Re: Re: [patch] error masks

Derick Rethans wrote: > And the suggestion? Yet another ini setting (some even advocate making > it INI_SYSTEM, which is obviously totally wacko)... This idea is an > ostrich method. Derick, I'm only trying to suggest ways of fixing problems, calling them "obviously wacko" is useless. I don't give a crap either way whether this exact feature gets into core, but it *is* in the notes, although not really in any recognizable language: 17. @operators. RFC it. (Stas, need more clarification) Let me be clear: what I am in support of is removing unnecessary performance hit when an error is completely hidden either by error_reporting or by @ (i.e. no track_errors, no user error handler, no other things that would expect to see the errors). That's all that matters to me, the suggestion on INI_SYSTEM was an example of a way to solve a specific problem with Stas's proposed patch, probably not the best solution, but calling it "wacko" is ridiculous. I suggest everyone grow up and if that isn't possible, at least stop pissing me off. I hear more than enough irrational screaming from my month-old, there's simply no need for it on internals@ as well. Love, Greg