Path: news.cs.au.dk!not-for-mail
From: "Sascha Kimmel" <kimmel@tricos.com>
Newsgroups: comp.lang.beta
Subject: RE: REGEXP not doing anything!
Date: 10 Jun 1999 09:05:24 -0000
Organization: University of Aarhus, Department of Computer Science (DAIMI)
Lines: 98
Approved: mailtonews@cs.au.dk
Distribution: world
Message-ID: <19990610090524.22112.qmail@noatun.mjolner.dk>
Reply-To: "Sascha Kimmel" <kimmel@tricos.com>
NNTP-Posting-Host: daimi.cs.au.dk
X-Trace: xinwen.cs.au.dk 929005543 4418594 255.255.255.255 (10 Jun 1999 09:05:43 GMT)
X-Complaints-To: news@cs.au.dk
NNTP-Posting-Date: 10 Jun 1999 09:05:43 GMT
Xref: news.cs.au.dk comp.lang.beta:11989

Hi!

Thank you for your quick reply.
But to say it directly: Mjolner has made this bug, too because the first
example is taken out of the Basic Libraries Documentation (mia90-08.pdf)
and there no t.reset can be found. Maybe Mjolner could fix this in newer
versions of the documentation? :)

Thanks again.

Regards,
Sascha Kimmel

> -----Original Message-----
> From: Jorgen Lindskov Knudsen [mailto:jlk@cs.au.dk]
> Sent: Thursday, June 10, 1999 10:17 AM
> To: usergroup@mjolner.dk
> Subject: Re: REGEXP not doing anything!
>
>
> You have made the same bug in both examples.  The problem is
> not relatet to
> regexp, but the fact, that text variables in BETA contains
> the notion of
> position within the text, and most text operations work 'from
> the current
> position' in the text.  When texts are initialized by statements like:
>     'Hello world'->t[]
> then the position will be at the very end of the text, and
> the following
> regexp_replace operation will try to replace in the string from that
> position - since thaat is at the end of the string, no match
> will be found
> (and the noMatch virtual will in fact be invoked).
>
> In your examples, just insert
>     t.reset
> as in:
>     'Hello world'->t[]; t.reset
> than everything will work as you expect.
>
> Regards,
>
> Jørgen Lindskov Knudsen
>
> Sascha Kimmel wrote in message
> <19990610073958.20891.qmail@noatun.mjolner.dk>...
> >Hi!
> >
> >I want to use the regexp fragment for parsing purposes, but even your
> >simplest demo program (p. 86 in mia90-08.pdf) does not work with BETA
> >4.1:
> >
> >ORIGIN '~beta/basiclib/v1.6/betaenv';
> >INCLUDE '~beta/basiclib/v1.6/regexp';
> >
> >-- program: descriptor --
> >
> >(#
> > t:^text;
> > do
> > 'Hello world' -> t[];
> > ('\\<w.*\\>', '\\0 champion') -> t.regexp_replace;
> > t[] -> putline; (* prints 'Hello world champion' *)
> >#)
> >
> >
> >This program just produces a "Hello world" and not "Hello world
> >champion".
> >The very very simple
> >
> >ORIGIN '~beta/basiclib/v1.6/betaenv';
> >INCLUDE '~beta/basiclib/v1.6/regexp';
> >
> >-- program: descriptor --
> >
> >(#
> > t:^text;
> > do
> > 'Hello BETA!' -> t[];
> > ('BETA', 'Mjolner') -> t.regexp_replace;
> > t[] -> putline; (* should print 'Hello Mjolner!' *)
> >#)
> >
> >only gives me a "Hello BETA!" and not the string that should
> be there.
> >
> >Is the regexp fragment working in some way?
> >
> >Any quick help is appreciated.
> >
> >Regards,
> >Sascha Kimmel
> >
>
>
>