Hi Paul, * Paul Fenwick <pjf@perltraining.com.au> [2008-06-04 05:00]: > I know which one I'd rather see. yes, sure. My point is, though, that I see no reason for anyone to use Fatal in the future, except in code that already uses it. But there are people who know Fatal, and who won’t know autodie, who may not feel any reason to switch. It would be an effective incentive for them if the POD says something like “use C<autodie> if you want prettier error messages.” At the same time, in code that already uses Fatal, there seems to be little gain in improving its error messages. Whoever wrote the code used Fatal in spite of its ugly output, so obviously they can at least live with it. > The advantage of making Fatal nicer overall is that people > *can* download the new module (we're dual-lifing it, right?), > drop it on their systems, and have error messages they'd > consider showing to their users, without having to go through > and make changes to their actual codebase. That's a big win if > your codebase is 100,000 lines, It seems pretty easy to substitute autodie for Fatal to me. You were considering returning exception objects from Fatal as well anyway, so in either case the breakage potential is identical. > or you need your code (which may be a module going to CPAN) to > still work with older Perls which don't have the new Fatal. That is a stronger argument, but it’s applicable only as long as it takes you to figure out how to make pragmata lexical in older perls (which you said appears to be possible), right? Regards, -- Aristotle Pagaltzis // <http://plasmasturm.org/>Thread Previous | Thread Next