Separate color from black-and-white pages for printing

Check out PDFsam: a free, open source utility for (you guessed it) splitting and merging PDFs. You may be able to script it to get it to do what you want.

wfaulk, how did you get on with this?

we have had this feature working for a long time.

its something that should be done in the print driver itself... changing automatically between mono & colour on the fly.

worth checking what options your manufacturer give you, different drivers perhaps.

As someone posted above, some of the print accounting sofwares do this.. and work out a charge but they are just using the print driver as provided.

You can use this utility from Microsoft to split PDF on two documents containing only colour and only grayscale (black and white) pages

You can create two printers on the server. One set to print Black only and the other colour. You can even add secrity to the colour printer so only approved users can print to thecolour device. One printer that appears as two to the users. Have also used this to set a certain page size or to force Duplex printing.

Have used this often in the past and it works well and costs nothing when running Windows Server 2003

You will need to find a 3rd party print server manager such as papercut which you have mentioned.

Some other solutions could inculde the following server applications, although I don't know if they specificity include what your looking for:

• Print Manager Plus
• Printer Peer
• Netop PrintLimit Pro
• Equitrack

Okay, it looks like I'm going to have to implement this myself. So far I have a crude perl script (that uses a lot of system calls to ghostscript and netpbm) that determines the color- or black-and-white-ness of each page of a PostScript (or PDF) document:

#!/usr/bin/perl

use File::Temp qw/ tempfile tempdir /;

my $pages=0;

my $tempdir = tempdir(CLEANUP=>1);
my ($psfh, $psfn) = tempfile(SUFFIX=>'.ps', DIR=>$tempdir);

while (<>) {
        # Write to temporary file
        print $psfh $_;
        # Count Pages
        $pages++ if ( /\%\%Page:/ );
}

if ( $pages == 0 ) {
        # Not DSC-conforming; count manually; ugh
}

my ($pdffh, $pdffn) = tempfile(SUFFIX=>'.pdf', DIR=>$tempdir, OPEN=>0);

# Convert to PDF
#print("ps2pdf ... ");
system("ps2pdf $psfn $pdffn");
#print("done\n");

# Count PDF pages
my $pdfpages = `pdfinfo $pdffn | egrep -e '^Pages'`;
$pdfpages =~ m/(\d+)/;
$pdfpages = $1;
#print("$pdfpages pages\n");

# Note: 1-based array!!
my @colorhist;
for ( my $i = 1; $i<=$pdfpages; $i++) {
        print("Page $i\n");
        my @colors = `gs -q -dFirstPage=$i -dLastPage=$i -dNOPAUSE -dBATCH -sDEVICE=ppmraw -sOutputFile=- $pdffn | ppmhist -noheader`;
        foreach (@colors) {
                m/^\s*(\d+)\s+(\d+)\s+(\d+)\s+\d+\s+(\d+)\s*$/;
                my $color = (($1<<16) + ($2<<8) + $3);
                my $instances = $4;
                $colorhist[$i]{$color} = $instances;
                #print "$color $instances\n";
        }
}

my @colorbw;

for ( my $i = 1; $i<=$pdfpages; $i++ ) {
        my %pagehist = %{$colorhist[$i]};
        #print("page $i:\n");
        my $numcolors = 0;
        my $totinstances = 0;
        foreach $key (keys(%pagehist)) {
                #print "$key\n";
                #print "\t$key $pagehist{$key}\n";
                $numcolors++;
                $totinstances += $pagehist{$key};
                my $r = $key >> 16;
                my $g = ( $key - ($r << 16) ) >> 8;
                my $b = $key - ($r << 16) - ($g << 8);
                if ( ( abs($r - $g) > 32 ) || ( abs($r - $b) > 32 ) || ( abs($g - $b) > 32 ) ) {
                        $colorbw[$i] = 2;
                }
        }
        if ( $numcolors <= 2 ) {
                $colorbw[$i] = 1;
        }
        print "$i\n" if ($colorbw[$i] > 1);
}

It has one optimization already, in that if a page contains only two colors, it marks it as B&W, even if one of the colors is nowhere near black, white, or grey. I also need to implement a filter to discard pages that have something like one pixel of another color, and also one to only mark the ones that are excessively color-y, to filter out things in color that happen on every page. (I figure that if someone is going to use this, they're intending to print at least some of the pages in black-and-white.)

After that, I have to implement printing each of those sets through the appropriate queue.

Hopefully I can figure out how to pull the Ghostscript and netpbm stuff back into the script, even if I have to change languages.

Is there a piece of software than can automatically find the color pages in a print job, send those pages to the color printer, and then print the rest of the job to a B&W printer?

Sure, wit this commercial solution. This video shows how to do it on a single printer to force black & white when there is only little color: https://www.youtube.com/watch?v=USogiHUqjTc

This video shows how to send color print jobs automatically to a different printer: https://www.youtube.com/watch?v=Q7546WqTFq8

Now it is easy to combine both for your request.

• Create a profile with two channels (one for the B&W printer, one for the color printer)
• Configure per channel the [Page selection]. For the channel for b&w configure it to only select the b&w pages (tip: even save more money and select pages that contain too little color and force it to print in b&w ;-))
• For the color printer configure it to only select color pages.

done :-) We heavily use Print&Share here, as we have profiles to force users to print in b&w. Trust me there is a big ROI with nifty piece of software.

On our printer we choose "Auto" instead of B & W or color. Auto charges us for only for the color part & then charges us differently for the B & W. We don't have to do anything else - just choose "Auto"