http://qs321.pair.com?node_id=186828

sdyates has asked for the wisdom of the Perl Monks concerning the following question:

This should be easy, but cannot quite do it.

here is the file:

9077/00:00///:8087
9079/00:00/////386/_

here is what I want

9077
9079

herre is my logic:
- remove from first / to end of line
- remove / charcter.

I am using system command. I am not sure how to tell to remove from the first character to end of line.

sed 's/\/*//g' pslog3


I've done it before, but I do not have my perl reference book here!!

please help

Replies are listed 'Best First'.
Re: Remove from character to end of line
by $name (Pilgrim) on Aug 01, 2002 at 16:47 UTC
    You could do somthing like

    my $filename = "9077/00:00///:8087";
my @file = split('/', $filename);

my $newname = $file[0];

    [download]

    MGW Aplications Developer QuinnTeam Inc.
[reply]
[d/l]
Re: Remove from character to end of line
by aufrank (Pilgrim) on Aug 01, 2002 at 16:57 UTC
    I would probably try a pattern like 
    s$\/+.*$$

    [download]
    which matches at least one forward slash and then any characters after it. note that perlstyle recommends that you not use a forward slash as the delimiter for any substitution or match involving slashes, but I don't know if that works from the command line, so use whatever delimiter is most friendly with your shell. I'm sure some regex guru will be along soon to provide a better answer, but this gets the job done at least (I think).

    hth

    --au
[reply]
[d/l]
Re: Remove from character to end of line
by particle (Vicar) on Aug 01, 2002 at 16:54 UTC 
    my( $answer ) = ( $data =~ m!^(.+?)/!);

    [download]

    ~Particle *accelerates*

[reply]
[d/l]
[select]
Re: Remove from character to end of line
by CukiMnstr (Deacon) on Aug 01, 2002 at 16:50 UTC
    you have the s/// operator in perl. this should do what you want: 
    while(<>) {
    s#^([^/]+).*#$1#;
    print;
}

    [download]

    hope this helps,

    Update: even if the s/// works, split() is faster: 

    while(<>) {
    print((split '/')[0], "\n");
}

    [download]

    Update: ++Abigail-II. I need to work on my golfing skills... ;)

[reply]
[d/l]
[select]
      But that has to capture. You can do without capture: 
          s!/.*!!;

      [download]
      It's smaller code too. ;-)

      Abigail

[reply]
[d/l]

        GREAT!!!

        Not only smaller code, but smart code too. I knew there had to be an easy way. It is amazing how dumb I feel with out my reference books.

        Thanks much,
        Simon
[reply]
Re: Remove from character to end of line
by Tomte (Priest) on Aug 01, 2002 at 16:56 UTC 
    my @lines = qw[ 9077/00:00///:8087 9079/00:00/////386/_];
my @erg;
foreach (@lines){
 s/([^\/]*)//;
 push @erg, $1;
}
# erg contains all segments you wanted

    [download]

    HTH
    regards,
    tomte

    update: damn, whenever I wrote and tested my examples, there are a bunch of so much cooler solutions....way to go tomte, way to go...:-)

    update_ii: s/$1 @erg/@erg, $1/

[reply]
[d/l]

Back to Seekers of Perl Wisdom