8 |
use Regexp::Common qw/balanced/; |
use Regexp::Common qw/balanced/; |
9 |
use Socialtext::Resting; |
use Socialtext::Resting; |
10 |
use Encode; |
use Encode; |
11 |
|
use HTTP::Date; |
12 |
use Data::Dump qw/dump/; |
use Data::Dump qw/dump/; |
13 |
|
|
14 |
|
my $max = 999; |
15 |
|
|
16 |
my $page; |
my $page; |
17 |
|
my $page_date; |
18 |
|
|
19 |
|
my @page_names; |
20 |
|
|
21 |
find({ |
find({ |
22 |
wanted => sub { |
wanted => sub { |
23 |
my $path = $File::Find::name; |
my $path = $File::Find::name; |
24 |
return unless -f $path; |
return unless -f $path; |
25 |
|
|
26 |
warn "# $path\n"; |
warn "# $path\n"; |
27 |
my $ref = XMLin( $path ) || die "can't open $path: $!"; |
my $ref = XMLin( $path ) || die "can't open $path: $!"; |
28 |
$page->{ $ref->{name} } = $ref->{widgets}->{widget}->{data}; |
|
29 |
warn dump( $ref->{widgets}->{widget}->{data} ); |
my $name = $ref->{name} || die "no name in $path"; |
30 |
|
|
31 |
|
return if $name =~ m/^TamSystem/; |
32 |
|
|
33 |
|
my $date = $ref->{meta}->{LastModified}->{value}; |
34 |
|
if ( ! $date ) { |
35 |
|
warn "SKIP: no LastModified in $path $name"; |
36 |
|
return; |
37 |
|
} |
38 |
|
my $data = |
39 |
|
$ref->{widgets}->{widget}->{data} || |
40 |
|
$ref->{widgets}->{widget}->{Body}->{data} || |
41 |
|
die "no data in $path ",dump( $ref ); |
42 |
|
|
43 |
|
$page->{ $name } = { |
44 |
|
content => convert_markup( $data ), |
45 |
|
date => time2str( $date ), |
46 |
|
}; |
47 |
|
|
48 |
|
# strip path from page name |
49 |
|
$name =~ s,^.+/([^/]+)$,$1,; |
50 |
|
push @page_names, $name; |
51 |
|
|
52 |
|
# warn dump( $ref ); |
53 |
}, |
}, |
54 |
}, shift @ARGV || '.'); |
}, shift @ARGV || '.'); |
55 |
|
|
56 |
|
my @pages = ( keys %$page ); |
57 |
|
|
58 |
warn "found following pages: ", join(", ", keys %$page),"\n"; |
warn "found following pages: ", join(", ", @page_names),"\n"; |
59 |
|
|
60 |
|
my $page_link_re = '\b(' . join('|', @page_names) . ')\b'; |
61 |
|
|
62 |
my $Rester = Socialtext::Resting->new( |
my $Rester = Socialtext::Resting->new( |
63 |
username => 'tamtam', |
username => 'tamtam', |
73 |
my $level = length($1); |
my $level = length($1); |
74 |
return "\n" . ( '^' x $level ) . " $2\n"; |
return "\n" . ( '^' x $level ) . " $2\n"; |
75 |
} else { |
} else { |
76 |
die "can't parse header: $h"; |
return $h; |
77 |
} |
} |
78 |
} |
} |
79 |
|
|
82 |
return $with . $what . $with; |
return $with . $what . $with; |
83 |
} |
} |
84 |
|
|
85 |
foreach my $name ( keys %$page ) { |
sub pre { |
86 |
my $body = $page->{$name} || die "no content for page $name"; |
my $text = shift; |
87 |
|
$text =~ s/^{{{//; |
88 |
|
$text =~ s/}}}$//; |
89 |
|
return '.pre' . $text . '.pre'; |
90 |
|
} |
91 |
|
|
92 |
|
sub convert_markup { |
93 |
|
my $body = shift; |
94 |
|
|
95 |
$body =~ s/\Q[[TableOfContents]]\E/{toc}/gs; |
$body =~ s/\Q[[TableOfContents]]\E/{toc}/gs; |
96 |
$body =~ s/\Q[[BR]]\E/\n/gs; |
$body =~ s/\Q[[BR]]\E/\n/gs; |
98 |
$body =~ s/''''(.+?)''''/surround('`',$1)/gse; |
$body =~ s/''''(.+?)''''/surround('`',$1)/gse; |
99 |
$body =~ s/'''(.+?)'''/surround('*',$1)/gse; |
$body =~ s/'''(.+?)'''/surround('*',$1)/gse; |
100 |
$body =~ s/''(.+?)''/surround('_',$1)/gse; |
$body =~ s/''(.+?)''/surround('_',$1)/gse; |
101 |
|
$body =~ s/$RE{balanced}{-begin => "{{{"}{-end => "}}}"}{-keep}/pre($1)/gse; |
102 |
|
|
103 |
|
# fix bullets |
104 |
|
$body =~ s/^\s+([\*])/$1/gm; |
105 |
|
|
106 |
|
# fix links |
107 |
|
$body =~ s/\["([^"]+)"\]/[$1]/gs; |
108 |
|
$body =~ s,\[(http://\S+)\s+([^\]]+)\],"$2"<$1>,gs; |
109 |
|
$body =~ s,\[(http://[^\]]+)\],$1,gs; |
110 |
|
|
111 |
|
# fix hr |
112 |
|
$body =~ s,(\S+)----,$1\n----,gs; |
113 |
|
$body =~ s,----(\S+),----\n$1,gs; |
114 |
|
|
115 |
|
return $body; |
116 |
|
} |
117 |
|
|
118 |
|
my $count = 0; |
119 |
|
|
120 |
|
foreach my $name ( keys %$page ) { |
121 |
|
last if $count++ == $max; |
122 |
|
|
123 |
|
my $p = $page->{$name}; |
124 |
|
my $body = $p->{content} || die "no content?"; |
125 |
|
my $date = $p->{date} || die "no date?"; |
126 |
|
|
127 |
|
my @tags = ( 'TamTam' ); |
128 |
|
|
129 |
|
if ( $name =~ m!/! ) { |
130 |
|
my @page_tags = split(m!/!, $name); |
131 |
|
$name = pop @page_tags; # remove page name |
132 |
|
push @tags, @page_tags; |
133 |
|
} |
134 |
|
|
135 |
|
# link named pages |
136 |
|
$body =~ s,$page_link_re,[$1],gs; |
137 |
|
|
138 |
|
$body .= qq{ |
139 |
|
---- |
140 |
|
|
141 |
|
Original: http://www.razmjenavjestina.org/$name |
142 |
|
}; |
143 |
|
|
144 |
Encode::_utf8_off( $body ); |
Encode::_utf8_off( $body ); |
145 |
|
|
146 |
$Rester->put_page( $name, $body ); |
$Rester->put_page( $name, { |
147 |
$Rester->put_pagetag( $name, 'TamTam' ); |
content => $body, |
148 |
|
date => $date, |
149 |
|
}); |
150 |
|
print "+ $name $date\n"; |
151 |
|
foreach ( @tags ) { |
152 |
|
$Rester->put_pagetag( $name, $_ ); |
153 |
|
print "+ $name [$_]\n"; |
154 |
|
} |
155 |
|
|
|
print "+ $name\n"; |
|
156 |
} |
} |
157 |
|
|