GlSpellChecker - Code Metrics - emmanuelroecker/php-spellchecker - Measure and Improve Code Quality continuously with Scrutinizer

GlSpellChecker A
last analyzed 2016-10-30 20:33 UTC

↳ Parent: Project

Complexity

Total Complexity

Size/Duplication

Total Lines	303
Duplicated Lines	0 %

Coupling/Cohesion

Components	1
Dependencies	9

Importance

Changes

Metric	Value
wmc	32
lcom	1
cbo	9
dl	0
loc	303
rs	9.6
c	0
b	0
f	0

8 Methods

Rating	Name	Size	Complexity
B	__construct()	30	4
A	__destruct()	8	2
B	convertToHtml()	47	5
C	checkYamlFiles()	47	7
B	checkHtmlFiles()	39	4
C	checkSentences()	47	7
A	startLanguageToolServer()	7	1
A	stopLanguageToolServer()	7	2

<?php
/**
 * Spell check html files
 *
 * PHP version 5.4
 *
 * @category  GLICER
 * @package   GlSpellChecker
 * @author    Emmanuel ROECKER
 * @author    Rym BOUCHAGOUR
 * @copyright 2015 GLICER
 * @license   MIT
 * @link      http://dev.glicer.com/
 *
 * Created : 04/05/15
 * File : GlSpellChecker.php
 *
 */

namespace GlSpellChecker;

use GlHtml\GlHtml;
use Symfony\Component\Process\Process;
use Symfony\Component\Finder\SplFileInfo;
use Symfony\Component\Finder\Finder;
use Symfony\Component\Yaml\Exception\ParseException;
use Symfony\Component\Yaml\Yaml;
use GuzzleHttp\Client;


/**
 * Class GlSpellChecker
 * @package GlSpellChecker
 */
class GlSpellChecker
{
    /**
     * @var int
     */
    private $languageToolServerPort = 8081;

    /**
     * @var string
     */
    private $languageToolLanguage = 'fr';

    /**
     * @var Process $languagetoolServer ;
     */
    private $languagetoolServer = null;

    /**
     * @var Client
     */
    private $languagetoolClientHttp;


    /**
     * @var string
     */
    private $enchantLanguage = "fr_FR";
    private $enchantDictionnary = null;
    private $enchantBroker = null;

    /**
     * @param string $languageToolDirectory
     * @param string $languageToolLanguage
     * @param string $enchantLanguage
     * @param string $languageToolServerIP
     * @param int    $languageToolServerPort
     *
     * @throws \Exception
     */
    public function __construct(
        $languageToolLanguage,
        $enchantLanguage,
        $languageToolDirectory = null,
        $languageToolServerIP = 'localhost',
        $languageToolServerPort = 8081
    ) {
        $this->languageToolLanguage   = $languageToolLanguage;
        $this->enchantLanguage        = $enchantLanguage;
        $this->languageToolServerPort = $languageToolServerPort;
        $this->languageToolServerIP   = $languageToolServerIP;


        if ($languageToolDirectory) {
''   == false // true
''   == null  // true
'ab' == false // false
'ab' == null  // false

// It is often better to use strict comparison
'' === false // false
'' === null  // false
            $this->startLanguageToolServer($languageToolDirectory);
        }

        $this->languagetoolClientHttp = new Client();

        if (extension_loaded('enchant')) {
            $this->enchantBroker = enchant_broker_init();

            enchant_broker_set_dict_path($this->enchantBroker, ENCHANT_MYSPELL, __DIR__ . '/../dicts');

            if (!enchant_broker_dict_exists($this->enchantBroker, $this->enchantLanguage)) {
                throw new \Exception("Cannot find dictionnaries for enchant");
            } else {
                $this->enchantDictionnary = enchant_broker_request_dict($this->enchantBroker, $this->enchantLanguage);
            }
        }
    }

    public function __destruct()
    {
        $this->stopLanguageToolServer();
        if ($this->enchantBroker) {
            enchant_broker_free_dict($this->enchantDictionnary);
            enchant_broker_free($this->enchantBroker);
        }
    }

    /**
     * @param string                   $title
     * @param GlSpellCheckerSentence[] $sentences
     *
     * @return string
     */
    public static function convertToHtml($title, $sentences)
    {
        $html = '<!DOCTYPE HTML>';
        $html .= '<html><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8">';
        $html .= '<title>' . $title . '</title>';
        $html .= '<style>';
        $html .= '.error {  color: red  }';
        $html .= '</style>';
        $html .= '</head><body>';

        foreach ($sentences as $sentence) {
            $html .= '<div class="sentence">';
            $text   = $sentence->getText();
            $errors = $sentence->mergeErrors();

            if (count($errors) <= 0) {
                $html .= $text;
                $html .= '</div>';
                continue;
            }

            $cons  = "";
            $start = 0;
            foreach ($errors as $error) {
                $offset = $error->getOffset();
                $length = $error->getLength();
                $cons .= mb_substr($text, $start, $offset - $start, 'UTF-8');

                $tooltip = $error->getMessage();
                $suggs   = $error->getSuggestions();
                if (count($suggs) > 0) {
                    $tooltip .= " : " . $suggs[0];
                }
                $zone = mb_substr($text, $offset, $length, 'UTF-8');
                $cons .= '<span class="error" title="' . $tooltip . '">' . $zone . '</span>';

                $start = $offset + $length;
            }
            $cons .= mb_substr($text, $start, mb_strlen($text) - $start, 'UTF-8');

            $html .= $cons;
            $html .= '</div>';
        }
        $html .= '<br><br><br></body></html>';

        return $html;
    }

    public function checkYamlFiles(
        Finder $files,
        array    $fields,
        callable $checkfilestart,
        callable $checksentence,
        callable $checkfileend
    ) {
        $results = [];
        /**
         * @var SplFileInfo $file
         */
        foreach ($files as $file) {
            try {
                $data = Yaml::parse(
                            file_get_contents(
                                $file->getRealPath()
                            )
                );
            } catch (ParseException $e) {
                throw new \Exception("Unable to parse YAML string: {$e->getMessage()}");
            }
            $sentences = [];
            foreach ($data as $item) {
$collection = json_decode($data, true);
if ( ! is_array($collection)) {
    throw new \RuntimeException('$collection must be an array.');
}

foreach ($collection as $item) { /** ... */ }
                foreach ($item as $key => $valueitem) {
                    foreach ($fields as $field) {
                        if ($key == $field) {
                            $sentences[] = $valueitem;
                        }
                    }
                }
            }
            $checkfilestart($file, count($sentences));
            $sentences = $this->checkSentences(
                              $sentences,
                                  $checksentence
            );
            $htmlcode  = $this->convertToHtml($file->getFilename(), $sentences);

            $checkerfile = sys_get_temp_dir() . "/" . uniqid("spellcheck") . ".html";
            file_put_contents($checkerfile, $htmlcode);
            $results[] = $checkerfile;

            $checkfileend();
        }

        return $results;
    }

    /**
     * @param Finder   $files
     * @param callable $checkfilestart
     * @param callable $checksentence
     * @param callable $checkfileend
     *
     * @return array
     */
    public function checkHtmlFiles(
        Finder $files,
        callable $checkfilestart,
        callable $checksentence,
        callable $checkfileend
    ) {
        $results = [];
        /**
         * @var SplFileInfo $file
         */
        foreach ($files as $file) {
            $html = file_get_contents($file->getRealPath());
            $html = new GlHtml($html);

            $title = $html->get("head title");

            if ($title && sizeof($title) > 0) {

                $title = $title[0]->getText();
            } else {
                $title = $file->getFilename();
            }

            $sentences = $html->getSentences();
            $checkfilestart($file, count($sentences));
            $sentences = $this->checkSentences(
                              $sentences,
                                  $checksentence
            );
            $htmlcode  = $this->convertToHtml($title, $sentences);

            $checkerfile = sys_get_temp_dir() . "/" . uniqid("spellcheck") . ".html";
            file_put_contents($checkerfile, $htmlcode);
            $results[] = $checkerfile;

            $checkfileend();
        }

        return $results;
    }

    /**
     * @param array    $sentences
     *
     * @param callable $closure
     *
     * @return GlSpellCheckerSentence[]
     */
    public
    function checkSentences(
        array $sentences,
        callable $closure
    ) {
        $url              = "http://{$this->languageToolServerIP}:{$this->languageToolServerPort}";

        $sentencesChecked = [];
        foreach ($sentences as $sentence) {
            $response = $this->languagetoolClientHttp->get(
                                                     $url,
                                                         [
                                                             'query' => [
                                                                 'language' => $this->languageToolLanguage,
                                                                 'text'     => $sentence
                                                             ]
                                                         ]
            );
            $xml      = $response->getBody()->getContents();
            $glxml           = new GlHtml($xml);
            $errors          = $glxml->get('error');
            $sentenceChecked = new GlSpellCheckerSentence($sentence);
            if (count($errors) > 0) {
                foreach ($errors as $error) {
                    $msg    = $error->getAttribute('msg');
                    $offset = (int)$error->getAttribute('offset');
                    $length = (int)$error->getAttribute('errorlength');
                    $suggs  = [];
                    $word   = null;
                    if ($error->getAttribute('locqualityissuetype') == 'misspelling') {
                        $word = mb_substr($sentence, $offset, $length, 'UTF-8');
                        if ($this->enchantDictionnary) {
                            $wordcorrect = enchant_dict_check($this->enchantDictionnary, $word);
                            if (!$wordcorrect) {
                                $suggs = enchant_dict_suggest($this->enchantDictionnary, $word);
                            }
                        }
                    }
                    $glerror = new GlSpellCheckerError($msg, $offset, $length, $word, $suggs);
                    $sentenceChecked->addError($glerror);
                }
            }
            $sentencesChecked[] = $sentenceChecked;
            $closure($sentence);
        }

        return $sentencesChecked;
    }

    /**
     * @param string $directory
     */
    private function startLanguageToolServer($directory)
    {
        $jar                      = $directory . "languagetool-server.jar";
        $command                  = "java -cp $jar org.languagetool.server.HTTPServer --port {$this->languageToolServerPort}";
        $this->languagetoolServer = new Process($command);
        $this->languagetoolServer->start();
    }

    private function stopLanguageToolServer()
    {
        if ($this->languagetoolServer) {
            $this->languagetoolServer->stop();
            $this->languagetoolServer = null;
        }
    }
} 


1			<?php
2			/**
3			* Spell check html files
4			*
5			* PHP version 5.4
6			*
7			* @category GLICER
8			* @package GlSpellChecker
9			* @author Emmanuel ROECKER
10			* @author Rym BOUCHAGOUR
11			* @copyright 2015 GLICER
12			* @license MIT
13			* @link http://dev.glicer.com/
14			*
15			* Created : 04/05/15
16			* File : GlSpellChecker.php
17			*
18			*/
19
20			namespace GlSpellChecker;
21
22			use GlHtml\GlHtml;
23			use Symfony\Component\Process\Process;
24			use Symfony\Component\Finder\SplFileInfo;
25			use Symfony\Component\Finder\Finder;
26			use Symfony\Component\Yaml\Exception\ParseException;
27			use Symfony\Component\Yaml\Yaml;
28			use GuzzleHttp\Client;
29
30
31			/**
32			* Class GlSpellChecker
33			* @package GlSpellChecker
34			*/
35			class GlSpellChecker
36			{
37			/**
38			* @var int
39			*/
40			private $languageToolServerPort = 8081;
41
42			/**
43			* @var string
44			*/
45			private $languageToolLanguage = 'fr';
46
47			/**
48			* @var Process $languagetoolServer ;
49			*/
50			private $languagetoolServer = null;
51
52			/**
53			* @var Client
54			*/
55			private $languagetoolClientHttp;
56
57
58			/**
59			* @var string
60			*/
61			private $enchantLanguage = "fr_FR";
62			private $enchantDictionnary = null;
63			private $enchantBroker = null;
64
65			/**
66			* @param string $languageToolDirectory
67			* @param string $languageToolLanguage
68			* @param string $enchantLanguage
69			* @param string $languageToolServerIP
70			* @param int $languageToolServerPort
71			*
72			* @throws \Exception
73			*/
74			public function __construct(
75			$languageToolLanguage,
76			$enchantLanguage,
77			$languageToolDirectory = null,
78			$languageToolServerIP = 'localhost',
79			$languageToolServerPort = 8081
80			) {
81			$this->languageToolLanguage = $languageToolLanguage;
82			$this->enchantLanguage = $enchantLanguage;
83			$this->languageToolServerPort = $languageToolServerPort;
84			$this->languageToolServerIP = $languageToolServerIP;
			0 ignored issues – show Bug introduced 2016-01-09 20:02 UTC by Report Bug Copy Issue Report The property `languageToolServerIP` does not seem to exist. Did you mean `languagetoolServer`? An attempt at access to an undefined property has been detected. This may either be a typographical error or the property has been renamed but there are still references to its old name. If you really want to allow access to undefined properties, you can define magic methods to allow access. See the php core documentation on Overloading. Loading history...
85
86			if ($languageToolDirectory) {
			0 ignored issues – show Bug Best Practice introduced 2016-01-09 20:02 UTC by Report Bug Copy Issue Report The expression `$languageToolDirectory` of type `string\|null` is loosely compared to `true`; this is ambiguous if the string can be empty. You might want to explicitly use `!== null` instead. In PHP, under loose comparison (like `==`, or `!=`, or `switch` conditions), values of different types might be equal. For `string` values, the empty string `''` is a special case, in particular the following results might be unexpected: '' == false // true '' == null // true 'ab' == false // false 'ab' == null // false // It is often better to use strict comparison '' === false // false '' === null // false Loading history...
87			$this->startLanguageToolServer($languageToolDirectory);
88			}
89
90			$this->languagetoolClientHttp = new Client();
91
92			if (extension_loaded('enchant')) {
93			$this->enchantBroker = enchant_broker_init();
94
95			enchant_broker_set_dict_path($this->enchantBroker, ENCHANT_MYSPELL, __DIR__ . '/../dicts');
96
97			if (!enchant_broker_dict_exists($this->enchantBroker, $this->enchantLanguage)) {
98			throw new \Exception("Cannot find dictionnaries for enchant");
99			} else {
100			$this->enchantDictionnary = enchant_broker_request_dict($this->enchantBroker, $this->enchantLanguage);
101			}
102			}
103			}
104
105			public function __destruct()
106			{
107			$this->stopLanguageToolServer();
108			if ($this->enchantBroker) {
109			enchant_broker_free_dict($this->enchantDictionnary);
110			enchant_broker_free($this->enchantBroker);
111			}
112			}
113
114			/**
115			* @param string $title
116			* @param GlSpellCheckerSentence[] $sentences
117			*
118			* @return string
119			*/
120			public static function convertToHtml($title, $sentences)
121			{
122			$html = '<!DOCTYPE HTML>';
123			$html .= '<html><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8">';
124			$html .= '<title>' . $title . '</title>';
125			$html .= '<style>';
126			$html .= '.error { color: red }';
127			$html .= '</style>';
128			$html .= '</head><body>';
129
130			foreach ($sentences as $sentence) {
131			$html .= '<div class="sentence">';
132			$text = $sentence->getText();
133			$errors = $sentence->mergeErrors();
134
135			if (count($errors) <= 0) {
136			$html .= $text;
137			$html .= '</div>';
138			continue;
139			}
140
141			$cons = "";
142			$start = 0;
143			foreach ($errors as $error) {
144			$offset = $error->getOffset();
145			$length = $error->getLength();
146			$cons .= mb_substr($text, $start, $offset - $start, 'UTF-8');
147
148			$tooltip = $error->getMessage();
149			$suggs = $error->getSuggestions();
150			if (count($suggs) > 0) {
151			$tooltip .= " : " . $suggs[0];
152			}
153			$zone = mb_substr($text, $offset, $length, 'UTF-8');
154			$cons .= '<span class="error" title="' . $tooltip . '">' . $zone . '</span>';
155
156			$start = $offset + $length;
157			}
158			$cons .= mb_substr($text, $start, mb_strlen($text) - $start, 'UTF-8');
159
160			$html .= $cons;
161			$html .= '</div>';
162			}
163			$html .= '<br><br><br></body></html>';
164
165			return $html;
166			}
167
168			public function checkYamlFiles(
169			Finder $files,
170			array $fields,
171			callable $checkfilestart,
172			callable $checksentence,
173			callable $checkfileend
174			) {
175			$results = [];
176			/**
177			* @var SplFileInfo $file
178			*/
179			foreach ($files as $file) {
180			try {
181			$data = Yaml::parse(
182			file_get_contents(
183			$file->getRealPath()
184			)
185			);
186			} catch (ParseException $e) {
187			throw new \Exception("Unable to parse YAML string: {$e->getMessage()}");
188			}
189			$sentences = [];
190			foreach ($data as $item) {
			0 ignored issues – show Bug introduced 2016-10-30 20:33 UTC by Report Bug Copy Issue Report The expression `$data` of type `string\|array\|object<stdClass>` is not guaranteed to be traversable. How about adding an additional type check? There are different options of fixing this problem. If you want to be on the safe side, you can add an additional type-check: $collection = json_decode($data, true); if ( ! is_array($collection)) { throw new \RuntimeException('$collection must be an array.'); } foreach ($collection as $item) { /** ... / } If you are sure that the expression is traversable, you might want to add a doc comment cast* to improve IDE auto-completion and static analysis: /** @var array $collection / $collection = json_decode($data, true); foreach ($collection as $item) { /* .. / } Mark the issue as a false-positive*: Just hover the remove button, in the top-right corner of this issue for more options. Loading history...
191			foreach ($item as $key => $valueitem) {
192			foreach ($fields as $field) {
193			if ($key == $field) {
194			$sentences[] = $valueitem;
195			}
196			}
197			}
198			}
199			$checkfilestart($file, count($sentences));
200			$sentences = $this->checkSentences(
201			$sentences,
202			$checksentence
203			);
204			$htmlcode = $this->convertToHtml($file->getFilename(), $sentences);
205
206			$checkerfile = sys_get_temp_dir() . "/" . uniqid("spellcheck") . ".html";
207			file_put_contents($checkerfile, $htmlcode);
208			$results[] = $checkerfile;
209
210			$checkfileend();
211			}
212
213			return $results;
214			}
215
216			/**
217			* @param Finder $files
218			* @param callable $checkfilestart
219			* @param callable $checksentence
220			* @param callable $checkfileend
221			*
222			* @return array
223			*/
224			public function checkHtmlFiles(
225			Finder $files,
226			callable $checkfilestart,
227			callable $checksentence,
228			callable $checkfileend
229			) {
230			$results = [];
231			/**
232			* @var SplFileInfo $file
233			*/
234			foreach ($files as $file) {
235			$html = file_get_contents($file->getRealPath());
236			$html = new GlHtml($html);
237
238			$title = $html->get("head title");
239
240			if ($title && sizeof($title) > 0) {
			0 ignored issues – show Bug Best Practice introduced 2016-01-09 19:58 UTC by Report Bug Copy Issue Report The expression `$title` of type `GlHtml\GlHtmlNode[]` is implicitly converted to a boolean; are you sure this is intended? If so, consider using `! empty($expr)` instead to make it clear that you intend to check for an array without elements. This check marks implicit conversions of arrays to boolean values in a comparison. While in PHP an empty array is considered to be equal (but not identical) to false, this is not always apparent. Consider making the comparison explicit by using `empty(..)` or `! empty(...)` instead. Loading history...
241			$title = $title[0]->getText();
242			} else {
243			$title = $file->getFilename();
244			}
245
246			$sentences = $html->getSentences();
247			$checkfilestart($file, count($sentences));
248			$sentences = $this->checkSentences(
249			$sentences,
250			$checksentence
251			);
252			$htmlcode = $this->convertToHtml($title, $sentences);
253
254			$checkerfile = sys_get_temp_dir() . "/" . uniqid("spellcheck") . ".html";
255			file_put_contents($checkerfile, $htmlcode);
256			$results[] = $checkerfile;
257
258			$checkfileend();
259			}
260
261			return $results;
262			}
263
264			/**
265			* @param array $sentences
266			*
267			* @param callable $closure
268			*
269			* @return GlSpellCheckerSentence[]
270			*/
271			public
272			function checkSentences(
273			array $sentences,
274			callable $closure
275			) {
276			$url = "http://{$this->languageToolServerIP}:{$this->languageToolServerPort}";
			0 ignored issues – show Bug introduced 2016-01-09 20:02 UTC by Report Bug Copy Issue Report The property `languageToolServerIP` does not seem to exist. Did you mean `languagetoolServer`? An attempt at access to an undefined property has been detected. This may either be a typographical error or the property has been renamed but there are still references to its old name. If you really want to allow access to undefined properties, you can define magic methods to allow access. See the php core documentation on Overloading. Loading history...
277			$sentencesChecked = [];
278			foreach ($sentences as $sentence) {
279			$response = $this->languagetoolClientHttp->get(
280			$url,
281			[
282			'query' => [
283			'language' => $this->languageToolLanguage,
284			'text' => $sentence
285			]
286			]
287			);
288			$xml = $response->getBody()->getContents();
289			$glxml = new GlHtml($xml);
290			$errors = $glxml->get('error');
291			$sentenceChecked = new GlSpellCheckerSentence($sentence);
292			if (count($errors) > 0) {
293			foreach ($errors as $error) {
294			$msg = $error->getAttribute('msg');
295			$offset = (int)$error->getAttribute('offset');
296			$length = (int)$error->getAttribute('errorlength');
297			$suggs = [];
298			$word = null;
299			if ($error->getAttribute('locqualityissuetype') == 'misspelling') {
300			$word = mb_substr($sentence, $offset, $length, 'UTF-8');
301			if ($this->enchantDictionnary) {
302			$wordcorrect = enchant_dict_check($this->enchantDictionnary, $word);
303			if (!$wordcorrect) {
304			$suggs = enchant_dict_suggest($this->enchantDictionnary, $word);
305			}
306			}
307			}
308			$glerror = new GlSpellCheckerError($msg, $offset, $length, $word, $suggs);
309			$sentenceChecked->addError($glerror);
310			}
311			}
312			$sentencesChecked[] = $sentenceChecked;
313			$closure($sentence);
314			}
315
316			return $sentencesChecked;
317			}
318
319			/**
320			* @param string $directory
321			*/
322			private function startLanguageToolServer($directory)
323			{
324			$jar = $directory . "languagetool-server.jar";
325			$command = "java -cp $jar org.languagetool.server.HTTPServer --port {$this->languageToolServerPort}";
326			$this->languagetoolServer = new Process($command);
327			$this->languagetoolServer->start();
328			}
329
330			private function stopLanguageToolServer()
331			{
332			if ($this->languagetoolServer) {
333			$this->languagetoolServer->stop();
334			$this->languagetoolServer = null;
335			}
336			}
337			}
338

emmanuelroecker / php-spellchecker

GlSpellChecker A last analyzed 2016-10-30 20:33 UTC

Complexity

Size/Duplication

Coupling/Cohesion

Importance

8 Methods

Duplication Side-by-Side

Filter issues like

GlSpellChecker A
last analyzed 2016-10-30 20:33 UTC