Search: use alias to link to search results of subprojects

docs/server-side-search.rst

@@ -95,6 +95,7 @@ This is ``https://docs.readthedocs.io/_/api/v2/search`` for the ``docs`` project
 
    :>json string type: The type of the result, currently page is the only type.
    :>json string project: The project slug
+   :>json string project_alias: Alias of the project if it's a subproject.
    :>json string version: The version slug
    :>json string title: The title of the page
    :>json string domain: Canonical domain of the resulting page
@@ -143,6 +144,7 @@ This is ``https://docs.readthedocs.io/_/api/v2/search`` for the ``docs`` project
               {
                   "type": "page",
                   "project": "docs",
+                  "project_alias": null,
                   "version": "latest",
                   "title": "Server Side Search",
                   "domain": "https://docs.readthedocs.io",

readthedocs/core/static-src/core/js/doc-embed/search.js

@@ -79,7 +79,7 @@ function attach_elastic_search_query_sphinx(data) {
 
                         // If the document is from a subproject, add extra information
                         if (result.project !== project) {
-                            var text = " (from project " + result.project + ")";
+                            var text = " (from project " + result.project_alias + ")";
                             var extra = $('<span>', {'text': text});
                             list_item.append(extra);
                         }
@@ -295,7 +295,7 @@ function attach_elastic_search_query_mkdocs(data) {
                         );
 
                         if (result.project !== project) {
-                            var text = '(from project ' + result.project + ')';
+                            var text = '(from project ' + result.project_alias + ')';
                             item.append($('<span>', {'text': text}));
                         }
 

readthedocs/search/api.py

@@ -15,7 +15,7 @@
 from readthedocs.search import tasks
 from readthedocs.search.faceted_search import PageSearch
 
-from .serializers import PageSearchSerializer, VersionData
+from .serializers import PageSearchSerializer, ProjectData, VersionData
 
 log = logging.getLogger(__name__)
 
@@ -183,15 +183,21 @@ def _get_all_projects_data(self):
         .. code::
 
            {
-               "requests": VersionData(
-                   "latest",
-                   "sphinx",
-                   "https://requests.readthedocs.io/en/latest/",
+               "requests": ProjectData(
+                   alias='alias',
+                   version=VersionData(
+                        "latest",
+                        "sphinx",
+                        "https://requests.readthedocs.io/en/latest/",
+                    ),
                ),
-               "requests-oauth": VersionData(
-                   "latest",
-                   "sphinx_htmldir",
-                   "https://requests-oauth.readthedocs.io/en/latest/",
+               "requests-oauth": ProjectData(
+                   alias=None,
+                   version=VersionData(
+                       "latest",
+                       "sphinx_htmldir",
+                       "https://requests-oauth.readthedocs.io/en/latest/",
+                   ),
                ),
            }
 
@@ -203,10 +209,13 @@ def _get_all_projects_data(self):
         main_project = self._get_project()
 
         projects_data = {
-            main_project.slug: VersionData(
-                slug=main_version.slug,
-                doctype=main_version.documentation_type,
-                docs_url=main_project.get_docs_url(version_slug=main_version.slug),
+            main_project.slug: ProjectData(
+                alias=None,
+                version=VersionData(
+                    slug=main_version.slug,
+                    doctype=main_version.documentation_type,
+                    docs_url=main_project.get_docs_url(version_slug=main_version.slug),
+                ),
             )
         }
 
@@ -232,11 +241,16 @@ def _get_all_projects_data(self):
 
             if version and self._has_permission(self.request.user, version):
                 url = subproject.get_docs_url(version_slug=version.slug)
-                projects_data[subproject.slug] = VersionData(
+                project_alias = subproject.superprojects.values_list('alias', flat=True).first()
+                version_data = VersionData(
                     slug=version.slug,
                     doctype=version.documentation_type,
                     docs_url=url,
                 )
+                projects_data[subproject.slug] = ProjectData(
+                    alias=project_alias,
+                    version=version_data,
+                )
 
         return projects_data
 
@@ -299,8 +313,8 @@ def get_queryset(self):
 
         if main_project.has_feature(Feature.SEARCH_SUBPROJECTS_ON_DEFAULT_VERSION):
             projects = {
-                project: version.slug
-                for project, version in self._get_all_projects_data().items()
+                project: project_data.version.slug
+                for project, project_data in self._get_all_projects_data().items()
             }
             # Check to avoid searching all projects in case it's empty.
             if not projects:

readthedocs/search/documents.py

@@ -69,6 +69,7 @@ class PageDocument(RTDDocTypeMixin, Document):
     # Metadata
     project = fields.KeywordField(attr='project.slug')
     version = fields.KeywordField(attr='version.slug')
+    doctype = fields.KeywordField(attr='version.documentation_type')
     path = fields.KeywordField(attr='processed_json.path')
     full_path = fields.KeywordField(attr='path')
     rank = fields.IntegerField()

readthedocs/search/serializers.py

@@ -21,6 +21,7 @@
 
 
 # Structure used for storing cached data of a version mostly.
+ProjectData = namedtuple('ProjectData', ['version', 'alias'])
 VersionData = namedtuple('VersionData', ['slug', 'docs_url', 'doctype'])
 
 
@@ -58,13 +59,51 @@ class PageSearchSerializer(serializers.Serializer):
 
     type = serializers.CharField(default='page', source=None, read_only=True)
     project = serializers.CharField()
+    project_alias = serializers.SerializerMethodField()
     version = serializers.CharField()
     title = serializers.CharField()
     path = serializers.SerializerMethodField()
     domain = serializers.SerializerMethodField()
     highlights = PageHighlightSerializer(source='meta.highlight', default=dict)
     blocks = serializers.SerializerMethodField()
 
+    def _get_project_data(self, obj):
+        """
+        Get and cache the project data.
+
+        Try to get the data from the ``projects_data`` context,
+        and fallback to get it from the database.
+        If the result is fetched from the database,
+        it's cached into ``projects_data``.
+        """
+        project_data = self.context.get('projects_data', {}).get(obj.project)
+        if project_data:
+            return project_data
+
+        project = Project.objects.filter(slug=obj.project).first()
+        if project:
+            docs_url = project.get_docs_url(version_slug=obj.version)
+            project_alias = project.superprojects.values_list('alias', flat=True).first()
+
+            projects_data = self.context.setdefault('projects_data', {})
+            version_data = VersionData(
+                slug=obj.version,
+                docs_url=docs_url,
+                doctype=None,
+            )
+            projects_data[obj.project] = ProjectData(
+                alias=project_alias,
+                version=version_data,
+            )
+            return projects_data[obj.project]
+        return None
+
+    def get_project_alias(self, obj):
+        project_data = self._get_project_data(obj)
+        if project_data:
+            return project_data.alias
+        return None
+
     def get_domain(self, obj):
         full_path = self._get_full_path(obj)
         if full_path:
@@ -80,40 +119,17 @@ def get_path(self, obj):
         return None
 
     def _get_full_path(self, obj):
-        """
-        Get the page link.
-
-        Try to get the link from the ``project_data`` context,
-        and fallback to get it from the database.
-        If the result is fetched from the database,
-        it's cached into ``project_data``.
-        """
-        # First try to build the URL from the context.
-        version_data = self.context.get('projects_data', {}).get(obj.project)
-        if version_data:
-            docs_url = version_data.docs_url
+        project_data = self._get_project_data(obj)
+        if project_data:
+            docs_url = project_data.version.docs_url
             path = obj.full_path
 
             # Generate an appropriate link for the doctypes that use htmldir,
             # and always end it with / so it goes directly to proxito.
-            if version_data.doctype in {SPHINX_HTMLDIR, MKDOCS}:
+            if project_data.version.doctype in {SPHINX_HTMLDIR, MKDOCS}:
                 path = re.sub('(^|/)index.html$', '/', path)
 
             return docs_url.rstrip('/') + '/' + path.lstrip('/')
-
-        # Fallback to build the URL querying the db.
-        project = Project.objects.filter(slug=obj.project).first()
-        if project:
-            docs_url = project.get_docs_url(version_slug=obj.version)
-            # cache the project URL
-            projects_data = self.context.setdefault('projects_data', {})
-            projects_data[obj.project] = VersionData(
-                slug=obj.version,
-                docs_url=docs_url,
-                doctype=None,
-            )
-            return docs_url + obj.full_path
-
         return None
 
     def get_blocks(self, obj):

readthedocs/search/tests/test_api.py

@@ -182,6 +182,7 @@ def test_doc_search_filter_by_version(self, api_client, project):
         data = resp.data['results']
         assert len(data) == 1
         assert data[0]['project'] == project.slug
+        assert data[0]['project_alias'] is None
 
     def test_doc_search_pagination(self, api_client, project):
         """Test Doc search result can be paginated"""
@@ -264,6 +265,7 @@ def test_doc_search_subprojects(self, api_client, all_projects):
         # First result should be the subproject
         first_result = data[0]
         assert first_result['project'] == subproject.slug
+        assert first_result['project_alias'] == subproject.slug
         # The result is from the same version as the main project.
         assert first_result['version'] == version.slug
         # Check the link is the subproject document link

readthedocs/search/views.py

@@ -15,6 +15,7 @@
 
 from .serializers import (
     PageSearchSerializer,
+    ProjectData,
     ProjectSearchSerializer,
     VersionData,
 )
@@ -62,11 +63,15 @@ def _get_project_data(self, project, version_slug):
             .get(slug=version_slug)
         )
         docs_url = project.get_docs_url(version_slug=version_slug)
+        version_data = VersionData(
+            slug=version_slug,
+            docs_url=docs_url,
+            doctype=version_doctype,
+        )
         project_data = {
-            project.slug: VersionData(
-                slug=version_slug,
-                docs_url=docs_url,
-                doctype=version_doctype,
+            project.slug: ProjectData(
+                alias=None,
+                version=version_data,
             )
         }
         return project_data