""" |
|
926
|
|
|
back = ( |
|
927
|
|
|
'<a href="#fnref-%s" class="footnote">↩</a>' |
|
928
|
|
|
) % escape(key) |
|
929
|
|
|
text = text.rstrip() |
|
930
|
|
|
if text.endswith('</p>'): |
|
931
|
|
|
text = re.sub(r'<\/p>$', r'%s</p>' % back, text) |
|
932
|
|
|
else: |
|
933
|
|
|
text = '%s<p>%s</p>' % (text, back) |
|
934
|
|
|
html = '<li id="fn-%s">%s</li>\n' % (escape(key), text) |
|
935
|
|
|
return html |
|
936
|
|
|
|
|
937
|
|
|
def footnotes(self, text): |
|
938
|
|
|
"""Wrapper for all footnotes. |
|
939
|
|
|
|
|
940
|
|
|
:param text: contents of all footnotes. |
|
941
|
|
|
""" |
|
942
|
|
|
html = '<div class="footnotes">\n%s<ol>%s</ol>\n</div>\n' |
|
943
|
|
|
return html % (self.hrule(), text) |
|
944
|
|
|
|
|
945
|
|
|
|
|
946
|
|
|
class Markdown(object): |
|
947
|
|
|
"""The Markdown parser. |
|
948
|
|
|
|
|
949
|
|
|
:param renderer: An instance of ``Renderer``. |
|
950
|
|
|
:param inline: An inline lexer class or instance. |
|
951
|
|
|
:param block: A block lexer class or instance. |
|
952
|
|
|
""" |
|
953
|
|
|
def __init__(self, renderer=None, inline=None, block=None, **kwargs): |
|
954
|
|
|
if not renderer: |
|
955
|
|
|
renderer = Renderer(**kwargs) |
|
956
|
|
|
else: |
|
957
|
|
|
kwargs.update(renderer.options) |
|
958
|
|
|
|
|
959
|
|
|
self.renderer = renderer |
|
960
|
|
|
|
|
961
|
|
|
if inline and inspect.isclass(inline): |
|
962
|
|
|
inline = inline(renderer, **kwargs) |
|
963
|
|
|
if block and inspect.isclass(block): |
|
964
|
|
|
block = block(**kwargs) |
|
965
|
|
|
|
|
966
|
|
|
if inline: |
|
967
|
|
|
self.inline = inline |
|
968
|
|
|
else: |
|
969
|
|
|
self.inline = InlineLexer(renderer, **kwargs) |
|
970
|
|
|
|
|
971
|
|
|
self.block = block or BlockLexer(BlockGrammar()) |
|
972
|
|
|
self.footnotes = [] |
|
973
|
|
|
self.tokens = [] |
|
974
|
|
|
|
|
975
|
|
|
# detect if it should parse text in block html |
|
976
|
|
|
self._parse_block_html = kwargs.get('parse_block_html') |
|
977
|
|
|
|
|
978
|
|
|
def __call__(self, text): |
|
979
|
|
|
return self.parse(text) |
|
980
|
|
|
|
|
981
|
|
|
def render(self, text): |
|
982
|
|
|
"""Render the Markdown text. |
|
983
|
|
|
|
|
984
|
|
|
:param text: markdown formatted text content. |
|
985
|
|
|
""" |
|
986
|
|
|
return self.parse(text) |
|
987
|
|
|
|
|
988
|
|
|
def parse(self, text): |
|
989
|
|
|
out = self.output(preprocessing(text)) |
|
990
|
|
|
|
|
991
|
|
|
keys = self.block.def_footnotes |
|
992
|
|
|
|
|
993
|
|
|
# reset block |
|
994
|
|
|
self.block.def_links = {} |
|
995
|
|
|
self.block.def_footnotes = {} |
|
996
|
|
|
|
|
997
|
|
|
# reset inline |
|
998
|
|
|
self.inline.links = {} |
|
999
|
|
|
self.inline.footnotes = {} |
|
1000
|
|
|
|
|
1001
|
|
|
if not self.footnotes: |
|
1002
|
|
|
return out |
|
1003
|
|
|
|
|
1004
|
|
|
footnotes = filter(lambda o: keys.get(o['key']), self.footnotes) |
|
1005
|
|
|
self.footnotes = sorted( |
|
1006
|
|
|
footnotes, key=lambda o: keys.get(o['key']), reverse=True |
|
1007
|
|
|
) |
|
1008
|
|
|
|
|
1009
|
|
|
body = self.renderer.placeholder() |
|
1010
|
|
|
while self.footnotes: |
|
1011
|
|
|
note = self.footnotes.pop() |
|
1012
|
|
|
body += self.renderer.footnote_item( |
|
1013
|
|
|
note['key'], note['text'] |
|
1014
|
|
|
) |
|
1015
|
|
|
|
|
1016
|
|
|
out += self.renderer.footnotes(body) |
|
1017
|
|
|
return out |
|
1018
|
|
|
|
|
1019
|
|
|
def pop(self): |
|
1020
|
|
|
if not self.tokens: |
|
1021
|
|
|
return None |
|
1022
|
|
|
self.token = self.tokens.pop() |
|
1023
|
|
|
return self.token |
|
1024
|
|
|
|
|
1025
|
|
|
def peek(self): |
|
1026
|
|
|
if self.tokens: |
|
1027
|
|
|
return self.tokens[-1] |
|
1028
|
|
|
return None # pragma: no cover |
|
1029
|
|
|
|
|
1030
|
|
|
def output(self, text, rules=None): |
|
1031
|
|
|
self.tokens = self.block(text, rules) |
|
1032
|
|
|
self.tokens.reverse() |
|
1033
|
|
|
|
|
1034
|
|
|
self.inline.setup(self.block.def_links, self.block.def_footnotes) |
|
1035
|
|
|
|
|
1036
|
|
|
out = self.renderer.placeholder() |
|
1037
|
|
|
while self.pop(): |
|
1038
|
|
|
out += self.tok() |
|
1039
|
|
|
return out |
|
1040
|
|
|
|
|
1041
|
|
|
def tok(self): |
|
1042
|
|
|
t = self.token['type'] |
|
1043
|
|
|
|
|
1044
|
|
|
# sepcial cases |
|
1045
|
|
|
if t.endswith('_start'): |
|
1046
|
|
|
t = t[:-6] |
|
1047
|
|
|
|
|
1048
|
|
|
return getattr(self, 'output_%s' % t)() |
|
1049
|
|
|
|
|
1050
|
|
|
def tok_text(self): |
|
1051
|
|
|
text = self.token['text'] |
|
1052
|
|
|
while self.peek()['type'] == 'text': |
|
1053
|
|
|
text += '\n' + self.pop()['text'] |
|
1054
|
|
|
return self.inline(text) |
|
1055
|
|
|
|
|
1056
|
|
|
def output_newline(self): |
|
1057
|
|
|
return self.renderer.newline() |
|
1058
|
|
|
|
|
1059
|
|
|
def output_hrule(self): |
|
1060
|
|
|
return self.renderer.hrule() |
|
1061
|
|
|
|
|
1062
|
|
|
def output_heading(self): |
|
1063
|
|
|
return self.renderer.header( |
|
1064
|
|
|
self.inline(self.token['text']), |
|
1065
|
|
|
self.token['level'], |
|
1066
|
|
|
self.token['text'], |
|
1067
|
|
|
) |
|
1068
|
|
|
|
|
1069
|
|
|
def output_code(self): |
|
1070
|
|
|
return self.renderer.block_code( |
|
1071
|
|
|
self.token['text'], self.token['lang'] |
|
1072
|
|
|
) |
|
1073
|
|
|
|
|
1074
|
|
|
def output_table(self): |
|
1075
|
|
|
aligns = self.token['align'] |
|
1076
|
|
|
aligns_length = len(aligns) |
|
1077
|
|
|
cell = self.renderer.placeholder() |
|
1078
|
|
|
|
|
1079
|
|
|
# header part |
|
1080
|
|
|
header = self.renderer.placeholder() |
|
1081
|
|
|
for i, value in enumerate(self.token['header']): |
|
1082
|
|
|
align = aligns[i] if i < aligns_length else None |
|
1083
|
|
|
flags = {'header': True, 'align': align} |
|
1084
|
|
|
cell += self.renderer.table_cell(self.inline(value), **flags) |
|
1085
|
|
|
|
|
1086
|
|
|
header += self.renderer.table_row(cell) |
|
1087
|
|
|
|
|
1088
|
|
|
# body part |
|
1089
|
|
|
body = self.renderer.placeholder() |
|
1090
|
|
|
for i, row in enumerate(self.token['cells']): |
|
1091
|
|
|
cell = self.renderer.placeholder() |
|
1092
|
|
|
for j, value in enumerate(row): |
|
1093
|
|
|
align = aligns[j] if j < aligns_length else None |
|
1094
|
|
|
flags = {'header': False, 'align': align} |
|
1095
|
|
|
cell += self.renderer.table_cell(self.inline(value), **flags) |
|
1096
|
|
|
body += self.renderer.table_row(cell) |
|
1097
|
|
|
|
|
1098
|
|
|
return self.renderer.table(header, body) |
|
1099
|
|
|
|
|
1100
|
|
|
def output_block_quote(self): |
|
1101
|
|
|
body = self.renderer.placeholder() |
|
1102
|
|
|
while self.pop()['type'] != 'block_quote_end': |
|
1103
|
|
|
body += self.tok() |
|
1104
|
|
|
return self.renderer.block_quote(body) |
|
1105
|
|
|
|
|
1106
|
|
|
def output_list(self): |
|
1107
|
|
|
ordered = self.token['ordered'] |
|
1108
|
|
|
body = self.renderer.placeholder() |
|
1109
|
|
|
while self.pop()['type'] != 'list_end': |
|
1110
|
|
|
body += self.tok() |
|
1111
|
|
|
return self.renderer.list(body, ordered) |
|
1112
|
|
|
|
|
1113
|
|
|
def output_list_item(self): |
|
1114
|
|
|
body = self.renderer.placeholder() |
|
1115
|
|
|
while self.pop()['type'] != 'list_item_end': |
|
1116
|
|
|
if self.token['type'] == 'text': |
|
1117
|
|
|
body += self.tok_text() |
|
1118
|
|
|
else: |
|
1119
|
|
|
body += self.tok() |
|
1120
|
|
|
|
|
1121
|
|
|
return self.renderer.list_item(body) |
|
1122
|
|
|
|
|
1123
|
|
|
def output_loose_item(self): |
|
1124
|
|
|
body = self.renderer.placeholder() |
|
1125
|
|
|
while self.pop()['type'] != 'list_item_end': |
|
1126
|
|
|
body += self.tok() |
|
1127
|
|
|
return self.renderer.list_item(body) |
|
1128
|
|
|
|
|
1129
|
|
|
def output_footnote(self): |
|
1130
|
|
|
self.inline._in_footnote = True |
|
1131
|
|
|
body = self.renderer.placeholder() |
|
1132
|
|
|
key = self.token['key'] |
|
1133
|
|
|
while self.pop()['type'] != 'footnote_end': |
|
1134
|
|
|
body += self.tok() |
|
1135
|
|
|
self.footnotes.append({'key': key, 'text': body}) |
|
1136
|
|
|
self.inline._in_footnote = False |
|
1137
|
|
|
return self.renderer.placeholder() |
|
1138
|
|
|
|
|
1139
|
|
|
def output_close_html(self): |
|
1140
|
|
|
text = self.token['text'] |
|
1141
|
|
|
return self.renderer.block_html(text) |
|
1142
|
|
|
|
|
1143
|
|
|
def output_open_html(self): |
|
1144
|
|
|
text = self.token['text'] |
|
1145
|
|
|
tag = self.token['tag'] |
|
1146
|
|
|
if self._parse_block_html and tag not in _pre_tags: |
|
1147
|
|
|
text = self.inline(text, rules=self.inline.inline_html_rules) |
|
1148
|
|
|
extra = self.token.get('extra') or '' |
|
1149
|
|
|
html = '<%s%s>%s</%s>' % (tag, extra, text, tag) |
|
1150
|
|
|
return self.renderer.block_html(html) |
|
1151
|
|
|
|
|
1152
|
|
|
def output_paragraph(self): |
|
1153
|
|
|
return self.renderer.paragraph(self.inline(self.token['text'])) |
|
1154
|
|
|
|
|
1155
|
|
|
def output_text(self): |
|
1156
|
|
|
return self.renderer.paragraph(self.tok_text()) |
|
1157
|
|
|
|
|
1158
|
|
|
|
|
1159
|
|
|
def markdown(text, escape=True, **kwargs): |
|
1160
|
|
|
"""Render markdown formatted text to html. |
|
1161
|
|
|
|
|
1162
|
|
|
:param text: markdown formatted text content. |
|
1163
|
|
|
:param escape: if set to False, all html tags will not be escaped. |
|
1164
|
|
|
:param use_xhtml: output with xhtml tags. |
|
1165
|
|
|
:param hard_wrap: if set to True, it will use the GFM line breaks feature. |
|
1166
|
|
|
:param parse_block_html: parse text only in block level html. |
|
1167
|
|
|
:param parse_inline_html: parse text only in inline level html. |
|
1168
|
|
|
""" |
|
1169
|
|
|
return Markdown(escape=escape, **kwargs)(text) |
|
1170
|
|
|
|
MikeCoder /
markdown-preview.vim
