DakuTree /
manga-tracker
Complex classes like Base_Site_Model often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes. You can also have a look at the cohesion graph to spot any un-connected, or weakly-connected components.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use Base_Site_Model, and based on these observations, apply Extract Interface, too.
| 1 | <?php declare(strict_types=1); defined('BASEPATH') OR exit('No direct script access allowed'); |
||
| 28 | abstract class Base_Site_Model extends CI_Model { |
||
| 29 | public $site = ''; |
||
| 30 | public $titleFormat = '//'; |
||
| 31 | public $chapterFormat = '//'; |
||
| 32 | |||
| 33 | /** |
||
| 34 | * 0: No custom updater. |
||
| 35 | * 1: Uses following page. |
||
| 36 | * 2: Uses latest releases page. |
||
| 37 | */ |
||
| 38 | public $customType = 0; |
||
| 39 | |||
| 40 | 16 | public function __construct() { |
|
| 47 | |||
| 48 | /** |
||
| 49 | * Generates URL to the title page of the requested series. |
||
| 50 | * |
||
| 51 | * NOTE: In some cases, we are required to store more data in the title_string than is needed to generate the URL. (Namely as the title_string is our unique identifier for that series) |
||
| 52 | * When storing additional data, we use ':--:' as a delimiter to separate the data. Make sure to handle this as needed. |
||
| 53 | * |
||
| 54 | * Example: |
||
| 55 | * return "http://mangafox.me/manga/{$title_url}/"; |
||
| 56 | * |
||
| 57 | * Example (with extra data): |
||
| 58 | * $title_parts = explode(':--:', title_url); |
||
| 59 | * return "https://bato.to/comic/_/comics/-r".$title_parts[0]; |
||
| 60 | * |
||
| 61 | * @param string $title_url |
||
| 62 | * @return string |
||
| 63 | */ |
||
| 64 | abstract public function getFullTitleURL(string $title_url) : string; |
||
| 65 | |||
| 66 | /** |
||
| 67 | * Generates chapter data from given $title_url and $chapter. |
||
| 68 | * |
||
| 69 | * Chapter must be in a (v[0-9]+/)?c[0-9]+(\..+)? format. |
||
| 70 | * |
||
| 71 | * NOTE: In some cases, we are required to store the chapter number, and the segment required to generate the chapter URL separately. |
||
| 72 | * Much like when generating the title URL, we use ':--:' as a delimiter to separate the data. Make sure to handle this as needed. |
||
| 73 | * |
||
| 74 | * Example: |
||
| 75 | * return [ |
||
| 76 | * 'url' => $this->getFullTitleURL($title_url).'/'.$chapter, |
||
| 77 | * 'number' => "c{$chapter}" |
||
| 78 | * ]; |
||
| 79 | * |
||
| 80 | * @param string $title_url |
||
| 81 | * @param string $chapter |
||
| 82 | * @return array [url, number] |
||
| 83 | */ |
||
| 84 | abstract public function getChapterData(string $title_url, string $chapter) : array; |
||
| 85 | |||
| 86 | /** |
||
| 87 | * Used to get the latest chapter of given $title_url. |
||
| 88 | * |
||
| 89 | * This <should> utilize both get_content and parseTitleDataDOM functions when possible, as these can both reduce a lot of the code required to set this up. |
||
| 90 | * |
||
| 91 | * $titleData params must be set accordingly: |
||
| 92 | * * `title` should always be used with html_entity_decode. |
||
| 93 | * * `latest_chapter` must match $this->chapterFormat. |
||
| 94 | * * `last_updated` should always be in date("Y-m-d H:i:s") format. |
||
| 95 | * * `followed` should never be set within via getTitleData, with the exception of via a array_merge with doCustomFollow. |
||
| 96 | * |
||
| 97 | * $firstGet is set to true when the series is first added to the DB, and is used to follow the series on given site (if possible). |
||
| 98 | * |
||
| 99 | * @param string $title_url |
||
| 100 | * @param bool $firstGet |
||
| 101 | * @return array|null [title,latest_chapter,last_updated,followed?] |
||
| 102 | */ |
||
| 103 | abstract public function getTitleData(string $title_url, bool $firstGet = FALSE) : ?array; |
||
| 104 | |||
| 105 | /** |
||
| 106 | * Validates given $title_url against titleFormat. |
||
| 107 | * |
||
| 108 | * Failure to match against titleFormat will stop the series from being added to the DB. |
||
| 109 | * |
||
| 110 | * @param string $title_url |
||
| 111 | * @return bool |
||
| 112 | */ |
||
| 113 | 2 | final public function isValidTitleURL(string $title_url) : bool { |
|
| 118 | |||
| 119 | /** |
||
| 120 | * Validates given $chapter against chapterFormat. |
||
| 121 | * |
||
| 122 | * Failure to match against chapterFormat will stop the chapter being updated. |
||
| 123 | * |
||
| 124 | * @param string $chapter |
||
| 125 | * @return bool |
||
| 126 | */ |
||
| 127 | 2 | final public function isValidChapter(string $chapter) : bool { |
|
| 132 | |||
| 133 | /** |
||
| 134 | * Used by getTitleData (& similar functions) to get the requested page data. |
||
| 135 | * |
||
| 136 | * @param string $url |
||
| 137 | * @param string $cookie_string |
||
| 138 | * @param string $cookiejar_path |
||
| 139 | * @param bool $follow_redirect |
||
| 140 | * @param bool $isPost |
||
| 141 | * @param array $postFields |
||
| 142 | * |
||
| 143 | * @return array|bool |
||
| 144 | */ |
||
| 145 | final protected function get_content(string $url, string $cookie_string = "", string $cookiejar_path = "", bool $follow_redirect = FALSE, bool $isPost = FALSE, array $postFields = []) { |
||
| 146 | $ch = curl_init(); |
||
| 147 | curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1); |
||
| 148 | curl_setopt($ch, CURLOPT_ENCODING , "gzip"); |
||
| 149 | //curl_setopt($ch, CURLOPT_VERBOSE, 1); |
||
| 150 | curl_setopt($ch, CURLOPT_HEADER, 1); |
||
| 151 | |||
| 152 | if($follow_redirect) curl_setopt($ch, CURLOPT_FOLLOWLOCATION, TRUE); |
||
| 153 | |||
| 154 | if(!empty($cookie_string)) curl_setopt($ch, CURLOPT_COOKIE, $cookie_string); |
||
| 155 | if(!empty($cookiejar_path)) curl_setopt($ch, CURLOPT_COOKIEFILE, $cookiejar_path); |
||
| 156 | |||
| 157 | //Some sites check the useragent for stuff, use a pre-defined user-agent to avoid stuff. |
||
| 158 | curl_setopt($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'); |
||
| 159 | |||
| 160 | //NOTE: This is required for SSL URLs for now. Without it we tend to get error code 60. |
||
| 161 | curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, FALSE); //FIXME: This isn't safe, but it allows us to grab SSL URLs |
||
| 162 | |||
| 163 | curl_setopt($ch, CURLOPT_URL, $url); |
||
| 164 | |||
| 165 | if($isPost) { |
||
| 166 | curl_setopt($ch,CURLOPT_POST, count($postFields)); |
||
| 167 | curl_setopt($ch,CURLOPT_POSTFIELDS, http_build_query($postFields)); |
||
| 168 | } |
||
| 169 | |||
| 170 | $response = curl_exec($ch); |
||
| 171 | if($response === FALSE) { |
||
| 172 | log_message('error', "curl failed with error: ".curl_errno($ch)." | ".curl_error($ch)); |
||
| 173 | //FIXME: We don't always account for FALSE return |
||
| 174 | return FALSE; |
||
| 175 | } |
||
| 176 | |||
| 177 | $status_code = curl_getinfo($ch, CURLINFO_HTTP_CODE); |
||
| 178 | $header_size = curl_getinfo($ch, CURLINFO_HEADER_SIZE); |
||
| 179 | $headers = http_parse_headers(substr($response, 0, $header_size)); |
||
| 180 | $body = substr($response, $header_size); |
||
| 181 | curl_close($ch); |
||
| 182 | |||
| 183 | return [ |
||
| 184 | 'headers' => $headers, |
||
| 185 | 'status_code' => $status_code, |
||
| 186 | 'body' => $body |
||
| 187 | ]; |
||
| 188 | } |
||
| 189 | |||
| 190 | /** |
||
| 191 | * Used by getTitleData to get the title, latest_chapter & last_updated data from the data returned by get_content. |
||
| 192 | * |
||
| 193 | * parseTitleDataDOM checks if the data returned by get_content is valid via a few simple checks. |
||
| 194 | * * If the request was actually successful, had a valid status code & data wasn't empty. We also do an additional check on an optional $failure_string param, which will throw a failure if it's matched. |
||
| 195 | * |
||
| 196 | * Data is cleaned by cleanTitleDataDOM prior to being passed to DOMDocument. |
||
| 197 | * |
||
| 198 | * All $node_* params must be XPath to the requested node, and must only return 1 result. Anything else will throw a failure. |
||
| 199 | * |
||
| 200 | * @param array $content |
||
| 201 | * @param string $title_url |
||
| 202 | * @param string $node_title_string |
||
| 203 | * @param string $node_row_string |
||
| 204 | * @param string $node_latest_string |
||
| 205 | * @param string $node_chapter_string |
||
| 206 | * @param string $failure_string |
||
| 207 | * @return DOMElement[]|false [nodes_title,nodes_chapter,nodes_latest] |
||
| 208 | */ |
||
| 209 | final protected function parseTitleDataDOM( |
||
| 264 | |||
| 265 | /** |
||
| 266 | * Used by parseTitleDataDOM to clean the data prior to passing it to DOMDocument & DOMXPath. |
||
| 267 | * This is mostly done as an (assumed) speed improvement due to the reduced amount of DOM to parse, or simply just making it easier to parse with XPath. |
||
| 268 | * |
||
| 269 | * @param string $data |
||
| 270 | * @return string |
||
| 271 | */ |
||
| 272 | public function cleanTitleDataDOM(string $data) : string { |
||
| 275 | |||
| 276 | /** |
||
| 277 | * Used to follow a series on given site if supported. |
||
| 278 | * |
||
| 279 | * This is called by getTitleData if $firstGet is true (which occurs when the series is first being added to the DB). |
||
| 280 | * |
||
| 281 | * Most of the actual following is done by handleCustomFollow. |
||
| 282 | * |
||
| 283 | * @param string $data |
||
| 284 | * @param array $extra |
||
| 285 | * @return array |
||
| 286 | */ |
||
| 287 | final public function doCustomFollow(string $data = "", array $extra = []) : array { |
||
| 314 | |||
| 315 | /** |
||
| 316 | * Used by doCustomFollow to handle following series on sites. |
||
| 317 | * |
||
| 318 | * Uses get_content to get data. |
||
| 319 | * |
||
| 320 | * $callback must return ($content, $id, closure $successCallback = NULL). |
||
| 321 | * * $content is simply just the get_content data. |
||
| 322 | * * $id is the dbID. This should be passed by the $extra arr. |
||
| 323 | * * $successCallback is an optional success check to make sure the series was properly followed. |
||
| 324 | * |
||
| 325 | * @param callable $callback |
||
| 326 | * @param string $data |
||
| 327 | * @param array $extra |
||
| 328 | */ |
||
| 329 | public function handleCustomFollow(callable $callback, string $data = "", array $extra = []) {} |
||
| 330 | |||
| 331 | /** |
||
| 332 | * Used to check the sites following page for new updates (if supported). |
||
| 333 | * This should work much like getTitleData, but instead checks the following page. |
||
| 334 | * |
||
| 335 | * This must return an array containing arrays of each of the chapters data. |
||
| 336 | */ |
||
| 337 | public function doCustomUpdate() {} |
||
| 338 | |||
| 339 | /** |
||
| 340 | * Used by the custom updater to check if a chapter looks newer than the current one. |
||
| 341 | * |
||
| 342 | * This calls doCustomCheckCompare which handles the majority of the checking. |
||
| 343 | * NOTE: Depending on the site, you may need to call getChapterData to get the chapter number to be used with this. |
||
| 344 | * |
||
| 345 | * @param string $oldChapter |
||
| 346 | * @param string $newChapter |
||
| 347 | */ |
||
| 348 | public function doCustomCheck(string $oldChapter, string $newChapter) {} |
||
| 349 | |||
| 350 | /** |
||
| 351 | * Used by doCustomCheck to check if a chapter looks newer than the current one. |
||
| 352 | * Chapter must be in a (v[0-9]+/)?c[0-9]+(\..+)? format. |
||
| 353 | * |
||
| 354 | * To avoid issues with the occasional off case, this will only ever return true if we are 100% sure that the new chapter is newer than the old one. |
||
| 355 | * |
||
| 356 | * @param array $oldChapterSegments |
||
| 357 | * @param array $newChapterSegments |
||
| 358 | * @return bool |
||
| 359 | */ |
||
| 360 | 12 | final public function doCustomCheckCompare(array $oldChapterSegments, array $newChapterSegments) : bool { |
|
| 410 | } |
||
| 411 | |||
| 520 |
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
$aand$care used. There was no need to assign$b.Instead, the list call could have been.