1
|
|
|
""" |
2
|
|
|
POST-PROCESSORS |
3
|
|
|
============================================================================= |
4
|
|
|
|
5
|
|
|
Markdown also allows post-processors, which are similar to preprocessors in |
6
|
|
|
that they need to implement a "run" method. However, they are run after core |
7
|
|
|
processing. |
8
|
|
|
|
9
|
|
|
""" |
10
|
|
|
|
11
|
|
|
from __future__ import absolute_import |
12
|
|
|
from __future__ import unicode_literals |
13
|
|
|
from collections import OrderedDict |
14
|
|
|
from . import util |
15
|
|
|
from . import odict |
16
|
|
|
import re |
17
|
|
|
|
18
|
|
|
|
19
|
|
|
def build_postprocessors(md_instance, **kwargs): |
20
|
|
|
""" Build the default postprocessors for Markdown. """ |
21
|
|
|
postprocessors = odict.OrderedDict() |
22
|
|
|
postprocessors["raw_html"] = RawHtmlPostprocessor(md_instance) |
23
|
|
|
postprocessors["amp_substitute"] = AndSubstitutePostprocessor() |
24
|
|
|
postprocessors["unescape"] = UnescapePostprocessor() |
25
|
|
|
return postprocessors |
26
|
|
|
|
27
|
|
|
|
28
|
|
|
class Postprocessor(util.Processor): |
29
|
|
|
""" |
30
|
|
|
Postprocessors are run after the ElementTree it converted back into text. |
31
|
|
|
|
32
|
|
|
Each Postprocessor implements a "run" method that takes a pointer to a |
33
|
|
|
text string, modifies it as necessary and returns a text string. |
34
|
|
|
|
35
|
|
|
Postprocessors must extend markdown.Postprocessor. |
36
|
|
|
|
37
|
|
|
""" |
38
|
|
|
|
39
|
|
|
def run(self, text): |
40
|
|
|
""" |
41
|
|
|
Subclasses of Postprocessor should implement a `run` method, which |
42
|
|
|
takes the html document as a single text string and returns a |
43
|
|
|
(possibly modified) string. |
44
|
|
|
|
45
|
|
|
""" |
46
|
|
|
pass # pragma: no cover |
47
|
|
|
|
48
|
|
|
|
49
|
|
|
class RawHtmlPostprocessor(Postprocessor): |
50
|
|
|
""" Restore raw html to the document. """ |
51
|
|
|
|
52
|
|
|
def run(self, text): |
53
|
|
|
""" Iterate over html stash and restore "safe" html. """ |
54
|
|
|
replacements = OrderedDict() |
55
|
|
|
for i in range(self.markdown.htmlStash.html_counter): |
56
|
|
|
html, safe = self.markdown.htmlStash.rawHtmlBlocks[i] |
57
|
|
|
if self.markdown.safeMode and not safe: |
58
|
|
|
if str(self.markdown.safeMode).lower() == 'escape': |
59
|
|
|
html = self.escape(html) |
60
|
|
|
elif str(self.markdown.safeMode).lower() == 'remove': |
61
|
|
|
html = '' |
62
|
|
|
else: |
63
|
|
|
html = self.markdown.html_replacement_text |
64
|
|
|
if (self.isblocklevel(html) and |
65
|
|
|
(safe or not self.markdown.safeMode)): |
66
|
|
|
replacements["<p>%s</p>" % |
67
|
|
|
(self.markdown.htmlStash.get_placeholder(i))] = \ |
68
|
|
|
html + "\n" |
69
|
|
|
replacements[self.markdown.htmlStash.get_placeholder(i)] = html |
70
|
|
|
|
71
|
|
|
if replacements: |
72
|
|
|
pattern = re.compile("|".join(re.escape(k) for k in replacements)) |
73
|
|
|
text = pattern.sub(lambda m: replacements[m.group(0)], text) |
74
|
|
|
|
75
|
|
|
return text |
76
|
|
|
|
77
|
|
|
def escape(self, html): |
78
|
|
|
""" Basic html escaping """ |
79
|
|
|
html = html.replace('&', '&') |
80
|
|
|
html = html.replace('<', '<') |
81
|
|
|
html = html.replace('>', '>') |
82
|
|
|
return html.replace('"', '"') |
83
|
|
|
|
84
|
|
|
def isblocklevel(self, html): |
85
|
|
|
m = re.match(r'^\<\/?([^ >]+)', html) |
86
|
|
|
if m: |
87
|
|
|
if m.group(1)[0] in ('!', '?', '@', '%'): |
88
|
|
|
# Comment, php etc... |
89
|
|
|
return True |
90
|
|
|
return util.isBlockLevel(m.group(1)) |
91
|
|
|
return False |
92
|
|
|
|
93
|
|
|
|
94
|
|
|
class AndSubstitutePostprocessor(Postprocessor): |
95
|
|
|
""" Restore valid entities """ |
96
|
|
|
|
97
|
|
|
def run(self, text): |
98
|
|
|
text = text.replace(util.AMP_SUBSTITUTE, "&") |
99
|
|
|
return text |
100
|
|
|
|
101
|
|
|
|
102
|
|
|
class UnescapePostprocessor(Postprocessor): |
103
|
|
|
""" Restore escaped chars """ |
104
|
|
|
|
105
|
|
|
RE = re.compile(r'%s(\d+)%s' % (util.STX, util.ETX)) |
106
|
|
|
|
107
|
|
|
def unescape(self, m): |
108
|
|
|
return util.int2str(int(m.group(1))) |
109
|
|
|
|
110
|
|
|
def run(self, text): |
111
|
|
|
return self.RE.sub(self.unescape, text) |
112
|
|
|
|