Dispositif /
Wikibot
| Total Complexity | 45 |
| Total Lines | 224 |
| Duplicated Lines | 0 % |
| Changes | 1 | ||
| Bugs | 0 | Features | 0 |
Complex classes like RefWebTransformer 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.
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 RefWebTransformer, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 29 | class RefWebTransformer implements TransformerInterface |
||
| 30 | { |
||
| 31 | |||
| 32 | public $skipUnauthorised = true; |
||
| 33 | /** |
||
| 34 | * @var Logger |
||
| 35 | */ |
||
| 36 | protected $log; |
||
| 37 | private $config; |
||
| 38 | /** |
||
| 39 | * @var string|string[] |
||
| 40 | */ |
||
| 41 | private $domain; |
||
| 42 | /** |
||
| 43 | * @var string |
||
| 44 | */ |
||
| 45 | private $url; |
||
| 46 | /** |
||
| 47 | * @var WebMapper |
||
| 48 | */ |
||
| 49 | private $mapper; |
||
| 50 | /** |
||
| 51 | * @var array |
||
| 52 | */ |
||
| 53 | private $data = []; |
||
| 54 | /** |
||
| 55 | * @var array |
||
| 56 | */ |
||
| 57 | public $summaryLog = []; |
||
| 58 | |||
| 59 | /** |
||
| 60 | * RefWebTransformer constructor. |
||
| 61 | * |
||
| 62 | * @param LoggerInterface $log |
||
| 63 | */ |
||
| 64 | public function __construct(LoggerInterface $log) |
||
| 65 | { |
||
| 66 | $this->log = $log; |
||
|
1 ignored issue
–
show
|
|||
| 67 | |||
| 68 | $this->config = Yaml::parseFile(__DIR__.'/resources/config_presse.yaml'); |
||
| 69 | $this->data['newspaper'] = json_decode(file_get_contents(__DIR__.'/resources/data_newspapers.json'), true); |
||
| 70 | $this->data['scientific domain'] = json_decode( |
||
| 71 | file_get_contents(__DIR__.'/resources/data_scientific_domain.json'), |
||
| 72 | true |
||
| 73 | ); |
||
| 74 | $this->data['scientific wiki'] = json_decode( |
||
| 75 | file_get_contents(__DIR__.'/resources/data_scientific_wiki.json'), |
||
| 76 | true |
||
| 77 | ); |
||
| 78 | |||
| 79 | $this->mapper = new WebMapper(new Logger()); |
||
| 80 | } |
||
| 81 | |||
| 82 | public function process(string $string) |
||
| 83 | { |
||
| 84 | if (!$this->isTransformAutorized($string)) { |
||
| 85 | return $string; |
||
| 86 | } |
||
| 87 | |||
| 88 | $publish = new PublisherAction($this->url); |
||
| 89 | try { |
||
| 90 | sleep(3); |
||
| 91 | $html = $publish->getHTMLSource(); |
||
| 92 | $htmlData = $publish->extractWebData($html); |
||
| 93 | $this->log->debug('htmlData', $htmlData); |
||
| 94 | } catch (Throwable $e) { |
||
| 95 | // ne pas générer de lien brisé !! |
||
| 96 | $this->log->notice('erreur sur extractWebData'); |
||
| 97 | |||
| 98 | return $string; |
||
| 99 | } |
||
| 100 | |||
| 101 | $mapData = $this->mapper->process($htmlData); |
||
| 102 | |||
| 103 | // check dataValide |
||
| 104 | if (empty($mapData) || empty($mapData['url']) || empty($mapData['titre'])) { |
||
| 105 | $this->log->info('Mapping incomplet'); |
||
| 106 | |||
| 107 | return $string; |
||
| 108 | } |
||
| 109 | |||
| 110 | $this->tagAndLog($mapData); |
||
| 111 | $this->addSummaryLog($mapData); |
||
| 112 | |||
| 113 | $template = $this->chooseTemplateByData($mapData); |
||
| 114 | |||
| 115 | $mapData = $this->replaceSitenameByConfig($mapData, $template); |
||
| 116 | |||
| 117 | $template->hydrate($mapData, true); |
||
| 118 | |||
| 119 | $serialized = $template->serialize(true); |
||
| 120 | $this->log->info($serialized."\n"); |
||
| 121 | |||
| 122 | return $serialized; |
||
| 123 | } |
||
| 124 | |||
| 125 | protected function isTransformAutorized(string $string): bool |
||
| 126 | { |
||
| 127 | if (!preg_match('#^http?s://[^ ]+$#i', $string)) { |
||
| 128 | return false; |
||
| 129 | } |
||
| 130 | $this->url = $string; |
||
| 131 | $parseURL = parse_url($this->url); |
||
| 132 | $this->domain = str_replace('www.', '', $parseURL['host']); |
||
| 133 | |||
| 134 | if (!isset($this->config[$this->domain])) { |
||
| 135 | $this->log->info("Domain ".$this->domain." non configuré\n"); |
||
| 136 | if ($this->skipUnauthorised) { |
||
| 137 | return false; |
||
| 138 | } |
||
| 139 | echo "> Domaine ".Color::LIGHT_RED.$this->domain.Color::NORMAL." non configuré\n"; |
||
| 140 | } |
||
| 141 | |||
| 142 | $this->config[$this->domain] = $this->config[$this->domain] ?? []; |
||
| 143 | $this->config[$this->domain] = is_array($this->config[$this->domain]) ? $this->config[$this->domain] : []; |
||
| 144 | |||
| 145 | if ($this->config[$this->domain] === 'desactived' || isset($this->config[$this->domain]['desactived'])) { |
||
| 146 | $this->log->info("Domain ".$this->domain." desactivé\n"); |
||
| 147 | |||
| 148 | return false; |
||
| 149 | } |
||
| 150 | |||
| 151 | return true; |
||
| 152 | } |
||
| 153 | |||
| 154 | private function tagAndLog(array $mapData) |
||
| 155 | { |
||
| 156 | $this->log->debug('mapData', $mapData); |
||
| 157 | |||
| 158 | if (isset($mapData['DATA-ARTICLE']) && $mapData['DATA-ARTICLE']) { |
||
| 159 | $this->log->notice("Article OK"); |
||
| 160 | } |
||
| 161 | if (isset($this->data['newspaper'][$this->domain])) { |
||
| 162 | $this->log->notice('PRESSE'); |
||
| 163 | } |
||
| 164 | if ($this->isScientificDomain()) { |
||
| 165 | $this->log->notice('SCIENCE'); |
||
| 166 | } |
||
| 167 | } |
||
| 168 | |||
| 169 | private function isScientificDomain(): bool |
||
| 170 | { |
||
| 171 | if (isset($this->data['scientific domain'][$this->domain])) { |
||
| 172 | return true; |
||
| 173 | } |
||
| 174 | if (strpos('.revues.org', $this->domain) > 0) { |
||
| 175 | return true; |
||
| 176 | } |
||
| 177 | |||
| 178 | return false; |
||
| 179 | } |
||
| 180 | |||
| 181 | private function addSummaryLog(array $mapData) |
||
| 182 | { |
||
| 183 | $this->summaryLog[] = $mapData['site'] ?? $mapData['périodique'] ?? '?'; |
||
| 184 | } |
||
| 185 | |||
| 186 | /** |
||
| 187 | * todo refac lisible |
||
| 188 | */ |
||
| 189 | private function chooseTemplateByData(array $mapData): AbstractWikiTemplate |
||
| 209 | } |
||
| 210 | |||
| 211 | /** |
||
| 212 | * Logique : remplacement titre périodique ou nom du site |
||
| 213 | * |
||
| 214 | * @param array $mapData |
||
| 215 | * @param $template |
||
| 216 | * |
||
| 217 | * @return array |
||
| 218 | */ |
||
| 219 | private function replaceSitenameByConfig(array $mapData, $template): array |
||
| 256 |
Our type inference engine has found a suspicous assignment of a value to a property. This check raises an issue when a value that can be of a given class or a super-class is assigned to a property that is type hinted more strictly.
Either this assignment is in error or an instanceof check should be added for that assignment.