-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 G'day Vincent / All, Vincent Pit wrote: > * Fixing the bug in the core, and making autodie depend on > B::Hooks::EndOfScope conditionnally on perl 5.10.0. This gets my vote. I'm happy with %^H being preserved when a string eval is compiled; I expect that 'use feature' pragma will break without it. But I consider it a bug that the contents of %^H are being stringified as soon as the string eval has been compiled. This would get my vote as well. > * The Variable::Magic part of B::Hooks::EndOfScope can be replaced by > tieing %^H to add it a destructor. The only drawback that I know of is > that dieing at compile-time in the scope of the pragma can then generate > some spurious "(in cleanup)" messages. If it's convenient, could you provide an example? I'm happy with the idea that %^H can be tied. I presume it could then trigger a destructor at the end of lexical scope. I guess that means that string eval actually copies %^H into an entirely new structure, rather than increasing the ref-count of the old one? All the best, Paul - -- Paul Fenwick <pjf@perltraining.com.au> | http://perltraining.com.au/ Director of Training | Ph: +61 3 9354 6001 Perl Training Australia | Fax: +61 3 9354 2681 -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.5 (MingW32) iD8DBQFKUGrxx5N6j7FHnlURAlQrAJwLOvCQbql6CSnoLy/pI6tmxJbzggCdHHZJ PmILdkVt51dpY3SDK3Ctn7U= =7qk2 -----END PGP SIGNATURE-----Thread Previous | Thread Next