| 1 |
#!/usr/bin/perl -w |
| 2 |
|
| 3 |
package main; |
| 4 |
|
| 5 |
use warnings; |
| 6 |
use strict; |
| 7 |
use CAM::PDF; |
| 8 |
use Getopt::Long; |
| 9 |
use Pod::Usage; |
| 10 |
|
| 11 |
our $VERSION = '1.60'; |
| 12 |
|
| 13 |
my %opts = ( |
| 14 |
template => '/usr/share/libcam-pdf-perl/crunchjpg_tmpl.pdf', |
| 15 |
|
| 16 |
verbose => 0, |
| 17 |
help => 0, |
| 18 |
version => 0, |
| 19 |
skip => {}, |
| 20 |
|
| 21 |
# Temporary values: |
| 22 |
skipval => [], |
| 23 |
); |
| 24 |
|
| 25 |
Getopt::Long::Configure('bundling'); |
| 26 |
GetOptions('S|skip=s' => \@{$opts{skipval}}, |
| 27 |
'v|verbose' => \$opts{verbose}, |
| 28 |
'h|help' => \$opts{help}, |
| 29 |
'V|version' => \$opts{version}, |
| 30 |
) or pod2usage(1); |
| 31 |
if ($opts{help}) |
| 32 |
{ |
| 33 |
pod2usage(-exitstatus => 0, -verbose => 2); |
| 34 |
} |
| 35 |
if ($opts{version}) |
| 36 |
{ |
| 37 |
print "CAM::PDF v$CAM::PDF::VERSION\n"; |
| 38 |
exit 0; |
| 39 |
} |
| 40 |
|
| 41 |
foreach my $flag (qw( skip )) |
| 42 |
{ |
| 43 |
foreach my $val (@{$opts{$flag.'val'}}) |
| 44 |
{ |
| 45 |
foreach my $key (split /\D+/xms, $val) |
| 46 |
{ |
| 47 |
$opts{$flag}->{$key} = 1; |
| 48 |
} |
| 49 |
} |
| 50 |
} |
| 51 |
if (!-f $opts{template}) |
| 52 |
{ |
| 53 |
die "Cannot find the template pdf called $opts{template}\n"; |
| 54 |
} |
| 55 |
|
| 56 |
if (@ARGV < 2) |
| 57 |
{ |
| 58 |
pod2usage(1); |
| 59 |
} |
| 60 |
|
| 61 |
my $infile = shift; |
| 62 |
my $outdir = shift; |
| 63 |
|
| 64 |
my $doc = CAM::PDF->new($infile) || die "$CAM::PDF::errstr\n"; |
| 65 |
|
| 66 |
my $pages = $doc->numPages(); |
| 67 |
my $nimages = 0; |
| 68 |
my $rimages = 0; |
| 69 |
my %doneobjs; |
| 70 |
|
| 71 |
for my $p (1..$pages) |
| 72 |
{ |
| 73 |
my $c = $doc->getPageContent($p); |
| 74 |
my @parts = split /(\/[\w]+\s*Do)\b/xms, $c; |
| 75 |
foreach my $part (@parts) |
| 76 |
{ |
| 77 |
if ($part =~ m/\A(\/[\w]+)\s*Do\z/xms) |
| 78 |
{ |
| 79 |
my $ref = $1; |
| 80 |
my $xobj = $doc->dereference($ref, $p); |
| 81 |
my $objnum = $xobj->{objnum}; |
| 82 |
my $im = $doc->getValue($xobj); |
| 83 |
my $w = $im->{Width} || $im->{W} || 0; |
| 84 |
if ($w) |
| 85 |
{ |
| 86 |
$w = $doc->getValue($w); |
| 87 |
} |
| 88 |
my $h = $im->{Height} || $im->{H} || 0; |
| 89 |
if ($h) |
| 90 |
{ |
| 91 |
$h = $doc->getValue($h); |
| 92 |
} |
| 93 |
|
| 94 |
next if (exists $doneobjs{$objnum}); |
| 95 |
|
| 96 |
$nimages++; |
| 97 |
_inform("Image $nimages page $p, $ref = object $objnum, (w,h)=($w,$h)", $opts{verbose}); |
| 98 |
|
| 99 |
if (exists $opts{skip}->{$objnum}) |
| 100 |
{ |
| 101 |
_inform("Skipping object $objnum", $opts{verbose}); |
| 102 |
next; |
| 103 |
} |
| 104 |
|
| 105 |
my $isjpg = _isjpg($im); |
| 106 |
|
| 107 |
if (!$isjpg) |
| 108 |
{ |
| 109 |
_inform('Not a jpeg', $opts{verbose}); |
| 110 |
} |
| 111 |
else |
| 112 |
{ |
| 113 |
my $oldsize = $doc->getValue($im->{Length}); |
| 114 |
if (!$oldsize) |
| 115 |
{ |
| 116 |
die "PDF error: Failed to get size of image\n"; |
| 117 |
} |
| 118 |
|
| 119 |
my $tmpl = CAM::PDF->new($opts{template}) || die "$CAM::PDF::errstr\n"; |
| 120 |
|
| 121 |
# Get a handle on the needed data bits from the template |
| 122 |
my $media_array = $tmpl->getValue($tmpl->getPage(1)->{MediaBox}); |
| 123 |
my $rawpage = $tmpl->getPageContent(1); |
| 124 |
|
| 125 |
$media_array->[2]->{value} = $w; |
| 126 |
$media_array->[3]->{value} = $h; |
| 127 |
my $page = $rawpage; |
| 128 |
$page =~ s/xxx/$w/igxms; |
| 129 |
$page =~ s/yyy/$h/igxms; |
| 130 |
$tmpl->setPageContent(1, $page); |
| 131 |
$tmpl->replaceObject(9, $doc, $objnum, 1); |
| 132 |
|
| 133 |
my $ofile = "/tmp/crunchjpg.$$"; |
| 134 |
$tmpl->cleanoutput($ofile); |
| 135 |
|
| 136 |
if (!-d $outdir) |
| 137 |
{ |
| 138 |
require File::Path; |
| 139 |
File::Path::mkpath($outdir); |
| 140 |
} |
| 141 |
`convert -quality 50 -density 72x72 -page ${w}x$h pdf:$ofile jpg:$outdir/$objnum.jpg`; ## no critic (Backtick) |
| 142 |
|
| 143 |
$doneobjs{$objnum} = 1; |
| 144 |
$rimages++; |
| 145 |
} |
| 146 |
} |
| 147 |
} |
| 148 |
} |
| 149 |
|
| 150 |
_inform("Extracted $rimages of $nimages images", $opts{verbose}); |
| 151 |
|
| 152 |
sub _isjpg |
| 153 |
{ |
| 154 |
my $im = shift; |
| 155 |
return if (!$im->{Filter}); |
| 156 |
|
| 157 |
my $f = $im->{Filter}; |
| 158 |
my @names = $f->{type} eq 'array' ? @{$f->{value}} : $f; |
| 159 |
for my $e (@names) |
| 160 |
{ |
| 161 |
my $name = $doc->getValue($e); |
| 162 |
if (ref $name) |
| 163 |
{ |
| 164 |
$name = $name->{value}; |
| 165 |
} |
| 166 |
#warn "Checking $name\n"; |
| 167 |
if ($name eq 'DCTDecode') |
| 168 |
{ |
| 169 |
return 1; |
| 170 |
} |
| 171 |
} |
| 172 |
return; |
| 173 |
} |
| 174 |
|
| 175 |
sub _inform |
| 176 |
{ |
| 177 |
my $str = shift; |
| 178 |
my $verbose = shift; |
| 179 |
|
| 180 |
if ($verbose) |
| 181 |
{ |
| 182 |
print STDERR $str, "\n"; |
| 183 |
} |
| 184 |
return; |
| 185 |
} |
| 186 |
|
| 187 |
|
| 188 |
__END__ |
| 189 |
|
| 190 |
=for stopwords extractjpgs.pl ImageMagick JPG |
| 191 |
|
| 192 |
=head1 NAME |
| 193 |
|
| 194 |
extractjpgs.pl - Save copies of all PDF JPG images to a directory |
| 195 |
|
| 196 |
=head1 SYNOPSIS |
| 197 |
|
| 198 |
extractjpgs.pl [options] infile.pdf outdirectory |
| 199 |
|
| 200 |
Options: |
| 201 |
-S --skip=imnum don't output the specified images (can be used mutliple times) |
| 202 |
-v --verbose print diagnostic messages |
| 203 |
-h --help verbose help message |
| 204 |
-V --version print CAM::PDF version |
| 205 |
|
| 206 |
C<imnum> is a comma-separated list of integers indicating the images |
| 207 |
in order that they appear in the PDF. Use F<listimages.pl> to retrieve |
| 208 |
the image numbers. |
| 209 |
|
| 210 |
=head1 DESCRIPTION |
| 211 |
|
| 212 |
Requires the ImageMagick B<convert> program to be available |
| 213 |
|
| 214 |
Searches the PDF for JPG images and saves them as individual files in the |
| 215 |
specified directory. The files are named C<E<lt>imnumE<gt>.jpg>. |
| 216 |
|
| 217 |
=head1 SEE ALSO |
| 218 |
|
| 219 |
CAM::PDF |
| 220 |
|
| 221 |
F<crunchjpgs.pl> |
| 222 |
|
| 223 |
F<listimages.pl> |
| 224 |
|
| 225 |
F<extractallimages.pl> |
| 226 |
|
| 227 |
F<uninlinepdfimages.pl> |
| 228 |
|
| 229 |
=head1 AUTHOR |
| 230 |
|
| 231 |
See L<CAM::PDF> |
| 232 |
|
| 233 |
=cut |