|
|
@@ -1,249 +0,0 @@
|
|
1
|
|
-<?php
|
|
2
|
|
-
|
|
3
|
|
-/* load config. You normally don't want to edit something here */
|
|
4
|
|
-require_once 'config.php';
|
|
5
|
|
-$instance = $config['mastodon-instance'];
|
|
6
|
|
-$uid = $config['user-id'];
|
|
7
|
|
-$searchurl = $config['search-url'];
|
|
8
|
|
-$search = isset($_GET['search']) ? strtolower($_GET['search']) : '';
|
|
9
|
|
-$debug_on = $config['debug'];
|
|
10
|
|
-/* cache files */
|
|
11
|
|
-$ctt = $config['cache_toots'];
|
|
12
|
|
-$dbt = "cache-toots.json";
|
|
13
|
|
-$ctc = $config['cache_comments'];
|
|
14
|
|
-$dbc = "cache-comments_%id.json";
|
|
15
|
|
-
|
|
16
|
|
-/* Exit if search empty */
|
|
17
|
|
-if (empty($search)) {
|
|
18
|
|
- debug("No proper search given");
|
|
19
|
|
- die();
|
|
20
|
|
-}
|
|
21
|
|
-
|
|
22
|
|
-/* MISC FUNCTIONS */
|
|
23
|
|
-function debug($data) {
|
|
24
|
|
- global $debug_on;
|
|
25
|
|
- if ($debug_on === true) {
|
|
26
|
|
- error_log("[getcomments.php] " . print_r($data, TRUE));
|
|
27
|
|
- }
|
|
28
|
|
-}
|
|
29
|
|
-
|
|
30
|
|
-/* CACHE FUNCTIONS */
|
|
31
|
|
-/* write data to file */
|
|
32
|
|
-function write_db($db, $data, $id) {
|
|
33
|
|
- // if $id is given, it's a comments file. Replace placeholder in filename
|
|
34
|
|
- if ($id) {
|
|
35
|
|
- $db = str_replace('%id', $id, $db);
|
|
36
|
|
- }
|
|
37
|
|
- $file['toots'] = $data;
|
|
38
|
|
- $file['timestamp'] = time();
|
|
39
|
|
- // encode and write file
|
|
40
|
|
- $encoded = json_encode($file, JSON_PRETTY_PRINT);
|
|
41
|
|
- file_put_contents($db, $encoded, LOCK_EX);
|
|
42
|
|
-}
|
|
43
|
|
-/* delete file */
|
|
44
|
|
-function delete_db($db, $id) {
|
|
45
|
|
- // if $id is given, it's a comments file. Replace placeholder in filename
|
|
46
|
|
- if ($id) {
|
|
47
|
|
- $db = str_replace('%id', $id, $db);
|
|
48
|
|
- }
|
|
49
|
|
- unlink($db);
|
|
50
|
|
-}
|
|
51
|
|
-/* access data from file */
|
|
52
|
|
-function read_db($db, &$data, $cachetime, &$cachebreak, $id) {
|
|
53
|
|
- // if $id is given, it's a comments file. Replace placeholder in filename
|
|
54
|
|
- if ($id) {
|
|
55
|
|
- $db = str_replace('%id', $id, $db);
|
|
56
|
|
- }
|
|
57
|
|
- // if DB does not exist, create it with empty array
|
|
58
|
|
- if (! file_exists($db)) {
|
|
59
|
|
- // if $data empty (usually with $toots, not with comment's $result), populate with empty array
|
|
60
|
|
- if (empty($data)) {
|
|
61
|
|
- $data = array();
|
|
62
|
|
- }
|
|
63
|
|
- touch($db);
|
|
64
|
|
- write_db($db, $data, $id);
|
|
65
|
|
- $cachebreak = true;
|
|
66
|
|
- }
|
|
67
|
|
- $file = file_get_contents($db, true);
|
|
68
|
|
- $data = json_decode($file, true);
|
|
69
|
|
-
|
|
70
|
|
- // check if timestamp in cache file too old
|
|
71
|
|
- if (empty($data['timestamp']) || ($data['timestamp'] + $cachetime < time())) {
|
|
72
|
|
- $cachebreak = true;
|
|
73
|
|
- }
|
|
74
|
|
-
|
|
75
|
|
- $data = $data['toots'];
|
|
76
|
|
-}
|
|
77
|
|
-
|
|
78
|
|
-/* TOOT FUNCTIONS */
|
|
79
|
|
-function collectToots($instance, $uid, $min_id, $searchurl) {
|
|
80
|
|
- $raw = file_get_contents("$instance/api/v1/accounts/$uid/statuses?exclude_reblogs=true&exclude_replies=true&limit=50&min_id=$min_id");
|
|
81
|
|
- $json_complete = json_decode($raw, true);
|
|
82
|
|
- $json = array();
|
|
83
|
|
- foreach ($json_complete as $toot) {
|
|
84
|
|
- $json[] = array('id' => $toot['id'], 'date' => $toot['created_at'] ,'url' => analyzeToot($instance, $toot['id'], $searchurl));
|
|
85
|
|
- }
|
|
86
|
|
- return($json);
|
|
87
|
|
-}
|
|
88
|
|
-/* Find out if a toot contains the searched URL */
|
|
89
|
|
-function analyzeToot($instance, $id, $searchurl) {
|
|
90
|
|
- debug("Searching for $searchurl in $id");
|
|
91
|
|
- $raw = file_get_contents("$instance/api/v1/statuses/$id");
|
|
92
|
|
- $json = json_decode($raw, true);
|
|
93
|
|
-
|
|
94
|
|
- // search for $searchurl inside of <a> tags, until (and excluding) a "
|
|
95
|
|
- preg_match("|$searchurl.+?(?=\")|i", $json['content'], $matches);
|
|
96
|
|
-
|
|
97
|
|
- if(!empty($matches)) {
|
|
98
|
|
- return(strtolower($matches[0])); // take first match inside toot
|
|
99
|
|
- } else {
|
|
100
|
|
- return("");
|
|
101
|
|
- }
|
|
102
|
|
-}
|
|
103
|
|
-/* of context, extract the interesting bits */
|
|
104
|
|
-function filterComments($descendants, $root, &$result) {
|
|
105
|
|
- // go through each comment
|
|
106
|
|
- foreach ($descendants as $d) {
|
|
107
|
|
- $result['comments'][$d['id']] = [
|
|
108
|
|
- 'author' => [
|
|
109
|
|
- 'display_name' => $d['account']['display_name'] ? $d['account']['display_name'] : $d['account']['username'],
|
|
110
|
|
- 'avatar' => $d['account']['avatar_static'],
|
|
111
|
|
- 'url' => $d['account']['url']
|
|
112
|
|
- ],
|
|
113
|
|
- 'toot' => $d['content'],
|
|
114
|
|
- 'date' => $d['created_at'],
|
|
115
|
|
- 'url' => $d['uri'],
|
|
116
|
|
- 'reply_to' => $d['in_reply_to_id'],
|
|
117
|
|
- 'root' => $root,
|
|
118
|
|
- ];
|
|
119
|
|
- }
|
|
120
|
|
- return $result;
|
|
121
|
|
-}
|
|
122
|
|
-/* get /context of toot */
|
|
123
|
|
-function tootContext($instance, $id, &$result) {
|
|
124
|
|
- $raw = file_get_contents("$instance/api/v1/statuses/$id/context");
|
|
125
|
|
- $json = json_decode($raw, true);
|
|
126
|
|
- filterComments($json['descendants'], $id, $result);
|
|
127
|
|
-}
|
|
128
|
|
-/* extract stats info from toot */
|
|
129
|
|
-function filterStats($stats) {
|
|
130
|
|
- $result = [
|
|
131
|
|
- 'reblogs' => (int)$stats['reblogs_count'],
|
|
132
|
|
- 'favs' => (int)$stats['favourites_count'],
|
|
133
|
|
- 'replies' => (int)$stats['replies_count'],
|
|
134
|
|
- 'url' => $stats['url']
|
|
135
|
|
- ];
|
|
136
|
|
- return $result;
|
|
137
|
|
-}
|
|
138
|
|
-/* for toot, extract interesting statistics */
|
|
139
|
|
-function tootStats($instance, $id, &$result) {
|
|
140
|
|
- debug("Checking ID $id");
|
|
141
|
|
- $raw = file_get_contents("$instance/api/v1/statuses/$id");
|
|
142
|
|
- $json = json_decode($raw, true);
|
|
143
|
|
- $newStats = filterStats($json);
|
|
144
|
|
- $result['stats']['reblogs'] += $newStats['reblogs'];
|
|
145
|
|
- $result['stats']['favs'] += $newStats['favs'];
|
|
146
|
|
- $result['stats']['replies'] += $newStats['replies'];
|
|
147
|
|
- if (empty($result['stats']['url'])) {
|
|
148
|
|
- $result['stats']['url'] = $newStats['url'];
|
|
149
|
|
- }
|
|
150
|
|
-}
|
|
151
|
|
-
|
|
152
|
|
-/***************
|
|
153
|
|
- * START PROGRAM
|
|
154
|
|
- ***************/
|
|
155
|
|
-
|
|
156
|
|
-/* check whether the cached file containing all toots is older than max. cache time */
|
|
157
|
|
-// this at the same time loads the cached DB, either way
|
|
158
|
|
-$cachebreak = false;
|
|
159
|
|
-read_db($dbt, $toots, $ctt, $cachebreak, false);
|
|
160
|
|
-
|
|
161
|
|
-if ($cachebreak) {
|
|
162
|
|
- /* Collect all the toots */
|
|
163
|
|
- /* get id of latest cached toot, and set as $min_id */
|
|
164
|
|
- debug("Toots cache outdated. Checking for new toots");
|
|
165
|
|
- if (!empty($toots['0']['id'])) {
|
|
166
|
|
- $min_id_cached = $toots['0']['id'];
|
|
167
|
|
- $min_id = $min_id_cached;
|
|
168
|
|
- } else {
|
|
169
|
|
- /* if cached toots do not exist, start from oldest toot */
|
|
170
|
|
- $min_id = "0";
|
|
171
|
|
- $min_id_cached = "0";
|
|
172
|
|
- }
|
|
173
|
|
-
|
|
174
|
|
- /* test whether there are new toots available */
|
|
175
|
|
- // Search for toots older than the cached latest toot ID ($min_id)
|
|
176
|
|
- $uptodate = false;
|
|
177
|
|
- while ($uptodate === false) {
|
|
178
|
|
- $toots = array_merge(collectToots($instance, $uid, $min_id, $searchurl), $toots);
|
|
179
|
|
- $min_id_new = $toots['0']['id']; // the latest ID of the recent search
|
|
180
|
|
-
|
|
181
|
|
- if ($min_id_new === $min_id) {
|
|
182
|
|
- // min_id is the latest, let's write the new DB and end this loop
|
|
183
|
|
- $uptodate = true;
|
|
184
|
|
- debug("Toots up-to-date. Rewrite cache DB.");
|
|
185
|
|
- write_db($dbt, $toots, false);
|
|
186
|
|
- } else {
|
|
187
|
|
- // next round looks for toots newer than the newly found ID
|
|
188
|
|
- debug("Newer toots than in cache found. Starting another search for new toots");
|
|
189
|
|
- $min_id = $min_id_new;
|
|
190
|
|
- }
|
|
191
|
|
- }
|
|
192
|
|
-} else {
|
|
193
|
|
- debug("Toots cache is up-to-date");
|
|
194
|
|
-}
|
|
195
|
|
-
|
|
196
|
|
-// create empty $result
|
|
197
|
|
-$result_empty = ['comments' => [], 'stats' => ['reblogs' => 0, 'favs' => 0, 'replies' => 0, 'url' => '', 'root' => 0]];
|
|
198
|
|
-$result = $result_empty;
|
|
199
|
|
-
|
|
200
|
|
-/* check if URL from $search exists in $toots */
|
|
201
|
|
-$id = array_keys(
|
|
202
|
|
- array_filter(
|
|
203
|
|
- array_column($toots, 'url'),
|
|
204
|
|
- function ($value) use ($search) {
|
|
205
|
|
- return (strpos($value, $search) !== false);
|
|
206
|
|
- }
|
|
207
|
|
- )
|
|
208
|
|
-);
|
|
209
|
|
-if (empty($id)) {
|
|
210
|
|
- debug("Blog URL \"$search\" has not been found");
|
|
211
|
|
-} else {
|
|
212
|
|
- // if multiple toots with the searched URL exist, take the oldest one (largest array index)
|
|
213
|
|
- $id = $toots[end($id)]['id'];
|
|
214
|
|
-
|
|
215
|
|
- /* read cached comments, or reload new comments if cached data too old */
|
|
216
|
|
- $cachebreak = false;
|
|
217
|
|
- read_db($dbc, $result, $ctc, $cachebreak, $id);
|
|
218
|
|
-
|
|
219
|
|
- if ($cachebreak) {
|
|
220
|
|
- debug("Comments cache for $id outdated. Checking for new comments");
|
|
221
|
|
- // delete old cache file, otherwise the stats would add up
|
|
222
|
|
- delete_db($dbc, $id);
|
|
223
|
|
- // re-create empty $result and new cache file
|
|
224
|
|
- $result = $result_empty;
|
|
225
|
|
- read_db($dbc, $result, $ctc, $cachebreak, $id);
|
|
226
|
|
- /* Extract comments and stats from toot */
|
|
227
|
|
- tootContext($instance, $id, $result);
|
|
228
|
|
- tootStats($instance, $id, $result);
|
|
229
|
|
- // FIXME: At the moment the API doesn't return the correct replies count so I count it manually
|
|
230
|
|
- $result['stats']['replies'] = count($result['comments']);
|
|
231
|
|
- $result['stats']['root'] = $id;
|
|
232
|
|
-
|
|
233
|
|
- write_db($dbc, $result, $id);
|
|
234
|
|
- } else {
|
|
235
|
|
- debug("Comments cache for $id up-to-date. Returning cached comments");
|
|
236
|
|
- }
|
|
237
|
|
-}
|
|
238
|
|
-
|
|
239
|
|
-// headers for not caching the results
|
|
240
|
|
-header('Cache-Control: no-cache, must-revalidate');
|
|
241
|
|
-header('Expires: Mon, 26 Jul 1997 05:00:00 GMT');
|
|
242
|
|
-
|
|
243
|
|
-// headers to tell that result is JSON
|
|
244
|
|
-header('Content-type: application/json');
|
|
245
|
|
-
|
|
246
|
|
-// actually output result as JSON, to be digested by getcomments.js
|
|
247
|
|
-echo json_encode($result);
|
|
248
|
|
-
|
|
249
|
|
-?>
|
max.mehl
3 weeks ago