in reply to Re^5: Renaming an image file
in thread Renaming an image file

At this point I already read the file in, is this looks ok based in what you have here? 
  for(my $i=0;$i<4;$i++) {
  
     my $file = $q->param('pic_'.$i);
     my @full = split(/(\\)|(\/)/,$file);
     my $filename = pop(@full);
        $filename =~ s/\n//;
        $filename =~ s/\r//;
        $filename = $username_pic."_".$filename;
        #$filename=~s/^([^_]+_)(.*)(\.\w{3})\z/$1.md5_hex($2).$3/eg;
        my $md5 = md5_hex( $filename );
        $filename =s[_(\d+).jpg$][_$md5.jpg];
}

[download]

Replies are listed 'Best First'.
Re^7: Renaming an image file
by BrowserUk (Patriarch) on Nov 28, 2010 at 18:09 UTC

    Sorry, I don't know much about cgi, but it still looks to me that you are md5'ing the filename and not the contents of the file.

[reply]
      OK I will add another string of numbers right after the MD5 if this file already exists: 
              if ( -e "$path/$filename" ) { 
           my $unique = unique();
           $filename=~s/^([^_]+_)(.*)(\.\w{3})\z/$1.$2.$unique.$3/eg;
           rename( "$path/$filename", "$path/$filename");
           unlink($filename);
         }
}
sub unique {
my ($sec, $min, $hour, $mday, $mon, $year, $wday, $yday, $isdst) = loc
+altime(time);
$mon++;
$year = $year + 1900;
my $timestamp = sprintf "%4d%02d%02d%02d%02d%02d", $year,$mon,$mday,$h
+our,$min,$sec;
my $rand = (rand 10000);
$rand = int($rand);
my $unique = "$timestamp$rand";
}

      [download]
[reply]
[d/l]

        Well, okay. But all you've done is move the goal posts a little.

        Webservers are inherently multitasking, so adding all those digits of date and time is pretty useless as there could be many uploads completing within each given second.

        And depending upon the quality of your rand() function, the rand(10000) bit does almost nothing:

        perl -E"
$t=time; 
++$h{ rand(10000) } while time()==$t; 
my $n=0; 
$_ > 1 and $n+=$_ for values %h; 
say $n, ' duplicates in 1 second';
"
13793 duplicates in 1 second

perl -E"$t=time; ++$h{ rand(10000) } while time()==$t; my $n=0; $_ > 1
+ and $n+=$_ for values %h; say $n, ' duplicates in 1 second';"
400896 duplicates in 1 second

>perl -E"$t=time; ++$h{ rand(10000) } while time()==$t; my $n=0; $_ > 
+1 and $n+=$_ for values %h; say $n, ' duplicates in 1 second';"
504998 duplicates in 1 second

        [download]

        Even with what is probably the best PRNG available, you can still generate duplicates:

        c:\test>perl -MMath::Random::MT=rand -E"$t=time; 1 while time==$t; ++$
+h{ rand(10000) } while time()==$t+1; my $n=0; $_ > 1 and $n+=$_ for v
+alues %h; say $n, ' duplicates in 1 second';"
4 duplicates in 1 second

c:\test>perl -MMath::Random::MT=rand -E"$t=time; 1 while time==$t; ++$
+h{ rand(10000) } while time()==$t+1; my $n=0; $_ > 1 and $n+=$_ for v
+alues %h; say $n, ' duplicates in 1 second';"
4 duplicates in 1 second

>perl -MMath::Random::MT=rand 
-E"
$t=time; 
1 while time==$t; 
++$h{ rand(10000) } while time()==$t+1; 
my $n=0; 
$_ > 1 and $n+=$_ for values %h; 
say $n, ' duplicates in 1 second';
"
6 duplicates in 1 second

        [download]

        It takes only a few milliseconds to read and md5 the contents of a medium sized image file.

        Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
        "Science is about questioning the status quo. Questioning authority".
        In the absence of evidence, opinion is indistinguishable from prejudice.
[reply]
[d/l]
[select]
Re^7: Renaming an image file
by Anonymous Monk on Nov 28, 2010 at 21:23 UTC
    This is wrong: $filename =s[_(\d+).jpg$][_$md5.jpg];

    You mean =~, not =. That fixes the error mentioning $_ (s/// works on $_ by default).

    Also, the Perlish way to loop over an interval like that is for my $i (0..3) {.

[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom