Merge pull request #743 from ybai08/fix/header-ids-from-rendered-text

Add a strip-syntax option to header-ids to keep markdown syntax out of ids (#192)

Author: nicholasserra

CHANGES.md

@@ -24,6 +24,7 @@
 - [pull #734] Fix empty list items being parsed as nested lists (#167)
 - [pull #735] Fix `smarty-pants` extra converting quotes inside inline HTML tags and comments (#150)
 - [pull #737] Fix XSS from code spans that the span gamut doesn't treat as code spans in safe mode (#726)
+- [pull #743] Add a `strip-syntax` option to the `header-ids` extra that leaves link URLs, image paths, inline HTML tags and emphasis markers out of header ids (#192)
 - [pull #747] Fix `link-patterns` extra linking text inside Markdown links and images whose label wraps onto another line
 
 

lib/markdown2.py

@@ -56,7 +56,9 @@
 * footnotes: Support footnotes as in use on daringfireball.net and
   implemented in other Markdown processors (tho not in Markdown.pl v1.0.1).
 * header-ids: Adds "id" attributes to headers. The id value is a slug of
-  the header text.
+  the header text. With `{"strip-syntax": True}`, the slug is made from the
+  rendered header, so link URLs, image paths, inline HTML tags and emphasis
+  markers are left out of it.
 * highlightjs-lang: Allows specifying the language which used for syntax
   highlighting when using fenced-code-blocks and highlightjs.
 * html-classes: Takes a dict mapping html tag names (lowercase) to a
@@ -1667,6 +1669,8 @@ def _do_links(self, text: str) -> str:
             text = link_processor.run(text)
         return text
 
+    _html_tag_re = re.compile(r'<[^>]*>')
+
     def header_id_from_text(self,
         text: str,
         prefix: str,
@@ -1751,16 +1755,26 @@ def _h_sub(self, match: re.Match[str]) -> str:
         if demote_headers:
             n = min(n + demote_headers, 6)
         header_id_attr = ""
+        html = self._run_span_gamut(header_group)
         if "header-ids" in self.extras:
             # in safe mode, inline HTML is hashed by now. Unhash it so the
             # sanitized HTML doesn't get swapped back into the id attribute
-            header_text = self._unhash_html_spans(header_group)
+            if self.extras["header-ids"].get('strip-syntax'):
+                # build the id from the rendered header so that markdown syntax
+                # (link URLs, image paths, emphasis markers) doesn't end up in it.
+                header_text = self._html_tag_re.sub('', self._unhash_html_spans(html))
+                if "footnotes" in self.extras:
+                    # footnote refs are numbered later on. Drop the placeholder
+                    # so it doesn't end up in the id
+                    marker = self._footnote_marker + '-'
+                    header_text = header_text.replace(marker, '')
+            else:
+                header_text = self._unhash_html_spans(header_group)
             header_id = self.header_id_from_text(
                 self._unescape_special_chars(header_text),
                 self.extras["header-ids"].get('prefix'), n)
             if header_id:
                 header_id_attr = ' id="%s"' % header_id
-        html = self._run_span_gamut(header_group)
         if "toc" in self.extras and header_id:
             self._toc_add_entry(n, header_id, html)
         return "<h%d%s>%s</h%d>\n\n" % (n, header_id_attr, html, n)

test/tm-cases/header_ids_footnote_ref.html

@@ -0,0 +1,16 @@
+<h1 id="header-with-a-footnote1">Header with a footnote<sup class="footnote-ref" id="fnref-1"><a href="#fn-1">1</a></sup></h1>
+
+<h1 id="linked-headernote"><a href="http://example.com">Linked</a> header<sup class="footnote-ref" id="fnref-note"><a href="#fn-note">2</a></sup></h1>
+
+<div class="footnotes">
+<hr />
+<ol>
+<li id="fn-1">
+<p>The first note.&#160;<a href="#fnref-1" class="footnoteBackLink" title="Jump back to footnote 1 in the text.">&#8617;</a></p>
+</li>
+
+<li id="fn-note">
+<p>The second note.&#160;<a href="#fnref-note" class="footnoteBackLink" title="Jump back to footnote 2 in the text.">&#8617;</a></p>
+</li>
+</ol>
+</div>

test/tm-cases/header_ids_footnote_ref.opts

@@ -0,0 +1 @@
+{"extras": {"header-ids": {"strip-syntax": True}, "footnotes": None}}

test/tm-cases/header_ids_footnote_ref.tags

@@ -0,0 +1 @@
+extra header-ids footnotes issue192

test/tm-cases/header_ids_footnote_ref.text

@@ -0,0 +1,6 @@
+# Header with a footnote[^1]
+
+# [Linked](http://example.com) header[^note]
+
+[^1]: The first note.
+[^note]: The second note.

test/tm-cases/header_ids_footnote_ref_default.html

@@ -0,0 +1,16 @@
+<h1 id="header-with-a-footnote1">Header with a footnote<sup class="footnote-ref" id="fnref-1"><a href="#fn-1">1</a></sup></h1>
+
+<h1 id="linkedhttpexamplecom-headernote"><a href="http://example.com">Linked</a> header<sup class="footnote-ref" id="fnref-note"><a href="#fn-note">2</a></sup></h1>
+
+<div class="footnotes">
+<hr />
+<ol>
+<li id="fn-1">
+<p>The first note.&#160;<a href="#fnref-1" class="footnoteBackLink" title="Jump back to footnote 1 in the text.">&#8617;</a></p>
+</li>
+
+<li id="fn-note">
+<p>The second note.&#160;<a href="#fnref-note" class="footnoteBackLink" title="Jump back to footnote 2 in the text.">&#8617;</a></p>
+</li>
+</ol>
+</div>

test/tm-cases/header_ids_footnote_ref_default.opts

@@ -0,0 +1 @@
+{"extras": ["header-ids", "footnotes"]}

test/tm-cases/header_ids_footnote_ref_default.tags

@@ -0,0 +1 @@
+extra header-ids footnotes issue192

test/tm-cases/header_ids_footnote_ref_default.text

@@ -0,0 +1,6 @@
+# Header with a footnote[^1]
+
+# [Linked](http://example.com) header[^note]
+
+[^1]: The first note.
+[^note]: The second note.