Re: Internal error with gawk-5.3.0

[Miguel Pineiro Jr.]

On Sun, Jan 7, 2024, at 4:34 PM, Miguel Pineiro Jr. wrote:
> I don't see any reason to zero when every byte that's allocated will be 
> written to by the populating loop (and by the fix in the patch).

Disregard. Nothing to see here ;) That analysis is wrong. If there's a decoding 
error in a non-utf8 multibyte stream, str2wstr will not map the byte at which 
the error began. In which case, the zero prevents a random value from indexing 
beyond the bounds of the array.

Miguel