this repo has no description
Merge pull request #19 from websages/refactor
Refactor for easier future work
authored by
Michael Stahnke
and committed by
GitHub
8c02eb92
ee5de529
+304
-148
+25
.github/workflows/test.yml
+25
.github/workflows/test.yml
···
1
+
name: Run Tests
2
+
3
+
on:
4
+
push:
5
+
branches: [ "master", "main" ]
6
+
pull_request:
7
+
branches: [ "master", "main" ]
8
+
9
+
jobs:
10
+
test:
11
+
runs-on: ubuntu-latest
12
+
13
+
steps:
14
+
- uses: actions/checkout@v4
15
+
16
+
- name: Setup Perl environment
17
+
uses: shogo82148/actions-setup-perl@v1
18
+
with:
19
+
perl-version: '5.38'
20
+
21
+
- name: Install Dependencies
22
+
run: cpanm JSON LWP::Simple Test::More
23
+
24
+
- name: Run Tests
25
+
run: make test
+44
-148
htdocs/lib/tumble.pm
+44
-148
htdocs/lib/tumble.pm
···
5
5
use lsrfsh::MySQL;
6
6
7
7
use DBI;
8
-
use POSIX qw( strftime );
9
-
use Time::Local qw( timelocal timegm );
10
-
use Cwd qw( abs_path getcwd );
11
-
use File::Spec;
12
-
13
-
use YAML qw( LoadFile );
14
-
15
8
use strict;
16
9
use warnings;
10
+
11
+
use tumble::Content;
12
+
use YAML qw( LoadFile );
13
+
use Cwd qw( abs_path getcwd );
14
+
use File::Spec;
17
15
18
16
my $CONFIG = LoadFile( 'config.yaml' );
19
17
···
21
19
sub setup {
22
20
my $self = shift;
23
21
22
+
# Ensure STDOUT handles UTF-8 to prevent "Wide character in print" errors
23
+
binmode(STDOUT, ':encoding(UTF-8)');
24
+
24
25
$self->run_modes([qw/
25
26
displayTumble
26
27
/]);
···
40
41
}
41
42
42
43
$self->{'dbh'} = lsrfsh::MySQL->new( config => 'config.yaml' );
44
+
$self->{'content_processor'} = tumble::Content->new( config => $CONFIG );
43
45
44
46
$self->start_mode( 'displayTumble' );
45
47
···
77
79
78
80
my ( $c, $d, $date );
79
81
80
-
foreach my $item ( reverse sort { $a cmp $b } keys %{$data} ) {
81
-
my ( $content );
82
+
foreach my $item_id ( reverse sort { $a cmp $b } keys %{$data} ) {
83
+
my $item = $data->{$item_id};
84
+
85
+
# Delegate processing to Tumble::Content
86
+
# This handles date formatting, twitter/youtube embeds, title truncation etc.
87
+
my $processed = $self->{'content_processor'}->process_item( $item );
88
+
89
+
# Update the data hash with processed values
90
+
$data->{$item_id} = $processed;
91
+
my $formatted_timestamp = $processed->{'timestamp'};
92
+
my $content = $processed->{'content'};
82
93
83
-
if (
84
-
$data->{$item}->{'timestamp'} =~
85
-
/(\d{4})-(\d{2})-(\d{2})\s(\d{2}):(\d{2}):(\d{2})/
86
-
) {
87
-
# Parse the timestamp: $1=year, $2=month, $3=day, $4=hour, $5=minute, $6=second
88
-
my ($year, $month, $day, $hour, $minute, $second) = ($1, $2, $3, $4, $5, $6);
89
-
90
-
# Convert to epoch time (month is 0-based in timelocal)
91
-
# Note: timelocal interprets time as local time based on server timezone
92
-
my $epoch = timelocal($second, $minute, $hour, $day, $month - 1, $year - 1900);
93
-
94
-
# Format as RFC 822 date (RSS pubDate format)
95
-
# Get localtime components for the timestamp
96
-
my @lt = localtime($epoch);
97
-
98
-
# Calculate timezone offset by comparing GMT and local time
99
-
# timegm returns GMT epoch for given local time components
100
-
my $gmt_epoch = timegm(@lt);
101
-
my $local_epoch = timelocal(@lt);
102
-
my $tz_offset_seconds = $local_epoch - $gmt_epoch;
103
-
my $tz_offset_minutes = $tz_offset_seconds / 60;
104
-
105
-
# Format timezone offset as +HHMM or -HHMM
106
-
my $tz_sign = $tz_offset_minutes >= 0 ? '+' : '-';
107
-
my $tz_hours = abs(int($tz_offset_minutes / 60));
108
-
my $tz_mins = abs(int($tz_offset_minutes % 60));
109
-
my $tz_offset = sprintf("%s%02d%02d", $tz_sign, $tz_hours, $tz_mins);
94
+
# Date header logic
95
+
if ( defined $processed->{_date_components} ) {
96
+
my $comps = $processed->{_date_components};
110
97
111
-
$data->{$item}->{'timestamp'} =
112
-
POSIX::strftime("%a, %d %b %Y %H:%M:%S $tz_offset", @lt);
113
-
114
-
if ( ( !$d ) || ( $3 ne $d ) ) {
115
-
$d = $3;
116
-
117
-
$date->{'day'} = POSIX::strftime(
118
-
"%a", 0, $5, $4, $3, $2 - 1, $1 - 1900
119
-
);
120
-
$date->{'mon'} = POSIX::strftime(
121
-
"%b", 0, $5, $4, $3, $2 - 1, $1 - 1900
122
-
);
123
-
124
-
$c .= $self->wrap(
125
-
wrapper => 'tumble_date',
126
-
month => $date->{'mon'},
127
-
day => $date->{'day'},
128
-
date => $d
129
-
);
130
-
}
131
-
}
132
-
133
-
for ( $data->{$item}->{'type'} ) {
134
-
/ircLink/ && do {
135
-
if ( $data->{$item}->{'title'} =~ /^(http:\/\/.*)/ ) {
136
-
if ( length( $1 ) > 40 ) {
137
-
$data->{$item}->{'title'} = substr( $1, 0, 40 ) . '...';
138
-
}
139
-
}
140
-
141
-
my $link_filler = $data->{$item}->{'title'};
142
-
143
-
# fall back to normal linking of images if they could be nsfw
144
-
if (($data->{$item}->{'content_type'} =~ /image/) and ($data->{$item}->{'user'} !~ /nsfw|otd/)) {
145
-
$link_filler = '<img src="' . $data->{$item}->{'url'} . '">';
146
-
}
147
-
148
-
if ($data->{$item}->{'url'} =~ /twitter/) {
149
-
use LWP::Simple;
150
-
use JSON;
151
-
my @parts = split('/' , $data->{$item}->{'url'});
152
-
my $id = $parts[-1];
153
-
# This is so URIs like id/photos/1 don't try to call json
154
-
next if $id !~ /[0-9]+/;
155
-
next if $#parts > 6;
156
-
my $tw_uri = "https://api.twitter.com/1/statuses/oembed.json?id=" . $id;
157
-
my $tw_j = get( $tw_uri );
158
-
next unless $tw_j;
159
-
my $stuff = from_json($tw_j);
160
-
$link_filler = $stuff->{'html'};
161
-
}
162
-
163
-
# Handle YouTube URLs - extract video ID and create embed
164
-
my $is_youtube = 0;
165
-
if ($data->{$item}->{'url'} =~ /youtube\.com|youtu\.be/i) {
166
-
my $video_id;
167
-
my $url = $data->{$item}->{'url'};
168
-
169
-
# Handle various YouTube URL formats (case-insensitive, with or without www/protocol)
170
-
if ($url =~ /(?:youtube\.com\/watch\?v=|youtube\.com\/embed\/|youtu\.be\/)([a-zA-Z0-9_-]{11})/i) {
171
-
$video_id = $1;
172
-
} elsif ($url =~ /youtube\.com\/watch\?.*[&?]v=([a-zA-Z0-9_-]{11})/i) {
173
-
$video_id = $1;
174
-
}
175
-
176
-
if ($video_id) {
177
-
# Create responsive YouTube embed (standalone, not wrapped in link)
178
-
$content = '<div class="youtube-embed-wrapper">' .
179
-
'<iframe width="560" height="315" ' .
180
-
'src="https://www.youtube.com/embed/' . $video_id . '?rel=0" ' .
181
-
'frameborder="0" ' .
182
-
'allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" ' .
183
-
'allowfullscreen></iframe>' .
184
-
'</div>';
185
-
$is_youtube = 1;
186
-
}
187
-
}
188
-
189
-
unless ($is_youtube) {
190
-
$content =
191
-
'<a href="http://' . $CONFIG->{'baseurl'} .
192
-
qq{/irclink/?} .
193
-
$data->{$item}->{'ircLinkID'} .
194
-
qq{">} .
195
-
$link_filler .
196
-
qq{</a>};
197
-
}
98
+
if ( ( !$d ) || ( $comps->{'raw_day'} ne $d ) ) {
99
+
$d = $comps->{'raw_day'};
198
100
199
-
};
101
+
$date->{'day'} = $comps->{'day'};
102
+
$date->{'mon'} = $comps->{'mon'};
200
103
201
-
/image/ && do {
202
-
$content =
203
-
qq{<img src="} .
204
-
$data->{$item}->{'url'} .
205
-
qq{" alt="image" />};
206
-
};
207
-
208
-
/quote/ && do {
209
-
# For quote items, build description text
210
-
my $quote_text = $data->{$item}->{'quote'} || '';
211
-
my $author_text = $data->{$item}->{'author'} || '';
212
-
# Build the description text - will be escaped in wrap() function
213
-
$content = '"' . $quote_text . '" --' . $author_text;
214
-
};
104
+
$c .= $self->wrap(
105
+
wrapper => 'tumble_date',
106
+
month => $date->{'mon'},
107
+
day => $date->{'day'},
108
+
date => $d
109
+
);
110
+
}
215
111
}
216
112
113
+
my %template_vars = (
114
+
wrapper => 'tumble_item_' . $processed->{'type'},
115
+
author => $processed->{'user'},
116
+
baseurl => $CONFIG->{'baseurl'},
117
+
%{$processed}
118
+
);
119
+
120
+
# Add content or description depending on item type
217
121
# For XML/RSS feeds, wrap HTML content in CDATA sections (for ircLink and image items)
218
122
my $xml_content = $content;
219
-
if ( $self->{'arg'}->{'dtype'} =~ /xml|rss/ && defined $content && $content ne '' && $data->{$item}->{'type'} ne 'quote' ) {
123
+
if ( $self->{'arg'}->{'dtype'} =~ /xml|rss/ && defined $content && $content ne '' && $processed->{'type'} ne 'quote' ) {
220
124
# Wrap HTML content in CDATA for RSS descriptions (quote already has CDATA)
221
125
$xml_content = '<![CDATA[' . $content . ']]>';
222
126
}
223
127
224
-
my %template_vars = (
225
-
wrapper => 'tumble_item_' . $data->{$item}->{'type'},
226
-
author => $data->{$item}->{'user'},
227
-
baseurl => $CONFIG->{'baseurl'},
228
-
%{$data->{$item}}
229
-
);
230
-
231
-
# Add content or description depending on item type
232
-
if ( $data->{$item}->{'type'} eq 'quote' ) {
128
+
if ( $processed->{'type'} eq 'quote' ) {
233
129
# For quote items, pass description (will be escaped in wrap() function)
234
130
$template_vars{'description'} = $content if defined $content;
235
131
} elsif ( defined $xml_content ) {
+144
htdocs/lib/tumble/Content.pm
+144
htdocs/lib/tumble/Content.pm
···
1
+
package tumble::Content;
2
+
3
+
use strict;
4
+
use warnings;
5
+
6
+
use POSIX qw( strftime );
7
+
use Time::Local qw( timelocal timegm );
8
+
use LWP::Simple qw(get);
9
+
use JSON qw(from_json);
10
+
11
+
sub new {
12
+
my ($class, %args) = @_;
13
+
my $self = {
14
+
config => $args{config} || {},
15
+
fetcher => $args{fetcher} || \&LWP::Simple::get,
16
+
};
17
+
bless $self, $class;
18
+
return $self;
19
+
}
20
+
21
+
sub process_item {
22
+
my ($self, $item_in) = @_;
23
+
my $item = { %$item_in };
24
+
25
+
$self->_process_dates($item);
26
+
$self->_process_content($item);
27
+
28
+
return $item;
29
+
}
30
+
31
+
sub _process_dates {
32
+
my ($self, $item) = @_;
33
+
if ($item->{timestamp} && $item->{timestamp} =~ /(\d{4})-(\d{2})-(\d{2})\s(\d{2}):(\d{2}):(\d{2})/) {
34
+
my ($year, $month, $day, $hour, $minute, $second) = ($1, $2, $3, $4, $5, $6);
35
+
36
+
my $epoch = timelocal($second, $minute, $hour, $day, $month - 1, $year - 1900);
37
+
my @lt = localtime($epoch);
38
+
39
+
my $gmt_epoch = timegm(@lt);
40
+
my $local_epoch = timelocal(@lt);
41
+
my $tz_offset_seconds = $local_epoch - $gmt_epoch;
42
+
my $tz_offset_minutes = $tz_offset_seconds / 60;
43
+
44
+
my $tz_sign = $tz_offset_minutes >= 0 ? '+' : '-';
45
+
my $tz_hours = abs(int($tz_offset_minutes / 60));
46
+
my $tz_mins = abs(int($tz_offset_minutes % 60));
47
+
my $tz_offset = sprintf("%s%02d%02d", $tz_sign, $tz_hours, $tz_mins);
48
+
49
+
$item->{timestamp} = POSIX::strftime("%a, %d %b %Y %H:%M:%S $tz_offset", @lt);
50
+
51
+
# Capture date parts for grouping if needed
52
+
$item->{_date_components} = {
53
+
day => POSIX::strftime("%a", 0, $5, $4, $3, $2 - 1, $1 - 1900),
54
+
mon => POSIX::strftime("%b", 0, $5, $4, $3, $2 - 1, $1 - 1900),
55
+
raw_day => $3
56
+
};
57
+
}
58
+
}
59
+
60
+
sub _process_content {
61
+
my ($self, $item) = @_;
62
+
63
+
if ($item->{type} eq 'ircLink') {
64
+
$self->_process_irclink($item);
65
+
} elsif ($item->{type} eq 'image') {
66
+
$item->{content} = qq{<img src="} . $item->{url} . qq{" alt="image" />};
67
+
} elsif ($item->{type} eq 'quote') {
68
+
my $quote_text = $item->{quote} || '';
69
+
my $author_text = $item->{author} || '';
70
+
$item->{content} = '"' . $quote_text . '" --' . $author_text;
71
+
}
72
+
}
73
+
74
+
sub _process_irclink {
75
+
my ($self, $item) = @_;
76
+
77
+
# Title truncation
78
+
if ($item->{title} =~ /^(http:\/\/.*)/) {
79
+
if (length($1) > 40) {
80
+
$item->{title} = substr($1, 0, 40) . '...';
81
+
}
82
+
}
83
+
84
+
my $link_filler = $item->{title};
85
+
86
+
# Image content type check
87
+
if (($item->{content_type} && $item->{content_type} =~ /image/) &&
88
+
($item->{user} && $item->{user} !~ /nsfw|otd/)) {
89
+
$link_filler = '<img src="' . $item->{url} . '">';
90
+
}
91
+
92
+
my $is_youtube = 0;
93
+
94
+
# Twitter
95
+
if ($item->{url} && $item->{url} =~ /twitter/) {
96
+
my @parts = split('/', $item->{url});
97
+
my $id = $parts[-1];
98
+
# basic check
99
+
if ($id =~ /[0-9]+/) {
100
+
my $tw_uri = "https://api.twitter.com/1/statuses/oembed.json?id=" . $id;
101
+
my $tw_j = $self->{fetcher}->($tw_uri);
102
+
if ($tw_j) {
103
+
my $stuff = eval { from_json($tw_j) };
104
+
if ($stuff && $stuff->{html}) {
105
+
$link_filler = $stuff->{html};
106
+
}
107
+
}
108
+
}
109
+
}
110
+
111
+
# YouTube
112
+
if ($item->{url} && $item->{url} =~ /youtube\.com|youtu\.be/i) {
113
+
my $video_id;
114
+
my $url = $item->{url};
115
+
if ($url =~ /(?:youtube\.com\/watch\?v=|youtube\.com\/embed\/|youtu\.be\/)([a-zA-Z0-9_-]{11})/i) {
116
+
$video_id = $1;
117
+
} elsif ($url =~ /youtube\.com\/watch\?.*[&?]v=([a-zA-Z0-9_-]{11})/i) {
118
+
$video_id = $1;
119
+
}
120
+
121
+
if ($video_id) {
122
+
$item->{content} = '<div class="youtube-embed-wrapper">' .
123
+
'<iframe width="560" height="315" ' .
124
+
'src="https://www.youtube.com/embed/' . $video_id . '?rel=0" ' .
125
+
'frameborder="0" ' .
126
+
'allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" ' .
127
+
'allowfullscreen></iframe>' .
128
+
'</div>';
129
+
$is_youtube = 1;
130
+
}
131
+
}
132
+
133
+
unless ($is_youtube) {
134
+
my $baseurl = $self->{config}->{baseurl} || '';
135
+
$item->{content} = '<a href="http://' . $baseurl .
136
+
qq{/irclink/?} .
137
+
($item->{ircLinkID} || '') .
138
+
qq{">} .
139
+
$link_filler .
140
+
qq{</a>};
141
+
}
142
+
}
143
+
144
+
1;
+86
t/content.t
+86
t/content.t
···
1
+
use strict;
2
+
use warnings;
3
+
use Test::More;
4
+
use FindBin;
5
+
use lib "$FindBin::Bin/../htdocs/lib";
6
+
7
+
BEGIN { use_ok('tumble::Content') };
8
+
9
+
# Mock fetcher for Twitter
10
+
sub mock_fetcher {
11
+
my ($url) = @_;
12
+
if ($url =~ /api.twitter.com/) {
13
+
return '{"html": "<blockquote>Mock Tweet</blockquote>"}';
14
+
}
15
+
return undef;
16
+
}
17
+
18
+
my $config = { baseurl => 'tumble.example.com' };
19
+
my $processor = tumble::Content->new(
20
+
config => $config,
21
+
fetcher => \&mock_fetcher
22
+
);
23
+
24
+
subtest 'process_item: date formatting' => sub {
25
+
my $item = {
26
+
timestamp => '2023-10-27 10:00:00',
27
+
type => 'text',
28
+
title => 'Test Title',
29
+
};
30
+
31
+
my $processed = $processor->process_item($item);
32
+
33
+
ok($processed->{timestamp}, 'Timestamp converted');
34
+
like($processed->{timestamp}, qr/^\w+, \d+ \w+ \d{4}/, 'Timestamp looks like RFC 822');
35
+
};
36
+
37
+
subtest 'process_item: youtube embed' => sub {
38
+
my $item = {
39
+
type => 'ircLink',
40
+
url => 'https://www.youtube.com/watch?v=dQw4w9WgXcQ',
41
+
title => 'Rick Roll',
42
+
};
43
+
44
+
my $processed = $processor->process_item($item);
45
+
46
+
like($processed->{content}, qr/iframe/, 'Contains iframe for YouTube');
47
+
like($processed->{content}, qr/dQw4w9WgXcQ/, 'Contains video ID');
48
+
};
49
+
50
+
subtest 'process_item: twitter embed' => sub {
51
+
my $item = {
52
+
type => 'ircLink',
53
+
url => 'https://twitter.com/user/status/1234567890',
54
+
title => 'Tweet',
55
+
};
56
+
57
+
my $processed = $processor->process_item($item);
58
+
59
+
# logic: if twitter, content is the oembed html wrapped in a link?
60
+
# Original code: $link_filler = $stuff->{'html'};
61
+
# Then: $content = <a ...>$link_filler</a>
62
+
# Wait, the original code sets $link_filler.
63
+
64
+
# We need to verify that our new module produces the 'content' field similarly.
65
+
# But wait, original code constructs the <a href...> wrapper around $link_filler.
66
+
# So we expect the content to contain the mocked HTML "<blockquote>Mock Tweet</blockquote>"
67
+
68
+
like($processed->{content}, qr/Mock Tweet/, 'Contains mocked tweet content');
69
+
};
70
+
71
+
subtest 'process_item: normal link construction' => sub {
72
+
my $item = {
73
+
type => 'ircLink',
74
+
url => 'http://example.com',
75
+
title => 'Example',
76
+
ircLinkID => 123,
77
+
};
78
+
79
+
my $processed = $processor->process_item($item);
80
+
# original: <a href="http://$baseurl/irclink/?$id">$title</a>
81
+
82
+
like($processed->{content}, qr/href="http:\/\/tumble.example.com\/irclink\/\?123"/, 'Link constructed with baseurl');
83
+
like($processed->{content}, qr/>Example<\/a>/, 'Link text is title');
84
+
};
85
+
86
+
done_testing();