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 |
|
11 |
my $collection; # name which will be inserted |
12 |
my $path_add; # add additional info in path |
13 |
my $verbose; |
14 |
my $exclude; |
15 |
|
16 |
#$verbose = 1; |
17 |
my $debug = 0; |
18 |
my $force = 0; |
19 |
|
20 |
my $result = GetOptions( |
21 |
"collection=s" => \$collection, |
22 |
"path=s" => \$path_add, |
23 |
"verbose!" => \$verbose, |
24 |
"debug!" => \$debug, |
25 |
"exclude=s" => \$exclude, |
26 |
"force!" => \$force, |
27 |
); |
28 |
|
29 |
my ($node_url,$dir) = @ARGV; |
30 |
|
31 |
die <<"_END_OF_USAGE_" if (! $node_url || ! $dir); |
32 |
usage: $0 http://localhost:1978/node/my_dir /path/to/directory |
33 |
|
34 |
options: |
35 |
--collection="name of collection" |
36 |
--path=/path/to/add/at/end |
37 |
--exclude=regex_to_exclude |
38 |
--verbose |
39 |
--force |
40 |
--debug |
41 |
_END_OF_USAGE_ |
42 |
|
43 |
if (! -e $dir) { |
44 |
warn "directory $dir doesn't exist, skipping\n"; |
45 |
exit 1; |
46 |
} |
47 |
|
48 |
#my $basedir = $0; |
49 |
#$basedir =~ s,/[^/]+$,/,; |
50 |
#require "$basedir/filter.pm"; |
51 |
|
52 |
my $pdftotext = which('pdftotext'); |
53 |
|
54 |
#my $mm = new File::MMagic('/usr/share/misc/file/magic'); |
55 |
my $mm = new File::MMagic::XS(); |
56 |
|
57 |
my $iconv = new Text::Iconv('iso-8859-2', 'utf-8'); |
58 |
|
59 |
select(STDERR); $|=1; |
60 |
select(STDOUT); $|=1; |
61 |
|
62 |
print STDERR "using $pdftotext to convert pdf into html\n" if ($pdftotext && $verbose); |
63 |
|
64 |
my $db = new Search::Estraier::Node; |
65 |
$db->set_url($node_url); |
66 |
$db->set_auth('admin', 'admin'); |
67 |
|
68 |
find({ wanted => \&file, |
69 |
follow => 1, |
70 |
follow_skip => 2, |
71 |
no_chdir => 1, |
72 |
}, $dir); |
73 |
|
74 |
|
75 |
exit; |
76 |
|
77 |
sub dump_contents($$$$) { |
78 |
my ($db,$contents,$mtime,$path) = @_; |
79 |
|
80 |
return unless ($contents); # don't die on empty files |
81 |
|
82 |
if ($exclude && $path =~ m/$exclude/i) { |
83 |
print STDERR "skip: $path\n" if ($verbose); |
84 |
return; |
85 |
} |
86 |
|
87 |
use bytes; |
88 |
my $size = length $contents; |
89 |
|
90 |
print STDERR " [$size]" if ($verbose); |
91 |
|
92 |
# create a document object |
93 |
my $doc = new Search::Estraier::Document; |
94 |
|
95 |
my $title = $1 if ($contents =~ m#<title>(.+?)</title>#is); |
96 |
|
97 |
# chop long titles to 100 chars |
98 |
$title = substr($title, 0, 100) . '...' if ($title && length($title) > 100); |
99 |
# use path if no title is found |
100 |
$title ||= $path; |
101 |
|
102 |
# add attributes to the document object |
103 |
$doc->add_attr('@uri', "file:///$path"); |
104 |
$doc->add_attr('@title', $iconv->convert($title)); |
105 |
$doc->add_attr('@size', $size); |
106 |
$doc->add_attr('@mtime', $mtime); |
107 |
|
108 |
# html2text |
109 |
$contents =~ s#<[^>]+/*>##gs; |
110 |
$contents =~ s#\s\s+# #gs; |
111 |
|
112 |
$doc->add_text($iconv->convert($contents)); |
113 |
|
114 |
# print $doc->dump_draft if ($verbose); |
115 |
|
116 |
# register the document object to the database |
117 |
$db->put_doc($doc); |
118 |
|
119 |
} |
120 |
|
121 |
sub file { |
122 |
|
123 |
my $path = $_; |
124 |
my $contents; |
125 |
|
126 |
return if (! $force && -l $path || $path =~ m#/.svn# || $path =~ m/(~|.bak)$/); |
127 |
|
128 |
my $mtime = (stat($path))[9] || -1; |
129 |
my $mtime_db = $db->get_doc_attr_by_uri("file:///$path", '@mtime') || -2; |
130 |
|
131 |
if ($mtime == $mtime_db) { |
132 |
print STDERR "# same: $path $mtime\n" if ($verbose); |
133 |
return unless($force); |
134 |
} else { |
135 |
print STDERR "# changed: $path $mtime != $mtime_db\n" if ($debug); |
136 |
} |
137 |
|
138 |
# skip files on which File::MMagic::XS croaks |
139 |
return if ($path =~ m#\.au$#); |
140 |
|
141 |
my $type = $mm->checktype_filename($path); |
142 |
$type =~ s/\s+/ /gs; |
143 |
|
144 |
print STDERR "# $path $type\n" if ($debug); |
145 |
|
146 |
if ($pdftotext && -f $path && $type =~ m/pdf/i) { |
147 |
|
148 |
print STDERR "$path {converting}" if ($verbose); |
149 |
|
150 |
open(F,"$pdftotext -htmlmeta \"$path\" - |") || die "can't open $pdftotext with '$path'"; |
151 |
my $html; |
152 |
while(<F>) { |
153 |
# XXX why pdftotext barks if I try to use this is beyond me. |
154 |
#$contents .= $_; |
155 |
|
156 |
$html .= $_; |
157 |
} |
158 |
close(F); |
159 |
|
160 |
return if (! $html); |
161 |
|
162 |
my $file_only = $path; |
163 |
$file_only =~ s/^.*\/([^\/]+)$/$1/g; |
164 |
|
165 |
my ($pre_html,$pages,$post_html) = ('<html><head><title>$path :: page ##page_nr##</title></head><body><pre>',$html,'</pre></body></html>'); |
166 |
|
167 |
($pre_html,$pages,$post_html) = ($1,$2,$3) if ($html =~ m/^(<html>.+?<pre>)(.+)(<\/pre>.+?)$/si); |
168 |
|
169 |
if ($collection) { |
170 |
$pre_html =~ s/<title>(.+?)<\/title>/<title>$collection :: page ##page_nr##<\/title>/si; |
171 |
} else { |
172 |
$pre_html =~ s/<title>(.+?)<\/title>/<title>$1 :: page ##page_nr##<\/title>/si || |
173 |
$pre_html =~ s/<title><\/title>/<title>$file_only :: page ##page_nr##<\/title>/si; |
174 |
} |
175 |
|
176 |
# save empty entry as a placeholder |
177 |
dump_contents($db, ' ', $mtime, "$path"); |
178 |
|
179 |
my $page_nr = 1; |
180 |
foreach my $page (split(/\f/s,$pages)) { |
181 |
print STDERR " $page_nr" if ($verbose); |
182 |
my $pre_tmp = $pre_html; |
183 |
$pre_tmp =~ s/##page_nr##/$page_nr<\/title>/s; |
184 |
dump_contents($db, $pre_tmp . $page . $post_html, $mtime, "$path#$page_nr") if ($page !~ m/^\s*$/s); |
185 |
$page_nr++; |
186 |
} |
187 |
|
188 |
} else { |
189 |
|
190 |
# return if (! -f $path || ! m/\.(html*|php|pl|txt|info|log|text)$/i); |
191 |
if (-f $path && $type =~ m/html/ || |
192 |
($type !~ m#text# && $path =~ m/\.(php|pl|txt|info|log|text)$/io) |
193 |
) { |
194 |
dump_contents($db, "<title>$path</title> $type", $mtime, $path); |
195 |
} |
196 |
|
197 |
# skip index files |
198 |
return if (m/index_[a-z]\.html*/i || m/index_symbol\.html*/i); |
199 |
|
200 |
open(F,"$path") || die "can't open file: $path"; |
201 |
print STDERR "$path ($type)" if ($verbose); |
202 |
while(<F>) { |
203 |
$contents .= "$_"; |
204 |
} |
205 |
$contents .= "\n\n"; |
206 |
|
207 |
#$contents = filter($contents,$collection); |
208 |
|
209 |
# add optional components to path |
210 |
$path .= " $path_add" if ($path_add); |
211 |
|
212 |
dump_contents($db, $contents, $mtime, $path); |
213 |
} |
214 |
|
215 |
print STDERR "\n" if ($verbose); |
216 |
# die "zero size content in '$path'" if (! $contents); |
217 |
|
218 |
} |
219 |
|