Keep path around, add full_path

Author: stsewd

readthedocs/projects/models.py

@@ -1254,7 +1254,8 @@ def get_processed_json(self):
         return {
             'headers': [],
             'content': '',
-            'path': self.path,
+            'path': file_path,
+            'full_path': self.path,
             'title': '',
             'sections': [],
         }

readthedocs/search/api.py

@@ -42,15 +42,14 @@ def get_link(self, obj):
         projects_url = self.context.get('projects_url')
         if projects_url:
             docs_url = projects_url[obj.project]
-            path = os.path.splitext(obj.path)[0]
-            return docs_url + path
+            return docs_url + obj.path
 
     def get_url(self, obj):
         """Gets the full url."""
         projects_url = self.context.get('projects_url')
         if projects_url:
             docs_url = projects_url[obj.project]
-            return docs_url + obj.path
+            return docs_url + obj.full_path
 
     def get_highlight(self, obj):
         highlight = getattr(obj.meta, 'highlight', None)

readthedocs/search/documents.py

@@ -116,6 +116,7 @@ class PageDocument(RTDDocTypeMixin, DocType):
     project = fields.KeywordField(attr='project.slug')
     version = fields.KeywordField(attr='version.slug')
     path = fields.KeywordField(attr='processed_json.path')
+    full_path = fields.KeywordField(attr='processed_json.full_path')
 
     # Searchable content
     title = fields.TextField(attr='processed_json.title')
@@ -153,7 +154,7 @@ def faceted_search(
 
     def get_queryset(self):
         """Overwrite default queryset to filter certain files to index."""
-        queryset = super(PageDocument, self).get_queryset()
+        queryset = super().get_queryset()
 
         # Do not index files that belong to non sphinx project
         # Also do not index certain files

readthedocs/search/parse_json.py

@@ -69,9 +69,15 @@ def process_file(fjson_filename, filename):
         raise
     data = json.loads(file_contents)
     sections = []
+    path = ''
     title = ''
     body_content = ''
 
+    if 'current_page_name' in data:
+        path = data['current_page_name']
+    else:
+        log.info('Unable to index file due to no name %s', filename)
+
     if data.get('body'):
         body = PyQuery(data['body'])
         body_content = body.text().replace('¶', '')
@@ -89,7 +95,8 @@ def process_file(fjson_filename, filename):
     return {
         'headers': process_headers(data, fjson_filename),
         'content': body_content,
-        'path': filename,
+        'path': path,
+        'full_path': filename,
         'title': title,
         'sections': sections,
     }