Re: gawk: other double free(_wstr)

[Karel Zak]

On Mon, Jan 15, 2007 at 05:07:31PM -0500, Andrew J. Schorr wrote:
> I agree, it shouldn't matter, as long as the NODE is of the correct type.

 Hey.. we all expect that NODE is correct thing :-) It's too
 critical part of gawk.

> But zeroing in all cases is just a (small) waste of CPU time that serves
> to obfuscate the correctness of the code.

 Frankly, it's not so important for me. I'd like to see robust and
 stable gawk -- both versions of free_wstr() works now. (Maybe in
 future we will see that one version is better. I don't think that we
 have to decide it right now.) More tests, more assert() and more
 paranoid options is always better from my point of view. And a
 performance optimalization should be based on performance tests.

  Karel

-- 
 Karel Zak  <address@hidden>