Release DeaDBeeF
- Thread starter ptitSeb
- Start date
levi
Still fresh, damnit!
I've also not managed to get the shoulders working as shotcuts, because they're keyboard modifiers - deadbeef will accept them as a valid control if combined with a letter, but not on their own.
Who said it could work? I can't see the posting.
Personally I've configured the lower row of keyboard keys to do my shortcuts. Even more accessible than the shoulders in the case I use, as it goes.
Who said it could work? I can't see the posting.
Personally I've configured the lower row of keyboard keys to do my shortcuts. Even more accessible than the shoulders in the case I use, as it goes.
Eight Bit
Hardcore Member
Try rockbox instead if you like to skip tracks with the shoulders
levi
Still fresh, damnit!
I've just got round to trying this out, and after a bit of trial and error I got it to work. The trick seems to be that the format of the config file has changed since the posting on page 4 was made.
You now need lines looking like this:
hotkey.key1 "Shift_R" 0 0 prev
hotkey.key2 "Control_R" 0 0 next
However, there are a few other options in the file that seem to relate to hotkeys, so to get them all set right, I'd suggest you run deadbeef and set a couple of nonsense shortcuts to the functions you want (prev and next in this case). Then quit it completely, and open up the config file. Find the lines that define those hot keys, and modify the bit in quotes to be the same as the example above. The key number may or may not be the same, but leave that as the program created it.
You now need lines looking like this:
hotkey.key1 "Shift_R" 0 0 prev
hotkey.key2 "Control_R" 0 0 next
However, there are a few other options in the file that seem to relate to hotkeys, so to get them all set right, I'd suggest you run deadbeef and set a couple of nonsense shortcuts to the functions you want (prev and next in this case). Then quit it completely, and open up the config file. Find the lines that define those hot keys, and modify the bit in quotes to be the same as the example above. The key number may or may not be the same, but leave that as the program created it.
levi
Still fresh, damnit!
I've been getting the occasionally segfault with this recently, while playing ogg encoded album tracks. It usually makes its way through about an album's worth of tracks before crashing, and it never crashes during a track, only when moving to the next track. I'm not sure if it's due to the update this received earlier in the summer - since then I've mostly been playing mp3ed radio programmes, and since those are multi-hour shows, only a two or three per playlist, but now I've started listening to my ogg-encoded music files also. Last year I used it to listen to FLAC encoded album files, and I never experienced a crash with it then.
I've kept the logged crash snippet, but it seems singularly unhelpful to me. I can pull it off my Pandora if needed though.
I've kept the logged crash snippet, but it seems singularly unhelpful to me. I can pull it off my Pandora if needed though.
ptitSeb
Serial Porter
the segfault logs in pndrun out file are most of the time useless indeed. I'll try to refresh the build soon, maybe it will help.
notaz
Certified Guru
The kernel is modified to log crash addresses, you can use that for addr2line if ptitSeb hasn't stripped his binaries and if it crashes in the program and not in some common library like glibc.
ptitSeb
Serial Porter
Oh, in dmesg ? Handy!
levi
Still fresh, damnit!
Ah, okay, I'll have to check my dmesg after it crashes again. Here's the pndrun log, if you're interested:
INFO: loading playlist New Playlist
INFO: from file /mnt/utmp/deadbeef/deadbeef/playlists/1.dbpl
gtkui plugin compiled for gtk version: 2.24.24
** (deadbeef:3259): WARNING **: Widget not found: Skip to_menu
Segmentation Fault
connecting button tray signals
backtrace() returned 1 addresses
./bin/deadbeef [0xaec4]
[SUCCESS]--- Starting the application ( scripts/deadbeef.sh ) ----------
The lines I've marked in italics appear in the log as soon as you start the program, while the lines after that seem to be related to the crash; it doesn't appear to make any further log entries before crashing in this case at least. I take it that address in the backtrace isn't what your looking for, being only 16-bits long.
INFO: loading playlist New Playlist
INFO: from file /mnt/utmp/deadbeef/deadbeef/playlists/1.dbpl
gtkui plugin compiled for gtk version: 2.24.24
** (deadbeef:3259): WARNING **: Widget not found: Skip to_menu
Segmentation Fault
connecting button tray signals
backtrace() returned 1 addresses
./bin/deadbeef [0xaec4]
[SUCCESS]--- Starting the application ( scripts/deadbeef.sh ) ----------
The lines I've marked in italics appear in the log as soon as you start the program, while the lines after that seem to be related to the crash; it doesn't appear to make any further log entries before crashing in this case at least. I take it that address in the backtrace isn't what your looking for, being only 16-bits long.
ptitSeb
Serial Porter
Mmmm, yes, that address is a little bit short...
levi
Still fresh, damnit!
Okay, I've observed two crashes now in the dmesg; one by scrolling back through, and one live. Here's the dmesg output:
Running addr2line:
But looking at the dmesg again, a read at 0x00 looks like it might be a null pointer access to me, and perhaps it's even somewhere in initfini.c.
So that would suggest it is limited to ogg vorbis files in my use cases. When I was listening to FLAC files, those were so called 'native FLAC' encoded, not FLAC vorbis.
Running addr2line:
So I found vorbis.so, and tried again:
Hmm, line 0? That doesn't look right. I tried also with the 64-bit address but that gave be ??:0 after a newline.
But looking at the dmesg again, a read at 0x00 looks like it might be a null pointer access to me, and perhaps it's even somewhere in initfini.c.
Last edited by a moderator:
ptitSeb
Serial Porter
levi
Still fresh, damnit!
Thanks, I find it tends to crash reliably after playing 10-15 files in a playlist (so about once per CD's worth), and also I had one crash caused by clearing the playlist while it was playing a file, although so far that's been unreproducible, and is the only case where it's crashed while playing a track, not just at the end.
I should have time to try to reproduce the issue on a new build if you want me to try before release.
I should have time to try to reproduce the issue on a new build if you want me to try before release.
ptitSeb
Serial Porter
Ok, I'll send you PM when it's ready.
pmprog
DNF (Did Not Finish)
- Joined
- Apr 25, 2011
- Messages
- 4,150
Could I make a change request, not sure how easy it'd be for you?
Basically, when you quit DeaDBeeF, it saves which was the last file to be played - could you get this to happen on every track change?
The reason I ask is that I frequently listen to music as I go to sleep, and I use the ShutdownTimer to turn the Pandora off after (n) minutes, but because DeaDBeeF doesn't close "nicely", when I restart it, it's always back at the same track.
Basically, when you quit DeaDBeeF, it saves which was the last file to be played - could you get this to happen on every track change?
The reason I ask is that I frequently listen to music as I go to sleep, and I use the ShutdownTimer to turn the Pandora off after (n) minutes, but because DeaDBeeF doesn't close "nicely", when I restart it, it's always back at the same track.
ptitSeb
Serial Porter
Mmm, that may not be easy indeed. I'll see if I can do it (but that will make the risk of file corruption higher).
