Front page | perl.perl5.porters |
Postings from November 2010
>On 11/27/10 12:36 PM, Rafael Garcia-Suarez wrote:
>>> I expect STDERR to be delocalized before final exeption handling kicks in.
>>> It's as if the lexical scope isn't being left.
>>
>> Not a bug, in my opinion (local() being dynamic scoping, not lexical scoping).
> Yes, the effect on STDERR is dynamic, but that dynamic effect should be <SNIP>
> undone as part of leaving that lexical scope.
But it *hasn't* left that scope:
% perl -d /tmp/e
[./.perldb] setting stdout to autoflush
Die handler removed.
warn set to 0, die set to 0
Loading DB routines from perl5db.pl version 1.32
Editor support available.
Enter h or `h h' for help, or `man perldebug' for more help.
main::(/tmp/e:12): bar();
DB<1> l 1-20
1 sub foo {
2: die q<why doesn't this show up?>;
3 }
4
5 sub bar {
6: my $x;
7: open my $h, '>', \$x;
8: local *STDERR = $h;
9: foo();
10 }
11
12==> bar();
DB<2> b 2
DB<3> c
main::foo(/tmp/e:2): die q<why doesn't this show up?>;
DB<3> T
. = main::foo() called from file `/tmp/e' line 9
. = main::bar() called from file `/tmp/e' line 12
DB<3>
It is still within scope.
--tom
Thread Previous
|
Thread Next