Move process_text_links back to the client app. Skip related tests. Convert it to BeautifulSoup. Remove unused imports.

federation/entities/mixins.py

@@ -10,7 +10,7 @@ from marshmallow import missing
 
 from federation.entities.activitypub.enums import ActivityType
 from federation.entities.utils import get_name_for_profile, get_profile
-from federation.utils.text import process_text_links, find_elements, find_tags, MENTION_PATTERN
+from federation.utils.text import find_elements, find_tags, MENTION_PATTERN
 
 
 class BaseEntity:

federation/tests/entities/activitypub/test_entities.py

@@ -10,7 +10,6 @@ from federation.entities.activitypub.models import context_manager
 from federation.entities.activitypub.models import Accept
 from federation.tests.fixtures.keys import PUBKEY
 from federation.types import UserType
-from federation.utils.text import process_text_links
 
 
 class TestEntitiesConvertToAS2:
@@ -65,6 +64,8 @@ class TestEntitiesConvertToAS2:
             'published': '2019-04-27T00:00:00',
         }
 
+    # Now handled by the client app
+    @pytest.mark.skip
     def test_comment_to_as2__url_in_raw_content(self, activitypubcomment):
         activitypubcomment.raw_content = 'raw_content http://example.com'
         activitypubcomment.rendered_content = process_text_links(

federation/tests/utils/test_text.py

@@ -1,4 +1,6 @@
-from federation.utils.text import decode_if_bytes, encode_if_text, validate_handle, process_text_links, find_tags
+import pytest
+
+from federation.utils.text import decode_if_bytes, encode_if_text, validate_handle, find_tags
 
 
 def test_decode_if_bytes():
@@ -63,6 +65,8 @@ class TestFindTags:
         assert tags == {"foobar", "barfoo"}
 
 
+# TODO: move these tests to the client app
+@pytest.mark.skip
 class TestProcessTextLinks:
     def test_link_at_start_or_end(self):
         assert process_text_links('https://example.org example.org\nhttp://example.org') == \

federation/utils/text.py

@@ -2,8 +2,6 @@ import re
 from typing import Set, List
 from urllib.parse import urlparse
 
-import bleach
-from bleach import callbacks
 from bs4 import BeautifulSoup
 from bs4.element import NavigableString
 from commonmark import commonmark
@@ -12,6 +10,7 @@ ILLEGAL_TAG_CHARS = "!#$%^&*+.,@£/()=?`'\\{[]}~;:\"’”—\xa0"
 TAG_PATTERN = re.compile(r'(#[\w]+)', re.UNICODE)
 # This will match non matching braces. I don't think it's an issue.
 MENTION_PATTERN = re.compile(r'(@\{?(?:[\w\-. \u263a-\U0001f645]*; *)?[\w]+@[\w\-.]+\.[\w]+}?)', re.UNICODE)
+URL_PATTERN = re.compile(r'(https?://[\w_\-.#?&/]+)', re.UNICODE)
 
 def decode_if_bytes(text):
     try:
@@ -65,28 +64,6 @@ def get_path_from_url(url: str) -> str:
     return parsed.path
 
 
-def process_text_links(text):
-    """Process links in text, adding some attributes and linkifying textual links."""
-    link_callbacks = [callbacks.nofollow, callbacks.target_blank]
-
-    def link_attributes(attrs, new=False):
-        """Run standard callbacks except for internal links."""
-        href_key = (None, "href")
-        if attrs.get(href_key, "").startswith("/"):
-            return attrs
-
-        # Run the standard callbacks
-        for callback in link_callbacks:
-            attrs = callback(attrs, new)
-        return attrs
-
-    return bleach.linkify(
-        text,
-        callbacks=[link_attributes],
-        parse_email=False,
-        skip_tags=["code"],
-    )
-
 
 def test_tag(tag: str) -> bool:
     """Test a word whether it could be accepted as a tag."""