Closed luizffgv closed 2 years ago
I followed the suggestions and updated the pull request.
Also, the max_artist_len
setting was removed since the details field shortens it if necessary. max_title_len
is kept, so large titles don't hide the artists' names
Well, if the limit is 128 bytes and not 128 characters, then we don't save anything by using …
, since that needs 3 bytes in UTF-8. I was under the assumption that we were limited to characters.
I'll have to test this locally some time, so I'm holding off the merge until then.
Well, if the limit is 128 bytes and not 128 characters, then we don't save anything by using
…
, since that needs 3 bytes in UTF-8. I was under the assumption that we were limited to characters.
If I remember correctly the docs say 128 characters but that was not the behavior I observed. That's weird, since the minimum length seems to be actually measured in characters.
That's probably something you will want to test later.
Rebased as f902f6aa88d19b3d9c7c14eb6fa8a91f7454b588 and reworked with 7aefbd9cbcdeafbd40172fa92e07d85e65bb59a4.
Turns out it's indeed characters and not bytes, as I confirmed with a few experiments. Furthermore, I added some dynamic weighting based on the format used, so that we don't need to hard-code the maximum title length to 64. See the commit for details.
I also tagged 0.3.1 with the changes.
Using textwrap.shorten
I'm not experienced with Python so this might be far from ideal.
Fixes #20