Small class, which will try to extract JSON strings from a string. There are two functions which will achieve that.
composer require ronolo/json-extract
If that does not work, you may have to add the repository to the top level composer.json like this:
{
"repositories": [
{
"type": "vcs",
"url": "https://github.com/ronolo/json-extract.git"
}
]
}
<?php
use RoNoLo\JsonExtractor\JsonExtractorService;
$jsonExtractor = new JsonExtractorService();
// $html contains a HTML full page string (as example)
$html = file_get_content('foo/bar.html');
$json = $jsonExtractor->extractJsonAfterIdentifier("foobar", $html);
This will expect the identifier somewhere in the given string. The identifier position will act as starting point and the next valid JSON object or array will be returned as PHP array.
<?php
use RoNoLo\JsonExtractor\JsonExtractorService;
$jsonExtractor = new JsonExtractorService();
// $html contains a HTML full page string
$html = file_get_content('foo/bar.html');
$vars = $jsonExtractor->extractAllJsonData($html);
This will look for any JSON objects or arrays in the string and returns them as an array of arrays of JSON data. You may have to check the list of JSON data for data you was looking for.
It is recommend to break HTML down into smaller parts with a DOM parser, like symfony/dom-crawler or similar. The smaller the portion is which shall be parsed the better will be the result.
For example, if you parse a HTML string and expect the JSON inside the
... tags, feed only the text inside the tags to the json-extract functions.Correct JSON
{
"paging": {
"pageNum": 1,
"pageSize": 25,
"numFound": 1,
"last": 1,
"lastUncapped": 1,
"display": [1]
}
}
Incorrect Single-Quotes JSON
{
'paging': {
'pageNum': 1,
'pageSize': 25,
'numFound': 1,
'last': 1,
'lastUncapped': 1,
'display': [1]
}
}
Javascript Objects (Thanks to the CJSON.php)
{
paging: {
pageNum: 1,
pageSize: 25,
numFound: 1,
last: 1,
lastUncapped: 1,
display: [1]
}
}
EOF