Refactor up the embedded image recognition code from ActivityPub AttachImagesMixin to the base RawContentMixin so Matrix can use that too to upload images pre sending.

federation/entities/activitypub/entities.py

@@ -1,5 +1,4 @@
 import logging
-import re
 import uuid
 from typing import Dict, List
 
@@ -26,16 +25,11 @@ class AttachImagesMixin(RawContentMixin):
         Attach any embedded images from raw_content.
         """
         super().pre_send()
-        if self._media_type != "text/markdown":
-            return
-        regex = r"!\[([\w ]*)\]\((https?://[\w\d\-\./]+\.[\w]*((?<=jpg)|(?<=gif)|(?<=png)|(?<=jpeg)))\)"
-        matches = re.finditer(regex, self.raw_content, re.MULTILINE | re.IGNORECASE)
-        for match in matches:
-            groups = match.groups()
+        for image in self.embedded_images:
             self._children.append(
                 ActivitypubImage(
-                    url=groups[1],
-                    name=groups[0] or "",
+                    url=image[0],
+                    name=image[1],
                     inline=True,
                 )
             )

federation/entities/mixins.py

@@ -2,7 +2,7 @@ import datetime
 import importlib
 import re
 import warnings
-from typing import List, Set, Union, Dict
+from typing import List, Set, Union, Dict, Tuple
 
 from commonmark import commonmark
 
@@ -202,6 +202,24 @@ class RawContentMixin(BaseEntity):
         super().__init__(*args, **kwargs)
         self._required += ["raw_content"]
 
+    @property
+    def embedded_images(self) -> List[Tuple[str, str]]:
+        """
+        Returns a list of images from the raw_content.
+        Currently only markdown supported.
+
+        Returns a Tuple of (url, filename).
+        """
+        images = []
+        if self._media_type != "text/markdown":
+            return images
+        regex = r"!\[([\w ]*)\]\((https?://[\w\d\-\./]+\.[\w]*((?<=jpg)|(?<=gif)|(?<=png)|(?<=jpeg)))\)"
+        matches = re.finditer(regex, self.raw_content, re.MULTILINE | re.IGNORECASE)
+        for match in matches:
+            groups = match.groups()
+            images.append((groups[1], groups[0] or ""))
+        return images
+
     @property
     def rendered_content(self) -> str:
         """Returns the rendered version of raw_content, or just raw_content."""