1
|
|
|
<?php declare(strict_types=1); defined('BASEPATH') OR exit('No direct script access allowed'); |
2
|
|
|
|
3
|
|
|
class Tracker_Sites_Model extends CI_Model { |
4
|
119 |
|
public function __construct() { |
5
|
119 |
|
parent::__construct(); |
6
|
119 |
|
} |
7
|
|
|
|
8
|
|
|
public function __get($name) { |
9
|
|
|
//TODO: Is this a good idea? There wasn't a good consensus on if this is good practice or not.. |
10
|
|
|
// It's probably a minor speed reduction, but that isn't much of an issue. |
11
|
|
|
// An alternate solution would simply have a function which generates a PHP file with code to load each model. Similar to: https://github.com/shish/shimmie2/blob/834bc740a4eeef751f546979e6400fd089db64f8/core/util.inc.php#L1422 |
12
|
|
|
if(!class_exists($name) || !(in_array(get_parent_class($name), ['Base_Site_Model', 'Base_FoolSlide_Site_Model']))) { |
13
|
|
|
return get_instance()->{$name}; |
14
|
|
|
} else { |
15
|
|
|
$this->loadSite($name); |
16
|
|
|
return $this->{$name}; |
17
|
|
|
} |
18
|
|
|
} |
19
|
|
|
|
20
|
|
|
private function loadSite(string $siteName) { |
21
|
|
|
$this->{$siteName} = new $siteName(); |
22
|
|
|
} |
23
|
|
|
} |
24
|
|
|
|
25
|
|
|
abstract class Base_Site_Model extends CI_Model { |
26
|
|
|
public $site = ''; |
27
|
|
|
public $titleFormat = ''; |
28
|
|
|
public $chapterFormat = ''; |
29
|
|
|
|
30
|
|
|
/** |
31
|
|
|
* 0: No custom updater. |
32
|
|
|
* 1: Uses following page. |
33
|
|
|
* 2: Uses latest releases page. |
34
|
|
|
*/ |
35
|
|
|
public $customType = 0; |
36
|
|
|
|
37
|
16 |
|
public function __construct() { |
38
|
16 |
|
parent::__construct(); |
39
|
|
|
|
40
|
16 |
|
$this->load->database(); |
41
|
|
|
|
42
|
16 |
|
$this->site = get_class($this); |
43
|
16 |
|
} |
44
|
|
|
|
45
|
|
|
abstract public function getFullTitleURL(string $title_url) : string; |
46
|
|
|
|
47
|
|
|
abstract public function getChapterData(string $title_url, string $chapter) : array; |
48
|
|
|
|
49
|
|
|
abstract public function getTitleData(string $title_url, bool $firstGet = FALSE) : ?array; |
50
|
|
|
|
51
|
2 |
|
final public function isValidTitleURL(string $title_url) : bool { |
52
|
2 |
|
$success = (bool) preg_match($this->titleFormat, $title_url); |
53
|
2 |
|
if(!$success) log_message('error', "Invalid Title URL ({$this->site}): {$title_url}"); |
54
|
2 |
|
return $success; |
55
|
|
|
} |
56
|
2 |
|
final public function isValidChapter(string $chapter) : bool { |
57
|
2 |
|
$success = (bool) preg_match($this->chapterFormat, $chapter); |
58
|
2 |
|
if(!$success) log_message('error', "Invalid Chapter ({$this->site}): {$chapter}"); |
59
|
2 |
|
return $success; |
60
|
|
|
} |
61
|
|
|
|
62
|
|
|
final protected function get_content(string $url, string $cookie_string = "", string $cookiejar_path = "", bool $follow_redirect = FALSE, bool $isPost = FALSE, array $postFields = []) { |
63
|
|
|
$ch = curl_init(); |
64
|
|
|
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1); |
65
|
|
|
curl_setopt($ch, CURLOPT_ENCODING , "gzip"); |
66
|
|
|
//curl_setopt($ch, CURLOPT_VERBOSE, 1); |
67
|
|
|
curl_setopt($ch, CURLOPT_HEADER, 1); |
68
|
|
|
|
69
|
|
|
if($follow_redirect) curl_setopt($ch, CURLOPT_FOLLOWLOCATION, TRUE); |
70
|
|
|
|
71
|
|
|
if(!empty($cookie_string)) curl_setopt($ch, CURLOPT_COOKIE, $cookie_string); |
72
|
|
|
if(!empty($cookiejar_path)) curl_setopt($ch, CURLOPT_COOKIEFILE, $cookiejar_path); |
73
|
|
|
|
74
|
|
|
//Some sites check the useragent for stuff, use a pre-defined user-agent to avoid stuff. |
75
|
|
|
curl_setopt($ch, CURLOPT_USERAGENT, 'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/54.0.2824.0 Safari/537.36'); |
76
|
|
|
|
77
|
|
|
//TODO: Check in a while if this being enabled still causes issues |
78
|
|
|
//curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, FALSE); //FIXME: This isn't safe, but it allows us to grab SSL URLs |
79
|
|
|
|
80
|
|
|
curl_setopt($ch, CURLOPT_URL, $url); |
81
|
|
|
|
82
|
|
|
if($isPost) { |
83
|
|
|
curl_setopt($ch,CURLOPT_POST, count($postFields)); |
84
|
|
|
curl_setopt($ch,CURLOPT_POSTFIELDS, http_build_query($postFields)); |
85
|
|
|
} |
86
|
|
|
|
87
|
|
|
$response = curl_exec($ch); |
88
|
|
|
if($response === FALSE) { |
89
|
|
|
log_message('error', "curl failed with error: ".curl_errno($ch)." | ".curl_error($ch)); |
90
|
|
|
//FIXME: We don't always account for FALSE return |
91
|
|
|
return FALSE; |
92
|
|
|
} |
93
|
|
|
|
94
|
|
|
$status_code = curl_getinfo($ch, CURLINFO_HTTP_CODE); |
95
|
|
|
$header_size = curl_getinfo($ch, CURLINFO_HEADER_SIZE); |
96
|
|
|
$headers = http_parse_headers(substr($response, 0, $header_size)); |
97
|
|
|
$body = substr($response, $header_size); |
98
|
|
|
curl_close($ch); |
99
|
|
|
|
100
|
|
|
return [ |
101
|
|
|
'headers' => $headers, |
102
|
|
|
'status_code' => $status_code, |
103
|
|
|
'body' => $body |
104
|
|
|
]; |
105
|
|
|
} |
106
|
|
|
|
107
|
|
|
/** |
108
|
|
|
* @param array $content |
109
|
|
|
* @param string $title_url |
110
|
|
|
* @param string $node_title_string |
111
|
|
|
* @param string $node_row_string |
112
|
|
|
* @param string $node_latest_string |
113
|
|
|
* @param string $node_chapter_string |
114
|
|
|
* @param string $failure_string |
115
|
|
|
* |
116
|
|
|
* @return DOMElement[]|false |
117
|
|
|
*/ |
118
|
|
|
final protected function parseTitleDataDOM( |
119
|
|
|
$content, string $title_url, |
120
|
|
|
string $node_title_string, string $node_row_string, |
121
|
|
|
string $node_latest_string, string $node_chapter_string, |
122
|
|
|
string $failure_string = "") { |
123
|
|
|
|
124
|
|
|
if(!is_array($content)) { |
125
|
|
|
log_message('error', "{$this->site} : {$title_url} | Failed to grab URL (See above curl error)"); |
126
|
|
|
} else { |
127
|
|
|
list('headers' => $headers, 'status_code' => $status_code, 'body' => $data) = $content; |
|
|
|
|
128
|
|
|
|
129
|
|
|
if(!($status_code >= 200 && $status_code < 300)) { |
130
|
|
|
log_message('error', "{$this->site} : {$title_url} | Bad Status Code ({$status_code})"); |
131
|
|
|
} else if(empty($data)) { |
132
|
|
|
log_message('error', "{$this->site} : {$title_url} | Data is empty? (Status code: {$status_code})"); |
133
|
|
|
} else if($failure_string !== "" && strpos($data, $failure_string) !== FALSE) { |
134
|
|
|
log_message('error', "{$this->site} : {$title_url} | Failure string matched"); |
135
|
|
|
} else { |
136
|
|
|
$data = $this->cleanTitleDataDOM($data); //This allows us to clean the DOM prior to parsing. It's faster to grab the only part we need THEN parse it. |
137
|
|
|
|
138
|
|
|
$dom = new DOMDocument(); |
139
|
|
|
libxml_use_internal_errors(TRUE); |
140
|
|
|
$dom->loadHTML('<?xml encoding="utf-8" ?>' . $data); |
141
|
|
|
libxml_use_internal_errors(FALSE); |
142
|
|
|
|
143
|
|
|
$xpath = new DOMXPath($dom); |
144
|
|
|
$nodes_title = $xpath->query($node_title_string); |
145
|
|
|
$nodes_row = $xpath->query($node_row_string); |
146
|
|
|
if($nodes_title->length === 1 && $nodes_row->length === 1) { |
147
|
|
|
$firstRow = $nodes_row->item(0); |
148
|
|
|
$nodes_latest = $xpath->query($node_latest_string, $firstRow); |
149
|
|
|
|
150
|
|
|
if($node_chapter_string !== '') { |
151
|
|
|
$nodes_chapter = $xpath->query($node_chapter_string, $firstRow); |
152
|
|
|
} else { |
153
|
|
|
$nodes_chapter = $nodes_row; |
154
|
|
|
} |
155
|
|
|
|
156
|
|
|
if($nodes_latest->length === 1 && $nodes_chapter->length === 1) { |
157
|
|
|
return [ |
158
|
|
|
'nodes_title' => $nodes_title->item(0), |
159
|
|
|
'nodes_latest' => $nodes_latest->item(0), |
160
|
|
|
'nodes_chapter' => $nodes_chapter->item(0) |
161
|
|
|
]; |
162
|
|
|
} else { |
163
|
|
|
log_message('error', "{$this->site} : {$title_url} | Invalid amount of nodes (LATEST: {$nodes_latest->length} | CHAPTER: {$nodes_chapter->length})"); |
164
|
|
|
} |
165
|
|
|
} else { |
166
|
|
|
log_message('error', "{$this->site} : {$title_url} | Invalid amount of nodes (TITLE: {$nodes_title->length} | ROW: {$nodes_row->length})"); |
167
|
|
|
} |
168
|
|
|
} |
169
|
|
|
} |
170
|
|
|
|
171
|
|
|
return FALSE; |
172
|
|
|
} |
173
|
|
|
|
174
|
|
|
public function cleanTitleDataDOM(string $data) : string { |
175
|
|
|
return $data; |
176
|
|
|
} |
177
|
|
|
|
178
|
|
|
final public function doCustomFollow(string $data = "", array $extra = []) : array { |
179
|
|
|
$titleData = []; |
180
|
|
|
$this->handleCustomFollow(function($content, $id, closure $successCallback = NULL) use(&$titleData) { |
181
|
|
|
if(is_array($content)) { |
182
|
|
|
if(array_key_exists('status_code', $content)) { |
183
|
|
|
$statusCode = $content['status_code']; |
184
|
|
|
if($statusCode === 200) { |
185
|
|
|
$isCallable = is_callable($successCallback); |
186
|
|
|
if(($isCallable && $successCallback($content['body'])) || !$isCallable) { |
187
|
|
|
$titleData['followed'] = 'Y'; |
188
|
|
|
|
189
|
|
|
log_message('info', "doCustomFollow succeeded for {$id}"); |
190
|
|
|
} else { |
191
|
|
|
log_message('error', "doCustomFollow failed (Invalid response?) for {$id}"); |
192
|
|
|
} |
193
|
|
|
} else { |
194
|
|
|
log_message('error', "doCustomFollow failed (Invalid status code ({$statusCode})) for {$id}"); |
195
|
|
|
} |
196
|
|
|
} else { |
197
|
|
|
log_message('error', "doCustomFollow failed (Missing status code?) for {$id}"); |
198
|
|
|
} |
199
|
|
|
} else { |
200
|
|
|
log_message('error', "doCustomFollow failed (Failed request) for {$id}"); |
201
|
|
|
} |
202
|
|
|
}, $data, $extra); |
203
|
|
|
return $titleData; |
204
|
|
|
} |
205
|
|
|
public function handleCustomFollow(callable $callback, string $data = "", array $extra = []) {} |
206
|
|
|
public function doCustomUpdate() {} |
207
|
|
|
public function doCustomCheck(string $oldChapter, string $newChapter) {} |
208
|
12 |
|
final public function doCustomCheckCompare(array $oldChapterSegments, array $newChapterSegments) : bool { |
209
|
|
|
//FIXME: Make this more generic when we have more site support for it. MangaFox and Batoto have similar chapter formats. |
210
|
|
|
|
211
|
|
|
//NOTE: We only need to check against the new chapter here, as that is what is used for confirming update. |
212
|
12 |
|
$status = FALSE; |
213
|
|
|
|
214
|
|
|
//Make sure we have a volume element |
215
|
12 |
|
if(count($oldChapterSegments) === 1) array_unshift($oldChapterSegments, 'v0'); |
216
|
12 |
|
if(count($newChapterSegments) === 1) array_unshift($newChapterSegments, 'v0'); |
217
|
|
|
|
218
|
12 |
|
$oldCount = count($oldChapterSegments); |
219
|
12 |
|
$newCount = count($newChapterSegments); |
220
|
12 |
|
if($newCount === $oldCount) { |
221
|
|
|
//Make sure chapter format looks correct. |
222
|
|
|
//NOTE: We only need to check newCount as we know oldCount is the same count. |
223
|
12 |
|
if($newCount === 2) { |
224
|
|
|
//FIXME: Can we loop this? |
225
|
12 |
|
$oldVolume = substr(array_shift($oldChapterSegments), 1); |
226
|
12 |
|
$newVolume = substr(array_shift($newChapterSegments), 1); |
227
|
|
|
|
228
|
|
|
//Forcing volume to 0 as TBD might not be the latest (although it can be, but that is covered by other checks) |
229
|
12 |
|
if(in_array($oldVolume, ['TBD', 'TBA', 'NA', 'LMT'])) $oldVolume = 0; |
230
|
12 |
|
if(in_array($newVolume, ['TBD', 'TBA', 'NA', 'LMT'])) $newVolume = 0; |
231
|
|
|
|
232
|
12 |
|
$oldVolume = floatval($oldVolume); |
233
|
12 |
|
$newVolume = floatval($newVolume); |
234
|
|
|
} else { |
235
|
|
|
$oldVolume = 0; |
236
|
|
|
$newVolume = 0; |
237
|
|
|
} |
238
|
12 |
|
$oldChapter = floatval(substr(array_shift($oldChapterSegments), 1)); |
239
|
12 |
|
$newChapter = floatval(substr(array_shift($newChapterSegments), 1)); |
240
|
|
|
|
241
|
12 |
|
if($newChapter > $oldChapter && ($oldChapter >= 10 && $newChapter >= 10)) { |
242
|
|
|
//$newChapter is higher than $oldChapter AND $oldChapter and $newChapter are both more than 10 |
243
|
|
|
//This is intended to cover the /majority/ of valid updates, as we technically shouldn't have to check volumes. |
244
|
|
|
|
245
|
4 |
|
$status = TRUE; |
246
|
8 |
|
} elseif($newVolume > $oldVolume && ($oldChapter < 10 && $newChapter < 10)) { |
247
|
|
|
//This is pretty much just to match a one-off case where the site doesn't properly increment chapter numbers across volumes, and instead does something like: v1/c1..v1/c5, v2/c1..v1/c5 (and so on). |
248
|
1 |
|
$status = TRUE; |
249
|
7 |
|
} elseif($newVolume > $oldVolume && $newChapter >= $oldChapter) { |
250
|
|
|
//$newVolume is higher, and chapter is higher so no need to check chapter. |
251
|
2 |
|
$status = TRUE; |
252
|
5 |
|
} elseif($newChapter > $oldChapter) { |
253
|
|
|
//$newVolume isn't higher, but chapter is. |
254
|
|
|
$status = TRUE; |
255
|
|
|
} |
256
|
|
|
} |
257
|
|
|
|
258
|
12 |
|
return $status; |
259
|
|
|
} |
260
|
|
|
} |
261
|
|
|
|
262
|
|
|
abstract class Base_FoolSlide_Site_Model extends Base_Site_Model { |
263
|
|
|
public $titleFormat = '/^[a-z0-9_-]+$/'; |
264
|
|
|
public $chapterFormat = '/^en\/[0-9]+(?:\/[0-9]+(?:\/[0-9]+(?:\/[0-9]+)?)?)?$/'; |
265
|
|
|
public $customType = 2; |
266
|
|
|
|
267
|
|
|
public $baseURL = ''; |
268
|
|
|
|
269
|
|
|
public function getFullTitleURL(string $title_url) : string { |
270
|
|
|
return "{$this->baseURL}/series/{$title_url}"; |
271
|
|
|
} |
272
|
|
|
|
273
|
|
|
public function getChapterData(string $title_url, string $chapter) : array { |
274
|
|
|
$chapter_parts = explode('/', $chapter); //returns #LANG#/#VOLUME#/#CHAPTER#/#CHAPTER_EXTRA#(/#PAGE#/) |
275
|
|
|
return [ |
276
|
|
|
'url' => "{$this->baseURL}/read/{$title_url}/{$chapter}/", |
277
|
|
|
'number' => ($chapter_parts[1] !== '0' ? "v{$chapter_parts[1]}/" : '') . "c{$chapter_parts[2]}" . (isset($chapter_parts[3]) ? ".{$chapter_parts[3]}" : '')/*)*/ |
278
|
|
|
]; |
279
|
|
|
} |
280
|
|
|
|
281
|
|
|
public function getTitleData(string $title_url, bool $firstGet = FALSE) : ?array { |
282
|
|
|
$titleData = []; |
283
|
|
|
|
284
|
|
|
$jsonURL = "{$this->baseURL}/api/reader/comic/stub/{$title_url}/format/json"; |
285
|
|
|
if($content = $this->get_content($jsonURL)) { |
286
|
|
|
$json = json_decode($content['body'], TRUE); |
287
|
|
|
if($json && isset($json['chapters']) && count($json['chapters']) > 0) { |
288
|
|
|
$titleData['title'] = trim($json['comic']['name']); |
289
|
|
|
|
290
|
|
|
//FoolSlide title API doesn't appear to let you sort (yet every other API method which has chapters does, so we need to sort ourselves.. |
291
|
|
|
usort($json['chapters'], function($a, $b) { |
292
|
|
|
return floatval("{$b['chapter']['chapter']}.{$b['chapter']['subchapter']}") <=> floatval("{$a['chapter']['chapter']}.{$a['chapter']['subchapter']}"); |
293
|
|
|
}); |
294
|
|
|
$latestChapter = reset($json['chapters'])['chapter']; |
295
|
|
|
|
296
|
|
|
$latestChapterString = "en/{$latestChapter['volume']}/{$latestChapter['chapter']}"; |
297
|
|
|
if($latestChapter['subchapter'] !== '0') { |
298
|
|
|
$latestChapterString .= "/{$latestChapter['subchapter']}"; |
299
|
|
|
} |
300
|
|
|
$titleData['latest_chapter'] = $latestChapterString; |
301
|
|
|
|
302
|
|
|
//No need to use date() here since this is already formatted as such. |
303
|
|
|
$titleData['last_updated'] = ($latestChapter['updated'] !== '0000-00-00 00:00:00' ? $latestChapter['updated'] : $latestChapter['created']); |
304
|
|
|
} |
305
|
|
|
} |
306
|
|
|
|
307
|
|
|
return (!empty($titleData) ? $titleData : NULL); |
308
|
|
|
} |
309
|
|
|
|
310
|
|
|
//Since we're just checking the latest updates page and not a following page, we just need to simulate a follow. |
311
|
|
|
//TODO: It would probably be better to have some kind of var which says that the custom update uses a following page.. |
312
|
|
|
public function handleCustomFollow(callable $callback, string $data = "", array $extra = []) { |
313
|
|
|
$content = ['status_code' => 200]; |
314
|
|
|
$callback($content, $extra['id']); |
315
|
|
|
} |
316
|
|
|
public function doCustomUpdate() { |
317
|
|
|
$titleDataList = []; |
318
|
|
|
|
319
|
|
|
//NOTE: desc_created might not work well for sites that delay their releases (HelveticaScans for example) |
320
|
|
|
// There shouldn't be much issue unless a site has a delay on a chapter long than it takes for 20 chapters to release (unlikely). |
321
|
|
|
$jsonURL = "{$this->baseURL}/api/reader/chapters/orderby/desc_created/format/json"; |
322
|
|
|
if(($content = $this->get_content($jsonURL)) && $content['status_code'] == 200) { |
323
|
|
|
$json = json_decode($content['body'], TRUE); |
324
|
|
|
|
325
|
|
|
$parsedTitles = []; |
326
|
|
|
foreach($json['chapters'] as $chapterData) { |
327
|
|
|
if(!in_array($chapterData['comic']['stub'], $parsedTitles)) { |
328
|
|
|
$parsedTitles[] = $chapterData['comic']['stub']; |
329
|
|
|
|
330
|
|
|
$titleData = []; |
331
|
|
|
$titleData['title'] = trim($chapterData['comic']['name']); |
332
|
|
|
|
333
|
|
|
$latestChapter = $chapterData['chapter']; |
334
|
|
|
|
335
|
|
|
$latestChapterString = "en/{$latestChapter['volume']}/{$latestChapter['chapter']}"; |
336
|
|
|
if($latestChapter['subchapter'] !== '0') { |
337
|
|
|
$latestChapterString .= "/{$latestChapter['subchapter']}"; |
338
|
|
|
} |
339
|
|
|
$titleData['latest_chapter'] = $latestChapterString; |
340
|
|
|
|
341
|
|
|
//No need to use date() here since this is already formatted as such. |
342
|
|
|
$titleData['last_updated'] = ($latestChapter['updated'] !== '0000-00-00 00:00:00' ? $latestChapter['updated'] : $latestChapter['created']); |
343
|
|
|
|
344
|
|
|
$titleDataList[$chapterData['comic']['stub']] = $titleData; |
345
|
|
|
} else { |
346
|
|
|
//We already have title data for this title. |
347
|
|
|
continue; |
348
|
|
|
} |
349
|
|
|
} |
350
|
|
|
} else { |
351
|
|
|
log_message('error', "Custom updating failed for {$this->baseURL}."); |
352
|
|
|
} |
353
|
|
|
|
354
|
|
|
return $titleDataList; |
355
|
|
|
} |
356
|
|
|
public function doCustomCheck(string $oldChapterString, string $newChapterString) { |
357
|
|
|
$oldChapterSegments = explode('/', $this->getChapterData('', $oldChapterString)['number']); |
358
|
|
|
$newChapterSegments = explode('/', $this->getChapterData('', $newChapterString)['number']); |
359
|
|
|
|
360
|
|
|
$status = $this->doCustomCheckCompare($oldChapterSegments, $newChapterSegments); |
361
|
|
|
|
362
|
|
|
return $status; |
363
|
|
|
} |
364
|
|
|
} |
365
|
|
|
|
This checks looks for assignemnts to variables using the
list(...)
function, where not all assigned variables are subsequently used.Consider the following code example.
Only the variables
$a
and$c
are used. There was no need to assign$b
.Instead, the list call could have been.