-
Notifications
You must be signed in to change notification settings - Fork 1
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
There were a bit changes to pass SensioLabs test.
- Loading branch information
Ivanov Dmitry
committed
Jun 19, 2017
1 parent
9b9f7b5
commit ce32936
Showing
8 changed files
with
482 additions
and
486 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,190 +1,190 @@ | ||
| <?php | ||
| /** | ||
| * Content parser and detector from different webservices | ||
| */ | ||
| namespace Deepslam\ContentParser; | ||
|
|
||
| abstract class ContentParser { | ||
| /** | ||
| * Error constants | ||
| * | ||
| * @const string | ||
| */ | ||
| const ERR_WRONG_URL = 'No URL has been received'; | ||
|
|
||
| /** | ||
| * Parsing constants | ||
| * | ||
| * @const boolean | ||
| */ | ||
| const PARSING_DONE = true; | ||
| const PARSING_FAIL = false; | ||
|
|
||
| /** | ||
| * Registered parsers | ||
| * | ||
| * @var array | ||
| */ | ||
| private static $parsers = []; | ||
|
|
||
| /** | ||
| * Url for parse | ||
| * | ||
| * @var string | ||
| */ | ||
| private $url = ''; | ||
|
|
||
| /** | ||
| * These are received URL params. | ||
| * | ||
| * @var array | ||
| */ | ||
| private $urlParams = []; | ||
|
|
||
| /** | ||
| * Stored latest parsing result | ||
| * | ||
| * @var array | ||
| */ | ||
| private $parsingResult = []; | ||
|
|
||
| /** | ||
| * Object constructor | ||
| * | ||
| * @return ContentParser | ||
| * @throw new Exception if no url has been received | ||
| */ | ||
| public function __construct() { | ||
| $this->parsingResult = new ParsingResult(); | ||
| } | ||
|
|
||
| /** | ||
| * Returns URL | ||
| * | ||
| * @return string Working URL | ||
| */ | ||
| public function getURL() { | ||
| return $this->url; | ||
| } | ||
|
|
||
| /** | ||
| * Returns parsed URL params such as host, path etc... | ||
| * | ||
| * @return array | ||
| */ | ||
| public function getURLParams() { | ||
| return $this->urlParams; | ||
| } | ||
|
|
||
| /** | ||
| * Returns parsing result | ||
| * | ||
| * @return ParsingResult Link to the instance of parsing result | ||
| */ | ||
| public function getResult() { | ||
| return $this->parsingResult; | ||
| } | ||
|
|
||
| /*** PROTECTED LAYER ***/ | ||
|
|
||
| /** | ||
| * Checks whether have to do clean code or not. | ||
| * | ||
| * @return bool True - need to clean code, false - needn't to clean code | ||
| */ | ||
| final protected function needsCodeClean() { | ||
| return ((bool)config('deepslam.parser.clean_code')); | ||
| } | ||
|
|
||
| /** | ||
| * Checks whether have to strip tags or not. | ||
| * | ||
| * @return bool True - need to strip tags, false - needn't to strip tags | ||
| */ | ||
| final protected function needsCodeStrip() { | ||
| return ((bool)config('deepslam.parser.strip_tags')); | ||
| } | ||
|
|
||
| /*** STATIC LAYER ***/ | ||
|
|
||
| /** | ||
| * Factory method for creation needle parser class | ||
| * | ||
| * @param String $url | ||
| * @param $parser | ||
| * | ||
| * @return ContentParserMercury The link to the created instance | ||
| */ | ||
| public static function create(String $parser = '') { | ||
| if (isset(self::$parsers[$parser])) { | ||
| $workParser = self::$parsers[$parser]; | ||
| } else { | ||
| $workParser = ContentParserGraby::class; | ||
| } | ||
| return new $workParser(); | ||
| } | ||
|
|
||
| /** | ||
| * Registers new grabber | ||
| * | ||
| * @param String $alias | ||
| * @param String $className | ||
| * | ||
| * @return nothing | ||
| */ | ||
| public static function register(Array $params) { | ||
| if (count($params) == 0) { | ||
| return ; | ||
| } | ||
| foreach ($params as $alias => $className) { | ||
| self::$parsers[$alias] = $className; | ||
| } | ||
| } | ||
|
|
||
| /** | ||
| * Parses the needle URL | ||
| * Throws exception if has been received wrong URL | ||
| * | ||
| * @param String $url URL for parsing | ||
| * | ||
| * @return @ParsingResult Link to the parsing result | ||
| */ | ||
| public function parse($url) { | ||
| if (!filter_var($url, FILTER_VALIDATE_URL)) { | ||
| throw new Exception(self::ERR_WRONG_URL); | ||
| } | ||
| $this->url = $url; | ||
| $this->urlParams = parse_url($this->getURL()); | ||
| if ($this->needsCodeStrip()) { | ||
| $this->parsingResult->stripContent(); | ||
| } | ||
| if ($this->needsCodeClean()) { | ||
| $this->parsingResult->cleanContent(); | ||
| } | ||
| $this->resetParsingResult(); | ||
| $this->getData(); | ||
| return $this->getResult(); | ||
| } | ||
|
|
||
| /*** PRIVATE LAYER ***/ | ||
| /** | ||
| * Resets parsing result | ||
| * | ||
| * @return nothing | ||
| */ | ||
| private function resetParsingResult() { | ||
| if ($this->parsingResult instanceof ParsingResult) { | ||
| unset($this->parsingResult); | ||
| } | ||
| $this->parsingResult = new ParsingResult(); | ||
| } | ||
|
|
||
| /*** ABSTRACT LAYER ***/ | ||
| /** | ||
| * Abstract function of parsing content | ||
| * | ||
| * @return mixed | ||
| */ | ||
| abstract protected function getData():bool; | ||
| } | ||
| <?php | ||
| /** | ||
| * Content parser and detector from different webservices | ||
| */ | ||
| namespace Deepslam\ContentParser; | ||
|
|
||
| abstract class ContentParser { | ||
| /** | ||
| * Error constants | ||
| * | ||
| * @const string | ||
| */ | ||
| const ERR_WRONG_URL = 'No URL has been received'; | ||
|
|
||
| /** | ||
| * Parsing constants | ||
| * | ||
| * @const boolean | ||
| */ | ||
| const PARSING_DONE = true; | ||
| const PARSING_FAIL = false; | ||
|
|
||
| /** | ||
| * Registered parsers | ||
| * | ||
| * @var array | ||
| */ | ||
| private static $parsers = []; | ||
|
|
||
| /** | ||
| * Url for parse | ||
| * | ||
| * @var string | ||
| */ | ||
| private $url = ''; | ||
|
|
||
| /** | ||
| * These are received URL params. | ||
| * | ||
| * @var array | ||
| */ | ||
| private $urlParams = []; | ||
|
|
||
| /** | ||
| * Stored latest parsing result | ||
| * | ||
| * @var array | ||
| */ | ||
| private $parsingResult = []; | ||
|
|
||
| /** | ||
| * Object constructor | ||
| * | ||
| * @return ContentParser | ||
| * @throw new Exception if no url has been received | ||
| */ | ||
| public function __construct() { | ||
| $this->parsingResult = new ParsingResult(); | ||
| } | ||
|
|
||
| /** | ||
| * Returns URL | ||
| * | ||
| * @return string Working URL | ||
| */ | ||
| public function getURL() { | ||
| return $this->url; | ||
| } | ||
|
|
||
| /** | ||
| * Returns parsed URL params such as host, path etc... | ||
| * | ||
| * @return array | ||
| */ | ||
| public function getURLParams() { | ||
| return $this->urlParams; | ||
| } | ||
|
|
||
| /** | ||
| * Returns parsing result | ||
| * | ||
| * @return ParsingResult Link to the instance of parsing result | ||
| */ | ||
| public function getResult() { | ||
| return $this->parsingResult; | ||
| } | ||
|
|
||
| /*** PROTECTED LAYER ***/ | ||
|
|
||
| /** | ||
| * Checks whether have to do clean code or not. | ||
| * | ||
| * @return bool True - need to clean code, false - needn't to clean code | ||
| */ | ||
| final protected function needsCodeClean() { | ||
| return ((bool)config('deepslam.parser.clean_code')); | ||
| } | ||
|
|
||
| /** | ||
| * Checks whether have to strip tags or not. | ||
| * | ||
| * @return bool True - need to strip tags, false - needn't to strip tags | ||
| */ | ||
| final protected function needsCodeStrip() { | ||
| return ((bool)config('deepslam.parser.strip_tags')); | ||
| } | ||
|
|
||
| /*** STATIC LAYER ***/ | ||
|
|
||
| /** | ||
| * Factory method for creation needle parser class | ||
| * | ||
| * @param String $url | ||
| * @param $parser | ||
| * | ||
| * @return ContentParserMercury The link to the created instance | ||
| */ | ||
| public static function create(String $parser = '') { | ||
| if (isset(self::$parsers[$parser])) { | ||
| $workParser = self::$parsers[$parser]; | ||
| } else { | ||
| $workParser = ContentParserGraby::class; | ||
| } | ||
| return new $workParser(); | ||
| } | ||
|
|
||
| /** | ||
| * Registers new grabber | ||
| * | ||
| * @param String $alias | ||
| * @param String $className | ||
| * | ||
| * @return nothing | ||
| */ | ||
| public static function register(Array $params) { | ||
| if (count($params) == 0) { | ||
| return ; | ||
| } | ||
| foreach ($params as $alias => $className) { | ||
| self::$parsers[$alias] = $className; | ||
| } | ||
| } | ||
|
|
||
| /** | ||
| * Parses the needle URL | ||
| * Throws exception if has been received wrong URL | ||
| * | ||
| * @param String $url URL for parsing | ||
| * | ||
| * @return @ParsingResult Link to the parsing result | ||
| */ | ||
| public function parse($url) { | ||
| $this->resetParsingResult(); | ||
| if (!filter_var($url, FILTER_VALIDATE_URL)) { | ||
| return $this->getResult(); | ||
| } | ||
| $this->url = $url; | ||
| $this->urlParams = parse_url($this->getURL()); | ||
| $this->getData(); | ||
| if ($this->needsCodeStrip()) { | ||
| $this->parsingResult->stripContent(); | ||
| } | ||
| if ($this->needsCodeClean()) { | ||
| $this->parsingResult->cleanContent(); | ||
| } | ||
| return $this->getResult(); | ||
| } | ||
|
|
||
| /*** PRIVATE LAYER ***/ | ||
| /** | ||
| * Resets parsing result | ||
| * | ||
| * @return nothing | ||
| */ | ||
| private function resetParsingResult() { | ||
| if ($this->parsingResult instanceof ParsingResult) { | ||
| unset($this->parsingResult); | ||
| } | ||
| $this->parsingResult = new ParsingResult(); | ||
| } | ||
|
|
||
| /*** ABSTRACT LAYER ***/ | ||
| /** | ||
| * Abstract function of parsing content | ||
| * | ||
| * @return mixed | ||
| */ | ||
| abstract protected function getData():bool; | ||
| } | ||
| ?> |
Oops, something went wrong.