Merge pull request #179 from canonical/handle-images

Handle images in links

canonicalwebteam/discourse/parsers/base_parser.py

@@ -531,6 +531,7 @@ def _process_topic_soup(self, soup):
 
         soup = self._replace_notifications(soup)
         soup = self._replace_notes_to_editors(soup)
+        soup = self._replace_image_src(soup)
         soup = self._replace_links(soup)
         soup = self._replace_polls(soup)
 
@@ -556,8 +557,20 @@ def _replace_links(self, soup, topics=[]):
             full_link = a.get("href", "")
             self._replace_text_link(a, topics)
 
+            # For images, link to the uploaded file
+            if a.find("img") and full_link.startswith("/uploads/"):
+                a["href"] = os.path.join(
+                    self.api.base_url, full_link.lstrip("/")
+                )
+                continue
+
             # For user references link to discourse profile pages
-            if full_link.startswith("/u") and a.string.startswith("@"):
+            if (
+                full_link
+                and full_link.startswith("/u/")
+                and a.string
+                and a.string.startswith("@")
+            ):
                 a["href"] = os.path.join(
                     self.api.base_url, full_link.lstrip("/")
                 )
@@ -592,6 +605,19 @@ def _replace_links(self, soup, topics=[]):
 
         return soup
 
+    def _replace_image_src(self, soup):
+        """
+        Given some HTML soup, replace relative image srcs
+        """
+        for img in soup.findAll("img"):
+            src = img.get("src", "")
+            if src and src.startswith("/"):
+                img["src"] = f"{self.api.base_url}{src}"
+            if img["srcset"]:
+                del img["srcset"]
+
+        return soup
+
     @cached_property
     def _notification_template(self):
         notification_html = (

setup.py

@@ -4,7 +4,7 @@
 
 setup(
     name="canonicalwebteam.discourse",
-    version="5.4.5",
+    version="5.4.6",
     author="Canonical webteam",
     author_email="[email protected]",
     url="https://github.com/canonical/canonicalwebteam.discourse",

tests/test_parser.py

@@ -106,6 +106,48 @@ def test_parser_username_link(self):
             parsed_topic["body_html"],
         )
 
+    def test_parser_upload(self):
+        discourse_api = DiscourseAPI("https://base.url", session=MagicMock())
+
+        parser = BaseParser(
+            api=discourse_api,
+            index_topic_id=1,
+            url_prefix="/",
+        )
+
+        parsed_topic = parser.parse_topic(
+            {
+                "id": 1,
+                "category_id": 1,
+                "title": "Sample",
+                "slug": "sample—text",
+                "post_stream": {
+                    "posts": [
+                        {
+                            "id": 11,
+                            "cooked": (
+                                "<a href='/uploads/test.png'>"
+                                "<img src='test.png' srcset='test.png' />"
+                                "</a>"
+                                "<a>No Link</a>"
+                                "<a></a>"
+                            ),
+                            "updated_at": "2018-10-02T12:45:44.259Z",
+                        }
+                    ],
+                },
+            }
+        )
+
+        self.assertIn(
+            (
+                '<a href="https://base.url/uploads/test.png">'
+                '<img src="test.png"/>'
+                "</a>"
+            ),
+            parsed_topic["body_html"],
+        )
+
     def test_emdash_in_slug(self):
         discourse_api = DiscourseAPI("https://base.url", session=MagicMock())