RFT (request for tests) on variable length lookbehind branch

The branch at

https://perl5.git.perl.org/perl.git/shortlog/refs/heads/smoke-me/khw-vlb

contains a preliminary implementation of variable length look behind.

I'm looking for people to cook up tests for it, as my imagination seems 
to be limited here.
0
public
2/6/2019 2:09:56 AM
perl.perl5.porters 47509 articles. 0 followers. Follow

2 Replies
9 Views

Similar Articles

[PageSpeed] 9

On 2/5/19 9:09 PM, Karl Williamson wrote:
> The branch at
> 
> https://perl5.git.perl.org/perl.git/shortlog/refs/heads/smoke-me/khw-vlb
> 
> contains a preliminary implementation of variable length look behind.
> 
> I'm looking for people to cook up tests for it, as my imagination seems 
> to be limited here.

Looking at the diff between that branch and blead, I don't see anything 
that specifically documents variable length look-behind.  Am I missing 
something?

Normally, I would expect to write tests exercising the functionality as 
documented.  But if there is no documentation, I'd just be guessing as 
to how to test.

Thank you very much.
Jim Keenan
0
jkeenan
2/6/2019 2:28:59 AM
On 2/5/19 7:28 PM, James E Keenan wrote:
> On 2/5/19 9:09 PM, Karl Williamson wrote:
>> The branch at
>>
>> https://perl5.git.perl.org/perl.git/shortlog/refs/heads/smoke-me/khw-v=
lb
>>
>> contains a preliminary implementation of variable length look behind.
>>
>> I'm looking for people to cook up tests for it, as my imagination=20
>> seems to be limited here.
>=20
> Looking at the diff between that branch and blead, I don't see anything=
=20
> that specifically documents variable length look-behind.=C2=A0 Am I mis=
sing=20
> something?

It's documented in perlre.   There are no syntax changes.  Any perl book=20
will talk about lookbehind assertions.  It's just that now you don't get=20
a compilation error if your assertion isn't fixed length.
>=20
> Normally, I would expect to write tests exercising the functionality as=
=20
> documented.=C2=A0 But if there is no documentation, I'd just be guessin=
g as=20
> to how to test.
>=20
> Thank you very much.
> Jim Keenan
>=20
0
public
2/6/2019 3:14:49 AM
Reply: