1 |
#!/usr/bin/perl |
2 |
|
3 |
use warnings; |
4 |
use strict; |
5 |
|
6 |
use XML::Simple; |
7 |
use File::Find; |
8 |
use Regexp::Common qw/balanced/; |
9 |
use Socialtext::Resting; |
10 |
use Encode; |
11 |
use HTTP::Date; |
12 |
use POSIX qw/strftime/; |
13 |
use File::Slurp; |
14 |
use File::MMagic::XS; |
15 |
use Getopt::Long; |
16 |
use Data::Dump qw/dump/; |
17 |
|
18 |
my $debug = 0; |
19 |
my $max = 999; |
20 |
my $attachments = 0; |
21 |
|
22 |
GetOptions( |
23 |
'debug+' => \$debug, |
24 |
'max=i' => \$max, |
25 |
'attachments' => \$attachments, |
26 |
); |
27 |
|
28 |
my $page; |
29 |
my $page_date; |
30 |
|
31 |
my @page_names; |
32 |
|
33 |
print "Collecting pages...\n"; |
34 |
|
35 |
find({ |
36 |
wanted => sub { |
37 |
my $path = $File::Find::name; |
38 |
return unless -f $path; |
39 |
|
40 |
warn "+ $path\n"; |
41 |
my $ref = XMLin( $path, |
42 |
KeyAttr => { |
43 |
'attachment' => '+name', |
44 |
'meta' => 'name', |
45 |
}, |
46 |
ForceArray => [ 'attachment', 'meta', 'widget' ], |
47 |
) || die "can't open $path: $!"; |
48 |
|
49 |
warn "## $path = ",dump( $ref ) if $debug; |
50 |
|
51 |
my $name = $ref->{name} || die "no name in $path"; |
52 |
|
53 |
return if $name =~ m/^TamSystem/; |
54 |
|
55 |
my $date = $ref->{meta}->{LastModified}->{value}; |
56 |
if ( ! $date ) { |
57 |
warn "SKIP: no LastModified in $path $name"; |
58 |
return; |
59 |
} |
60 |
|
61 |
my $data; |
62 |
|
63 |
foreach my $w ( @{ $ref->{widgets}->{widget} } ) { |
64 |
|
65 |
warn "## w = ",dump( $w ) if $debug; |
66 |
|
67 |
$data .= "\n----\n" if $data; |
68 |
$data .= $w->{data} || die "no data?"; |
69 |
} |
70 |
|
71 |
my $attachments; |
72 |
|
73 |
if ( my $a = $ref->{attachment} ) { |
74 |
foreach my $name ( keys %$a ) { |
75 |
my $full_path = $path; |
76 |
$full_path =~ s,pages/,attachments/,; |
77 |
$full_path .= '.' . $name; |
78 |
die "$full_path doesn't exist" unless -e $full_path; |
79 |
push @$attachments, { |
80 |
full_path => $full_path, |
81 |
name => ( $name || $a->{$name}->{desc} || 'noname' ), |
82 |
}; |
83 |
} |
84 |
} |
85 |
|
86 |
$page->{ $name } = { |
87 |
content => convert_markup( $data ), |
88 |
date => convert_date( $date ), |
89 |
attachments => $attachments, |
90 |
}; |
91 |
|
92 |
$name =~ s,^.+/([^/]+)$,$1,; |
93 |
push @page_names, $name; |
94 |
|
95 |
}, |
96 |
no_chdir=>1, |
97 |
}, shift @ARGV || '.'); |
98 |
|
99 |
my @pages = ( keys %$page ); |
100 |
|
101 |
warn "found following pages: ", join(", ", @page_names),"\n"; |
102 |
|
103 |
my $page_link_re = '\b(' . join('|', @page_names) . ')\b'; |
104 |
|
105 |
my $Rester = Socialtext::Resting->new( |
106 |
username => 'tamtam', |
107 |
password => 'import', |
108 |
server => 'http://saturn.ffzg.hr/', |
109 |
workspace => 'razmjenavjestina', |
110 |
); |
111 |
$Rester->put_workspacetag('TamTam'); |
112 |
|
113 |
sub convert_date { |
114 |
my $date = shift; |
115 |
# return time2str( $date ); |
116 |
return strftime('%F %T %z', gmtime( $date )); |
117 |
} |
118 |
|
119 |
sub header { |
120 |
my $h = shift; |
121 |
if ( $h =~ m/^(=+)\s+(.+?)\s+\1$/ ) { |
122 |
my $level = length($1); |
123 |
return "\n" . ( '^' x $level ) . " $2\n"; |
124 |
} else { |
125 |
return $h; |
126 |
} |
127 |
} |
128 |
|
129 |
sub surround { |
130 |
my ( $with, $what ) = @_; |
131 |
return $with . $what . $with; |
132 |
} |
133 |
|
134 |
sub pre { |
135 |
my $text = shift; |
136 |
$text =~ s/^{{{//; |
137 |
$text =~ s/}}}$//; |
138 |
return '.pre' . $text . '.pre'; |
139 |
} |
140 |
|
141 |
sub convert_markup { |
142 |
my $body = shift; |
143 |
|
144 |
$body =~ s/\Q[[TableOfContents]]\E/{toc}/gs; |
145 |
$body =~ s/\Q[[BR]]\E/\n/gs; |
146 |
$body =~ s/$RE{balanced}{-begin => "= |== |=== |==== |===== |===== "}{-end => " =| ==| ===| ====| ====="}{-keep}/header($1)/gse; |
147 |
$body =~ s/''''(.+?)''''/surround('`',$1)/gse; |
148 |
$body =~ s/'''(.+?)'''/surround('*',$1)/gse; |
149 |
$body =~ s/''(.+?)''/surround('_',$1)/gse; |
150 |
$body =~ s/$RE{balanced}{-begin => "{{{"}{-end => "}}}"}{-keep}/pre($1)/gse; |
151 |
|
152 |
# fix bullets |
153 |
$body =~ s/^\s+([\*])/$1/gm; |
154 |
|
155 |
# fix links |
156 |
$body =~ s/\["([^"]+)"\]/[$1]/gs; |
157 |
$body =~ s,\[(http://\S+)\s+([^\]]+)\],"$2"<$1>,gs; |
158 |
$body =~ s,\[(http://[^\]]+)\],$1,gs; |
159 |
|
160 |
# fix hr |
161 |
$body =~ s,(\S+)----,$1\n----,gs; |
162 |
$body =~ s,----(\S+),----\n$1,gs; |
163 |
|
164 |
# attachments |
165 |
$body =~ s,\[attachment:([^\]]+)(gif|png|jpg|jpeg)\],{image: $1$2},gis; |
166 |
$body =~ s,\[attachment:([^\]]+)\],{file: $1},gs; |
167 |
|
168 |
return $body; |
169 |
} |
170 |
|
171 |
my $count = 0; |
172 |
|
173 |
my $m = File::MMagic::XS->new; |
174 |
|
175 |
foreach my $name ( keys %$page ) { |
176 |
last if $count++ == $max; |
177 |
|
178 |
my $p = $page->{$name}; |
179 |
|
180 |
warn "## $name = ",dump( $p ) if $debug; |
181 |
|
182 |
my $body = $p->{content} || die "no content?"; |
183 |
my $date = $p->{date} || die "no date?"; |
184 |
|
185 |
my @tags = ( 'TamTam' ); |
186 |
|
187 |
my $full_name = $name; |
188 |
|
189 |
if ( $name =~ m!/! ) { |
190 |
my @page_tags = split(m!/!, $name); |
191 |
$name = pop @page_tags; # remove page name |
192 |
push @tags, @page_tags; |
193 |
} |
194 |
|
195 |
# link named pages |
196 |
$body =~ s,\b$page_link_re\b,[$1],gs; |
197 |
$body =~ s,``,,gs; |
198 |
|
199 |
$body .= qq{ |
200 |
|
201 |
---- |
202 |
|
203 |
"original"<http://www.razmjenavjestina.org/$full_name> {date: $date} |
204 |
}; |
205 |
|
206 |
Encode::_utf8_off( $body ); |
207 |
|
208 |
$Rester->put_page( $name, { content => $body, date => $date }); |
209 |
print "$name $date\n"; |
210 |
foreach ( @tags ) { |
211 |
$Rester->put_pagetag( $name, $_ ); |
212 |
print "+ tag $_\n"; |
213 |
} |
214 |
|
215 |
if ( $attachments ) { |
216 |
foreach my $a ( @{ $p->{attachments} } ) { |
217 |
my $type = $m->get_mime( $a->{full_path} ); |
218 |
my $content = read_file( $a->{full_path} ); |
219 |
print "+ attachment ", $a->{name}," $type ", length($content), " bytes\n"; |
220 |
$Rester->post_attachment($name, $a->{name}, $content, $type ); |
221 |
} |
222 |
} |
223 |
|
224 |
} |
225 |
|