Re: Is negative lookbehind behaving here?
[email protected] (Timo Paulssen) Fri, 4 May 2018 18:40:20 +0200
| Newsgroups | perl.perl6.language |
|---|---|
| Message-ID | <[email protected]> |
--------------5B4ACBDF8F79BD3A8250EF63
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Yes, you've encountered a bug. It's got these two tickets:
https://rt.perl.org/Public/Bug/Display.html?id=3D124898
https://rt.perl.org/Public/Bug/Display.html?id=3D131964
I've got a branch in nqp and rakudo that I'll merge very soon that fixes
both of these bugs.
Until then you can switch $ and ^, but when the version lands that fixes
the bug, that'll "misbehave" again. I think having ^|$ in the
before/after assertions, i.e. literally match either the beginning or
the end, should work with pre-fix and post-fix versions, as you cannot
have text after the end of the string for example.
Hope that helps!
=C2=A0 - Timo
On 04/05/18 07:38, yary wrote:
> I want to match something anywhere but at the end of a string in one
> example, or anywhere but at the start of a string in another example.
> The "except at start" one has me stumped. Not sure if it's me or if
> I've tickled a bug.
>
> perl6 --version
> This is Rakudo Star version 2018.01 built on MoarVM version 2018.01
> implementing Perl 6.c.
>
> # Match all but at end- this works as I like
> > 'abcd' ~~ m/.+<!before $>/
> =EF=BD=A2abc=EF=BD=A3
>
> # Match all but at start- this puzzles me, 'a' is after the start of
> string but still matches
> > 'abcd' ~~ m/<!after ^>.+/
> =EF=BD=A2abcd=EF=BD=A3
>
> # This is a workaround
> > 'abcd' ~~ m/<!before ^>.+/
> =EF=BD=A2bcd=EF=BD=A3
>
> Using a perl5 debugger session, roughly translated negative
> lookbehind/lookahead work as I expect.
>
> =C2=A0 DB<1> p 'abcd' =3D~ /(.+(?!$))/
> abc
> =C2=A0 DB<2> p 'abcd' =3D~ /((?<!^).*)/
> bcd
>
> -y
--------------5B4ACBDF8F79BD3A8250EF63
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: quoted-printable
<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dutf=
-8">
</head>
<body text=3D"#000000" bgcolor=3D"#FFFFFF">
<p>Yes, you've encountered a bug. It's got these two tickets:</p>
<p><a class=3D"moz-txt-link-freetext" href=3D"https://rt.perl.org/Pub=
lic/Bug/Display.html?id=3D124898">https://rt.perl.org/Public/Bug/Display.=
html?id=3D124898</a><br>
<a class=3D"moz-txt-link-freetext" href=3D"https://rt.perl.org/Publ=
ic/Bug/Display.html?id=3D131964">https://rt.perl.org/Public/Bug/Display.h=
tml?id=3D131964</a></p>
<p>I've got a branch in nqp and rakudo that I'll merge very soon
that fixes both of these bugs.</p>
<p>Until then you can switch $ and ^, but when the version lands
that fixes the bug, that'll "misbehave" again. I think having ^|$
in the before/after assertions, i.e. literally match either the
beginning or the end, should work with pre-fix and post-fix
versions, as you cannot have text after the end of the string for
example.</p>
<p>Hope that helps!<br>
=C2=A0 - Timo<br>
</p>
<br>
<div class=3D"moz-cite-prefix">On 04/05/18 07:38, yary wrote:<br>
</div>
<blockquote type=3D"cite"
cite=3D"mid:[email protected]=
l.com">
<div dir=3D"ltr">
<div>I want to match something anywhere but at the end of a
string in one example, or anywhere but at the start of a
string in another example. The "except at start" one has me
stumped. Not sure if it's me or if I've tickled a bug.<br>
</div>
<div><br>
perl6 --version<br>
This is Rakudo Star version 2018.01 built on MoarVM version
2018.01<br>
implementing Perl 6.c.<br>
<br>
</div>
<div># Match all but at end- this works as I like<br>
</div>
<div>> 'abcd' ~~ m/.+<!before $>/<br>
=EF=BD=A2abc=EF=BD=A3<br>
<br>
</div>
<div># Match all but at start- this puzzles me, 'a' is after the
start of string but still matches<br>
</div>
<div>> 'abcd' ~~ m/<!after ^>.+/<br>
=EF=BD=A2abcd=EF=BD=A3<br>
<br>
</div>
<div># This is a workaround<br>
> 'abcd' ~~ m/<!before ^>.+/<br>
=EF=BD=A2bcd=EF=BD=A3<br>
</div>
<div><br>
Using a perl5 debugger session, roughly translated negative
lookbehind/lookahead work as I expect.<br>
<br>
=C2=A0 DB<1> p 'abcd' =3D~ /(.+(?!$))/<br>
abc<br>
=C2=A0 DB<2> p 'abcd' =3D~ /((?<!^).*)/<br>
bcd<br>
<br clear=3D"all">
<div>
<div class=3D"gmail_signature">-y<br>
</div>
</div>
</div>
</div>
</blockquote>
<br>
</body>
</html>
--------------5B4ACBDF8F79BD3A8250EF63--