| 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 |
import util |
| 12 |
import odict |
| 13 |
|
| 14 |
def build_postprocessors(md_instance, **kwargs): |
| 15 |
""" Build the default postprocessors for Markdown. """ |
| 16 |
postprocessors = odict.OrderedDict() |
| 17 |
postprocessors["raw_html"] = RawHtmlPostprocessor(md_instance) |
| 18 |
postprocessors["amp_substitute"] = AndSubstitutePostprocessor() |
| 19 |
return postprocessors |
| 20 |
|
| 21 |
|
| 22 |
class Postprocessor(util.Processor): |
| 23 |
""" |
| 24 |
Postprocessors are run after the ElementTree it converted back into text. |
| 25 |
|
| 26 |
Each Postprocessor implements a "run" method that takes a pointer to a |
| 27 |
text string, modifies it as necessary and returns a text string. |
| 28 |
|
| 29 |
Postprocessors must extend markdown.Postprocessor. |
| 30 |
|
| 31 |
""" |
| 32 |
|
| 33 |
def run(self, text): |
| 34 |
""" |
| 35 |
Subclasses of Postprocessor should implement a `run` method, which |
| 36 |
takes the html document as a single text string and returns a |
| 37 |
(possibly modified) string. |
| 38 |
|
| 39 |
""" |
| 40 |
pass |
| 41 |
|
| 42 |
|
| 43 |
class RawHtmlPostprocessor(Postprocessor): |
| 44 |
""" Restore raw html to the document. """ |
| 45 |
|
| 46 |
def run(self, text): |
| 47 |
""" Iterate over html stash and restore "safe" html. """ |
| 48 |
for i in range(self.markdown.htmlStash.html_counter): |
| 49 |
html, safe = self.markdown.htmlStash.rawHtmlBlocks[i] |
| 50 |
html = self.unescape(html) |
| 51 |
if self.markdown.safeMode and not safe: |
| 52 |
if str(self.markdown.safeMode).lower() == 'escape': |
| 53 |
html = self.escape(html) |
| 54 |
elif str(self.markdown.safeMode).lower() == 'remove': |
| 55 |
html = '' |
| 56 |
else: |
| 57 |
html = self.markdown.html_replacement_text |
| 58 |
if safe or not self.markdown.safeMode: |
| 59 |
text = text.replace("<p>%s</p>" % |
| 60 |
(self.markdown.htmlStash.get_placeholder(i)), |
| 61 |
html + "\n") |
| 62 |
text = text.replace(self.markdown.htmlStash.get_placeholder(i), |
| 63 |
html) |
| 64 |
return text |
| 65 |
|
| 66 |
def unescape(self, html): |
| 67 |
""" Unescape any markdown escaped text within inline html. """ |
| 68 |
for k, v in self.markdown.treeprocessors['inline'].stashed_nodes.items(): |
| 69 |
ph = util.INLINE_PLACEHOLDER % k |
| 70 |
html = html.replace(ph, '\%s' % v) |
| 71 |
return html |
| 72 |
|
| 73 |
def escape(self, html): |
| 74 |
""" Basic html escaping """ |
| 75 |
html = html.replace('&', '&') |
| 76 |
html = html.replace('<', '<') |
| 77 |
html = html.replace('>', '>') |
| 78 |
return html.replace('"', '"') |
| 79 |
|
| 80 |
|
| 81 |
class AndSubstitutePostprocessor(Postprocessor): |
| 82 |
""" Restore valid entities """ |
| 83 |
def __init__(self): |
| 84 |
pass |
| 85 |
|
| 86 |
def run(self, text): |
| 87 |
text = text.replace(util.AMP_SUBSTITUTE, "&") |
| 88 |
return text |
