bpa wrote:
> This wasn't the case back on XP where I also spent time adding in code
> to socketwrapper.
> Have you tried this with socketwrapper where named pipes are used to
> commuicate between processes which may not be the case with cmd.exe
Well I've tried the fixed version in LMS and it works on this stream
whereas the old one doesn't - the symptoms exactly fit what I've seen
running faad standalone in a cygwin shell as I described earlier.
Looking at the relevant bit of code in main.c:
Code:
--------------------
retval = fseek(b.infile, 0, SEEK_END);
#ifdef _WIN32
if (0 == strcmp(aacfile, "-")) {
retval = -1;
}
#endif
if (retval )
{
faad_fprintf(stderr, "Input not seekable %s\n", aacfile);
fileread = -1;
streaminput = 1;
} else {
fileread = ftell(b.infile);
fseek(b.infile, 0, SEEK_SET);
};
--------------------
It looks like someone already noticed this happening and put in the
'forced' failure via the bit of code inside the #ifdef _WIN32. But that
isn't a great fix - the trial seek has already messed things up!
I think the fixed code just wants to be the following for all
platforms:
Code:
--------------------
if (0 == strcmp(aacfile, "-")) {
retval = -1;
} else {
retval = fseek(b.infile, 0, SEEK_END);
}
if (retval )
{
faad_fprintf(stderr, "Input not seekable %s\n", aacfile);
fileread = -1;
streaminput = 1;
} else {
fileread = ftell(b.infile);
fseek(b.infile, 0, SEEK_SET);
};
--------------------
------------------------------------------------------------------------
utgg's Profile: http://forums.slimdevices.com/member.php?userid=40900
View this thread: http://forums.slimdevices.com/showthread.php?t=53229
_______________________________________________
plugins mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/plugins
