RadioCaster Metadata from non-latin charsets displays incorrectly

suiyonari

New member
Hello I purchased RadioCaster after finding that it does everything* I had hoped it would do, however I ran into an issue where, after setting the audio source to a network stream which plays MP3 encoded audio with metadata which features from time to time non-western character sets (i.e., Japanese), as well as setting the metadata to be read from stream via the dropdown menu, it will display "garbage" characters that do not match the original stream's metadata. This is before I even attempt to serve or re-encode the audio, the characters that display next to the "Source" field, directly under the toolbar are incorrect. Of course, serving the re-encoded stream when such tracks play will also result in the same "garbage" characters being displayed on my client.
Hoping this was a simple matter of obsolete or mismatched codepages, I thought it would be a good idea to sidestep the whole issue by fetching the properly formatted UTF8 metadata from the original stream's caster page using the option to "Read from URL" under Settings > Metadata dropdown menu, however I found very little documentation regarding how to utilize this option properly.

With this post, I am hoping someone can help me with either forcing the program to decode the origin stream's metadata using the correct codepage, or failing that, point me towards a full explanation on how to properly use the "Read from URL" metadata setting.
 
Can you please specify the stream's URL so we can check? If the stream properly served the metadata, there should be no problem with displaying it. If you don't want to publicly share the URL you can send it to support@djsoft.net
 
Sure, here:
hxxps://asiadreamradio.torontocast.stream/stations/en/index.html (wouldn't let me paste the proper http link as the forum considered it spam)
Any of the streams on this page will occasionally play tracks where some or all characters are Japanese kana.
I've noticed that when it's only a single kana, the program will decode it correctly and my icecast server will also receive the correct character, as well as my player will show the correct metadata. Problems only seem to arise once more than two or three Japanese characters appear together, which is why I am led to believe it's an issue with there not being any way to specify what codepage the program ought to use.
Normally the system locale indicates to applications which codepage to use, but even after switching my system to japanese locale, RadioCaster still has trouble. Compare it to applications like VLC or MPV, they don't seem to have any issue displaying the correct characters on their window and taskbar titles.

If it's too complicated to modify RadioCaster to accommodate different character sets, then may I have a guide on how to use the "Read from URL" metadata option? It might be the easier fix in this situation.
 
The latest RadioCaster version seems to be handling this properly, please see the screenshot attached.

may I have a guide on how to use the "Read from URL" metadata option?
This merely makes a request to a specified URL and uses whatever's returned in response as a title. I don't think it will work in your case. And I'm not sure it's needed as Japanese seems to be properly handled by RaidoCaster.
 

Attachments

  • image_2024-09-19_141350570.png
    image_2024-09-19_141350570.png
    21.6 KB · Views: 39
Ah I see you're right. In my case, i had been using latest version of Radiocaster but not latest version of Windows (had been running Radiocaster on an old unused machine with Windows 7 still on it). Once I installed it on my Windows 10 machine the metadata started displaying properly. Sorry for the confusion.
While we're on the subject, considering the functionality of the "Read from URL" option, I'd like to submit a feature request if I could, for an advanced version of it, which would fetch the given URL's html source and do a user-editable regex search for the starting position and ending position of the artist/title strings and pass it onto the metadata.

I appreciate all the help, and I certainly appreciate this wonderful application.
 
While we're on the subject, considering the functionality of the "Read from URL" option, I'd like to submit a feature request if I could, for an advanced version of it, which would fetch the given URL's html source and do a user-editable regex search for the starting position and ending position of the artist/title strings and pass it onto the metadata.
If you have your web site, you can create a script there that will do the necessary parsing and call this script from RadioCaster.
 
Back
Top