Ilst: Ignore invalid covr data types when not strict

Serial-ATA · Nov 20, 2024 · 013b17d · 013b17d
1 parent 3d291d1
commit 013b17d
Show file tree

Hide file tree

Showing 2 changed files with 21 additions and 9 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -27,14 +27,16 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   - `AtomData::data_type()` to get the data type code of the atom content.
 
 ### Changed
-- **Ilst**: Add new rules for `gnre` atom upgrades ([issue](https://github.com/Serial-ATA/lofty-rs/issues/409)) ([PR](https://github.com/Serial-ATA/lofty-rs/pull/485))
-  - In the case that a `©gen` and `gnre` atom are present in a file, there was no way to tell which `©gen` atoms were upgraded.
-    the new rules are:
-    - `gnre` present + no `©gen` present, `gnre` gets upgraded as normal
-    - `gnre` present + `©gen` present, `©gen` takes precedence and `gnre` is discarded
-      - With [ParsingOptions::implicit_conversions](https://docs.rs/lofty/latest/lofty/config/struct.ParseOptions.html#method.implicit_conversions)
-        set to `false`, `gnre` will be retained as an atom of type `Unknown`.
-- **RIFF INFO**: Ignore text decoding errors when not using `ParsingMode::Strict` ([issue](https://github.com/Serial-ATA/lofty-rs/issues/373))
+- **Ilst**:
+  - Add new rules for `gnre` atom upgrades ([issue](https://github.com/Serial-ATA/lofty-rs/issues/409)) ([PR](https://github.com/Serial-ATA/lofty-rs/pull/485))
+    - In the case that a `©gen` and `gnre` atom are present in a file, there was no way to tell which `©gen` atoms were upgraded.
+      the new rules are:
+      - `gnre` present + no `©gen` present, `gnre` gets upgraded as normal
+      - `gnre` present + `©gen` present, `©gen` takes precedence and `gnre` is discarded
+        - With [ParsingOptions::implicit_conversions](https://docs.rs/lofty/latest/lofty/config/struct.ParseOptions.html#method.implicit_conversions)
+          set to `false`, `gnre` will be retained as an atom of type `Unknown`.
+  - Ignore invalid `covr` data types when not using `ParsingMode::Strict` ([issue](https://github.com/Serial-ATA/lofty-rs/issues/482)) ([PR](https://github.com/Serial-ATA/lofty-rs/pull/486))
+- **RIFF INFO**: Ignore text decoding errors when not using `ParsingMode::Strict` ([issue](https://github.com/Serial-ATA/lofty-rs/issues/373)) ([PR](https://github.com/Serial-ATA/lofty-rs/pull/486))
   - RIFF INFO tags may be encoded with a non UTF-8 system encoding, that we have no way of knowing. It's no longer an error to read these files,
     it's just unlikely that anything useful come out of the RIFF INFO tags.
 

diff --git a/lofty/src/mp4/ilst/read.rs b/lofty/src/mp4/ilst/read.rs
@@ -378,7 +378,17 @@ where
 				DataType::Jpeg => Some(MimeType::Jpeg),
 				DataType::Png => Some(MimeType::Png),
 				DataType::Bmp => Some(MimeType::Bmp),
-				_ => err!(BadAtom("\"covr\" atom has an unknown type")),
+				_ => {
+					if parsing_mode == ParsingMode::Strict {
+						err!(BadAtom("\"covr\" atom has an unknown type"))
+					}
+
+					log::warn!(
+						"Encountered \"covr\" atom with an unknown type of `{}`, discarding",
+						Into::<u32>::into(data_type)
+					);
+					return Ok(());
+				},
 			};
 
 			let picture_data = AtomData::Picture(Picture {