Fix Tidal 404 handling: share links, missing lyrics, and the crashing handler

streamrip/client/tidal.py

@@ -9,7 +9,7 @@
 import aiohttp
 
 from ..config import Config
-from ..exceptions import NonStreamableError
+from ..exceptions import ItemNotFoundError, NonStreamableError
 from .client import Client
 from .downloadable import TidalDownloadable
 
@@ -124,7 +124,13 @@ async def get_metadata(self, item_id: str, media_type: str) -> dict:
                     item["lyrics"] = resp.get("lyrics") or ""
                 else:
                     item["lyrics"] = resp.get("subtitles") or resp.get("lyrics") or ""
-            except TypeError as e:
+            except ItemNotFoundError:
+                # Most tracks simply have no lyrics. That is the expected
+                # answer, not a problem worth reporting.
+                logger.debug("No lyrics available for track %s", item_id)
+            except (NonStreamableError, TypeError) as e:
+                # Lyrics that should have been there but could not be
+                # fetched -- worth knowing about.
                 logger.warning(f"Failed to get lyrics for {item_id}: {e}")
 
         logger.debug(item)
@@ -353,7 +359,10 @@ async def _api_request(self, path: str, params=None, base: str = BASE) -> dict:
         async with self.rate_limiter:
             async with self.session.get(f"{base}/{path}", params=params) as resp:
                 if resp.status == 404:
-                    logger.warning("TIDAL: track not found", resp)
-                    raise NonStreamableError("TIDAL: Track not found")
+                    # Logged at debug, not warning: some callers ask for
+                    # optional things (lyrics) where a 404 is the normal
+                    # answer. Callers that do care log it themselves.
+                    logger.debug("TIDAL: item not found (404): %s", resp.url)
+                    raise ItemNotFoundError(f"TIDAL: item not found: {resp.url}")
                 resp.raise_for_status()
                 return await resp.json()

streamrip/exceptions.py

@@ -66,5 +66,14 @@ def print_msg(self, item) -> str:
         return " ".join(base_msg)
 
 
+class ItemNotFoundError(NonStreamableError):
+    """The API returned 404 for an item.
+
+    A subclass of NonStreamableError so existing handlers are unaffected, but
+    distinguishable for callers fetching something optional -- "this does not
+    exist" and "this failed to download" deserve different log levels.
+    """
+
+
 class ConversionError(Exception):
     """ConversionError."""

streamrip/rip/parse_url.py

@@ -20,6 +20,7 @@
 URL_REGEX = re.compile(
     r"https?://(?:www|open|play|listen)?\.?(qobuz|tidal|deezer)\.com?(?:(?:/(album|artist|track|playlist|video|label))|(?:\/[-\w]+?))+\/([-\w]+)",
 )
+TIDAL_SHARE_SUFFIX_REGEX = re.compile(r"^(https?://[^/]*tidal\.com/.+?)/u/?$")
 SOUNDCLOUD_URL_REGEX = re.compile(r"https://soundcloud.com/[-\w:/]+")
 LASTFM_URL_REGEX = re.compile(r"https://www.last.fm/user/\w+/playlists/\w+")
 QOBUZ_INTERPRETER_URL_REGEX = re.compile(
@@ -54,6 +55,12 @@ async def into_pending(
 class GenericURL(URL):
     @classmethod
     def from_str(cls, url: str) -> URL | None:
+        # Tidal's share sheet produces links ending in "/u". URL_REGEX takes
+        # the last path segment as the item id -- it has to, because Qobuz
+        # album urls look like /<locale>/album/<slug>/<id> -- so that suffix
+        # would be parsed as an id of "u" and the API would 404.
+        url = TIDAL_SHARE_SUFFIX_REGEX.sub(r"\1", url)
+
         generic_url = URL_REGEX.match(url)
         if generic_url is None:
             return None

tests/test_parse_url.py

@@ -51,6 +51,28 @@ def test_tidal_track_url(self):
         self.assertEqual(groups[1], "track")  # media_type
         self.assertEqual(groups[2], "3083287")  # item_id
 
+    def test_tidal_share_url_with_u_suffix(self):
+        """Test that Tidal share links ending in /u parse to the real item id.
+
+        The share sheet appends "/u", which would otherwise be taken as the
+        item id and 404 against the API.
+        """
+        for url in (
+            "https://tidal.com/album/152697662/u",
+            "https://tidal.com/album/152697662/u/",
+            "https://tidal.com/browse/album/152697662/u",
+        ):
+            with self.subTest(url=url):
+                result = parse_url(url)
+
+                self.assertIsNotNone(result)
+                self.assertIsInstance(result, GenericURL)
+                self.assertEqual(result.source, "tidal")
+
+                groups = result.match.groups()
+                self.assertEqual(groups[1], "album")  # media_type
+                self.assertEqual(groups[2], "152697662")  # item_id
+
     def test_deezer_track_url(self):
         """Test that Deezer track URLs are matched correctly."""
         url = "https://www.deezer.com/track/4195713"