On Thu, Jan 5, 2012 at 10:48 AM, h.g. muller
<address@hidden> wrote:
I am not sure why all the option starting with a hyphen exist next to those starting with x,
e.g. -xcoords and --xcoords as opposites of -coords. Those starting with hyphen are neither
documented for XBoard or WinBoard. Can't we simply drop them all?
Perhaps Tim can shed some light on this.
Sure, let's get rid of those. I don't remember why I had both "-" and
"x" prefixes to turn off boolean options. The "-" prefix seems like a
bad idea.
{ "internetChessServerComPort", ArgString, (void *) &appData.icsCommPort, FALSE, INVALID },
{ "icscom", ArgString, (void *) &appData.icsCommPort, FALSE, INVALID },
I don't know what these do; might also be WB only.
Probably WB only. You could try to talk to the ICS through a serial port that way. I don't remember why one would ever want to do that. I don't think anyone would miss the feature if it were deleted.
These options are potentially damaging, as they specify features to be used as if
the engine had sent them, which normally would thoroughly wreck operation.
I think I added the on request as a kludge to run Windows engines under wine,
as a wine bug causes a SIG_INT to kill such binaries even when they ignore SIG_INT,
so that you have to fool XBoard into thinking the engine sent 'feature sigint=0' to
make it work. I could have made explicit options for that (e.g. -firstNoSigInt etc.),
but it seemed more versatile (but also potentially more damaging) to allow
-firstFeatures "sigint=0". I am a bit hesitant to document such hacks.
The sigint thing is a bit of a botch. It is on by default because it was part of the protocol version 1 standard, but I believe only certain versions of GNU Chess ever needed it. For all other engines it's just a problem to be worked around.