1 | #!/usr/bin/perl -w
|
---|
2 |
|
---|
3 |
|
---|
4 | ###########################################################################
|
---|
5 | #
|
---|
6 | # pdftohtml.pl -- convert documents to HTML ot TEXT format
|
---|
7 | #
|
---|
8 | # A component of the Greenstone digital library software
|
---|
9 | # from the New Zealand Digital Library Project at the
|
---|
10 | # University of Waikato, New Zealand.
|
---|
11 | #
|
---|
12 | # Copyright (C) 1999 New Zealand Digital Library Project
|
---|
13 | #
|
---|
14 | # This program is free software; you can redistribute it and/or modify
|
---|
15 | # it under the terms of the GNU General Public License as published by
|
---|
16 | # the Free Software Foundation; either version 2 of the License, or
|
---|
17 | # (at your option) any later version.
|
---|
18 | #
|
---|
19 | # This program is distributed in the hope that it will be useful,
|
---|
20 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
21 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
22 | # GNU General Public License for more details.
|
---|
23 | #
|
---|
24 | # You should have received a copy of the GNU General Public License
|
---|
25 | # along with this program; if not, write to the Free Software
|
---|
26 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
27 | #
|
---|
28 | ###########################################################################
|
---|
29 |
|
---|
30 | # pdftohtml.pl is a wrapper for running pdftohtml utility which converts
|
---|
31 | # PDF documents to HTML, and converts images to PNG format for display in
|
---|
32 | # the HTML pages generated
|
---|
33 |
|
---|
34 | BEGIN {
|
---|
35 | die "GSDLHOME not set\n" unless defined $ENV{'GSDLHOME'};
|
---|
36 | unshift (@INC, "$ENV{'GSDLHOME'}/perllib");
|
---|
37 | }
|
---|
38 |
|
---|
39 | use parsargv;
|
---|
40 | use util;
|
---|
41 | use Cwd;
|
---|
42 | use File::Basename;
|
---|
43 |
|
---|
44 | sub print_usage {
|
---|
45 | print STDERR
|
---|
46 | ("pdftohtml version 0.22\n",
|
---|
47 | "Usage: pdftohtml [options] <PDF-file> <html-file>\n",
|
---|
48 | " -f <int> : first page to convert\n",
|
---|
49 | " -l <int> : last page to convert\n",
|
---|
50 | " -d <dir> : target directory (default: basename of pdf-file)\n",
|
---|
51 | " -o <file> : name of output file; - means stdout (default index.html)\n",
|
---|
52 | " -q : don't print any messages or errors\n",
|
---|
53 | " -h : print this usage information\n",
|
---|
54 | " -p : exchange .pdf links by .html\n",
|
---|
55 | " -c : generate complex HTML document\n",
|
---|
56 | " -F : don't use frames in HTML document\n",
|
---|
57 | " -i : ignore images\n",
|
---|
58 | " -e <string> : set extension for images (in the Html-file) (default png)\n"
|
---|
59 | );
|
---|
60 | exit (1);
|
---|
61 | }
|
---|
62 |
|
---|
63 | sub main {
|
---|
64 | my (@ARGV) = @_;
|
---|
65 | my ($first,$last,$target_dir,$out_file,$img_ext,
|
---|
66 | $optq,$opth,$optp,$optc,$optF,$opti);
|
---|
67 |
|
---|
68 | # read command-line arguments so that
|
---|
69 | # you can change the command in this script
|
---|
70 | if (!parsargv::parse(\@ARGV,
|
---|
71 | 'f/\d+/1', \$first,
|
---|
72 | 'l/\d+/1', \$last,
|
---|
73 | 'd/[\S]*/', \$target_dir,
|
---|
74 | 'o/[\S]*/', \$out_file,
|
---|
75 | 'e/[\S]*/', \$img_ext,
|
---|
76 | 'q', \$optq,
|
---|
77 | 'h', \$opth,
|
---|
78 | 'p', \$optp,
|
---|
79 | 'c', \$optc,
|
---|
80 | 'F', \$optF,
|
---|
81 | 'i', \$opti
|
---|
82 | ))
|
---|
83 | {
|
---|
84 | print_usage();
|
---|
85 | }
|
---|
86 |
|
---|
87 | # Make sure the input file exists and can be opened for reading
|
---|
88 | if (scalar(@ARGV!=2)) {
|
---|
89 | print_usage();
|
---|
90 | }
|
---|
91 |
|
---|
92 | my $input_filename = $ARGV[0];
|
---|
93 | my $output_filestem = $ARGV[1];
|
---|
94 | $output_filestem =~ s/\.html$//; # pdftohtml adds this suffix
|
---|
95 |
|
---|
96 | my @dir = split (/(\/|\\)/, $input_filename);
|
---|
97 | pop(@dir);
|
---|
98 | my $dir = join ("", @dir);
|
---|
99 |
|
---|
100 | if (!-r $input_filename) {
|
---|
101 | print STDERR "Error: unable to open $input_filename for reading\n";
|
---|
102 | exit(1);
|
---|
103 | }
|
---|
104 |
|
---|
105 | # Heuristical code added by John McPherson to attempt to reject
|
---|
106 | # PDF's with no text in them.... based entirely on observation. We
|
---|
107 | # should really read the PDF specifications someday...
|
---|
108 | open (PDFIN, $input_filename) ||
|
---|
109 | die "Error: unable to open $input_filename for reading\n";
|
---|
110 |
|
---|
111 | my $found_text_object=0;
|
---|
112 | my $num_objects=0;
|
---|
113 | my $non_text_objects=0;
|
---|
114 | my $unenc_stream_objects=0;
|
---|
115 | my $line;
|
---|
116 | while (!$found_text_object && ($_=<PDFIN>)) {
|
---|
117 | s/\r/\n/g;
|
---|
118 | if (/^\d+ \d+ obj/ms) {
|
---|
119 | # start of new object
|
---|
120 | my $object="";
|
---|
121 | $num_objects++;
|
---|
122 | while (! eof && ! /(>>\s*)?endobj/) {
|
---|
123 | $object.=$_;
|
---|
124 | $_=<PDFIN>;
|
---|
125 | }
|
---|
126 | if (!defined $_) {$_="";} # we've hit end of file in a funny place.
|
---|
127 | # we've got to the end of the current PDF object.
|
---|
128 | $object.=$_;
|
---|
129 |
|
---|
130 | # remove newline chars, to help our pattern matching for whitespace
|
---|
131 | $object =~ s/\n/ /gs;
|
---|
132 |
|
---|
133 | #determine object type...
|
---|
134 | $_=$object;
|
---|
135 |
|
---|
136 | # for PDFWriter , and pdflatex and distill. Eg:
|
---|
137 | # "12 0 obj << /Length 13 0 R /Filter /LZWDecode >> stream ..."
|
---|
138 | # Ie this looks like compressed text....
|
---|
139 | if (/\d+\s+\d+\s+obj\s+<<\s+\/Length\s+\d+\s+\d+\s*.\s*\/Filter/) {
|
---|
140 | $found_text_object=1;
|
---|
141 | }
|
---|
142 | # For pdflatex or ps2pdf from dvi->ps:
|
---|
143 | # if we are setting a font, then following object is probably text
|
---|
144 | # Eg "obj << /Font" or "obj << /ProcSet [...] /Font"
|
---|
145 | elsif (/obj\s*<<\s*(\/ProcSet \[.+?\]\s*)?\/Font /s) {
|
---|
146 | $found_text_object=1;
|
---|
147 | }
|
---|
148 | # Unencoded streams. Eg
|
---|
149 | # "<< /Length 45 0 R >> stream BT /R43 8.96638 Tf 1..."
|
---|
150 | elsif (/<<\s+\/Length\s+\d+\s+\d+\s+R\s+>>\s+stream\s+(q\s)?BT\s/s)
|
---|
151 | {
|
---|
152 | $unenc_stream_objects++;
|
---|
153 | }
|
---|
154 | # (some) non-text objects
|
---|
155 | elsif (/<<.*\/(Type).*>>/s) {
|
---|
156 | $non_text_objects++;
|
---|
157 | }
|
---|
158 |
|
---|
159 | } else { # not in an object...
|
---|
160 | # header? footer?
|
---|
161 | # print $_;
|
---|
162 | }
|
---|
163 | if ($found_text_object) {close PDFIN;}
|
---|
164 |
|
---|
165 | } # end of while
|
---|
166 | close PDFIN;
|
---|
167 |
|
---|
168 | # decide whether to accept or reject...
|
---|
169 | # some of these numbers are completely arbitrary based on a few .pdfs.
|
---|
170 | if ( ($found_text_object > 0) ||
|
---|
171 | ($num_objects<=1500 && $unenc_stream_objects > 5)
|
---|
172 | )
|
---|
173 | {
|
---|
174 | # accept this .pdf. Currently do nothing except fall through...
|
---|
175 | } else {
|
---|
176 | # reject this .pdf.
|
---|
177 | print STDERR "pdftohtml.pl: $input_filename appears to have no ";
|
---|
178 | print STDERR "textual data. Aborting.\n";
|
---|
179 | # print STDERR "num: $unenc_stream_objects and $non_text_objects from $num_objects\n";
|
---|
180 | exit(1);
|
---|
181 | }
|
---|
182 |
|
---|
183 | # formulate the command
|
---|
184 | my $cmd = &util::filename_cat($ENV{'GSDLHOME'}, "bin", $ENV{'GSDLOS'}, "pdftohtml.bin");
|
---|
185 |
|
---|
186 | # don't include path on windows (to avoid having to play about
|
---|
187 | # with quoting when GSDLHOME might contain spaces) but assume
|
---|
188 | # that the PATH is set up correctly - note also that on windows
|
---|
189 | # we use pdftohtml.exe not pdftohtml.bin
|
---|
190 | $cmd = "pdftohtml" if ($ENV{'GSDLOS'} =~ /^windows$/);
|
---|
191 |
|
---|
192 | if ($timeout) {$cmd = "ulimit -t $timeout; $cmd";}
|
---|
193 | $cmd .= " -noframes \"$input_filename\" \"$output_filestem.html\"";
|
---|
194 | $cmd .= " > \"$output_filestem.out\"";
|
---|
195 |
|
---|
196 | # attempting to redirect STDERR on windows 95/98 is a bad idea
|
---|
197 | $cmd .= " 2> \"$output_filestem.err\""
|
---|
198 | if $ENV{'GSDLOS'} !~ /^windows$/i;
|
---|
199 |
|
---|
200 | # system() returns -1 if it can't run, otherwise it's $cmds ret val.
|
---|
201 | if (system($cmd)!=0) {
|
---|
202 | print STDERR "Error executing $cmd: $!\n";
|
---|
203 | &util::rm("$output_filestem.text") if (-e "$output_filestem.text");
|
---|
204 | &util::rm("$output_filestem.err") if (-e "$output_filestem.err");
|
---|
205 | return 0;
|
---|
206 | }
|
---|
207 |
|
---|
208 | # Need to convert images from PPM format to PNG format
|
---|
209 | my @images;
|
---|
210 |
|
---|
211 |
|
---|
212 | open (IMAGES, "images.log");
|
---|
213 | while (<IMAGES>) {
|
---|
214 | push (@images, $_);
|
---|
215 | }
|
---|
216 | close IMAGES;
|
---|
217 |
|
---|
218 | for $image (@images) {
|
---|
219 | chomp($image);
|
---|
220 | my $cmd = "";
|
---|
221 | if ($ENV{'GSDLOS'} =~ /^windows/i) {
|
---|
222 | $cmd = "pnmtopng $image";
|
---|
223 | if (system($cmd)!=0) {
|
---|
224 | print STDERR "Error executing $cmd\n";
|
---|
225 | return 0; # not sure about whether to leave this one in or take it out
|
---|
226 | }
|
---|
227 | } else {
|
---|
228 | my @nameparts = split(/\./, $image);
|
---|
229 | my $image_base = shift(@nameparts);
|
---|
230 |
|
---|
231 | $cmd = "pnmtopng $image > $image_base.png 2>/dev/null";
|
---|
232 | if (system($cmd)!=0) {
|
---|
233 | $cmd = "convert $image $image_base.png 2>/dev/null";
|
---|
234 | if (system($cmd)!=0) {
|
---|
235 | print STDERR "Cannot convert $image into PNG format (tried `pnmtopng' and `convert')...\n";
|
---|
236 | return 0; # not sure about whether to leave this one in or take it out
|
---|
237 | }
|
---|
238 | }
|
---|
239 | }
|
---|
240 | &util::rm($image);
|
---|
241 | }
|
---|
242 |
|
---|
243 | return 1;
|
---|
244 | }
|
---|
245 |
|
---|
246 | &main(@ARGV);
|
---|