[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [gnugo-devel] Evan's replacement pattern
From: |
Gunnar Farneback |
Subject: |
Re: [gnugo-devel] Evan's replacement pattern |
Date: |
Fri, 13 Sep 2002 08:45:08 +0200 |
User-agent: |
EMH/1.14.1 SEMI/1.14.3 (Ushinoya) FLIM/1.14.2 (Yagi-Nishiguchi) APEL/10.3 Emacs/20.7 (sparc-sun-solaris2.7) (with unibyte mode) |
Evan wrote:
> However, none of these apply to my patch. Is this the delta for something
> else I missed, or have I managed to screw up something else?
Those are new tests, with no relation to your patch.
> Also, attached is a new version of the pattern. It should be slightly
> more general, and it also passes the test. I have not run a regression
> delta for it yet.
Don't bother with the regressions. It's clearly way too general.
> +Pattern RE11
> +# evand new pattern (3.3.9)
> +
> +|..
> +|X*
> +|OX
> +|.O
> +
> +:8,-
> +
> +|..
> +|X*
> +|Oc
> +|ab
> +
> +; alive(b) && lib(c) == 2
> +> replace(a,*)
Consider e.g.
|....
|X.X.
|OX..
|.OXX
|.OOO
or
|...X
|..X.
|X..X
|OXXO
|.OOO
|....
or
|XXX.
|...X
|X..X
|OXXO
|.OOO
|....
You really can't do without a reading constraint for this pattern,
unless you make it extremely specific.
/Gunnar