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

Hello everyone,

If I have a file in which I am searching for any line beginning with "Open Square Bracket" and ending with "Close Square Bracket" and then saving all such matching lines in an Array called @BROKERS. How would I tell perl to go through the file again, matching on each element of @Brokers, and then after it found a line that matched that element push each subsidiary line that followed into an array named $BROKER[x] until a blank line is reached. At this point the loop should continue to the next element of @BROKERS and repeat the above proccess until there are no elements left.

I am very new to perl and languages in general so please excusive the crudity of the code I have written so far. 

#!/usr/bin/perl -w
#

my @BROKERS;

open (CONFIG, "ubroker.properties") || die "Can't Open ubroker file: $
+!";

while (<CONFIG>)
  {
    next if /^#/;
    if (/^\[UBroker\.WS\./)
    {
    push (@BROKERS,$_);
    }
  }
chomp @BROKERS;
while (<CONFIG>)
  {
    foreach $b(@BROKERS)
     {
       if (/$b/)
         {
# Here is where I run out of steam
# Begin Metacode
          read each line and push ($_,@"$b") continue to do this until
+ a newline is reached.  When a newline is reached intialize the next 
+$b as an array and repeat the above proccess until all elements of @B
+ROKERS are exhausted.

[download]

I hope the question ain't to crude, if it is I'll be glad to expand on it!
Thanks!
Pat

Edit by tye

Replies are listed 'Best First'.
Re: Appropriate application of readline
by blakem (Monsignor) on Sep 26, 2001 at 23:34 UTC
    It seems like you should be using a slightly different datastructure for this. A hash of arrays (HoA) would fit nicely... The hash is keyed on the broker's name and the values are arrays containing the matching lines. Something like this: 
    #!/usr/bin/perl -wT
use strict;

my @brokerlist = qw(broknum1 broknum2 broknum3);    # array of brokers
my $pattern = join('|',@brokerlist);    # create a ORed regex of the b
+rokers

my %brokers;       # will become a hash of arrays, keyed on broker nam
+e

while(<DATA>) {
  chomp;
  # if the line matches push it onto appropriate array
  push(@{$brokers{$1}},$_ ) if /\b($pattern)\b/o;   
}


### print out the datastructure to make sure it is correct
for my $broker (keys %brokers) {
  print "$broker\n";
  print "\t$_\n" for (@{$brokers{$broker}});
}

=OUTPUT
broknum2
        broknum2 is my favorite broker
broknum3
        i am a broknum3 client
        my uncle is broknum3
broknum1
        this line if for broknum1
        broknum1 is a fool
=cut

__DATA__
this line if for broknum1
i am a broknum3 client
broknum2 is my favorite broker
broknum1 is a fool
my uncle is broknum3

    [download]

    -Blake

[reply]
[d/l]
      Thanks for the reply's everbody!!!
      While I was waiting for some answers I decided to forge on by myself. Obviously the ideas you guys gave me are better, and I will be incorporating them into my current code. For posterity though here is how I worked around:
      #!/usr/bin/perl -w
#
my @BROKERS;

open (CONFIG, "ubroker.properties") || die "Can't Open ubroker file: $
+!";

while (<CONFIG>)
  {
    $/="";
    next if /^#/;
    if (/^\[UBroker\.WS\./)
    {
      push (@BROKERS,$_);
    }
  }
#chomp @BROKERS;
@FIRST=split(/\s+/,$BROKERS[1]);
#@SECOND=split(/\s+/,$BROKERS[2]);
#@THIRD=split(/\s+/,$BROKERS[3]);
#@FOURTH=split(/\s+/,$BROKERS[4]);;

foreach $i(@FIRST)
 {
   if ($i =~ /^initialSrvrInstance/i)
     {
       $max= $i;
       print "$max\n";
     }
 }

close (CONFIG) || die "Can't Close ubroker file: $!";
exit 0;

      [download]

      But like you said I really like the entire hash approach better then what I am doing now!!!
      Thanks for the Help!
      Pat+
[reply]
[d/l]
Re: Appropriate application of readline
by tachyon (Chancellor) on Sep 26, 2001 at 23:14 UTC

    You would probably be best to use a hash where the key is the broker you find and the values is the data. Here is an example: 

    #!/usr/bin/perl -w

use strict;
my (%brokers, $broker);

# get file contents into an array for convenience
open (CONFIG, "ubroker.properties") || die "Can't Open ubroker file: $
+!";

while (<CONFIG>) {
    next if /^#/;
    chomp;
    $broker = $_ if /^\[.*\]$/;  # found a broker
    $broker = '' if /^\s*$/;     # found a blank line
    # add line to current $broker if there is one
    $brokers{$broker} .= "$_\n" if $broker;
}

close CONFIG;

# keys %brokers returns an array of brokers to iterate over
for (keys %brokers) {
    print "\nBroker: $_\n"; # this is the broker name
    print $brokers{$_};     # this is the data (includes name)
}

    [download]

    If you want to do it with a while(<CONFIG>) you either need to do it in one loop (as you will reach the end of the file after the first loop) or use seek CONFIG, 0, 0; to go back to the begining of the file before you do a second while loop through it.

    In this example we use a hash to store both the brokers array (as the keys) and the data associated with each broker (as the values) which means that there is a logical link between the data. If you just wanted arrays then: 

    @brokers = keys %brokers;
@data    = values %brokers;

    [download]

    cheers

    tachyon

    s&&rsenoyhcatreve&&&s&n.+t&"$'$`$\"$\&"&ee&&y&srve&&d&&print

[reply]
[d/l]
[select]

Back to Seekers of Perl Wisdom