Reading CGI.pm file upload spools with Archive::Zip

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

Greetings monks! First, I'll start off with the code, so you can tell me if I'm even going in the right direction. Then I'll tell you my problem.

#!/usr/bin/perl -w

use strict;
use CGI;
use Archive::Zip;

##Initialize CGI
my $cgi = new CGI;
print $cgi->header();

##Ohnoes start over
unless( $cgi->param('step') ) {

    starthtml();
print <<HTML;
    <font face="Verdana, Arial, Helvetica, sans-serif">Please enter th
+e number of files you need to upload:
HTML

##FormStuff
print $cgi->start_form;
print $cgi->textfield('nfiles');
print $cgi->hidden('step','2');
print $cgi->submit;
print $cgi->end_form;

##Finish off Teh HTML
print <<HTML;
    </p></div>
</body>
</html>
HTML

}

if ( $cgi->param('step') == '2' ) {
    ##Clear the step num
    $cgi->delete('step');
    starthtml();

    ##Start Form and loop through to create number of upload fields we
+ need
    print $cgi->start_form;
    for(my $i = 1; $i <= $cgi->param('nfiles');$i++) {
        print $cgi->filefield(-name=>"file$i"),"<br />\n";
    }
    ##Next step, also send number of files again
    print $cgi->hidden('step','3');
    print $cgi->hidden('nfiles',$cgi->param('nfiles'));
    print $cgi->submit();
    print $cgi->end_form();

} elsif ( $cgi->param('step') == '3' ) {

    ##Here is teh problem##
    my @files;
    for (my $i = 1; $i <= $cgi->param('nfiles'); $i++) {
        my $filename = $cgi->param("file$i");
        my $upload_filehandle = $cgi->upload("file$i"); ## Probably no
+t the best way to do that but ohwell
        ###
        ### Archive::Zip Processing here
        ###
}


sub starthtml {
print <<START;
<html>
<head>
        <title>Update Tool</title>
</head>
<body bgcolor="#006633">
        <div align="center">
        <p><font size="+2" face="Verdana, Arial, Helvetica, sans-serif
+">Update Tool</font><br />
        <br />
        <br />
START
}
[download]

I read over the CGI.pm docs and calling the upload() method will return the filehandle to the temporary file that CGI.pm created. But looking over Archive::Zip, it looks as if it wants the name of the zip in the new() constructor. Are there any CGI.pm calls I'm missing that I can get the name of the file? Or will I have to write to disk ala:

open (OUTFILE,">>/tmp/file");
binmode(OUTFILE);
while ($bytesread=read($filename,$buffer,1024)) {
print OUTFILE $buffer;
}
[download]

then read the file with Archive::Zip, process what needs done, then unlink?

Comment on Reading CGI.pm file upload spools with Archive::Zip Select or Download Code

Replies are listed 'Best First'.
Re: Reading CGI.pm file upload spools with Archive::Zip by ikegami (Patriarch) on Sep 01, 2004 at 19:43 UTC
`$zip = Archive::Zip->new(...);` is the same as `$zip = Archive::Zip->new();` `my $status = $zip->read(...);` `$zip = undef unless ($status == AZ_OK);` so use `$zip = Archive::Zip->new();` `my $status = $zip->readFromFileHandle($upload_filehandle);` `$zip = undef unless ($status == AZ_OK);` Untested. Update: You know what, it looks like read() and (therefore) new() accept a filehandle as an alternative to a file name. So, `$zip = Archive::Zip->new($upload_filehandle);` should work fine.	[reply] [d/l] [select]
Re^2: Reading CGI.pm file upload spools with Archive::Zip by OverlordQ (Hermit) on Sep 01, 2004 at 20:31 UTC
Hmmm I tried your first code, but i got this: print "On file: $i\n<br>"; my $filename = $cgi->param("file1") or print "Cannot do that: $!\n"; print "Uploading . . ."; my $upload_filehandle = $cgi->upload("file$i") or print "cant upload: +$!\n"; print "done<br>\n"; print "Making zip . . ."; my $zip = Archive::Zip->new() or print "Canot make zippah: $!\n"; print "done<br>\n"; print "Reading Zip . . . "; my $status = $zip->readFromFilehandle($upload_filehandle); print "done<br>\n"; print "Mah Statuhs: $status\n"; $zip = undef unless ($status eq 'AZ_OK'); if(defined($zip)) { print $zip->members(); } else { print "Ooops something happened: $!\n"; } [download] But this is what I get for output: Mungeing! On file: 1 Uploading . . .done Making zip . . .done Reading Zip . . . Dunno what's up there :-(	[reply] [d/l]
Re^3: Reading CGI.pm file upload spools with Archive::Zip by siroskey (Initiate) on Sep 16, 2004 at 01:47 UTC
I really do not have a explination, and if someone has one I would like to know. You have to make a reference of the dereferenced file handler object. In your case you read line should look like this. my $status = $zip->readFromFilehandle(\$$upload_filehandle); Here a segement of code that I used to get this working... my $fh = $c->upload('zip'); my $zip = Archive::Zip->new(); $zip->read(\$$fh); my @members = $zip->members(); foreach my $member (@members){ my $filename = $member->fileName(); print "$filename\n"; }	[reply]


Perl: the Markov chain saw
	PerlMonks