Front page | perl.perl5.porters |
Postings from January 2017
On 01/25/2017 08:47 AM, Hugo van der Sanden via RT wrote:
> On Sun, 22 Jan 2017 04:48:49 -0800, randir wrote:
>> Before v5.18.0, this'd just die() during a regex compilation. It was
>> either exposed or caused by the following commit:
>>
>> 31c15ce5372b770c3ca899df6cf102f1ed6866ba is the first bad commit
>> commit 31c15ce5372b770c3ca899df6cf102f1ed6866ba
>> Author: Karl Williamson <public@khwilliamson.com>
>> Date: Sat Sep 15 12:27:22 2012 -0600
>>
>> PATCH: [perl #82954] Make "Can't do {n,m} with n > m into warning
>>
>> This commit now causes this situation to warn instead of dying. The
>> portion of the regular expression that can't match is optimized into
>> an
>> OPFAIL.
>
> Yes, this change falls foul of:
> /* Do setup, note this code has side effects beyond
> * the rest of this block. Specifically setting
> * RExC_recurse[] must happen at least once during
> * study_chunk(). */
> .. since it optimizes away the node we must still study.
>
> Karl, do you see any problem with just removing the optimizing aspect, per the attached patch? If not, I'll wrap in a test and commit.
I see no problems. Too bad that comment in the file is buried
>
> Hugo
>
> ---
> via perlbug: queue: perl5 status: new
> https://rt.perl.org/Ticket/Display.html?id=130561
>
Thread Previous
|
Thread Next