| 1 |
#!/usr/bin/perl -w |
| 2 |
use strict; |
| 3 |
use File::Find; |
| 4 |
use Getopt::Long; |
| 5 |
use File::Which; |
| 6 |
use Search::Estraier; |
| 7 |
use Text::Iconv; |
| 8 |
#use File::MMagic; |
| 9 |
use File::MMagic::XS qw/:compat/; |
| 10 |
use Time::HiRes qw/time/; |
| 11 |
|
| 12 |
my $collection; # name which will be inserted |
| 13 |
my $path_add; # add additional info in path |
| 14 |
my $verbose; |
| 15 |
my $exclude; |
| 16 |
|
| 17 |
#$verbose = 1; |
| 18 |
my $debug = 0; |
| 19 |
my $force = 0; |
| 20 |
my $all = 0; |
| 21 |
|
| 22 |
my $result = GetOptions( |
| 23 |
"collection=s" => \$collection, |
| 24 |
"path=s" => \$path_add, |
| 25 |
"verbose!" => \$verbose, |
| 26 |
"debug!" => \$debug, |
| 27 |
"exclude=s" => \$exclude, |
| 28 |
"force!" => \$force, |
| 29 |
"all!" => \$all, |
| 30 |
); |
| 31 |
|
| 32 |
my ($node_url,$dir) = @ARGV; |
| 33 |
|
| 34 |
die <<"_END_OF_USAGE_" if (! $node_url || ! $dir); |
| 35 |
usage: $0 http://localhost:1978/node/my_dir /path/to/directory |
| 36 |
|
| 37 |
options: |
| 38 |
--collection="name of collection" |
| 39 |
--path=/path/to/add/at/end |
| 40 |
--exclude=regex_to_exclude |
| 41 |
--verbose |
| 42 |
--force |
| 43 |
--debug |
| 44 |
--all save placeholders for all files |
| 45 |
_END_OF_USAGE_ |
| 46 |
|
| 47 |
if (! -e $dir) { |
| 48 |
warn "directory $dir doesn't exist, skipping\n"; |
| 49 |
exit 1; |
| 50 |
} |
| 51 |
|
| 52 |
#my $basedir = $0; |
| 53 |
#$basedir =~ s,/[^/]+$,/,; |
| 54 |
#require "$basedir/filter.pm"; |
| 55 |
|
| 56 |
my $docs = 0; |
| 57 |
my $start_t = time(); |
| 58 |
|
| 59 |
my $filter; |
| 60 |
foreach my $f (qw/pdftotext pstotext/) { |
| 61 |
my $w = which($f); |
| 62 |
if ($f) { |
| 63 |
$filter->{$f} = $w; |
| 64 |
print STDERR "using $f filter at $w\n" if ($verbose); |
| 65 |
} |
| 66 |
} |
| 67 |
|
| 68 |
#my $mm = new File::MMagic('/usr/share/misc/file/magic'); |
| 69 |
my $mm = new File::MMagic::XS(); |
| 70 |
|
| 71 |
my $iconv = new Text::Iconv('iso-8859-2', 'utf-8'); |
| 72 |
|
| 73 |
select(STDERR); $|=1; |
| 74 |
select(STDOUT); $|=1; |
| 75 |
|
| 76 |
my $db = new Search::Estraier::Node( |
| 77 |
url => $node_url, |
| 78 |
user => 'admin', |
| 79 |
passwd => 'admin', |
| 80 |
croak_on_error => 1, |
| 81 |
); |
| 82 |
|
| 83 |
find({ wanted => \&file, |
| 84 |
follow => 1, |
| 85 |
follow_skip => 2, |
| 86 |
no_chdir => 1, |
| 87 |
}, $dir); |
| 88 |
|
| 89 |
my $dur = (time() - $start_t) || 1; |
| 90 |
printf STDERR "%d documents in %.2fs [%.2f docs/s]\n", $docs, $dur, ($docs / $dur); |
| 91 |
|
| 92 |
$db->master( |
| 93 |
action => 'sync' |
| 94 |
); |
| 95 |
|
| 96 |
|
| 97 |
exit; |
| 98 |
|
| 99 |
sub dump_contents { |
| 100 |
my ($db,$contents,$mtime,$path,$size) = @_; |
| 101 |
|
| 102 |
return unless (defined($contents)); # don't die on empty files |
| 103 |
|
| 104 |
if ($exclude && $path =~ m/$exclude/i) { |
| 105 |
print STDERR "skip: $path\n" if ($verbose); |
| 106 |
return; |
| 107 |
} |
| 108 |
|
| 109 |
use bytes; |
| 110 |
if (! $size) { |
| 111 |
$size = length $contents; |
| 112 |
} |
| 113 |
|
| 114 |
print STDERR " [$size]" if ($verbose); |
| 115 |
|
| 116 |
# create a document object |
| 117 |
my $doc = new Search::Estraier::Document; |
| 118 |
|
| 119 |
my $title = $1 if ($contents =~ m#<title>(.+?)</title>#is); |
| 120 |
|
| 121 |
# chop long titles to 100 chars |
| 122 |
$title = substr($title, 0, 100) . '...' if ($title && length($title) > 100); |
| 123 |
# use path if no title is found |
| 124 |
$title ||= $path; |
| 125 |
|
| 126 |
# add attributes to the document object |
| 127 |
$doc->add_attr('@uri', "file:///$path"); |
| 128 |
$doc->add_attr('@title', $iconv->convert($title)); |
| 129 |
$doc->add_attr('@size', $size); |
| 130 |
$doc->add_attr('@mtime', $mtime); |
| 131 |
|
| 132 |
if ($contents) { |
| 133 |
# html2text |
| 134 |
$contents =~ s#<[^>]+/*>##gs; |
| 135 |
$contents =~ s#\s\s+# #gs; |
| 136 |
|
| 137 |
$doc->add_text($iconv->convert($contents)); |
| 138 |
} |
| 139 |
# store path |
| 140 |
$doc->add_hidden_text($path); |
| 141 |
# boost title |
| 142 |
$doc->add_hidden_text($title); |
| 143 |
|
| 144 |
print $doc->dump_draft if ($debug); |
| 145 |
|
| 146 |
# register the document object to the database |
| 147 |
$db->put_doc($doc); |
| 148 |
|
| 149 |
$docs++; |
| 150 |
|
| 151 |
} |
| 152 |
|
| 153 |
sub filter_to_pages { |
| 154 |
my ($path, $mtime, $command) = @_; |
| 155 |
|
| 156 |
print STDERR "$path {converting}" if ($verbose); |
| 157 |
|
| 158 |
open(F,"$command |") || die "can't open $command with '$path': $!"; |
| 159 |
my $html; |
| 160 |
while(<F>) { |
| 161 |
$html .= $_; |
| 162 |
} |
| 163 |
close(F); |
| 164 |
|
| 165 |
return if (! $html); |
| 166 |
|
| 167 |
my $file_only = $path; |
| 168 |
$file_only =~ s/^.*\/([^\/]+)$/$1/g; |
| 169 |
|
| 170 |
my ($pre_html,$pages,$post_html) = ('<html><head><title>$path :: page ##page_nr##</title></head><body><pre>',$html,'</pre></body></html>'); |
| 171 |
|
| 172 |
($pre_html,$pages,$post_html) = ($1,$2,$3) if ($html =~ m/^(<html>.+?<pre>)(.+)(<\/pre>.+?)$/si); |
| 173 |
|
| 174 |
if ($collection) { |
| 175 |
$pre_html =~ s/<title>(.+?)<\/title>/<title>$collection :: page ##page_nr##<\/title>/si; |
| 176 |
} else { |
| 177 |
$pre_html =~ s/<title>(.+?)<\/title>/<title>$1 :: page ##page_nr##<\/title>/si || |
| 178 |
$pre_html =~ s/<title><\/title>/<title>$file_only :: page ##page_nr##<\/title>/si; |
| 179 |
} |
| 180 |
|
| 181 |
# save empty entry as a placeholder |
| 182 |
dump_contents($db, ' ', $mtime, "$path"); |
| 183 |
|
| 184 |
my $page_nr = 1; |
| 185 |
foreach my $page (split(/\f/s,$pages)) { |
| 186 |
print STDERR " $page_nr" if ($verbose); |
| 187 |
my $pre_tmp = $pre_html; |
| 188 |
$pre_tmp =~ s/##page_nr##/$page_nr<\/title>/s; |
| 189 |
dump_contents($db, $pre_tmp . $page . $post_html, $mtime, "$path#$page_nr") if ($page !~ m/^\s*$/s); |
| 190 |
$page_nr++; |
| 191 |
} |
| 192 |
|
| 193 |
|
| 194 |
|
| 195 |
} |
| 196 |
|
| 197 |
sub file { |
| 198 |
|
| 199 |
my $path = $_; |
| 200 |
my $contents; |
| 201 |
|
| 202 |
return if (! $force && -l $path || $path =~ m#/.svn# || $path =~ m/(~|.bak|.gif)$/); |
| 203 |
return if (! $all && -d $path); |
| 204 |
|
| 205 |
my $mtime = (stat($path))[9] || -1; |
| 206 |
my $mtime_db = eval { $db->get_doc_attr_by_uri("file:///$path", '@mtime') }; |
| 207 |
$mtime_db ||= -2; |
| 208 |
|
| 209 |
if ($mtime == $mtime_db) { |
| 210 |
print STDERR "# same: $path $mtime\n" if ($verbose); |
| 211 |
return unless($force); |
| 212 |
} else { |
| 213 |
print STDERR "# changed: $path $mtime != $mtime_db\n" if ($debug); |
| 214 |
} |
| 215 |
|
| 216 |
# skip files on which File::MMagic::XS croaks |
| 217 |
if ($path =~ m#\.au$#) { |
| 218 |
warn "skipped '$path' to prevent File::MMagic::XS bug\n" if ($debug); |
| 219 |
return; |
| 220 |
} |
| 221 |
|
| 222 |
my $type = $mm->checktype_filename($path); |
| 223 |
$type =~ s/\s+/ /gs; |
| 224 |
|
| 225 |
print STDERR "# $path $type\n" if ($debug); |
| 226 |
|
| 227 |
if ($type =~ m/pdf/i) { |
| 228 |
if ($filter->{pdftotext}) { |
| 229 |
filter_to_pages($path, $mtime, qq( $filter->{pdftotext} -htmlmeta "$path" - )); |
| 230 |
} else { |
| 231 |
warn "skipping '$path', no pdftotext filter\n" if ($verbose); |
| 232 |
return; |
| 233 |
} |
| 234 |
} elsif ($type eq 'application/postscript') { |
| 235 |
if ($filter->{pstotext}) { |
| 236 |
filter_to_pages($path, $mtime, qq( $filter->{pstotext} "$path" )); |
| 237 |
} else { |
| 238 |
warn "skipping '$path', no pstotext filter\n" if ($verbose); |
| 239 |
return; |
| 240 |
} |
| 241 |
} else { |
| 242 |
|
| 243 |
# return if (! -f $path || ! m/\.(html*|php|pl|txt|info|log|text)$/i); |
| 244 |
if (-f $path && |
| 245 |
$type !~ m/html/ && |
| 246 |
$path !~ m/\.(php|pl|txt|info|log|text)$/io |
| 247 |
) { |
| 248 |
dump_contents($db, '', $mtime, $path, -s $path) if ($all); |
| 249 |
return; |
| 250 |
} |
| 251 |
|
| 252 |
# skip index files |
| 253 |
return if ($path =~ m/index_(?:[a-z]+|symbol)\.html*/i); |
| 254 |
|
| 255 |
open(F,"$path") || die "can't open file: $path"; |
| 256 |
print STDERR "$path ($type)" if ($verbose); |
| 257 |
while(<F>) { |
| 258 |
$contents .= "$_"; |
| 259 |
} |
| 260 |
$contents .= "\n\n"; |
| 261 |
|
| 262 |
#$contents = filter($contents,$collection); |
| 263 |
|
| 264 |
# add optional components to path |
| 265 |
$path .= " $path_add" if ($path_add); |
| 266 |
|
| 267 |
dump_contents($db, $contents, $mtime, $path); |
| 268 |
} |
| 269 |
|
| 270 |
print STDERR "\n" if ($verbose); |
| 271 |
# die "zero size content in '$path'" if (! $contents); |
| 272 |
|
| 273 |
} |
| 274 |
|
Parent Directory
| 
Revision Log