szyku/php-nltk-client

PHP client implementation for szyku/nltk-api

v1.1.2 2017-05-21 16:23 UTC

Requires

• php: ^5.6 || ^7.0
• beberlei/assert: ^2.7
• guzzlehttp/guzzle: ^6.2
• marc-mabe/php-enum: ^2.3
• symfony/serializer: ^3.2

Requires (Dev)

• phpunit/phpunit: ^5.0

MIT

• Szymon Szymański

README

This is a PHP client for szyku/nltk-api which exposes python's NLTK resources and WordNet database through a friendly API.

The aim of this client is to provide a comfortable and intuitive way of interacting with the NLTK API.

Installation

This package was tested for PHP >= 5.6.

Composer:

composer require szyku/php-nltk-client

Git clone:

git clone [e-mail]:szyku/php-nltk-client.git

Or download zip:

wget https://github.com/szyku/php-nltk-client/archive/master.zip 

Or tarball:

wget https://github.com/szyku/php-nltk-client/archive/master.tar.gz

How to use it

Easy to setup:

require __DIR__ . '/vendor/autoload.php';

use GuzzleHttp\Psr7\Uri;
use Szyku\NLTK\Client;
use GuzzleHttp\Client as Guzzle;

$uri = new Uri("http://localhost:5000");
$guzzle = new Guzzle(['base_uri' => $uri]);
// or just $guzzle = new Guzzle(['base_uri' => 'http://localhost:5000']);

$nltkClient = new Client($guzzle);

How to query definitions with handy methods:

use Szyku\NLTK\Request\Dictionary\DefinitionLookupRequest;
use Szyku\NLTK\Request\Dictionary\SimilarLookupRequest;
use Szyku\NLTK\Request\Lemma\LemmatizationRequestBuilder as Builder;
use Szyku\NLTK\Request\Tagger\TaggingRequestBuilder as TaggerBuilder;

$similarWordsToCastle = SimilarLookupRequest::noun('castle');
$definitionsForRA = DefinitionLookupRequest::noun('rheumatoid arthritis');

$sentenceTagging = TaggerBuilder::create()
    ->add('He is a cunning man.')
    ->addMany(['Before and after.', 'Ups and downs'])
    ->build();

$lemmatizeSentence = Builder::create()
    ->adjective('biggest')
    ->noun('dogs')
    ->verb('fought')
    ->adverb('loudly')
    ->findAllFor('at')
    ->build();

// results are hydrated to objects like WordLookupResponse or LemmatizationResponse
$castleResult = $nltkClient->dictionary($similarWordsToCastle);
$raResult = $nltkClient->dictionary($definitionsForRA);
$lemmatizationResult = $nltkClient->lemmatization($lemmatizeSentence);
$taggingResult = $nltkClient->tagging($sentenceTagging);

// easy to consume
echo "Time taken in seconds: " . $castleResult->lookupTime();
echo "Searched for: " . $castleResult->queriedPhrase();

foreach ($castleResult->results() as $result) {
    echo $result->phrase() . ": " . $result->definition();
}
// prints "palace: A large and stately mansion"