Join GitHub today
GitHub is home to over 50 million developers working together to host and review code, manage projects, and build software together.
Sign upoverloading 'x' is broken #13811
overloading 'x' is broken #13811
Comments
From @salvaCreated by @salvaCreate the following module: ------------overx.pm-------------------- package overx; require overload; sub new { sub _x { sub _length { sub _to_string { 1; ------------end-of-overx.pm-------------------- and run... $ perl -I. -Moverx -MData::Dumper -e '$a = overx->new(1,'a'); print Dumper [$a, scalar(($a x $a) x $a)]' I get... {1, a} x {1, a} -> {1, a, 1, a} when expecting... {1, a} x {1, a} -> {1, a, 1, a} Other operators work as expected. For instance, the following generates the expected result. perl -I. -Moverx -MData::Dumper -e '$a = overx->new(1,'a'); print Dumper [$a, scalar(($a * $a) * $a)]' While the following also fails: perl -I. -Moverx -MData::Dumper -e '$a = overx->new(1,'a'); print Dumper [$a, scalar(($a * $a) x $a)]' I can reproduce the bug also in 5.8.8. Perl Info |
From ambrus@math.bme.huOn Thu May 08 07:54:20 2014, salva wrote:
Could this be because the x operator in that case is parsed as list repetition, and only string repetition can be overloaded? |
|
The RT System itself - Status changed from 'new' to 'open' |
From @salva----- Original Message -----
No, the expression is wrapped by a "scalar" call. In any case, the real issue is that the return value from the first |
From @salvaThe issue is caused by a 'pushmark' OP being prepended to the left-hand tree of the 'x' binary operator by the parser when the left expression is inside parenthesis. When 'x' is not overloaded the code in pp_repeat undoes the pushmark and places the result on the parent frame, but when the operation is overloaded, the tryAMAGICbin_MG macro call just returns without removing the extra frame. |
From [Unknown Contact. See original ticket]The issue is caused by a 'pushmark' OP being prepended to the left-hand tree of the 'x' binary operator by the parser when the left expression is inside parenthesis. When 'x' is not overloaded the code in pp_repeat undoes the pushmark and places the result on the parent frame, but when the operation is overloaded, the tryAMAGICbin_MG macro call just returns without removing the extra frame. |
From @iabynOn Sun, May 11, 2014 at 11:30:44AM +0200, Salvador Fandiño wrote:
In the 'x' case, an extra pushmark is happening after the '[ $a, ', Whether the pushmark is extraneous, or whether its the fault of something Anyway, I haven't really got time to look any further into this yet, as it -- |
From @cpansproutOn Mon May 12 04:17:13 2014, salva wrote:
There were two stack bugs in pp_repeat, so I fixed them both in 3a100da. Then I added an optimisation in 82e4f30 that sidesteps the issue most of the time. -- Father Chrysostomos |
|
@cpansprout - Status changed from 'open' to 'pending release' |
From @khwilliamsonThanks for submitting this ticket The issue should be resolved with the release today of Perl v5.22. If you find that the problem persists, feel free to reopen this ticket -- |
|
@khwilliamson - Status changed from 'pending release' to 'resolved' |
Migrated from rt.perl.org#121827 (status was 'resolved')
Searchable as RT121827$