RE: perl 5.6.1 bug report

RE: perl 5.6.1 bug report

[Neil Lunn]

>Of course, prove me wrong 
> and check out
> a bleadperl build on cygwin (I haven't done this for over a 
> week -- sorry)
> to see how it works there. 
> 
> Neil

Should have checked my own records first. Oops sorry, it is fixed in
bleadperl. A pretty *stable* patch is available from Activestate.

http://downloads.activestate.com/ActivePerl/src/5.6/AP626_diff.txt

Neil

__________________________________________________________________________
Please Note :
Only  the intended recipient is authorised to access or use this e-mail.  If
you are not the intended recipient,
please delete this e-mail and notify the sender immediately.   The contents
of this e-mail are the writer's 
opinion and are not necessarily endorsed by the Gunz Companies unless
expressly stated.

We use virus scanning software but exclude all liability for viruses or
similar in any attachment.



--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

Re: perl 5.6.1 bug report

[e.j.pronk]

neillunn@gunz.com.au wrote:
> > > > > while ( $input =~ /(.*?)#/gsm ) {

> Actually it should be right in all perl 5.6.1 builds. Which leads me to say
> that I am actually not sure which one is the bug, but it is not a "cygwin"
> bug.

> Someone please correct me if I'm wrong on this but I thought (*?) with /s as
> a modifier would never match more than once. In fact, try the same string
> with no newlines and it should only match once.

This is from perlfaq6 - Regexes :

=====
I'm having trouble matching over more than one line. What's wrong?

Here's code that finds everything between START and END in a paragraph:

    undef $/;           # read in whole file, not just one line or
paragraph
    while ( <> ) {
        while ( /START(.*?)END/sm ) { # /s makes . cross line boundaries
            print "$1\n";
        }
    }
=====

Eddy

--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

RE: perl 5.6.1 bug report

[Neil Lunn]

> -----Original Message-----
> From: e.j.pronk@philips.com [ mailto:e.j.pronk@philips.com ]
> Sent: Wednesday, May 16, 2001 9:52 PM
> To: neillunn@gunz.com.au
> Subject: Re: perl 5.6.1 bug report
> neillunn@gunz.com.au wrote:
> 
> > > 1. This looks like a CR/LF problem.  Cygwin's perl-5.6.1 
> is known to
> > > have, er, issues with line endings.  Look for a new cygwin
> > > release when
> > > perl5-porters releases 5.6.2.
> > 
> > Nope.
> > 
> > > > while ( $input =~ /(.*?)#/gsm ) {
> > > >    print "[$1]";
> > > > }
> > 
> > I think you meant /(.+?)#/gsm)
> >                      ^
> 
> You're right. This makes it work with cygwin perl.

Actually it should be right in all perl 5.6.1 builds. Which leads me to say
that I am actually not sure which one is the bug, but it is not a "cygwin"
bug.

Someone please correct me if I'm wrong on this but I thought (*?) with /s as
a modifier would never match more than once. In fact, try the same string
with no newlines and it should only match once.

> This does not change the fact that there is different 
> behaviour between
> the activestate and the cygwin version.
> They are both 5.6.1 and should at least give the same output.

Yet as I state, I see no difference between cygwin, Linux and AIX on 5.6.1.
There is a difference in the Activestate release but it is a difference from
the standard distribution. There is a patch applied to the current ASPerl
from the standard perl distribution, thus meaning it is different from the
current standard.

Again, unless my regex interpretation is screwed up, the result you are
getting with ASPerl is not correct. Of course, prove me wrong and check out
a bleadperl build on cygwin (I haven't done this for over a week -- sorry)
to see how it works there. 

Neil

> 
> use re 'debug'; # to see where it goes wrong
> 
> (that'a perl and english at the same time once again. :-) )
> 
> > 
> > And you will need to strip new lines from the result 
> because you asked it to
> > with "s".
> 
> It's nothing useful anyway. It's a not working .dsw parser 
> cut down to a
> 5 liner to debug perl.
> 
> thanx anyway.
> 
> 
> Eddy
> 

__________________________________________________________________________
Please Note :
Only  the intended recipient is authorised to access or use this e-mail.  If
you are not the intended recipient,
please delete this e-mail and notify the sender immediately.   The contents
of this e-mail are the writer's 
opinion and are not necessarily endorsed by the Gunz Companies unless
expressly stated.

We use virus scanning software but exclude all liability for viruses or
similar in any attachment.



--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

RE: perl 5.6.1 bug report

[Neil Lunn]

> -----Original Message-----
> From: Charles S. Wilson [ mailto:cwilson@ece.gatech.edu ]
> Sent: Wednesday, May 16, 2001 12:35 AM
> To: e.j.pronk@philips.com
> Cc: cygwin@cygwin.com
> Subject: Re: perl 5.6.1 bug report
> 
> 
> 1. This looks like a CR/LF problem.  Cygwin's perl-5.6.1 is known to
> have, er, issues with line endings.  Look for a new cygwin 
> release when
> perl5-porters releases 5.6.2. 

Nope.

> > while ( $input =~ /(.*?)#/gsm ) {
> >    print "[$1]";
> > }

I think you meant /(.+?)#/gsm)
                     ^

And you will need to strip new lines from the result because you asked it to
with "s".

Neil

> 
> 2. Please don't send email to individuals on the cygwin list.  Use the
> list.  That's what it's for.  I have copied this message to 
> the list and
> reset the Reply-To address for your convenience.
> 
> 3. If you *were* going to individually mail somebody about a 
> cygwin-perl
> problem -- why me? I don't have anything to do with cygwin's perl
> package. 
> 
> --Chuck
> 
> e.j.pronk@philips.com wrote:
> > 
> > This gives me an unexpected result with 5.6.1
> > I used 5.005_03 on Linux and 5.6.1 from activestate on NT as
> > second/third oppinion.
> > 
> > cygwin perl only matches aaa while the others match both aaa and bbb
> > (with some newlines).
> > 
> > Eddy
> > 
> > ===
> > $input = <<EOF;
> > aaa
> > #
> > bbb
> > #
> > EOF
> > 
> > while ( $input =~ /(.*?)#/gsm ) {
> >    print "[$1]";
> > }
> 
> --
> Want to unsubscribe from this list?
> Check out: http://cygwin.com/ml/#unsubscribe-simple
> 

__________________________________________________________________________
Please Note :
Only  the intended recipient is authorised to access or use this e-mail.  If
you are not the intended recipient,
please delete this e-mail and notify the sender immediately.   The contents
of this e-mail are the writer's 
opinion and are not necessarily endorsed by the Gunz Companies unless
expressly stated.

We use virus scanning software but exclude all liability for viruses or
similar in any attachment.



--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

Re: perl 5.6.1 bug report

[Charles S. Wilson]

1. This looks like a CR/LF problem.  Cygwin's perl-5.6.1 is known to
have, er, issues with line endings.  Look for a new cygwin release when
perl5-porters releases 5.6.2. 

2. Please don't send email to individuals on the cygwin list.  Use the
list.  That's what it's for.  I have copied this message to the list and
reset the Reply-To address for your convenience.

3. If you *were* going to individually mail somebody about a cygwin-perl
problem -- why me? I don't have anything to do with cygwin's perl
package. 

--Chuck

e.j.pronk@philips.com wrote:
> 
> This gives me an unexpected result with 5.6.1
> I used 5.005_03 on Linux and 5.6.1 from activestate on NT as
> second/third oppinion.
> 
> cygwin perl only matches aaa while the others match both aaa and bbb
> (with some newlines).
> 
> Eddy
> 
> ===
> $input = <<EOF;
> aaa
> #
> bbb
> #
> EOF
> 
> while ( $input =~ /(.*?)#/gsm ) {
>    print "[$1]";
> }

--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple