Re: [Musicpd-dev-team] Fwd: null_encoder: optionally add pcm wave header

2009-11-10 Thread Max Kellermann
On 2009/11/10 22:36, Metyl Methylius wrote: > I've got it ! I've worked on big endian platform for about 3 years and never > realized there is such > a no difference from register point of view :) :) Looks good now, let's hope it really works on BE as I expect it. It's merged. -

[Musicpd-dev-team] Fwd: null_encoder: optionally add pcm wave header

2009-11-10 Thread Metyl Methylius
I've got it ! I've worked on big endian platform for about 3 years and never realized there is such a no difference from register point of view :) metyl -- Forwarded message -- From: Metyl Methylius Date: Tue, Nov 10, 2009 at 9:15 PM Subject: Fwd: [Musicpd-dev-team] null_encoder

[Musicpd-dev-team] Fwd: null_encoder: optionally add pcm wave header

2009-11-10 Thread Metyl Methylius
So you telling me that the code for both cases is the same ? cited from wiki: "Big-endian: The most significant byte (MSB) value, is stored at the memory location with the lowest address". Thats ((value) & 0xFF;) This value is set to zero an therefor is not used in expression. Next the LSB is stor

Re: [Musicpd-dev-team] null_encoder: optionally add pcm wave header

2009-11-10 Thread Max Kellermann
On 2009/11/10 20:06, Metyl Methylius wrote: > why pcm24_to_wave its wrong ? it's maybe not effective but should not be > wrong > it works for little endian and GUINT32_TO_LE() helps to use same masks as > with little endian. > As you pointed its very ineffective and i have rewritten it. That's st

Re: [Musicpd-dev-team] null_encoder: optionally add pcm wave header

2009-11-10 Thread Metyl Methylius
why pcm24_to_wave its wrong ? it's maybe not effective but should not be wrong it works for little endian and GUINT32_TO_LE() helps to use same masks as with little endian. As you pointed its very ineffective and i have rewritten it. I thought the pcm_convert.c does not handle Xbit to 8bit or 8bit

Re: [Musicpd-dev-team] null_encoder: optionally add pcm wave header

2009-11-10 Thread Max Kellermann
On 2009/11/10 19:22, Metyl Methylius wrote: > I think WAVe was designed by Micro$oft runining Intel and there is no way to > support other little endian. > I made convesions (glib based) from all possible sample_bits (8/16/24/32) > but i've tested only on little endian architecture. Your pcm24_to

Re: [Musicpd-dev-team] null_encoder: optionally add pcm wave header

2009-11-10 Thread Metyl Methylius
I think WAVe was designed by Micro$oft runining Intel and there is no way to support other little endian. I made convesions (glib based) from all possible sample_bits (8/16/24/32) but i've tested only on little endian architecture. I'll test it also on my router which is big endian when I get more

Re: [Musicpd-dev-team] null_encoder: optionally add pcm wave header

2009-11-10 Thread Max Kellermann
On 2009/11/10 00:16, Metyl Methylius wrote: > Ok I've separated it. Makes more sence but a also more work. +AC_ARG_ENABLE(wave-encoder, + AS_HELP_STRING([--enable-wave-encoder], + [enable the PCM wave encoder]),, + enable_wave_encoder=auto) Why default to "auto"

Re: [Musicpd-dev-team] IDLE and elapsed time

2009-11-10 Thread Max Kellermann
On 2009/11/10 12:33, Sardem FF7 wrote: > I need to write a (little) client for the MPD and there is a problem > with some data : > - - I need to be notify when the status change (the IDLE command) > - - I need to query status for elapsed time each second > > The only solution I found for the mome

[Musicpd-dev-team] IDLE and elapsed time

2009-11-10 Thread Sardem FF7
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hello, I need to write a (little) client for the MPD and there is a problem with some data : - - I need to be notify when the status change (the IDLE command) - - I need to query status for elapsed time each second The only solution I found for the m