iScrobbler 1.0.0 statistics window and queue problems

Edit caused dbl post. Weird.

Damaged,

OK, you posted your response while I posted my edit, which was that the 3rd cd played with everything submitted.

While the cd with 3 skipped tracks was playing, I was basically reading forum posts and some news sites, as I recall. Also, as I had mentioned before, I bought this cd new and it has only been played a half-dozen times and has been handled by its edges and stored in its case, so it is unlikely (but not definitely) undamaged.

I had long ago discovered that tunes won't submit if you have Get Info open, even for a different track. I was backing up my playlist and burning a lot of cds last week while playing tunes on my computer. I don't recall any tunes that were not submitted.

Tom

I've played another 3 cds or so using tests 2 & 3 with no skipped tracks. The track skipping was not all that common, when I think about the number of cds I play. The problem does appear to like certain cds, such as the first three I tested today and last night. I will try to remember to keep track of which cds have problems and see if I can find anything in common amongst them that might explain it.

Quoth DoctorT:
[Feb 06, 2005 15:33:54 -0500]-[WARN] Invalid submit interval '-38.00' for track 'Noah And The Rabbit, Best Of British Folk - Original Recordings - Original Artists, John Renbourn & Bert Jansch'. Track will not be submitted. Duration: 169, Position: 127.


169 / 2 = 84.5
84.5 - 127 = -42.5
-42.5 + 4.0 = -38.5

Negative submit intervals are not allowed. So the question is, why is iTunes notifying us of a brand new play (correct?) that is 127 seconds into the song?

Quoth DoctorT:
Damaged,

I just played the cd from track one through to the end. I didn't notice this one track was not submitted until I looked at my Recent Tracks. It was submitted successfully when I played it again later. All I was doing while playing the cd was reading posts and surfing the web.


There are some real problems with iTunes and the data it sends while playing CD's. Anway, what was the duration of the track played just before the missed one? I added some code to deal with iTunes setting the postion of a new track to the duration of the previous one, but it seems that this is not the case in this instance.


On another front, several of the tracks from this cd appeared in my list twice because the cache did not clear and, even though Russ says that the spam protocol is fixed, they were entered into my profile with doubled play counts.

Yes, they said it was fixed on Friday, and I had quite a few dupes yesterday. No dupes today, but I haven't been able to submit for an hour now, and I didn't play too many songs this morning. I think something is still broken.

Quoth DoctorT:
The duration of the song before was 126.

Ack! Fuckin' iTunes! 1 second off the reported postion of the new track. What a piece of shit!

Well, I guess I can change the code from checking equality to >= duration-1 and <= duration +1. But that may screw up something else (because we get play notifications when the track meta-data is changed too). So I'm not sure if this should be done.

The timing spam message is back - I just got it while listening to Last.fm.

I am in agreement with Simon. It is not simply an issue of crossfade or of Airport Express. Rather it appears to be the combination the combination of the two. Turning crossfading off altogether eliminated the problem, in my testing. Likewise, playing the tracks locally within iTunes, but not streaming to the Airport Express, also se4ms to have resolved the issue. Even without testing, this would seem to have been verified by the posts of John, Simon and myself as further discourse on the issue has revealed this to be the uniting factor with the three of us and the anomaly we have experienced.

Many thanks to Brian for the excellent work and the willingness to work with us to diagnose the issue and arrive at a fix — hack though it is. I am in complete agreement with him as well, with respect to iTunes. None of these problems appear to be the fault of iScrobbler; instead being the result of crappy coding at Apple. It would appear that they rushed the "Airtunes" compatible version out without complete testing...or they simply do not care that it munges the track lengths and such.

Though it does not appear to be related, I am interested to see what, if any, fix evolves from the problem Tom has been experiencing. It seems less straight forward in its cause. Nevertheless, is likely just as frustrating.

Thank you again, Brian, for your clever workaround and willingness to indulge our frustrations. :-)

Herman