Rockbox General > Rockbox General Discussion

Rejection of "ignore the" patch (split from the Evil_G unsupported build thread)

<< < (19/23) > >>

lalittle:

--- Quote from: Llorean on June 13, 2007, 05:23:33 PM ---I don't understand how it's a better idea not updating the files?
--- End quote ---

My thinking was that the process could be significantly faster since it would not have to re-write the files, and the possibility of any sort of hiccups or corruption during this process would be eliminated.

Also, another "potential" concern of mine in all this has been how the re-writing of files might effect the "native" database on the unit.  I'm not sure about other units, but at least with iPods, Rockbox has always maintained a clean separation from the native database so that when booting to the native firmware, the iPod functions just like it would normally have.  This is important since there are still reasons to maintain full "native" functionality in certain situations.  Are there any potential issues with using the native firmware after re-writing the files via Rockbox?

Thanks,

Larry

Llorean:
It would only make use of "legal" tags within the tagging specifications. This means the only time one could expect the original firmware to have problems is if it doesn't actually support the tagging specifications it claims to, in which case other "valid" files would break it just as easily.

All Rockbox would be doing is the same job many other taggers do. These tags aren't something we've made up, they're part of several specifications (though the only one for MP3 files that supports this, and is also supported by Rockbox, is ID3v2.4, it's also in Vorbis Comments, ApeV2, and I believe MP4 metadata though I could be wrong on the last one). In the case of the iPod for example, it would just mean more tags on the files that, because the Apple firmware doesn't support them, get ignored completely much like if you had additional comment tags or one of the other obscure tags that are part of the specification.

ell1ps1s:
Since version 7.1, iTunes does support sorting tags. As far as I can tell, this is not just within its own database; it does also write  TSOP etc. tags to the audio files. (At least on my Mac; I can't comment on the Windows version, although I would hope they behave in the same way.) It also seems to supplement its own behaviour of ignoring (English) articles by adding sort tags, which would thus be available to Rockbox if support for those tags were implemented.

Llorean:
I need to apologize, apparently there's no "official" way of doing it in Vorbis Comments. Which means I need to figure out where I "learned" that there was.

Either way, it looks like there's at least one mainstream program doing it this way now. And apparently, Apple has received complaints because "The Beatles" are sorted differently in the German version of iTunes than the English version because of differences in their list of articles... ;)

jurrie:

--- Quote from: lalittle on June 13, 2007, 05:17:39 PM ---Forgive me, but I'm unclear what this means.  How does it "update the sorting attributes"?  Could you clarify the details on what the process would involve?  I agree that it seems like a better idea to not alter the files, but I'm unclear how your suggestion would work.

--- End quote ---

I had envisioned that the plugin would have logic similar to:

* chug through all supported/recognized (by the plugin) audio content
* upon encountering an item requiring sorting modification:

* call (new) database API to locate the entry in the database (based on artist/album/track/title info)
* if the entry exists in the database, call another (new) database API to update the entry (presumably using a handle/identifier returned by the previous call)
* if the entry does not exist in the database, either add it to the database via another (new) database API call; or notify the user at completion that new content exists and suggest "General Settings -> Database -> Update Now" and then re-running the plugin

Navigation

[0] Message Index

[#] Next page

[*] Previous page

Go to full version