headers
Davy Durham | 12 Apr 06:40

Re: Markers in wave files

Okay.. this the 3rd time I've sent this email but haven't seen it show 
up on the list.. What is/has-been the deal?

My original question:   Does this look to be an acceptable patch?  If 
so, when might it show up in the CVS tree?

Davy Durham wrote:

> Hi.. I have checked out the sourcecode to libaudiofile and make added 
> support for reading and writing cues in the wave format.
>
> Wave already supported reading cue positions, but didn't read cue 
> names. The prescribed way of storing cue names in the format is with a 
> 'list'/'adtl' chunk according to 
> http://www.borg.com/~jglatt/tech/wave.htm .
>
> I added this reading support to wave.c  and also adding writing 
> marks/cues to wavewrite.c
>
> I have attached a patch... I have tested it with my audio editing 
> application and am able to load and save cues succefully.  Also, I 
> just threw in reading mark/cue comments since it was just a simple 
> added case to one section of code.  To write the comments would also 
> be pretty easy, but I'm not using comments so I didn't do it.  If you 
> want me to I could.
>
> I pretty much followed the way things were done according to the rest 
> of the wave I/O code.
>
> I'm wondering, that if the patch is accepted when the next release of
(Continue reading)

Permalink | Reply | 0 comments |
headers
Davy Durham | 12 Apr 06:51

Re: Markers in wave files

Oh.. um.. also, I just applied the patch to the latest cvs version and 
_af_free seems to have disappeared... If you apply the patch just change 
the two calls in wave.c to _af_free() to just free()

Davy Durham wrote:

> Hi.. I have checked out the sourcecode to libaudiofile and make added 
> support for reading and writing cues in the wave format.
>
> Wave already supported reading cue positions, but didn't read cue 
> names. The prescribed way of storing cue names in the format is with a 
> 'list'/'adtl' chunk according to 
> http://www.borg.com/~jglatt/tech/wave.htm .
>
> I added this reading support to wave.c  and also adding writing 
> marks/cues to wavewrite.c
>
> I have attached a patch... I have tested it with my audio editing 
> application and am able to load and save cues succefully.  Also, I 
> just threw in reading mark/cue comments since it was just a simple 
> added case to one section of code.  To write the comments would also 
> be pretty easy, but I'm not using comments so I didn't do it.  If you 
> want me to I could.
>
> I pretty much followed the way things were done according to the rest 
> of the wave I/O code.
>
> I'm wondering, that if the patch is accepted when the next release of 
> libaudiofile is scheduled for?  Because I want to debut my application 
> to the community in probably a month or so, and it would be nice if
(Continue reading)

Permalink | Reply | 0 comments |
headers
Davy Durham | 6 Apr 08:15

Re: Markers in wave files

Sooooo.....  Did anyone with cvs write access check out this patch? I've 
been trying to 'cvs update' for several days now, but it looks as if the 
cvs server is down.

-- Davy
Permalink | Reply | 2 comments |
headers
Daniel Kobras | 17 Apr 10:12
Picon

Re: Markers in wave files

On Sat, Apr 06, 2002 at 12:15:02AM -0600, Davy Durham wrote:
> Sooooo.....  Did anyone with cvs write access check out this patch? I've 
> been trying to 'cvs update' for several days now, but it looks as if the 
> cvs server is down.

I don't have access to the main CVS but I intend to include your patch
in the next Debian version of audiofile.  This will only be in a couple
of weeks though, when the new release is out.  I haven't found the time
to properly test your patch so far, but don't worry, it's not forgotten.

Regards,

Daniel.
Permalink | Reply | 1 comment |
headers
Davy Durham | 17 Apr 16:05

Re: Markers in wave files

Daniel Kobras wrote:

>On Sat, Apr 06, 2002 at 12:15:02AM -0600, Davy Durham wrote:
>
>>Sooooo.....  Did anyone with cvs write access check out this patch? I've 
>>been trying to 'cvs update' for several days now, but it looks as if the 
>>cvs server is down.
>>
>
>I don't have access to the main CVS but I intend to include your patch
>in the next Debian version of audiofile.  This will only be in a couple
>of weeks though, when the new release is out.  I haven't found the time
>to properly test your patch so far, but don't worry, it's not forgotten.
>
>Regards,
>
>Daniel.
>
>
Don't bother..... Michael and I are working to even improve it and it 
should be in cvs soon...

There are some questions about marker I/O because the 'standard' 
(whereever that is) isn't clear..

Thanx BTW
Permalink | Reply | 0 comments |
headers
Baker, Steve | 21 Apr 14:19
Favicon

virtual file support in audiofile

I am in the process of implementing an AFvirtualfile for our audiofile
GStreamer plugin. This will allow us to completely integrate audiofile into
the GStreamer system.

However afOpenVirtualFile isn't actually implemented even though it is in
the public header.

So below is my implementation - I know there is another patch floating
around so it would be good if something was commited to CVS.

Now I get a segfault the first time one of the vfile callbacks is called
(which just happens to be vfile->tell). My callback func isn't actually
reached but vfile->tell is non null.

I suspect linking voodoo is at fault. I would appreciate suggestions on what
I could look into.

cheers

Index: openclose.c
===================================================================
RCS file: /cvs/audiofile/libaudiofile/openclose.c,v
retrieving revision 1.8
diff -u -r1.8 openclose.c
--- openclose.c	2001/08/28 20:34:46	1.8
+++ openclose.c	2002/04/21 12:06:22
@@ -261,6 +263,44 @@

 	if (_afOpenFile(access, vf, filename, &filehandle, setup) !=
AF_SUCCEED)
(Continue reading)

Permalink | Reply | 2 comments |
headers
Daniel Kobras | 21 Apr 14:46
Picon

Re: virtual file support in audiofile

On Sun, Apr 21, 2002 at 02:19:33PM +0200, Baker, Steve wrote:
> Index: openclose.c
> ===================================================================
> RCS file: /cvs/audiofile/libaudiofile/openclose.c,v
> retrieving revision 1.8
> diff -u -r1.8 openclose.c
> --- openclose.c	2001/08/28 20:34:46	1.8
> +++ openclose.c	2002/04/21 12:06:22
[...]
> +	if (_afOpenFile(access, vf, NULL, &filehandle, setup) !=
> AF_SUCCEED){
                                ^^ make that 'vfile'.
				
> +		printf("_afOpenFile failed");
> +		af_fclose(vf);
                          ^^ dito.

Below's the tested patch from the Debian diff.

Regards,

Daniel.

---------------------
PatchSet 70
Date: 2002/03/01 21:05:11
Author: kobras
Log:
Implement afOpenVirtualFile(). [dk]
(Continue reading)

Permalink | Reply | 0 comments |
headers
Baker, Steve | 21 Apr 14:54
Favicon

RE: virtual file support in audiofile

> Below's the tested patch from the Debian diff.

This patch works for me - thanks.

My patch would have worked fine were it not for the fact that it was utterly
broken ;)

Is there any estimation on when this might be commited and a new audiofile
be released?

cheers
Permalink | Reply | 752 comments |
headers
Daniel Kobras | 21 Apr 15:35
Picon

Re: virtual file support in audiofile

On Sun, Apr 21, 2002 at 02:54:19PM +0200, Baker, Steve wrote:
> Is there any estimation on when this might be commited and a new audiofile
> be released?

Debian's current version in woody and sid ships with the patch.  Cannot
comment on the upstream version, of course.

Regards,

Daniel.
Permalink | Reply | 751 comments |

Gmane