The perl module Spreadsheet::ParseExcel called, works too slowly !

Hi evrybody,

Tanks for your help
my problems are below :
1-/ I'm calling Spreadsheet::ParseExcel v0.59 module in my script. It seems to work very slowly and i dont know why.
2-/ How to change the field separator from "," to ";" for Spreadsheet::ParseExcel v0.59 module.

I'm joining the code and input file.
Here is the command : livraison_ENV.pl -a $INR_PRM -r INT -f S2ID2.BULL.Demandes_livraisons_TMA.xls :
-a : input file directory
-r : INT (Integration environment) ==> no change
-f : Name of input file

#!/usr/bin/perl

### Mise en place de library ###
#use lib '/opt/hrdev/bao/dev_bao/e-sda/exploit/sources';
use lib '/opt/hrdev/bao/utils/lib';
use warnings;
use strict;
use Spreadsheet::ParseExcel;
use Spreadsheet::ParseExcel::Utility qw(xls2csv);
use Getopt::Std;
#use Getopt::Simple;
use File::Copy;
use MIME::Lite;
use POSIX qw(strftime);

### Mise a jour variables ###
my @ENTITES;
my $oExcel;
my $ORDRE;
my $ENTITES;
my $LIGNE;
my $oBook;
my $CODES;
my $FEUILLE;
my $oWkS;
our @CODES;
our @LIGNE;


our @champs;
our @cles;
my $dir_prm;
my $param;
my $rapport;
my $par;
my $lignes;
my $cod;
our $fich_xls;
our $param_rep;
our $param_ent;
our $param_man;
our $status;
our %opts=();
our @LECTURE;
our @CODE;
our @CODE_M;
our $i;

@ARGV or die "Usage $0 FICHIER\n";

### Mise en place des parametres ###

getopts( "a:f:d:r:e:m:c:", \%opts );

        if (defined $opts{a}) { $dir_prm = $opts{a} } else { $dir_prm=""; }
        if (defined $opts{f}) { $fich_xls = $opts{f} } else { $fich_xls=""; }
        if (defined $opts{d}) { $param_rep = $opts{d} } else { $param_rep=""; }
        if (defined $opts{r}) { $rapport = $opts{r} } else { $rapport="no"; }
        if (defined $opts{e}) { $param_ent = $opts{e} } else { $param_ent=""; }
        if (defined $opts{m}) { $param_man = $opts{m} } else { $param_man=""; }
        if (defined $opts{c}) { $status = $opts{c} } else { $status=""; }

     open my $out_fh, '>>', $dir_prm."/param_liv_rep_OUT.txt" or die "impossible de creer 'param_liv_rep_OUT.txt' $!";

# ------------------------- FIN OPTIONS -------------------------------------------

#print {$out_fh} strftime("%Y%m%d%H%M\n", localtime);
my $temps = strftime("%Y%m%d%H%M\n", localtime);

# ------------------ SAISIE MANUELLE DES DONNEES ET CREATION DES CARTES PARAMETRES HRACCESS ------------------------

@cles = keys(%opts);
#-------------- CREATION DU RAPPORT DE FIN DE LIVRAISON -------------------------

if ( ( grep /r/,@cles ) && ( grep /f/,@cles ) ) {
        $rapport =~ tr [a-z] [A-Z];

        #==================== EXTRACTION DU FICHIER EXCEL =======================

        if (( $rapport =~ "INT" ) || ( $rapport =~ "INTEGRATION" )) {

                        ### Extraction du fichier excel ###

          chdir "$dir_prm/fic" or die "\n--- ERR : Impossible de se deplacer dans le repertoire fic ---\n";
          ( -f $fich_xls ) or die " \n--- ERR : Le fichier Excel n\'est pas present --- \n";
          print "\n --- Extraction des donnees du fichier excel ...\n";

          $oExcel = new Spreadsheet::ParseExcel;
          $oBook = $oExcel->Parse($fich_xls);

          foreach $oWkS (@{$oBook->{Worksheet}}) {
            $ORDRE ++;
            ($FEUILLE=$oWkS->{Name}) =~ s/ //g ;
            if ( grep /Entites/,$FEUILLE) { $ENTITES=(xls2csv( ${fich_xls}, "${ORDRE}-A3:W500"))};
#           if ( grep /Codes/,$FEUILLE) { $CODES=(xls2csv( ${fich_xls}, "${ORDRE}-A1:W500"))};
          }

          @ENTITES=split /\n/,$ENTITES;
#         @CODES=split /\n/,$CODES;

#         join(';', @ENTITES);

          for $LIGNE ( grep /prochaine.livraison/i,@ENTITES) {
            @LIGNE=split ',',$LIGNE;
            if (( grep /prochaine.livraison/i,$LIGNE[10] ) && ( ! $LIGNE[13] ) && ( ! $LIGNE[14] ) && ( ! $LIGNE[15] )) {
              print $LIGNE;
              print "\n";
            }
          }

          print "\nFin de la saisie.\n";
        }
}
S2ID2.BULL.Demandes-livraisons-T.xls
Seydina FallIT ingineerAsked:
Who is Participating?
 
FishMongerConnect With a Mentor Commented:
Here's a short test script that I ran which only parses the worksheet and outputs the results.  My test run only took a few seconds to complete.  Note that I reduced the number of rows to match what was actually being used in your posted example file.

#!/usr/bin/perl

use warnings;
use strict;
use Spreadsheet::ParseExcel;
use Spreadsheet::ParseExcel::Utility qw(xls2csv);
use Data::Dumper;

print scalar localtime, $/;

my $xls_file = 'S2ID2.BULL.Demandes-livraisons-T.xls';
my $xls2csv  = xls2csv($xls_file, '1-A3:W348');

print scalar localtime, $/;
print Dumper $xls2csv;

Open in new window

0
 
Seydina FallIT ingineerAuthor Commented:
Hi,
When I try to send mail after the Spreadsheet module execution I get that error  :
open |/usr/lib/sendmail -t -oi -oem: Ne peut allouer de la mémoire (can't allow enough memory) at ./livraison_ENV.pl line 277.

I think that the module takes a lot of memory.

Thanks for some feedback

Seydina
0
 
FishMongerCommented:
You should profile your script to see where it's spending its time.  With that info you can begin to refactor/optimize those sections.

Devel::NYTProf - Powerful fast feature-rich Perl source code profiler
0
Upgrade your Question Security!

Your question, your audience. Choose who sees your identity—and your question—with question security.

 
Seydina FallIT ingineerAuthor Commented:
Hello,

I installed the Devil-NYTProf module and execute it well.
Spreadsheet module consumes 99% of the execution time, actually.
Is there any tool which can refactor/optimze official modules ?
0
 
FishMongerCommented:
Spreadsheet module consumes 99% of the execution time
That's pretty vague.

You need to determine which line(s) of code is being executed which is utilizing that amount time and how much time it takes for each execution of that statement.  Is the line being executed fairly quickly but a large number of times?  Is the statement making a bunch of system calls?

Taking a second look at your code, I'd say the first step might be to simply clean it up.  It's very messy.

Declaring all of your vars at the top is almost always a bad design.  You should declare your vars in the smallest scope that they require and as close as possible to where they are needed/used.  You should only use global vars when absolutely necessary and I don't see any need for them in this script.  So, change all of your 'our' declarations to 'my'.

Your initialization of your getops params (i.e., the if defined statements) is very kludgy.

Your use of grep on a scalar is odd and inefficient.  Use index() instead of grep().
0
 
Seydina FallIT ingineerAuthor Commented:
You're right,

My code is now running well. I've used the refactor module to identify the lines which were abnormally long.

I pulled out the pool parsing lines and simplify the code by translating all our vars on my vars ...

But when I make the sum of the second line of the ouput file form refactoring I get the double of the real execution time of the code. Is it right ?

Anyway this helped me to resolve my issues.

Thanks
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.