1 |
#!/usr/bin/perl |
2 |
|
3 |
use warnings; |
4 |
use strict; |
5 |
|
6 |
package Grep::Import::ScrapBook; |
7 |
|
8 |
=head1 NAME |
9 |
|
10 |
Grep::Import::ScrapBook - importer for local ScrapBook pages |
11 |
|
12 |
=head1 CONFIGURATION |
13 |
|
14 |
You can symlink your ScrapBook directory |
15 |
|
16 |
~/Grep/share/web/static$ ln -sf /home/dpavlin/private/ScrapBook scrapbook |
17 |
|
18 |
or modify L<ScrapBookDir> path (relative to Grep installation static root). |
19 |
|
20 |
=cut |
21 |
|
22 |
use XML::Simple; |
23 |
use File::Slurp; |
24 |
use HTML::ResolveLink; |
25 |
use HTML::TreeBuilder; |
26 |
use Data::Dump qw/dump/; |
27 |
|
28 |
sub import { |
29 |
my $self = shift; |
30 |
|
31 |
my $config = Jifty->config->app('Import')->{'ScrapBook'}; |
32 |
|
33 |
if (! $config ) { |
34 |
Jifty->log->warn("skipping ScrapBook importer, no application->Import->ScrapBook config"); |
35 |
return; |
36 |
} |
37 |
|
38 |
# required parametars in config.yml |
39 |
foreach my $param ( qw/Dir OwnerEmail/ ) { |
40 |
Jifty->log->die("can't find $param in Scrapbook config") unless defined ( $config->{$param} ); |
41 |
}; |
42 |
|
43 |
my $dir = |
44 |
Jifty::Util->app_root . '/' . |
45 |
Jifty->config->framework('Web')->{'StaticRoot'} . '/' . $config->{'Dir'}; |
46 |
|
47 |
my $path = $dir . '/scrapbook.rdf'; |
48 |
$path =~ s!//+!/!g; |
49 |
|
50 |
if ( ! -e $dir || ! -e $path ) { |
51 |
Jifty->log->warn("Skipping ScrapBook import $path: $!"); |
52 |
return 1; |
53 |
} |
54 |
|
55 |
my $rdf = XMLin( |
56 |
$path, |
57 |
# KeyAttr => [ qw/RDF:about/ ], |
58 |
) || die "can't open $path: $!"; |
59 |
|
60 |
# warn "## original rdf -> ", dump( $rdf ); |
61 |
|
62 |
my $owner = Grep::Model::User->new(); |
63 |
$owner->load_by_cols( email => $config->{OwnerEmail} ); |
64 |
die "can't find ScrapBookOwner ", $config->{OwnerEmail} unless ( $owner->id ); |
65 |
|
66 |
Jifty->log->info( "Using user ", $owner->id, " from ", $owner->email, " for import" ); |
67 |
|
68 |
my $feed = Grep::Model::Feed->new( current_user => $owner ); |
69 |
$feed->load_or_create( |
70 |
uri => 'file://' . $path, |
71 |
title => 'ScrapBook', |
72 |
#source => 'Grep::Source', |
73 |
owner => $owner, |
74 |
); |
75 |
|
76 |
my $search = Grep::Search->new; |
77 |
|
78 |
my $stats; |
79 |
|
80 |
foreach my $item ( @{ $rdf->{'RDF:Description'} } ) { |
81 |
|
82 |
$stats->{total}++; |
83 |
|
84 |
#warn "## item = ",dump( $item ); |
85 |
|
86 |
my $hash; |
87 |
foreach my $k ( keys %$item ) { |
88 |
next if $k =~ m/^RDF:/; |
89 |
next if ( $item->{$k} eq '' ); |
90 |
my $n = $k; |
91 |
$n =~ s/^\w+://; # strip namespace |
92 |
$hash->{$n} = $item->{$k}; |
93 |
} |
94 |
|
95 |
#warn "## hash = ", dump( $hash ); |
96 |
|
97 |
|
98 |
# fetch full-text content and import it |
99 |
|
100 |
my $rel_path = '/data/' . $hash->{id} . '/index.html'; |
101 |
|
102 |
my $content_path = $dir . $rel_path; |
103 |
if ( ! -r $content_path ) { |
104 |
Jifty->log->warn("can't import $content_path: $!"); |
105 |
$stats->{failure}++; |
106 |
next; |
107 |
} |
108 |
my $content = read_file( $content_path ) or |
109 |
die "can't read $content_path: $!"; |
110 |
|
111 |
my $tree = HTML::TreeBuilder->new or die "can't create html tree"; |
112 |
$tree->parse( $content ) or die "can't parse fetched content"; |
113 |
|
114 |
my $body = $tree->look_down( '_tag', 'body' ); |
115 |
|
116 |
my $resolver = HTML::ResolveLink->new( base => '/static/' . $config->{Dir} . $rel_path ); |
117 |
$content = $resolver->resolve( $body->as_HTML ); |
118 |
|
119 |
# create date from id |
120 |
|
121 |
my $dt; |
122 |
if ( $hash->{id} =~ m/^(\d{4})(\d\d)(\d\d)(\d\d)(\d\d)(\d\d)$/ ) { |
123 |
$dt = DateTime->new( |
124 |
year => $1, |
125 |
month => $2, |
126 |
day => $3, |
127 |
hour => $4, |
128 |
minute => $5, |
129 |
second => $6, |
130 |
#time_zone => 'UTC', |
131 |
); |
132 |
} else { |
133 |
warn "can't parse date from ", $hash->{id}; |
134 |
} |
135 |
|
136 |
my $i = Grep::Model::Item->new( current_user => $owner ); |
137 |
my ($ok,$msg) = $i->load_or_create( |
138 |
in_feed => $feed, |
139 |
title => $hash->{title}, |
140 |
link => $hash->{source}, |
141 |
content => $content, |
142 |
created_on => $hash->{id}, |
143 |
); |
144 |
|
145 |
if ( ! $ok ) { |
146 |
Jifty->log->error( $msg ); |
147 |
$stats->{failure}++; |
148 |
next; |
149 |
} |
150 |
|
151 |
if ( $msg && $msg =~ m/^Found/ ) { |
152 |
$stats->{old}++; |
153 |
} else { |
154 |
$stats->{new}++; |
155 |
Jifty->log->info("created ", $i->id ," ", $i->link, " ", length( $content ), " bytes"); |
156 |
$search->add( $i, $owner->id ); |
157 |
} |
158 |
|
159 |
} |
160 |
|
161 |
$search->finish; |
162 |
|
163 |
return $stats; |
164 |
} |
165 |
|
166 |
=head1 SEE ALSO |
167 |
|
168 |
L<http://amb.vis.ne.jp/mozilla/scrapbook/> - ScrapBook FireFox extension |
169 |
|
170 |
=cut |
171 |
|
172 |
1; |