On Thursday 18 September 2003 22:09, Larry Troxler wrote:
Was there a known problem with the MIDI input parser in the 0.36-devel CVS version?
Yes, sys_dispatchnextmidii seems very suspect to me, although it's a bit hard to read because for some bizarre reason the file I got from CVS uses both hard tabs and spaces combined to do indentation - I am not kidding - the file actually has 0x09's in it - why??? - do I need to give some flag to CVS to avoid this?
Anyhow, from what I could read of it, it looks like it's missing any special handling for song position pointer. It seems to go to the switch statement for any byte from 0x80 and 0xf8, but the only cases are for channel status bytes and there is no definition at all for song position pointer (0xf2).
Apologies in advance it this has been fixed for 0.37.
Larry