> I fear it will require me a lot of time to figure out how to do this...> Could anyone be interested in a bounty for successfully implementing this> feature?>> I was thinking about 50$ on e.g. bountysource.com or similar>> Let me know if someone is interested :-)
If you decide to put up a bounty, I would like to match $4.20.
It's not much, but I thought it'd be fun if it helps getting
some attention on the project.
BTW I've just started working on a Zig binding for OpenAL [0].
There's nothing yet, it might get done soon, it might never
get finished; but if you somehow find Zig [1] interesting and easier
(than C and C++) you might want to check the library out next year.
The rationale for it is that CMake is just a huge pain, building
with Python or not. With Zig (magically) bundle-able via Python
packaging tooling [2] and the existence of setuptools-zig [3],
I really hope for a PEP 517 [4] back-end for zig build sometime
in a near future.
> Hi Nguyễn, sorry, I've missed your reply.
No worries about the time, in fact, I am glad that you still keep
your interest. One side note though, as a Vietnamese, my family name
goes in front of the other parts and I happen to share it with 40%
of the population. In other words, while it's not wrong to call me
Nguyễn, nobody ever does it IRL and thus it's rather awkward.
It's a non-issue, I just thought you'd be interested to find out
about that cultural fact d-;
[0] https://git.sr.ht/~cnx/zeal
[1] https://ziglang.org
[2] https://pypi.org/project/ziglang
[3] https://pypi.org/project/setuptools-zig
[4] https://www.python.org/dev/peps/pep-0517
Thanks for answering! My replies below
On Fri, Aug 20, 2021 at 6:07 PM Nguyễn Gia Phong <mcsinyx@disroot.org> wrote:
> If you decide to put up a bounty, I would like to match $4.20.> It's not much, but I thought it'd be fun if it helps getting> some attention on the project.
Not sure about that figure meaning, but I'm surely ok with it XD
If you are willing to pick it up, just let me know what's your fav
collecting channel
> BTW I've just started working on a Zig binding for OpenAL [0].> There's nothing yet, it might get done soon, it might never> get finished; but if you somehow find Zig [1] interesting and easier> (than C and C++) you might want to check the library out next year.>> The rationale for it is that CMake is just a huge pain, building> with Python or not. With Zig (magically) bundle-able via Python> packaging tooling [2] and the existence of setuptools-zig [3],> I really hope for a PEP 517 [4] back-end for zig build sometime> in a near future.
I didn't know about Zig, I don't think I will be able to take a look at it
anytime soon though.
Out of curiosity, are you still backing palace or it's no longer
interesting to you?
> > Hi Nguyễn, sorry, I've missed your reply.>> No worries about the time, in fact, I am glad that you still keep> your interest. One side note though, as a Vietnamese, my family name> goes in front of the other parts and I happen to share it with 40%> of the population. In other words, while it's not wrong to call me> Nguyễn, nobody ever does it IRL and thus it's rather awkward.> It's a non-issue, I just thought you'd be interested to find out> about that cultural fact d-;
lol, I see, what's your preferred handle? ;D
> [0] https://git.sr.ht/~cnx/zeal> [1] https://ziglang.org> [2] https://pypi.org/project/ziglang> [3] https://pypi.org/project/setuptools-zig> [4] https://www.python.org/dev/peps/pep-0517
> Out of curiosity, are you still backing palace or it's no longer> interesting to you?
I am still maintaining palace, which includes trying to fix reported bugs
and ensure compatibility with updates from dependencies. Palace is not
under active development though, it has reached the original goal [0]
of covering alure2's API.
Aside from #1 which I still have no idea how to fix [1], there isn't
anything to do except for your feature request. However, IMHO alure2's
architecture is rather complex and gave me quite a fatigue trying to do
either. I have much better hope with the Zig one, which is a degree
simpler than alure2; in fact I've been able to replicate a similar aPI
and make it functional in just a few days, so until Zeal is more stable,
> If you are willing to pick it up, just let me know what's your fav> collecting channel
I don't think I will work on the loopback device for palace.
I don't need the bounty either and wish somebody will beat me to it.
I am currently financially supported by my parents and I am applying
for a fully-funded scholarship starting next year. Plus GSoC'20 left me
a decent saving compared to the local cost of living.
BTW, after working directly with OpenAL for a while, the loopback example
Chris Robinson (kcat) posted [2] looks pretty simple. More importantly,
alure2's relevance as a codec wrapper has fell as libsndfile is supporting
all format now (with the exception of MP3 in beta [3]). If you
are interested, I can drop you a simple C program with loopback device,
HRTF context, buffers loaded from audio files, etc. so all you need
is calling the alSource* and alEffect* functions to manipulate the sounds.
[0] https://drewdevault.com/2021/01/04/A-culture-of-stability-and-reliability.html
[1] https://todo.sr.ht/~cnx/palace/1
[2] https://github.com/kcat/openal-soft/issues/496
[3] https://github.com/libsndfile/libsndfile/pull/499
> what's your preferred handle? ;D
I use my real name Nguyễn Gia Phong and pseudonym McSinyx
interchangeably online. IRL conversations, I am often called
Gia Phong or just Phong.
Thanks Phong, I cannot tell if I could work on it now (my time for
this has shrunk again), but for you it's quick and easy, the sample C
program could be a useful future reference.
Note: I offered a symbolic bounty with no intention to allude that
anyone involved would need it, it was just a less unpolite way to ask
for a feature than a "hey, why don't you just spend your spare time
for me and for a priceless 'thank you'" :D
BTW kudos for your GSoC'20 and good luck for having a funded scholarship!
> good luck for having a funded scholarship!
Sorry for the delay, I had an interview last week
and it seemed to go quite well!
Anyway, I drafted a loopback example reading and writing files [0],
which I waived all copyright and related or neighboring rights to
(i.e. it's under public domain). With OpenAL Soft and libsndfile
installed, you can compiled it via
cc -std=c99 main.c -lopenal -lsndfile
For brevity most errors are left unhandled, although they can
be silently ignored if the produced file is OK. It might be useful
to refer to the docs of OpenAL [1] and libsndfile [2] while hacking.
As for the bounty, I am sorry for being unclear, I just wanted somebody
else to be able to claim it, since that would benefit the project better.
I really hope to see someone taking up the task one day.
[0] https://paste.sr.ht/~cnx/c275b45ca70c8efe0db85a9fa3b005b523f55d3c
[1] https://www.openal.org/documentation/OpenAL_Programmers_Guide.pdf
[2] https://libsndfile.github.io/libsndfile/api.html
Thanks a lot Phong, and great news for the interview, good luck!
On Mon, Aug 30, 2021 at 12:03 PM Nguyễn Gia Phong <mcsinyx@disroot.org> wrote:
>> > cc -std=c99 main.c -lopenal -lsndfile>> I forgot to link math here, the right command should have been>> cc -std=c99 main.c -lm -lopenal -lsndfile