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

A busy day ! (at least for me)

I have an array of strings and a file that is of the format: 
str1   value1
str2   value2
...

[download]
both the array and the strings in the file are not necessarly sorted.
I need a quick way to run on this file and get the values for the strings in the array (we shell assume they all exist), I thought about reading the file to an array, sort it and run on it again (I sorted the array too).
Anyone has a better way?

Thanks.

Hotshot

Replies are listed 'Best First'.
Re: searching a file
by tachyon (Chancellor) on Mar 19, 2002 at 17:13 UTC

    If you have key => value pairs then a hash is the way to go: 

    my %hash;
# get data as hash
open FILE, $file or die $!\n;
while (<FILE>) {
    my ($key, $value) = split;
    $hash{$key} = $value;
}
close FILE;

my %changes = ( 'str1' => 'val1', 'str2' => 'val2' );

# change data in hash (or add new elements)
$hash{$_} = $changes{$_} for keys %changes

# overwrite old file with modified hash
open FILE, ">$file" or die $!;
print FILE "$_\t$hash{$_}\n" for keys %hash;
close FILE;

    [download]

    cheers

    tachyon

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

[reply]
[d/l]
      Hi!

      Nice code - especially:

      $hash{$_} = $changes{$_} for keys %changes

      Greetz, Tom.
[reply]
[d/l]
        If assigning changes in a hash I prefer to do the following 
        @hash{keys %changes} = values %changes;

        [download]
        I believe it does the same thing (unless I'm missing something) as values() returns the hash values in the same order as keys().
        TIMTOWTDI

        broquaint

[reply]
[d/l]
Re: searching a file
by PrakashK (Pilgrim) on Mar 19, 2002 at 17:04 UTC
    I am assuming that you want to search using a strn to get the corresponding valuen.

    In that case, a hash is an ideal solution (assuming your file is not too big as to cause memory problems for you). once you have the hash populated, getting a value is as easy as: 

      my $value_n = $hash{$str_n};

    [download]

    /prakash

[reply]
[d/l]
Re: searching a file
by strat (Canon) on Mar 19, 2002 at 17:07 UTC 
    my @list = qw(value1 value2);
unless (open (FILE, $filename)){
  die("Error in opening $filename: $!\n");
} # if

else {
  while (<FILE>){
    chomp($_);
    # split up to maximal "two columns" by whitespaces
    my ($str, $val) = split(/\s+/, $_, 2);

    # search @list for $val
    if ( grep { $val eq $_ } @list ){
      print "$val found\n";
    }
  } # while
  close (FILE);
} # else

    [download]
    If you don't need $str in your code, write something like this: 
      ...
  while (<FILE>){
    chomp($_);
    # split up to maximal "two columns" by whitespaces
    my ($val) = ( split(/\s+/, $_, 2) )[1];

    # search @list for $val
    if ( grep { $val eq $_ } @list ){
      print "$val found\n";
    }
  } # while

} # else

    [download]

    I haven't tested these codes.

    Best regards,
    perl -le "s==*F=e=>y~\*martinF~stronat~=>s~[^\w]~~g=>chop,print"

[reply]
[d/l]
[select]

      If you find yourself doing a grep inside a loop you essentially have an inefficient solution as you have one loop inside another (n*n) Paying the one off price to genreate a hash lookup table (or better still storing your data in this form) makes for far greater efficiency. An alternative in some instances is to build and compile a matching regex (outside the loop!) and use this in place of the grep/inner loop.

      cheers

      tachyon

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

[reply]

Back to Seekers of Perl Wisdom