Re: [gnugo-devel] a patch from a newbie :)

[bump]

> Just a short note to confirm that changing values of WIN, KO_A and
> KO_B is now safe (at least, for regressions), provided that :
> - the symetry is being kept ( REVERSE_RESULT() )
> - gtp_print_code() is getting accordingly updated
> 
> The attached patch is intended for those who would like to verify
> it, as well as for me (testing my address@hidden&$ mail client)

After this patch, called portela_3_9.3, there are some unexpected test
results. There are NO unexpected results in the current CVS ouside of
nngs2.tst. But after the patch, there are the following unexpected results:

./regress.sh . nngs1.tst 
30 unexpected PASS!
./regress.sh . trevorb.tst 
380 unexpected FAIL: Correct 'K4', got 'L5'
./regress.sh . nicklas2.tst 
701 unexpected PASS!
./regress.sh . nicklas3.tst 
401 unexpected FAIL: Correct 'A5', got 'C6'
./regress.sh . trevor.tst 
262 unexpected PASS!
./regress.sh . nngs.tst 
500 unexpected FAIL: Correct 'J17', got 'Q1'
510 unexpected FAIL: Correct 'G14', got 'Q1'
600 unexpected PASS!
./regress.sh . strategy3.tst 
122 unexpected PASS!
./regress.sh . global.tst 
31 unexpected FAIL: Correct 'M8', got 'F11'
38 unexpected PASS!

Ostensibly portela_3_9.3 should have no effect on the regressions. 
Clearly we need to track these down.

Dan