setup.py

@@ -15,7 +15,6 @@ PACKAGE_NAME = 'socials_util'
 with open('README.md') as f:
     readme = f.read()
 
-
 def parse_version_file(text: str) -> str:
     match = re.match(r'^__version__\s*=\s*(["\'])([\d\.]+)\1$', text)
     if match is None:
@@ -23,11 +22,9 @@ def parse_version_file(text: str) -> str:
         raise Exception(msg)
     return match.group(2)
 
-
 with open(PACKAGE_NAME + '/_version.py') as f:
     version = parse_version_file(f.read())
 
-
 def parse_requirements(text: str) -> list[str]:
     return text.strip().split('\n')
 

socials_util/__init__.py

@@ -188,17 +188,12 @@ WIKIDATA_PROPERTIES: dict[SocialSiteId | int, WikidataInfo] = {
 
 
 def re_social_subdomain(main_domain: str) -> str:
-    return (
+    return r'^(?:https?:\/\/)?([\w_-]+)\.' + re.escape(main_domain) + r'(\/.*)?$'
-        r'^(?:https?:\/\/)?(?:www\.)?([\w_-]+)\.' + re.escape(main_domain) + r'(\/.*)?$'
-    )
 
 
-RE_ID = r'@?([^\s/]+)'
+RE_ID = r'@?([^/]+)'
-RE_DUAL_ID = r'@?([^\s/]+/[^\s/]+)'
+RE_DUAL_ID = r'@?([^/]+/[^/]+)'
-RE_ANY_SUBPATH = r'(|\/|\/\S*)$'
+RE_ANY_SUBPATH = r'(|\/|\/.*)$'
-
-SPECIAL_REGEX_LITERALS = frozenset({RE_ID, RE_DUAL_ID, RE_ANY_SUBPATH})
-DOES_NOT_NEED_AUTO_SLASH = frozenset({RE_ANY_SUBPATH})
 
 
 def re_social_path(main_domain: str) -> str:
@@ -217,13 +212,12 @@ def re_social_path_adv(main_domain: str, *path: str) -> str:
     ]
 
     for p in path:
-        if p not in DOES_NOT_NEED_AUTO_SLASH:
+        if p != RE_ANY_SUBPATH:
             regex_builder.append(r'\/')
         regex_builder.append(
-            p if p in SPECIAL_REGEX_LITERALS else re.escape(p),
+            p if p in {RE_ID, RE_DUAL_ID, RE_ANY_SUBPATH} else re.escape(p),
         )
-        del p
+    if path[-1] != RE_ANY_SUBPATH:
-    if path[-1] not in DOES_NOT_NEED_AUTO_SLASH:
         regex_builder.append(r'\/?$')
     return ''.join(regex_builder)
 
@@ -266,9 +260,7 @@ PIXIV_SKETCH_USER_NICKNAME_URL = re_social_path_adv('sketch.pixiv.net', RE_ID)
 
 URL_PARSE_CARRD_PAGE = re_social_subdomain('carrd.co')
 URL_PARSE_YOUTUBE_CHANNEL_HANDLE_1 = re_social_path_adv(
-    'youtube.com',
+    'youtube.com', RE_ID, RE_ANY_SUBPATH
-    RE_ID,
-    RE_ANY_SUBPATH,
 )
 URL_PARSE_YOUTUBE_CHANNEL_HANDLE_2 = re_social_path_adv('youtube.com', 'c', RE_ID)
 URL_PARSE_YOUTUBE_CHANNEL_ID = re_social_path_adv('youtube.com', 'channel', RE_ID)
@@ -421,7 +413,7 @@ WELL_KNOWN_MASTODON_INSTANCES: frozenset[str] = frozenset(
     },
 )
 
-DISALLOWED_IDENTIFIERS: frozenset[str] = frozenset({'www', 'intent', 'user'})
+DISALLOWED_IDENTIFIERS: frozenset[str] = frozenset({'www'})
 
 
 def determine_social_from_url_internally(

test/test_parsing.py

@@ -1,11 +1,10 @@
 import pytest
 
-from socials_util import SocialLink, SocialSiteId, determine_social_from_url
+from socials_util import *
 
 PARSABLE_SOCIAL_IDS_COMBINED: list[tuple[str, object, str]] = [
     # Tumblr formats
     ('https://triviallytrue.tumblr.com/', SocialSiteId.TUMBLR, 'triviallytrue'),
-    ('https://www.triviallytrue.tumblr.com/', SocialSiteId.TUMBLR, 'triviallytrue'),
     ('https://tumblr.com/triviallytrue', SocialSiteId.TUMBLR, 'triviallytrue'),
     ('https://tumblr.com/blog/triviallytrue', SocialSiteId.TUMBLR, 'triviallytrue'),
     (
@@ -180,18 +179,9 @@ PARSABLE_SOCIAL_IDS_COMBINED: list[tuple[str, object, str]] = [
     ('https://solquiet.deviantart.com/', SocialSiteId.DEVIANT_ART_ACCOUNT, 'solquiet'),
 ]
 
-NOT_PARSABLE = [
-    # Twitter intents are not supported
-    'twitter.com/intent/user?user_id=123',
-    'https://twitter.com/intent/user?user_id=123',
-    'https://twitter.com/intent/user',
-    'https://twitter.com/intent',
-]
-
 
 @pytest.mark.parametrize(
-    'url,expected_social_site_id,expected_social_id',
+    'url,expected_social_site_id,expected_social_id', PARSABLE_SOCIAL_IDS_COMBINED
-    PARSABLE_SOCIAL_IDS_COMBINED,
 )
 def test_parse_social_ids(url, expected_social_site_id, expected_social_id):
     social_link: SocialLink | None = determine_social_from_url(url)
@@ -200,8 +190,3 @@ def test_parse_social_ids(url, expected_social_site_id, expected_social_id):
         expected_social_id,
         expected_social_site_id,
     ), url
-
-
-@pytest.mark.parametrize('url', NOT_PARSABLE)
-def test_not_parsable(url: str):
-    assert determine_social_from_url(url) is None