mirrors/gstreamer
mirrors/gstreamer
1
1
Fork 0
mirror of https://gitlab.freedesktop.org/gstreamer/gstreamer.git synced 2024-07-03 05:05:52 +00:00
Code Issues 1 Projects Releases Wiki Activity

gst/audioconvert/audioconvert.c: More elegant and working temp buffer selection algo.

Browse source 
Original commit message from CVS:
* gst/audioconvert/audioconvert.c: (if), (float),
(audio_convert_get_func_index), (check_default),
(audio_convert_clean_fmt), (audio_convert_prepare_context),
(audio_convert_clean_context), (audio_convert_get_sizes),
(audio_convert_convert):
More elegant and working temp buffer selection algo.
This commit is contained in:
Wim Taymans 2005-08-26 18:43:02 +00:00
parent 123aa7de1a
commit ddec57c089
2 changed files with 47 additions and 52 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

9
ChangeLog
View file

@ -1,3 +1,12 @@
2005-08-26 Wim Taymans <wim@fluendo.com>
* gst/audioconvert/audioconvert.c: (if), (float),
(audio_convert_get_func_index), (check_default),
(audio_convert_clean_fmt), (audio_convert_prepare_context),
(audio_convert_clean_context), (audio_convert_get_sizes),
(audio_convert_convert):
More elegant and working temp buffer selection algo.
2005-08-26 Wim Taymans <wim@fluendo.com>
* gst/audioconvert/audioconvert.c: (if), (float),

90
gst/audioconvert/audioconvert.c
View file

@ -302,36 +302,13 @@ audio_convert_get_sizes (AudioConvertCtx * ctx, gint samples, gint * srcsize,
return TRUE;
}
static gpointer
get_temp_buffer (AudioConvertCtx * ctx, gpointer src, gint srcsize,
gboolean writable, gint tmpsize)
{
gpointer result;
if (srcsize >= tmpsize && writable) {
result = src;
} else {
if (ctx->tmpbufsize < tmpsize) {
ctx->tmpbuf = g_realloc (ctx->tmpbuf, tmpsize);
ctx->tmpbufsize = tmpsize;
}
result = ctx->tmpbuf;
}
return result;
}
gboolean
audio_convert_convert (AudioConvertCtx * ctx, gpointer src,
gpointer dst, gint samples, gboolean src_writable)
{
gint insize, outsize;
gboolean final;
gpointer buf;
gint bufsize;
gboolean bufwritable;
gpointer tmpdst;
gint tmpsize = 0;
gpointer inbuf, outbuf, tmpbuf;
g_return_val_if_fail (ctx != NULL, FALSE);
g_return_val_if_fail (src != NULL, FALSE);
@ -344,54 +321,63 @@ audio_convert_convert (AudioConvertCtx * ctx, gpointer src,
insize = ctx->in.unit_size * samples;
outsize = ctx->out.unit_size * samples;
/* find biggest temp buffer size */
{
gint biggest = 0;
if (!ctx->in_default)
biggest = insize * 32 / ctx->in.width;
if (!ctx->mix_passthrough)
biggest = MAX (biggest, outsize * 32 / ctx->out.width);
/* see if one of the buffers can be used as temp */
if (insize >= biggest && src_writable)
tmpbuf = src;
else if (outsize >= biggest)
tmpbuf = dst;
else {
if (biggest > ctx->tmpbufsize) {
ctx->tmpbuf = g_realloc (ctx->tmpbuf, biggest);
ctx->tmpbufsize = biggest;
}
tmpbuf = ctx->tmpbuf;
}
}
/* this is our source data, we start with the input src data. */
buf = src;
bufsize = insize;
bufwritable = src_writable;
inbuf = src;
outbuf = dst;
if (!ctx->in_default) {
/* check if final conversion */
final = (ctx->out_default && ctx->mix_passthrough);
if (final)
tmpdst = dst;
else {
tmpsize = insize * 32 / ctx->in.width;
tmpdst = get_temp_buffer (ctx, buf, bufsize, bufwritable, tmpsize);
}
if (!final)
outbuf = tmpbuf;
ctx->unpack (buf, tmpdst, ctx->scale, samples * ctx->in.channels);
ctx->unpack (inbuf, outbuf, ctx->scale, samples * ctx->in.channels);
if (!final) {
/* new source data, it is writable */
buf = tmpdst;
bufsize = tmpsize;
bufwritable = TRUE;
}
inbuf = outbuf;
}
if (!ctx->mix_passthrough) {
/* see if we need an intermediate step */
final = ctx->out_default;
if (final)
tmpdst = dst;
else {
tmpsize = outsize * 32 / ctx->out.width;
tmpdst = get_temp_buffer (ctx, buf, bufsize, bufwritable, tmpsize);
}
if (!final)
outbuf = tmpbuf;
else
outbuf = dst;
/* convert */
gst_channel_mix_mix (ctx, buf, tmpdst, samples);
gst_channel_mix_mix (ctx, inbuf, outbuf, samples);
if (!final) {
buf = tmpdst;
}
inbuf = outbuf;
}
if (!ctx->out_default) {
/* output always to dst buffer */
ctx->pack (buf, dst, ctx->depth, samples * ctx->out.channels);
ctx->pack (inbuf, dst, ctx->depth, samples * ctx->out.channels);
}
return TRUE;