Re: set.q

[John Darrington]

On Tue, Dec 06, 2005 at 07:37:29PM -0800, Ben Pfaff wrote:
     
     > Anyway if you really want to go down the 'custom' path I can do that.
     
     I agree that's a bad way to do it.  But there's no reason to do
     it that way.  In the case of BOXSTRING, for example, nothing
     needs to change at all: it is parsed only to give an error
     message anyhow, because we consider it obsolete.  This is also
     true of several other SET subcommands.
     
     Most of the others could be converted simply by replacing
     statements of the form "set_name = value;" with "set_name
     (value);" and writing a trivial set_name().
     
     SHOW would require a bit more work.  It would probably make sense
     to start from the code for it generated by q2c and then hack it
     by hand.  Some things don't make sense for us to support in SHOW
     anyhow, e.g. as far as I can tell there is no reason to support
     SHOWing the various obsolete settings we don't support.
     
     I'm going to check in a big set of clean-ups sometime this week;
     I could make this part of it if you like.  Let me know.

Perhaps it would be best if you took care of this.  If you can massage
it into a form which makes the get_* functions relatively independent,
then I can get rid of many of the nasty hacks I have in
psppire/src/stubs.c

Then psppire will soon be in a state that it can be merged into the
main branch.

Thanks,

J'
     

-- 
PGP Public key ID: 1024D/2DE827B3 
fingerprint = 8797 A26D 0854 2EAB 0285  A290 8A67 719C 2DE8 27B3
See http://pgp.mit.edu or any PGP keyserver for public key.

pgpoLMAZMbaFY.pgp
Description: PGP signature