Search: don't index permalinks

readthedocs/search/parsers.py

@@ -166,11 +166,13 @@ def _clean_body(self, body):
 
            This will mutate the original `body`.
         """
-        # Remove all navigation nodes
         nodes_to_be_removed = itertools.chain(
+            # Navigation nodes
             body.css('nav'),
             body.css('[role=navigation]'),
             body.css('[role=search]'),
+            # Permalinks
+            body.css('.headerlink'),
         )
         for node in nodes_to_be_removed:
             node.decompose()

readthedocs/search/tests/data/sphinx/in/page.html

@@ -17,6 +17,17 @@ <h2>Sub-title one<a class="headerlink" href="#sub-title-one" title="Permalink to
     <div class="section" id="subsub-title">
       <h3>Subsub title<a class="headerlink" href="#subsub-title" title="Permalink to this headline">¶</a></h3>
       <p>This is a H3 title.</p>
+
+      <!-- Figure with permalink and caption -->
+      <div class="figure align-center" id="id1">
+        <a class="reference internal image-reference" href="_images/search-analytics-demo.png">
+          <img alt="Search analytics demo" src="_images/figure.png" style="width: 50%;">
+        </a>
+        <p class="caption">
+          <span class="caption-number">Fig. 4 </span><span class="caption-text">I'm a figure!</span>
+          <a class="headerlink" href="#id1" title="Permalink to this image">¶</a>
+        </p>
+      </div>
     </div>
   </div>
 

readthedocs/search/tests/data/sphinx/out/page.json

@@ -28,7 +28,7 @@
       "content": "Sphinx configuration file used to build this docs: # -*- coding: utf-8 -*- # Default settings project = 'Test Builds' extensions = [ 'sphinx_autorun', ] latex_engine = 'xelatex' # allow us to build Unicode chars # Include all your settings here html_theme = 'sphinx_rtd_theme' >>> # Build at >>> import datetime >>> datetime.datetime.utcnow() # UTC datetime.datetime(2020, 5, 3, 16, 38, 11, 137311)"
     },
     {
-      "content": "This is a H3 title.",
+      "content": "This is a H3 title. Fig. 4 I'm a figure!",
       "id": "subsub-title",
       "title": "Subsub title"
     }