More typing

socials_util/__init__.py

@@ -118,7 +118,7 @@ class WikidataInfo(object):
     nickname_version_of: Optional[SocialSiteId] = None
 
 
-WIKIDATA_PROPERTIES = {
+WIKIDATA_PROPERTIES: dict[SocialSiteId, WikidataInfo] = {
     SocialSiteId.EMAIL: WikidataInfo(968, None),
     SocialSiteId.RSS_FEED: WikidataInfo(1079, None),
     SocialSiteId.FACEBOOK_PAGE: WikidataInfo(2013, None),
@@ -184,8 +184,7 @@ WIKIDATA_PROPERTIES = {
 }
 
 
-def re_social_subdomain(main_domain):
+def re_social_subdomain(main_domain: str) -> str:
-    # return r'^(?:https?:\/\/)?([\w_-]+)\.'+re.escape(main_domain)+'\/?$'
     return r'^(?:https?:\/\/)?([\w_-]+)\.' + re.escape(main_domain) + r'(\/.*)?$'
 
 
@@ -194,23 +193,23 @@ RE_DUAL_ID = r'@?([^/]+/[^/]+)'
 RE_ANY_SUBPATH = r'(|\/|\/.*)$'
 
 
-def re_social_path(main_domain):
+def re_social_path(main_domain: str) -> str:
-    # return r'^(?:https?:\/\/)?(?:www\.)?'+re.escape(main_domain)+'\/'+RE_ID+'\/?$'
     return re_social_path_adv(main_domain, RE_ID)
 
 
-def re_social_path_adv(main_domain, *path):
+def re_social_path_adv(main_domain: str, *path: str) -> str:
-    assert not main_domain.startswith('www.'), 'Redundant www.'
+    if main_domain.startswith('www.'):
-    l = [r'^', r'(?:https?:\/\/)?', r'(?:www\.)?', re.escape(main_domain)]
+        msg = f'Redundant www: {main_domain}'
+        raise ValueError(msg)
+    regex_builder: list[str] = [r'^', r'(?:https?:\/\/)?', r'(?:www\.)?', re.escape(main_domain)]
 
     for p in path:
         if p != RE_ANY_SUBPATH:
-            l.append(r'\/')
+            regex_builder.append(r'\/')
-        l.append(p if p in {RE_ID, RE_DUAL_ID, RE_ANY_SUBPATH} else re.escape(p))
+        regex_builder.append(p if p in {RE_ID, RE_DUAL_ID, RE_ANY_SUBPATH} else re.escape(p))
     if path[-1] != RE_ANY_SUBPATH:
-        l.append(r'\/?$')
+        regex_builder.append(r'\/?$')
-    regex = ''.join(l)
+    return ''.join(regex_builder)
-    return regex
 
 
 MAILTO_URL = r'^mailto:(?:[\w._.]+@[\w._.]+)$'
@@ -264,7 +263,7 @@ URL_PARSE_DANBOORU_ARTIST = re_social_path_adv('danbooru.donmai.us', 'artists',
 URL_PARSE_BANDCAMP = re_social_subdomain('bandcamp.com')
 URL_PARSE_BLUESKY = re_social_path_adv('bsky.app', 'profile', RE_ID)
 
-REGEXES = [
+REGEXES: list[tuple[str, SocialSiteId]] = [
     # Reddit
     (REDDIT_SUBREDDIT_URL, SocialSiteId.REDDIT_SUBREDDIT),
     (REDDIT_USER_URL, SocialSiteId.REDDIT_USER),
@@ -364,7 +363,7 @@ REGEXES = [
     (re_social_subdomain('blogspot.com'), SocialSiteId.GOOGLE_BLOGGER_PAGE),
 ]
 
-WELL_KNOWN_MASTODON_INSTANCES = frozenset(
+WELL_KNOWN_MASTODON_INSTANCES: frozenset[str] = frozenset(
     {
         # Includes all servers with 50 000+ users as of 6 july 2023.
         # based on https://mastodonservers.net/servers/top
@@ -388,7 +387,7 @@ WELL_KNOWN_MASTODON_INSTANCES = frozenset(
 )
 
 
-def determine_social_from_url_internally(url: str):
+def determine_social_from_url_internally(url: str) -> tuple[SocialSiteId | None, str | None]:
     assert isinstance(url, str)
 
     # Regexes
@@ -405,15 +404,23 @@ def determine_social_from_url_internally(url: str):
         return (SocialSiteId.MASTODON_PAGE, None)
 
     # Feed (?)
-    elif 'feed' in url or 'xml' in url or 'rss' in url or 'atom' in url:
+    if 'feed' in url or 'xml' in url or 'rss' in url or 'atom' in url:
         return (SocialSiteId.RSS_FEED, None)
 
     return (None, None)
 
-
+def to_parse_result(url: str | urllib.parse.ParseResult) -> urllib.parse.ParseResult:
-def determine_social_from_url(url):
     if isinstance(url, str):
-        url = urllib.parse.urlparse(url)
+        return urllib.parse.urlparse(url)
+    if isinstance(url, urllib.parse.ParseResult):
+        return url
+
+    # Throw error
+    msg = f'Expected {urllib.parse.ParseResult} or {str}'
+    raise TypeError(msg)
+
+def determine_social_from_url(url_not_normalized: str | urllib.parse.ParseResult) -> SocialLink | None:
+    url = to_parse_result(url_not_normalized)
     (social_site_id, social_id) = determine_social_from_url_internally(
         url._replace(query='', fragment='').geturl()
     )