draft: string processing

music_kraken/utils/string_processing.py

@@ -6,6 +6,7 @@ from functools import lru_cache
 from transliterate.exceptions import LanguageDetectionError
 from transliterate import translit
 from pathvalidate import sanitize_filename
+from urllib.parse import urlparse, ParseResult, parse_qs
 
 
 COMMON_TITLE_APPENDIX_LIST: Tuple[str, ...] = (
@@ -21,6 +22,7 @@ def unify(string: str) -> str:
     returns a unified str, to make comparisons easy.
     a unified string has the following attributes:
     - is lowercase
+    - is transliterated to Latin characters from e.g. Cyrillic
     """
 
     if string is None:
@@ -132,8 +134,27 @@ def unify_punctuation(to_unify: str) -> str:
         to_unify = to_unify.replace(char, UNIFY_TO)
     return to_unify
 
-def hash_url(url: str) -> int:
-    return url.strip().lower().lstrip("https://").lstrip("http://")
+def hash_url(url: Union[str, ParseResult]) -> str:
+    if isinstance(url, str): 
+        url = urlparse(url)
+
+    query = url.query
+    query_dict: Optional[dict] = None
+    try:
+        query_dict: dict = parse_qs(url.query, strict_parsing=True)
+    except ValueError:
+        # the query couldn't be parsed
+        pass
+
+    if isinstance(query_dict, dict):
+        # sort keys alphabetically
+        query = ""
+        for key, value in sorted(query_dict.items(), key=lambda i: i[0]):
+            query += f"_{key.strip()}_{''.join(i.strip() for i in value)}"
+
+    r = f"{url.netloc}_{url.path.replace('/', '_')}{query}"
+    r = r.lower().strip()
+    return r
 
 
 def remove_feature_part_from_track(title: str) -> str: