Sounds like a great place to use a hash instead of an array. It would be better if you had some code or some pseudo-code so I could be sure I understand the task.
| [reply] |
This is basically the code I'm talking about. Looping through a number of files, reading them (this routine is quick), and looping through all parameter names in each file, checking if they already exists in the global @names array and take action depending on the result. It is this check (find_element) that I want to speed up. Any ideas? If hash is a good alternative, could you please explain the difference compared to an array? New at perl so I'm not so familiar with all expressions... Thank you!
foreach $filename (@input_names) #Go through input files
{
($names_ref,$data_ref) = &read_file($filename);
@tmp_names = @$names_ref;
@tmp_data = @$data_ref;
foreach $name (@tmp_names) #Go through lines of current input
+file
{
$index = &find_element($name, @names); #Check if variable
+is already present in name array (time consuming!)
if ($index==-1) #Name not present, put both name and data
+in array
{
push(@names,$name);
push(@data,$tmp_data[$tmp_index]);
}
else #Name present, only replace data for the name
{
$data[$index] = $tmp_data[$tmp_index];
}
$tmp_index++;
}
}
| [reply]
[d/l] |
The main difference between a hash and an array is that an array is indexed by number while a hash is indexed by name. So where a lookup in an array is fast if you know the position of the value, a lookup in a hash is fast if you know the "name" of the value, that is, the (string) key the value is associated with.
| [reply] |
Yeah, you probably want something more like this, or something similar to it.
my %hash;
while(my $line = <>) {
my ($name, $value) = split m/something goes here/, $line;
die "error: value already defined!" if exists $hash{$name};
$hash{$name} = $value;
}
| [reply]
[d/l] |
| [reply]
[d/l]
[select] |
Thank you for your answer. And sorry for the lack of details in my description. The files I am reading are txt files with each row containing a parameter name and its value (separated by spaces). However, I don't have any difficulties reading the files, creating an array of names and an array of the corresponding values.
What does "do" mean in this context? As I'm sure you understand, I'm quite a beginner at Perl :)
| [reply] |
use warnings;
use strict;
my %vars;
while (<>) {
local @_ = split;
$vars{$_[0]} = $_[1];
}
which takes each line of each file given on the command line and updates the %vars hash, keyed on the first field & values of the 2nd field - as per postings here and elsewhere on this thread.
A user level that continues to overstate my experience :-))
| [reply]
[d/l]
[select] |
After creating the hash, using the parameter values as the key (see above), you can search for the parameter with:
if (exists $vars{'parameter'}) {
# It is there!
}
else {
# It is not there
}
This avoids an iterative search through the array. | [reply]
[d/l] |