Commit | Line | Data |
95148a72 |
1 | package SCSite::PageSet; |
2 | |
95148a72 |
3 | use Text::MultiMarkdown 'markdown'; |
4 | use HTML::Zoom; |
ebd4c292 |
5 | use Sub::Quote; |
6 | use Syntax::Keyword::Gather; |
7 | use SCSite::Page; |
d01538bc |
8 | use IO::All; |
2a19b2a6 |
9 | use Try::Tiny; |
d01538bc |
10 | use JSON; |
11 | use Moo; |
12 | |
13 | { |
14 | my $j = JSON->new; |
15 | sub _json { $j } |
16 | } |
95148a72 |
17 | |
34597fb2 |
18 | has top_dir => (is => 'ro', lazy => 1, builder => 'base_dir'); |
95148a72 |
19 | has base_dir => (is => 'ro', required => 1); |
ebd4c292 |
20 | has max_depth => (is => 'ro', default => quote_sub q{ 0 }); |
221c4151 |
21 | has min_depth => (is => 'ro', default => quote_sub q{ 1 }); |
95148a72 |
22 | |
34597fb2 |
23 | has rel_path => (is => 'lazy'); |
24 | |
25 | sub _build_rel_path { |
26 | my ($self) = @_; |
27 | io->dir('/') |
28 | ->catdir(File::Spec->abs2rel($self->base_dir->name, $self->top_dir->name)) |
29 | } |
30 | |
95148a72 |
31 | sub get { |
32 | my ($self, $spec) = @_; |
33 | $spec->{path} or die "path is required to get"; |
34 | my ($dir, $file) = $spec->{path} =~ m{^(?:(.*)/)?([^/]+)$}; |
35 | my $type; |
5cc6d9e2 |
36 | my @poss = io->dir($self->base_dir)->${\sub { |
95148a72 |
37 | my $io = shift; |
38 | defined($dir) ? $io->catdir($dir) : $io |
ebd4c292 |
39 | }}->filter(sub { |
40 | $_->filename =~ /^\Q${file}\E${\$self->_types_re}$/ and $type = $1 |
41 | }) |
221c4151 |
42 | ->${\sub { -e "$_[0]" ? $_[0]->all_files : () }}; |
95148a72 |
43 | die "multiple files found for ${\$spec->{path}}:\n".join "\n", @poss |
44 | if @poss > 1; |
ebd4c292 |
45 | return undef unless @poss; |
2a19b2a6 |
46 | $self->_inflate( |
d01538bc |
47 | $type, $self->rel_path->catdir($spec->{path}), $poss[0] |
34597fb2 |
48 | ); |
ebd4c292 |
49 | } |
50 | |
2a19b2a6 |
51 | sub _inflate { |
d01538bc |
52 | my ($self, $type, $path, $io) = @_; |
53 | (my $cache_name = $io->name) =~ s/\/([^\/]+)$/\/.htcache.$1.json/; |
54 | my $cache = io($cache_name); |
55 | if (-f $cache_name) { |
56 | if ($cache->mtime >= $io->mtime) { |
57 | return $self->_new_page($path, $self->_json->decode($cache->all)); |
58 | } |
59 | } |
60 | my $raw = $io->all; |
61 | try { |
62 | my $extracted = $self->${\"_extract_from_${type}"}($raw); |
63 | $cache->print($self->_json->encode($extracted)); |
64 | $self->_new_page($path, $extracted); |
65 | } catch { |
66 | die "Error inflating ${path} as ${type}: $_\n\nData was: ${raw}"; |
2a19b2a6 |
67 | } |
68 | } |
69 | |
ebd4c292 |
70 | sub map { |
71 | my ($self, $mapper) = @_; |
72 | [ map $mapper->($_), $self->flatten ] |
73 | } |
74 | |
fc436d2e |
75 | sub _depth_under_base { |
76 | my ($self, $path) = @_; |
77 | File::Spec->splitdir(File::Spec->abs2rel($path, $self->base_dir->name)) |
78 | } |
79 | |
ebd4c292 |
80 | sub flatten { |
81 | my ($self) = @_; |
fc436d2e |
82 | return unless (my $base = $self->base_dir)->exists; |
ebd4c292 |
83 | my %seen; |
34597fb2 |
84 | my $slash = io->dir('/'); |
fc436d2e |
85 | my $min = $self->min_depth; |
ebd4c292 |
86 | map { |
87 | my ($path, $type) = $_->name =~ /^(.*)${\$self->_types_re}$/; |
2a19b2a6 |
88 | $self->_inflate( |
89 | $type, |
fc436d2e |
90 | $slash->catdir(File::Spec->abs2rel($path, $self->top_dir->name)), |
d01538bc |
91 | $_ |
ebd4c292 |
92 | ); |
fc436d2e |
93 | } map { |
94 | $_->filter(sub { $_->filename =~ /${\$self->_types_re}$/ }) |
221c4151 |
95 | ->all_files($self->max_depth - ($min-1)) |
fc436d2e |
96 | } map |
221c4151 |
97 | $min > 1 |
fc436d2e |
98 | ? do { |
221c4151 |
99 | # can't use ->all_dirs($min-1) since we only want the final level |
100 | my @x = ($_); @x = map $_->all_dirs, @x for 1..$min-1; @x |
fc436d2e |
101 | } |
102 | : $_, |
221c4151 |
103 | $base; |
ebd4c292 |
104 | } |
105 | |
106 | sub latest { |
107 | my ($self, $max) = @_; |
108 | require SCSite::LatestPageSet; |
109 | SCSite::LatestPageSet->new( |
110 | parent => $self, |
111 | max_entries => $max, |
112 | ); |
95148a72 |
113 | } |
114 | |
115 | sub _new_page { |
ebd4c292 |
116 | SCSite::Page->new({ path => $_[1], page_set => $_[0], %{$_[2]} }) |
95148a72 |
117 | } |
118 | |
ebd4c292 |
119 | sub _types_re { qw/\.(html|md)/ } |
120 | |
95148a72 |
121 | sub _extract_from_html { |
122 | my ($self, $html) = @_; |
123 | HTML::Zoom->from_html($html) |
124 | ->select('title')->collect_content({ into => \my @title }) |
2a19b2a6 |
125 | ->select('meta[name=subtitle]')->collect({ into => \my @subtitle }) |
95148a72 |
126 | ->select('meta[name=description]')->collect({ into => \my @description }) |
127 | ->select('meta[name=keywords]')->collect({ into => \my @keywords }) |
ebd4c292 |
128 | ->select('meta[name=created]')->collect({ into => \my @created }) |
95148a72 |
129 | ->select('body')->collect_content({ into => \my @body }) |
130 | ->run; |
131 | +{ |
ebd4c292 |
132 | title => $title[0]->{raw}||'', |
2a19b2a6 |
133 | subtitle => $subtitle[0]->{attrs}{content}||'', |
ebd4c292 |
134 | description => $description[0]->{attrs}{content}||'', |
135 | keywords => $keywords[0]->{attrs}{content}||'', |
136 | created => $created[0]->{attrs}{content}||'', |
137 | body => HTML::Zoom->from_events(\@body)->to_html||'', |
95148a72 |
138 | } |
139 | } |
140 | |
95148a72 |
141 | sub _extract_from_md { |
142 | my ($self, $md) = @_; |
143 | $self->_extract_from_html(markdown($md, { document_format => 'complete' })); |
144 | } |
145 | |
146 | 1; |