README

Event-based parser for large JSON collections (consumes small amount of memory). Built on top of JSON Streaming Parser

This package is compliant with PSR-4 and PSR-12 code styles and supports parsing of PSR-7 message interfaces. If you notice compliance oversights, please send a patch via pull request.

Installation

You will need Composer to install the package

composer require maxakawizard/json-collection-parser:~1.0

Input data format

Data must be in one of following formats:

Array of objects (valid JSON)

[
    {
        "id": 78,
        "title": "Title",
        "dealType": "sale",
        "propertyType": "townhouse",
        "properties": {
            "bedroomsCount": 6,
            "parking": "yes"
        },
        "photos": [
            "1.jpg",
            "2.jpg"
        ],
        "agents": [
            {
                "name": "Joe",
                "email": "joe@realestate.email"
            },
            {
                "name": "Sally",
                "email": "sally@realestate.email"
            }
         ]
    },
    {
        "id": 729,
        "dealType": "rent_long",
        "propertyType": "villa"
    },
    {
        "id": 5165,
        "dealType": "rent_short",
        "propertyType": "villa"
    }
]

Sequence of object literals:

{
    "id": 78,
    "dealType": "sale",
    "propertyType": "townhouse"
}
{
    "id": 729,
    "dealType": "rent_long",
    "propertyType": "villa"
}
{
    "id": 5165,
    "dealType": "rent_short",
    "propertyType": "villa"
}

Sequence of object and array literals:

[[{
    "id": 78,
    "dealType": "sale",
    "propertyType": "townhouse"
}]]
{
    "id": 729,
    "dealType": "rent_long",
    "propertyType": "villa"
}
[{
    "id": 5165,
    "dealType": "rent_short",
    "propertyType": "villa"
}]

Sequence of object and array literals (some of objects in subarrays, comma-separated):

[
{
    "id": 78,
    "dealType": "sale",
    "propertyType": "townhouse"
},
{
    "id": 729,
    "dealType": "rent_long",
    "propertyType": "villa"
}
]
{
    "id": 5165,
    "dealType": "rent_short",
    "propertyType": "villa"
}

Usage

Function as callback:

function processItem(array $item)
{
    is_array($item); //true
    print_r($item);
}

$parser = new \JsonCollectionParser\Parser();
$parser->parse('/path/to/file.json', 'processItem');

Closure as callback:

$items = [];

$parser = new \JsonCollectionParser\Parser();
$parser->parse('/path/to/file.json', function (array $item) use (&$items) {
    $items[] = $item;
});

Static method as callback:

class ItemProcessor {
    public static function process(array $item)
    {
        is_array($item); //true
        print_r($item);
    }
}

$parser = new \JsonCollectionParser\Parser();
$parser->parse('/path/to/file.json', ['ItemProcessor', 'process']);

Instance method as callback:

class ItemProcessor {
    public function process(array $item)
    {
        is_array($item); //true
        print_r($item);
    }
}

$parser = new \JsonCollectionParser\Parser();
$processor = new \ItemProcessor();
$parser->parse('/path/to/file.json', [$processor, 'process']);

Receive items as objects:

function processItem(\stdClass $item)
{
    is_array($item); //false
    is_object($item); //true
    print_r($item);
}

$parser = new \JsonCollectionParser\Parser();
$parser->parseAsObjects('/path/to/file.json', 'processItem');

Receive chunks of items as arrays:

function processChunk(array $chunk)
{
    is_array($chunk);    //true
    count($chunk) === 5; //true

    foreach ($chunk as $item) {
        is_array($item);  //true
        is_object($item); //false
        print_r($item);
    }
}

$parser = new \JsonCollectionParser\Parser();
$parser->chunk('/path/to/file.json', 'processChunk', 5);

Receive chunks of items as objects:

function processChunk(array $chunk)
{
    is_array($chunk);    //true
    count($chunk) === 5; //true

    foreach ($chunk as $item) {
        is_array($item);  //false
        is_object($item); //true
        print_r($item);
    }
}

$parser = new \JsonCollectionParser\Parser();
$parser->chunkAsObjects('/path/to/file.json', 'processChunk', 5);

Pass stream as parser input:

$stream = fopen('/path/to/file.json', 'r');

$parser = new \JsonCollectionParser\Parser();
$parser->parseAsObjects($stream, 'processItem');

Pass PSR-7 MessageInterface as parser input:

use Psr\Http\Message\MessageInterface;

/** @var MessageInterface $resource */
$resource = $httpClient->get('https://httpbin.org/get');

$parser = new \JsonCollectionParser\Parser();
$parser->parseAsObjects($resource, 'processItem');

Pass PSR-7 StreamInterface as parser input:

use Psr\Http\Message\MessageInterface;

/** @var MessageInterface $resource */
$resource = $httpClient->get('https://httpbin.org/get');

$parser = new \JsonCollectionParser\Parser();
$parser->parseAsObjects($resource->getBody(), 'processItem');

Supported formats

• .json - raw JSON
• .gz - GZIP-compressed JSON (you will need zlib PHP extension installed)

Supported sources

• file
• string
• stream / resource
• HTTP message interface PSR-7

License

This library is released under MIT license.