[ilikenwf]

Constructive criticism is good. I have a feeling we'll do a 1.12.1 and possibly a 1.12.2 maintenance release in the future, anyway. I'll look at it because I might as well.

OCD sucks! I've dealt with it too.

[tibitts]

Is there any progress on the play count issue? This is the one problem which still prevents me from using Nightingale.
Thanks.

[rsjtdrjgfuzkfg]

(06-02-2013, 07:58 PM)tibitts Wrote: Is there any progress on the play count issue? This is the one problem which still prevents me from using Nightingale.
Thanks.

Sadly I never were able to reproduce the issue in a debug build. If you succeed in reproducing the issue in a debug build that would be great. However, I cannot do much about it as I don't get it properly reproduced on release builds, and not at all in debug builds.

Kinda hoping someone can track it down or give STR that work for me too...

[tibitts]

What does it take to install a debug build? And what am I supposed to do with a debug build?

[rsjtdrjgfuzkfg]

(06-05-2013, 04:28 AM)tibitts Wrote: What does it take to install a debug build? And what am I supposed to do with a debug build?

When building, you need to specify

Code:

ac_add_options --enable-debug=yes

in your nightingale.config and download and unzip debug dependencies to your debug folder. The resulting build can get started just as a regular build, a different profile might be a good idea to not mess in your main profile, though.
On windows (not sure about linux) you should also set the environment variable "XPCOM_DEBUG_BREAK" to "warn".

The issue I wrote about is that (on windows) I did not find a way to trigger the bug in debug builds. If you're able to reproduce the issue in a windows debug build, you can attatch a debugger and investigate how code is executed. That means if somebody (you?) finds a way to trigger the bug and posts what he did Step-by-Step (starting from a clean profile), I might have a chance to reproduce it... however it is possible that the issue lies somewhere in the dependencies or something, in that case it might be impossible to reproduce it in such a build on windows.

If you know how to debug, a starting point you might use is sbPlaybackHistoryService.cpp.

[tibitts]

I built Nightingale with debug enabled on Linux.Starting Nightingale from a shell prompt dumps tons of information.
From all that I believe this is relevant to the issue. The next three Warnings are appear very many times for every second song played, pretty much towards the end of the track. It's the first track played, and the play count gets updated:

WARNING: GetMediaItem called with a bad GUID!: file /home/christ1/Downloads/nightingale/nightingale-hacking/components/library/localdatabase/src/sbLocalDatabaseLibrary.cpp, line 3097
WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result 0x80040111: file /home/christ1/Downloads/nightingale/nightingale-hacking/components/mediacore/playback/history/src/sbPlaybackHistoryService.cpp, line 1242
WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result 0x80040111: file /home/christ1/Downloads/nightingale/nightingale-hacking/components/mediacore/playback/history/src/sbPlaybackHistoryService.cpp, line 880

These messages appear for every second song played, also pretty much towards the end of the track. This is the second track played, and the play count does not get updated:

###!!! ASSERTION: wrong thread: 'NS_IsMainThread()', file /mnt/ngale/nightingale-deps/xulrunner-1.9.2/mozilla/netwerk/base/src/nsIOService.cpp, line 492
###!!! ASSERTION: wrong thread: 'NS_IsMainThread()', file /mnt/ngale/nightingale-deps/xulrunner-1.9.2/mozilla/netwerk/base/src/nsIOService.cpp, line 492
###!!! ASSERTION: nsStandardURL not thread-safe: '_mOwningThread.GetThread() == PR_GetCurrentThread()', file /mnt/ngale/nightingale-deps/xulrunner-1.9.2/mozilla/netwerk/base/src/nsStandardURL.cpp, line 902
WARNING: NS_ENSURE_TRUE(mCurrentStartTime) failed: file /home/christ1/Downloads/nightingale/nightingale-hacking/components/mediacore/playback/history/src/sbPlaybackHistoryService.cpp, line 1606

As said before, using Linux there is nothing in the error console upon playing songs, while on Windows there is.

[rsjtdrjgfuzkfg]

Seems like this one is causing the issue (jump to line):

(06-22-2013, 09:08 PM)tibitts Wrote: WARNING: NS_ENSURE_TRUE(mCurrentStartTime) failed: file /home/christ1/Downloads/nightingale/nightingale-hacking/components/mediacore/playback/history/src/sbPlaybackHistoryService.cpp, line 1606

Thanks a lot for that logs!

[tibitts]

I've built Nightingale for Linux with the 'possible fix for play count increment, #142'. Not sure if this is the final fix yet, but it appears to work for me now, play count is incremented correctly. What a relief!
Thanks a lot.

[johnmurrayvi]

Glad to hear it's working! Sounds like it could finally be resolved then.
Thank you for your help and log!

[Willynux]

How do I get this "possible fix for play count increment, #142"?
I found this commit on github: https://github.com/nightingale-media-pla...b9c7c54f84

...but I don't have a file called sbPlaybackHistoryService.cpp, nor a folder called mediacore, nor any of the structure described. There are two files, sbPlaybackHistoryService.so and sbPlaybackHistoryService.xpt on the folder /usr/share/nightingale/components