JEmplode 41: Delete won't?

I have a saved search: refs=0
I pull it up and select everything, copy, then paste into a playlist and synchronize. Everything's good, I then have nothing with 0 refs.

I select everything in the list. I say delete, check delete completely, and say "yes to all". Then synchronize.

Now, go back to the refs=0 search. Fewer matches but there are still tracks there.

I'm confused.

If I delete em one track at a time, they do go away.

Yeah, I also reported this. If you look carefully, it looks like it's putting all the things you want deleted in order, and then only deleting the odd-indexed ones (or even, if your origin is zero).

Yeah, I just found your message. Normally I keep up but I happened to be away for a day and a half and have been catching up.

I think what's happening is that there's a loop somewhere in there that is counting up from 0 and deleting, which means it would do:

0: delete 0
1: since 0 is deleted, 1 went to 0, now you're deleting 2
2: same thing, so now you're deleting 4
.
.

What it's supposed to do is count down, so you don't have this problem. No worry, i'm throwing out most of that code anyway.

Mike

So, did I also then delete other random things?

photopc (the software I use for dealing with my digital camera) had a shockingly similar bug a while back.

It's a pretty typical sort of bug, so I'm not surprised other apps have had that problem ... I doubt you deleted random things. i think the way it would work is that it would go through every other _of the things you selected_.

So you have any suggestions for working around until there's a new version?

Delete one, sync, delete one, sync, gets really tedious

I don't think you have to sync inbetween deleting one at a time ... I haven't gone looking for the actual code that has the problem yet, so I don't have an immediate workaround to offer you. I have to decide if I want to spend time hunting that bug and put out a release that fixes just that or whether I want to just wait and put out the whole next release together ... It will probably be next weekend before I can get the next release out.

I don't think you have to sync inbetween deleting one at a time
Well, that's at least a little better. Will try it now.

I hand-deleted each file, and still ended up with "refs=0" files, so I guess it's not that simple.

This is more disturbing. I choose, say, about 10 albums to push. It does, but then it ends up needing to spend a lot of time rebuilding databases, and several of the playlists don't appear.I get a warning that corrected were several missing FIDs. The music appears to be all there, with some of the playlists, in my refs=0 search. Some of the uploads seem to succeed.

This is why i'm rewriting all this crap ... Each release has just gotten more complex and thus harder to debug. Ther upper 38 releases were not this bad ... I think a lot of this got nasty when I tried to put in a complex concept like mirroring on top of a system that was already too complex for its own good.

I would recommend going to a pre-mirror release... Unfortunately I don't have those jars up -- Maybe someone has one they can put up. Or go to the last official release.

I would recommend going to a pre-mirror release... Unfortunately I don't have those jars up -- Maybe someone has one they can put up.

Anyone? Bueller?

40 seems to have the same problem, and I managed to get 35 to trash all my playlists (though I recovered them from the refs=0 search easily)

So I guess I'll wait for the next version and try not to mess with it much until then

That's kind of interesting ... sorry about that

It's "experimental", and it only trashed the playlists, not the actual data. Not really a big deal at all.

Actually, I bet I still have a jar file from March which I know was stable, as long as my wife doesn't start caring about Wendy flags right now I can probably get by with that easily.

If you have anything you want tested later I'll be happy to volunteer also

Aren't all jEmplode users automatically volunteer alpha testers ?

No, sometimes we're beta testers