Improved extraction of usernames from links in personal data

maigret/utils.py

@@ -28,4 +28,30 @@ def enrich_link_str(link: str) -> str:
     link = link.strip()
     if link.startswith('www.') or (link.startswith('http') and '//' in link):
         return f'<a class="auto-link" href="{link}">{link}</a>'
-    return link
+    return link
+
+
+class URLMatcher:
+    _HTTP_URL_RE_STR = '^https?://(www.)?(.+)$'
+    HTTP_URL_RE = re.compile(_HTTP_URL_RE_STR)
+    UNSAFE_SYMBOLS = '.?'
+
+    @classmethod
+    def extract_main_part(self, url: str) -> str:
+        match = self.HTTP_URL_RE.search(url)
+        if match and match.group(2):
+            return match.group(2).rstrip('/')
+
+        return ''
+
+    @classmethod
+    def make_profile_url_regexp(self, url: str, username_regexp: str = '') -> re.Pattern:
+        url_main_part = self.extract_main_part(url)
+        for c in self.UNSAFE_SYMBOLS:
+            url_main_part = url_main_part.replace(c, f'\\{c}')
+        username_regexp = username_regexp or '.+?'
+
+        url_regexp = url_main_part.replace('{username}', f'({username_regexp})')
+        regexp_str = self._HTTP_URL_RE_STR.replace('(.+)', url_regexp)
+
+        return re.compile(regexp_str)