5 |
use File::Which; |
use File::Which; |
6 |
use HyperEstraier; |
use HyperEstraier; |
7 |
use Text::Iconv; |
use Text::Iconv; |
8 |
|
#use File::MMagic; |
9 |
|
use File::MMagic::XS qw/:compat/; |
10 |
|
|
11 |
# do we use Node API? |
# do we use Node API? |
12 |
my $node_url; |
my $node_url; |
17 |
my $exclude; |
my $exclude; |
18 |
|
|
19 |
#$verbose = 1; |
#$verbose = 1; |
20 |
|
my $debug = 0; |
21 |
|
my $force = 0; |
22 |
|
|
23 |
my $result = GetOptions( |
my $result = GetOptions( |
24 |
"collection=s" => \$collection, |
"collection=s" => \$collection, |
25 |
"path=s" => \$path_add, |
"path=s" => \$path_add, |
26 |
"verbose!" => \$verbose, |
"verbose!" => \$verbose, |
27 |
"debug!" => \$verbose, |
"debug!" => \$debug, |
28 |
"exclude=s" => \$exclude, |
"exclude=s" => \$exclude, |
29 |
"node=s" => \$node_url, |
"node=s" => \$node_url, |
30 |
|
"force!" => \$force, |
31 |
); |
); |
32 |
|
|
33 |
my $dir = shift @ARGV || die "usage: $0 [dir]"; |
my $dir = shift @ARGV || die "usage: $0 [dir]"; |
34 |
|
|
35 |
|
if (! -e $dir) { |
36 |
|
warn "directory $dir doesn't exist, skipping\n"; |
37 |
|
exit 1; |
38 |
|
} |
39 |
|
|
40 |
#my $basedir = $0; |
#my $basedir = $0; |
41 |
#$basedir =~ s,/[^/]+$,/,; |
#$basedir =~ s,/[^/]+$,/,; |
42 |
#require "$basedir/filter.pm"; |
#require "$basedir/filter.pm"; |
43 |
|
|
44 |
my $pdftotext = which('pdftotext'); |
my $pdftotext = which('pdftotext'); |
45 |
|
|
46 |
|
#my $mm = new File::MMagic('/usr/share/misc/file/magic'); |
47 |
|
my $mm = new File::MMagic::XS(); |
48 |
|
|
49 |
my $iconv = new Text::Iconv('iso-8859-2', 'utf-8'); |
my $iconv = new Text::Iconv('iso-8859-2', 'utf-8'); |
50 |
|
|
51 |
select(STDERR); $|=1; |
select(STDERR); $|=1; |
106 |
# create a document object |
# create a document object |
107 |
my $doc = HyperEstraier::Document->new; |
my $doc = HyperEstraier::Document->new; |
108 |
|
|
109 |
my $title = $1 if ($contents =~ m#<title>(.+)</title>#is); |
my $title = $1 if ($contents =~ m#<title>(.+?)</title>#is); |
110 |
|
|
111 |
|
# chop long titles to 100 chars |
112 |
|
$title = substr($title, 0, 100) . '...' if ($title && length($title) > 100); |
113 |
|
# use path if no title is found |
114 |
|
$title ||= $path; |
115 |
|
|
116 |
# add attributes to the document object |
# add attributes to the document object |
117 |
$doc->add_attr('@uri', "file:///$path"); |
$doc->add_attr('@uri', "file:///$path"); |
118 |
$doc->add_attr('@title', $title || $path); |
$doc->add_attr('@title', $iconv->convert($title)); |
119 |
$doc->add_attr('@size', $size); |
$doc->add_attr('@size', $size); |
120 |
$doc->add_attr('@mtime', $mtime); |
$doc->add_attr('@mtime', $mtime); |
121 |
|
|
141 |
my $path = $_; |
my $path = $_; |
142 |
my $contents; |
my $contents; |
143 |
|
|
144 |
return if (-l $path); |
return if (! $force && -l $path || $path =~ m#/.svn# || $path =~ m/(~|.bak)$/); |
145 |
|
|
146 |
|
my $mtime = (stat($path))[9] || -1; |
147 |
|
my $mtime_db = $db->get_doc_attr_by_uri("file:///$path", '@mtime') || -2; |
148 |
|
|
149 |
|
if ($mtime == $mtime_db) { |
150 |
|
print STDERR "# same: $path $mtime\n" if ($verbose); |
151 |
|
return unless($force); |
152 |
|
} else { |
153 |
|
print STDERR "# changed: $path $mtime != $mtime_db\n" if ($debug); |
154 |
|
} |
155 |
|
|
156 |
|
# skip files on which File::MMagic::XS croaks |
157 |
|
return if ($path =~ m#\.au$#); |
158 |
|
|
159 |
if ($pdftotext && -f $path && $path =~ m/\.pdf$/i) { |
my $type = $mm->checktype_filename($path); |
160 |
|
$type =~ s/\s+/ /gs; |
161 |
|
|
162 |
|
print STDERR "# $path $type\n" if ($debug); |
163 |
|
|
164 |
|
if ($pdftotext && -f $path && $type =~ m/pdf/i) { |
165 |
|
|
166 |
print STDERR "$path {converting}" if ($verbose); |
print STDERR "$path {converting}" if ($verbose); |
167 |
|
|
191 |
$pre_html =~ s/<title><\/title>/<title>$file_only :: page ##page_nr##<\/title>/si; |
$pre_html =~ s/<title><\/title>/<title>$file_only :: page ##page_nr##<\/title>/si; |
192 |
} |
} |
193 |
|
|
194 |
|
# save empty entry as a placeholder |
195 |
|
dump_contents($db, ' ', $mtime, "$path"); |
196 |
|
|
197 |
my $page_nr = 1; |
my $page_nr = 1; |
198 |
foreach my $page (split(/\f/s,$pages)) { |
foreach my $page (split(/\f/s,$pages)) { |
199 |
print STDERR " $page_nr" if ($verbose); |
print STDERR " $page_nr" if ($verbose); |
200 |
my $pre_tmp = $pre_html; |
my $pre_tmp = $pre_html; |
201 |
$pre_tmp =~ s/##page_nr##/$page_nr<\/title>/s; |
$pre_tmp =~ s/##page_nr##/$page_nr<\/title>/s; |
202 |
dump_contents($db, $pre_tmp . $page . $post_html,time(), $path) if ($page !~ m/^\s*$/s); |
dump_contents($db, $pre_tmp . $page . $post_html, $mtime, "$path#$page_nr") if ($page !~ m/^\s*$/s); |
203 |
$page_nr++; |
$page_nr++; |
204 |
} |
} |
205 |
|
|
206 |
} else { |
} else { |
207 |
|
|
208 |
return if (! -f $path || ! m/\.(html*|php|pl|txt|info|log|text)$/i); |
# return if (! -f $path || ! m/\.(html*|php|pl|txt|info|log|text)$/i); |
209 |
|
return unless (-f $path && $type =~ m/html/ || |
210 |
|
($type =~ m#text# && $path =~ m/\.(php|pl|txt|info|log|text)$/io) |
211 |
|
); |
212 |
|
|
213 |
# skip index files |
# skip index files |
214 |
return if (m/index_[a-z]\.html*/i || m/index_symbol\.html*/i); |
return if (m/index_[a-z]\.html*/i || m/index_symbol\.html*/i); |
215 |
|
|
216 |
open(F,"$path") || die "can't open file: $path"; |
open(F,"$path") || die "can't open file: $path"; |
217 |
print STDERR "$path" if ($verbose); |
print STDERR "$path ($type)" if ($verbose); |
218 |
while(<F>) { |
while(<F>) { |
219 |
$contents .= "$_"; |
$contents .= "$_"; |
220 |
} |
} |
225 |
# add optional components to path |
# add optional components to path |
226 |
$path .= " $path_add" if ($path_add); |
$path .= " $path_add" if ($path_add); |
227 |
|
|
228 |
dump_contents($db, $contents,time(), $path); |
dump_contents($db, $contents, $mtime, $path); |
229 |
} |
} |
230 |
|
|
231 |
print STDERR "\n" if ($verbose); |
print STDERR "\n" if ($verbose); |