Earlier I wrote: :I'll try to look a bit more deeply later in the weekend, and maybe see if :I can audit the rest of regcomp for similar issues. I'm happy with the general use of SSize_t_MAX; trying to audit is throwing up too many possibly dubious cases to look at, so I'll try to incorporate that into my more general cleanup of study_chunk - I suspect that at least some of the apparently dubious ones are actually safe, due to the REG_INFTY cap. HugoThread Previous | Thread Next