• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1"""
2Python Markdown
3
4A Python implementation of John Gruber's Markdown.
5
6Documentation: https://python-markdown.github.io/
7GitHub: https://github.com/Python-Markdown/markdown/
8PyPI: https://pypi.org/project/Markdown/
9
10Started by Manfred Stienstra (http://www.dwerg.net/).
11Maintained for a few years by Yuri Takhteyev (http://www.freewisdom.org).
12Currently maintained by Waylan Limberg (https://github.com/waylan),
13Dmitry Shachnev (https://github.com/mitya57) and Isaac Muse (https://github.com/facelessuser).
14
15Copyright 2007-2018 The Python Markdown Project (v. 1.7 and later)
16Copyright 2004, 2005, 2006 Yuri Takhteyev (v. 0.2-1.6b)
17Copyright 2004 Manfred Stienstra (the original version)
18
19License: BSD (see LICENSE.md for details).
20
21POST-PROCESSORS
22=============================================================================
23
24Markdown also allows post-processors, which are similar to preprocessors in
25that they need to implement a "run" method. However, they are run after core
26processing.
27
28"""
29
30from collections import OrderedDict
31from . import util
32import re
33
34
35def build_postprocessors(md, **kwargs):
36    """ Build the default postprocessors for Markdown. """
37    postprocessors = util.Registry()
38    postprocessors.register(RawHtmlPostprocessor(md), 'raw_html', 30)
39    postprocessors.register(AndSubstitutePostprocessor(), 'amp_substitute', 20)
40    return postprocessors
41
42
43class Postprocessor(util.Processor):
44    """
45    Postprocessors are run after the ElementTree it converted back into text.
46
47    Each Postprocessor implements a "run" method that takes a pointer to a
48    text string, modifies it as necessary and returns a text string.
49
50    Postprocessors must extend markdown.Postprocessor.
51
52    """
53
54    def run(self, text):
55        """
56        Subclasses of Postprocessor should implement a `run` method, which
57        takes the html document as a single text string and returns a
58        (possibly modified) string.
59
60        """
61        pass  # pragma: no cover
62
63
64class RawHtmlPostprocessor(Postprocessor):
65    """ Restore raw html to the document. """
66
67    BLOCK_LEVEL_REGEX = re.compile(r'^\<\/?([^ >]+)')
68
69    def run(self, text):
70        """ Iterate over html stash and restore html. """
71        replacements = OrderedDict()
72        for i in range(self.md.htmlStash.html_counter):
73            html = self.stash_to_string(self.md.htmlStash.rawHtmlBlocks[i])
74            if self.isblocklevel(html):
75                replacements["<p>{}</p>".format(
76                    self.md.htmlStash.get_placeholder(i))] = html
77            replacements[self.md.htmlStash.get_placeholder(i)] = html
78
79        def substitute_match(m):
80            key = m.group(0)
81
82            if key not in replacements:
83                if key[3:-4] in replacements:
84                    return f'<p>{ replacements[key[3:-4]] }</p>'
85                else:
86                    return key
87
88            return replacements[key]
89
90        if replacements:
91            base_placeholder = util.HTML_PLACEHOLDER % r'([0-9]+)'
92            pattern = re.compile(f'<p>{ base_placeholder }</p>|{ base_placeholder }')
93            processed_text = pattern.sub(substitute_match, text)
94        else:
95            return text
96
97        if processed_text == text:
98            return processed_text
99        else:
100            return self.run(processed_text)
101
102    def isblocklevel(self, html):
103        m = self.BLOCK_LEVEL_REGEX.match(html)
104        if m:
105            if m.group(1)[0] in ('!', '?', '@', '%'):
106                # Comment, php etc...
107                return True
108            return self.md.is_block_level(m.group(1))
109        return False
110
111    def stash_to_string(self, text):
112        """ Convert a stashed object to a string. """
113        return str(text)
114
115
116class AndSubstitutePostprocessor(Postprocessor):
117    """ Restore valid entities """
118
119    def run(self, text):
120        text = text.replace(util.AMP_SUBSTITUTE, "&")
121        return text
122
123
124@util.deprecated(
125    "This class will be removed in the future; "
126    "use 'treeprocessors.UnescapeTreeprocessor' instead."
127)
128class UnescapePostprocessor(Postprocessor):
129    """ Restore escaped chars """
130
131    RE = re.compile(r'{}(\d+){}'.format(util.STX, util.ETX))
132
133    def unescape(self, m):
134        return chr(int(m.group(1)))
135
136    def run(self, text):
137        return self.RE.sub(self.unescape, text)
138