使用Snoopy类解析html文件的方法

set_time_limit(0);

$user = array('20517', '20518');

header("content-Type: text/html; charset=utf-8");

require_once './Snoopy.php';

$snoopy = new Snoopy();

//$uri = 'http://www.juzimi.com/meitumeiju';

$uri = 'http://www.juzimi.com/meitumeiju?page=1';

$snoopy->fetch($uri);

$con = $snoopy->getResults();

$doc = new DOMDocument();

$doc->loadHTML($con);

/* make a result array ... */

$result = [];

/* go through all nodes which have class="baby" ... */

$i = 0;

foreach( ( new DOMXPath( $doc ) )->query( '//*[@id="bdshare"]' ) as $element ) {

  $row = $doc->saveHTML($element);

  $start = stripos($row, '{');

  $end = stripos($row, '}');

  $len = $end - $start + 1;

  $json = substr($row, $start, $len);

  $json = str_replace("'", '"', $json);

  $obj = json_decode($json);

  $pic = $obj->pic;

  $desc = $obj->desc;

  $filename = '/Uploads/meiju/card' . date('ymdhis') . $i . '.png';

  $file = '/alidata/www/test' . $filename;

  $image = file_get_contents($pic);

  file_put_contents($file, $image);

}