search

MaxFour / Music-Player

Lightweight and Material designed Music Player
GNU General Public License v3.0
283 stars 71 forks source link

Accented characters are not always correctly displayed #24

Open mirandir opened 5 years ago

mirandir commented 5 years ago

Hello,

I'm not sure why, but sometimes accented characters aren't looking good in Album, Artists and Play view, but are ok in the tag editor: char1 char2

Files in screenshots are correctly displayed in other music players (like Vanilla Music). I can send you files if you need help to reproduce the problem.

MaxFour commented 5 years ago

Hello! This may be because of the encoding. I'll see what can be done with this

stemy2 commented 5 years ago

I have the same problem. It happens with «ç» too.

MaxFour commented 5 years ago

@mirandir @stemy2 I recommend you this app: ID3Fixer

Malakai-DB commented 3 years ago

Hello @MaxFour,

First of all thank you for this application (I installed it from F-Droid), I really like it so far.

I just wanted to let you know that this issue is still relevant at the moment I write this comment. For instance, for the same album, some tracks are correctly shown in the app and others have a weird sign instead of the accented characters. This also happens for the album title, the app displaying 2 albums (one with the correct accented character and the other with the weird sign), when in fact it is the same album. Also as @mirandir stated in the original post, all is good in the tag editor. All files where tagged with Musicbrainz Picard (which is using by default the UTF-16 charset) and all other players have no issue correctly displaying the accented characters.

Is this fixable?

sjlongland commented 3 years ago

I've noticed this problem before… initially I noticed it with a song by Plastic Bertrand: "Ça Plane Pour Moi"… later with the band name Blue Öyster Cult and more recently with a couple of songs by Yothu Yindi ("Djäpana" and "Mätula").

The files in my case are Ogg/Vorbis files, for which the specification says they should be encoded UTF-8 (and they are in my case) so there's no ambiguity on how the characters should be interpreted. I played around with iconv this morning, wondering if I could reproduce the type of mojibake problem I'm seeing, couldn't get an exact match to the glyph that shows up. Taking the "Djäpana" example, it shows up as "Dj?ana"… the ? being some Chinese glyph (and clearly swallowing 3 bytes).

I wonder if there's some character encoding switch hidden in Android's UI that I've missed. My device is a ZTE T83 (so Chinese-made phone) running Android 4.1. Other screenshots look like UTF-8 being interpreted as ISO8859-1… but in my case it seems to be one of the CJK charsets (just haven't figured out which one).