Re: reg exp

• Previous message: Ken Chesak: "reg exp"
• In reply to: Ken Chesak: "reg exp"
• Next in thread: Joe Smith: "Re: reg exp"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

Date: Mon, 30 Aug 2004 17:15:54 GMT

Ken Chesak wrote:
> Perl scipt is formatting text for HTML page. It changes things like
> an & to &amp. But should not change &nbsp. It uses \ as an escape
> character. So \&nbsp will become &nbsp. The final results are
> correct, but is there a better way to do this?
>
> Input file test.txt
> \HOME & \  BORN \& FREE BORN FREE ' \' HELP " \" w\\\\\\\w
>
> 1st change
> 1a= \HOME & \  BORN \& FREE BORN FREE '' \' HELP " \"
> w\\\\\\\w
> 2nd changes
> 1b= HOME &   BORN & FREE BORN FREE '' ' HELP " "
> w\\\w
>
> #!/usr/local/bin/perl5
> #
> %encode = ( '&' => '&',
> '"' => '"',
> '\'' => '\'\'' );
>
> $data = `cat test.txt`;
> print "Oa= $data\n";
> $data =~ s/(?<!\\)(.)/defined($encode{$1})?$encode{$1}:$1/eg;
> print "1a= $data\n";
> $data =~ s/(\\)(.)/$2/g;
> print "1b= $data\n";

Don't know about better, but this does it with one substitution, and
does not require escaping of HTML entities in the original text:

     $data =~ s{(&#?\w+;)|\\(.)|([&"'])}
               { $1 ? $1 : $2 ? $2 : $encode{$3} }eg;

Another thing is that I'm a bit confused about the wider purpose with
the exercise...

-- 
Gunnar Hjalmarsson
Email: http://www.gunnar.cc/cgi-bin/contact.pl

• Previous message: Ken Chesak: "reg exp"
• In reply to: Ken Chesak: "reg exp"
• Next in thread: Joe Smith: "Re: reg exp"
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]