Annotation of /trunk/spider/progspider

#!/usr/bin/perl -w
use strict;
use File::Find;
use Getopt::Long;
use File::Which;

my $collection;         # name which will be inserted
my $path_add;           # add additional info in path
my $verbose;

#$verbose = 1;

my $result = GetOptions(
        "collection=s" => \$collection,
        "path=s" => \$path_add,
        "verbose!" => \$verbose,
        "debug!" => \$verbose,
);

my $dir = shift @ARGV || die "usage: $0 [dir]";

my $basedir = $0;
$basedir =~ s,/[^/]+$,/,;
require "$basedir/filter.pm";

my $pdftotext = which('pdftotext');

select(STDERR); $|=1;
select(STDOUT); $|=1;

print STDERR "using $pdftotext to convert pdf into html\n" if ($pdftotext && $verbose);

find({ wanted => \&file, 
        follow => 1,
        no_chdir => 1
}, $dir);

sub dump_contents($$$) {
        my ($contents,$mtime,$path) = @_;

        return if (! $contents);        # don't die on empty files

        use bytes;
        my $size = length $contents;

        print STDERR " [$size]" if ($verbose);

        # Output the document (to swish)
        print <<EOF;
Path-Name: $path
Content-Length: $size
Last-Mtime: $mtime
Document-Type: html*

EOF
        print $contents;

}

sub file {

        my $path = $_;
        my $contents;

        if ($pdftotext && -f $path && $path =~ m/\.pdf$/i) {

                print STDERR "$path {converting}" if ($verbose);

                open(F,"$pdftotext -htmlmeta \"$path\" - |") || die "can't open $pdftotext with '$path'";
                my $html;
                while(<F>) {
                        # XXX why pdftotext barks if I try to use this is beyond me.
                        #$contents .= $_;

                        $html .= $_;
                }
                close(F);

                return if (! $html);

                my ($pre_html,$pages,$post_html) = ('<html><head><title>$path :: page ##page_nr##</title></head><body><pre>',$html,'</pre></body></html>');

                ($pre_html,$pages,$post_html) = ($1,$2,$3) if ($html =~ m/^(<html>.+?<pre>)(.+)(<\/pre>.+?)$/si);

                if ($collection) {
                        $pre_html =~ s/<title>(.+?)<\/title>/<title>$collection :: page ##page_nr##<\/title>/si;
                } else {
                        $pre_html =~ s/<title>(.+?)<\/title>/<title>$1 :: page ##page_nr##<\/title>/si;
                }

                my $page_nr = 1;
                foreach my $page (split(/\f/s,$pages)) {
                        print STDERR " $page_nr" if ($verbose);
                        my $pre_tmp = $pre_html;
                        $pre_tmp =~ s/##page_nr##/$page_nr<\/title>/s;
                        dump_contents($pre_tmp . $page . $post_html,time(), $path) if ($page !~ m/^\s*$/s);
                        $page_nr++;
                }

        } else {

                return if (! -f $path || ! m/\.(html*|php|pl|txt|info|log|text)$/i);

                # skip index files
                return if (m/index_[a-z]\.html*/i || m/index_symbol\.html*/i);

                open(F,"$path") || die "can't open file: $path";
                print STDERR "$path" if ($verbose);
                while(<F>) {
                        $contents .= "$_";
                }
                $contents .= "\n\n";

                $contents = filter($contents,$collection);

                # add optional components to path
                $path .= " $path_add" if ($path_add);

                dump_contents($contents,time(), $path);
        }

        print STDERR "\n" if ($verbose);
#       die "zero size content in '$path'" if (! $contents);

}

1	dpavlin	81	#!/usr/bin/perl -w
2	dpavlin	46	use strict;
3			use File::Find;
4	dpavlin	56	use Getopt::Long;
5	dpavlin	63	use File::Which;
6	dpavlin	46
7	dpavlin	56	my $collection; # name which will be inserted
8			my $path_add; # add additional info in path
9			my $verbose;
10	dpavlin	46
11	dpavlin	57	#$verbose = 1;
12
13	dpavlin	56	my $result = GetOptions(
14			"collection=s" => \$collection,
15			"path=s" => \$path_add,
16			"verbose!" => \$verbose,
17			"debug!" => \$verbose,
18			);
19
20	dpavlin	46	my $dir = shift @ARGV \|\| die "usage: $0 [dir]";
21
22			my $basedir = $0;
23			$basedir =~ s,/[^/]+$,/,;
24			require "$basedir/filter.pm";
25
26	dpavlin	63	my $pdftotext = which('pdftotext');
27	dpavlin	56
28	dpavlin	66	select(STDERR); $\|=1;
29			select(STDOUT); $\|=1;
30
31	dpavlin	63	print STDERR "using $pdftotext to convert pdf into html\n" if ($pdftotext && $verbose);
32
33	dpavlin	46	find({ wanted => \&file,
34			follow => 1,
35			no_chdir => 1
36			}, $dir);
37
38	dpavlin	66	sub dump_contents($$$) {
39			my ($contents,$mtime,$path) = @_;
40
41			return if (! $contents); # don't die on empty files
42
43			use bytes;
44			my $size = length $contents;
45
46			print STDERR " [$size]" if ($verbose);
47
48			# Output the document (to swish)
49			print <<EOF;
50			Path-Name: $path
51			Content-Length: $size
52			Last-Mtime: $mtime
53	dpavlin	81	Document-Type: html*
54	dpavlin	66
55			EOF
56			print $contents;
57
58			}
59
60	dpavlin	46	sub file {
61
62	dpavlin	63	my $path = $_;
63			my $contents;
64	dpavlin	46
65	dpavlin	63	if ($pdftotext && -f $path && $path =~ m/\.pdf$/i) {
66	dpavlin	56
67	dpavlin	63	print STDERR "$path {converting}" if ($verbose);
68	dpavlin	46
69	dpavlin	66	open(F,"$pdftotext -htmlmeta \"$path\" - \|") \|\| die "can't open $pdftotext with '$path'";
70	dpavlin	63	my $html;
71			while(<F>) {
72			# XXX why pdftotext barks if I try to use this is beyond me.
73			#$contents .= $_;
74
75			$html .= $_;
76			}
77			close(F);
78
79	dpavlin	81	return if (! $html);
80
81	dpavlin	66	my ($pre_html,$pages,$post_html) = ('<html><head><title>$path :: page ##page_nr##</title></head><body><pre>',$html,'</pre></body></html>');
82	dpavlin	63
83	dpavlin	72	($pre_html,$pages,$post_html) = ($1,$2,$3) if ($html =~ m/^(<html>.+?<pre>)(.+)(<\/pre>.+?)$/si);
84	dpavlin	66
85	dpavlin	72	if ($collection) {
86			$pre_html =~ s/<title>(.+?)<\/title>/<title>$collection :: page ##page_nr##<\/title>/si;
87			} else {
88			$pre_html =~ s/<title>(.+?)<\/title>/<title>$1 :: page ##page_nr##<\/title>/si;
89			}
90	dpavlin	66
91			my $page_nr = 1;
92	dpavlin	72	foreach my $page (split(/\f/s,$pages)) {
93			print STDERR " $page_nr" if ($verbose);
94	dpavlin	66	my $pre_tmp = $pre_html;
95			$pre_tmp =~ s/##page_nr##/$page_nr<\/title>/s;
96	dpavlin	68	dump_contents($pre_tmp . $page . $post_html,time(), $path) if ($page !~ m/^\s*$/s);
97	dpavlin	66	$page_nr++;
98			}
99
100	dpavlin	63	} else {
101
102	dpavlin	81	return if (! -f $path \|\| ! m/\.(html*\|php\|pl\|txt\|info\|log\|text)$/i);
103	dpavlin	63
104			# skip index files
105			return if (m/index_[a-z]\.html/i \|\| m/index_symbol\.html/i);
106
107			open(F,"$path") \|\| die "can't open file: $path";
108			print STDERR "$path" if ($verbose);
109			while(<F>) {
110			$contents .= "$_";
111			}
112			$contents .= "\n\n";
113
114			$contents = filter($contents,$collection);
115	dpavlin	66
116			# add optional components to path
117			$path .= " $path_add" if ($path_add);
118
119			dump_contents($contents,time(), $path);
120	dpavlin	46	}
121
122	dpavlin	66	print STDERR "\n" if ($verbose);
123	dpavlin	50	# die "zero size content in '$path'" if (! $contents);
124
125	dpavlin	66	}
126	dpavlin	46