Don't linkify text with .py filenames in markdown (#274)

coaster/utils/markdown.py

@@ -23,7 +23,13 @@
 from markupsafe import Markup
 from pymdownx.emoji import to_alt as emoji_to_alt
 
-from .text import VALID_TAGS, normalize_spaces_multiline, sanitize_html
+from .text import (
+    LINKIFY_CALLBACKS,
+    LINKIFY_SKIP_TAGS,
+    VALID_TAGS,
+    normalize_spaces_multiline,
+    sanitize_html,
+)
 
 __all__ = ['markdown', 'MARKDOWN_HTML_TAGS']
 
@@ -157,6 +163,7 @@ def markdown(text, html=False, valid_tags=None):
                     extensions=extensions_text,
                     extension_configs=extension_configs,
                 ).convert(text),
-                skip_tags=['pre'],
+                callbacks=LINKIFY_CALLBACKS,
+                skip_tags=LINKIFY_SKIP_TAGS,
             )
         )

coaster/utils/text.py

@@ -30,6 +30,8 @@
 
 __all__ = [
     'VALID_TAGS',
+    'LINKIFY_SKIP_TAGS',
+    'LINKIFY_CALLBACKS',
     'deobfuscate_email',
     'normalize_spaces',
     'normalize_spaces_multiline',
@@ -119,6 +121,26 @@
     'ul': [],
 }
 
+LINKIFY_SKIP_TAGS = ['pre', 'code', 'kbd', 'samp', 'var']
+
+
+# Adapted from https://bleach.readthedocs.io/en/latest/linkify.html#preventing-links
+def dont_linkify_filenames(attrs, new=False):
+    # This is an existing link, so leave it be
+    if not new:
+        return attrs
+    # If the TLD is '.py', make sure it starts with http: or https:.
+    # Use _text because that's the original text
+    link_text = attrs['_text']
+    if link_text.endswith('.py') and not link_text.startswith(('http:', 'https:')):
+        # This looks like a Python file, not a URL. Don't make a link.
+        return None
+    # Everything checks out, keep going to the next callback.
+    return attrs
+
+
+LINKIFY_CALLBACKS = list(DEFAULT_CALLBACKS) + [dont_linkify_filenames]
+
 
 def sanitize_html(value, valid_tags=None, strip=True, linkify=False):
     """
@@ -128,7 +150,9 @@ def sanitize_html(value, valid_tags=None, strip=True, linkify=False):
         valid_tags = VALID_TAGS
     if linkify:
         filters = [
-            partial(LinkifyFilter, skip_tags=['pre'], callbacks=DEFAULT_CALLBACKS)
+            partial(
+                LinkifyFilter, callbacks=LINKIFY_CALLBACKS, skip_tags=LINKIFY_SKIP_TAGS
+            )
         ]
     else:
         filters = []

tests/test_utils_markdown.py

@@ -30,9 +30,13 @@
 6. Symbols (ignored): (tm) (c) (r) c/o
 7. Symbols (converted): +/- --> <-- <--> =/= 1/2 1/4 1st 2nd 3rd 4th 42nd
 
+An [inline link](https://www.example.com/)
+
 A naked link: https://www.example.com/
 
-An [inline link](https://www.example.com/)
+Un-prefixed naked links: hasgeek.in python.py
+
+Python/Paraguay link, prefixed: http://python.py
 
 ```python hl_lines="1"
 def foo():
@@ -68,8 +72,10 @@ def foo():
 <li>Symbols (ignored): (tm) (c) (r) c/o</li>
 <li>Symbols (converted): ± → ← ↔ ≠ ½ ¼ 1<sup>st</sup> 2<sup>nd</sup> 3<sup>rd</sup> 4<sup>th</sup> 42<sup>nd</sup></li>
 </ol>
-<p>A naked link: <a href="https://www.example.com/" rel="nofollow">https://www.example.com/</a></p>
 <p>An <a href="https://www.example.com/" rel="nofollow">inline link</a></p>
+<p>A naked link: <a href="https://www.example.com/" rel="nofollow">https://www.example.com/</a></p>
+<p>Un-prefixed naked links: <a href="http://hasgeek.in" rel="nofollow">hasgeek.in</a> python.py</p>
+<p>Python/Paraguay link, prefixed: <a href="http://python.py" rel="nofollow">http://python.py</a></p>
 <div class="highlight"><pre><span></span><code><span class="hll"><span class="k">def</span> <span class="nf">foo</span><span class="p">():</span>
 </span>    <span class="k">return</span> <span class="s1">'https://www.example.com/'</span>
 </code></pre></div>
@@ -133,8 +139,10 @@ def foo():
 <li>Symbols (ignored): (tm) (c) (r) c/o</li>
 <li>Symbols (converted): &plusmn; &rarr; &larr; &harr; &ne; &frac12; &frac14; 1<sup>st</sup> 2<sup>nd</sup> 3<sup>rd</sup> 4<sup>th</sup> 42<sup>nd</sup></li>
 </ol>
-<p>A naked link: <a href="https://www.example.com/" rel="nofollow">https://www.example.com/</a></p>
 <p>An <a href="https://www.example.com/" rel="nofollow">inline link</a></p>
+<p>A naked link: <a href="https://www.example.com/" rel="nofollow">https://www.example.com/</a></p>
+<p>Un-prefixed naked links: <a href="http://hasgeek.in" rel="nofollow">hasgeek.in</a> python.py</p>
+<p>Python/Paraguay link, prefixed: <a href="http://python.py" rel="nofollow">http://python.py</a></p>
 <pre><code>def foo():
     return &#39;https://www.example.com/&#39;
 </code></pre>