While subjecting oneself to heisenbugs and seg faults willingly is not necessarily dumb, it's definitely not wise.
| [reply] |
So you are declaring all the millions of line of pre-5.8 Perl that used signals, and all the billions of lines of C that still use "unsafe" signal handling, "definitely not wise.
Which according to your logic means that the authors of Perl::Unsafe::Signals, along with those p5pers that left the PERL_SIGNALS=unsafe workaround in place should be condemned for their actions?
Or maybe there is another reason for your continued stance.
| [reply]
[d/l] |
So you are declaring all the millions of line of pre-5.8 Perl that used signals [...] not wise
No. Where did you get the idea that it was possible to turn off safe signals pre 5.8? I can't speak as to how wise it is to do something impossible.
So you are declaring [...] all the billions of lines of C that still use "unsafe" signal handling, "definitely not wise.
No. What makes you say that all C code uses unsafe signal handling? The Perl library pre 5.8 wasn't signal-safe, but that doesn't mean that C code in general can't be made signal-safe. Perl 5.8(.?) is an example of such a feat being accomplished.
| [reply] |