SoFunction
Updated on 2025-04-04

Three solutions to crawl Gzip web page garbled code in php file_get_contents

Turn the captured content down and encode it ($content=iconv("GBK", "UTF-8//IGNORE", $content);). What we are discussing here is how to crawl the page with Gzip opened. How to judge? The content-Encoding: gzip is shown in the obtained header that the content is GZIP compressed. If you look at FireBug, you will know if the page is open with gzip. Below is the header information of my blog using firebug. Gzip is open.

Copy the codeThe code is as follows:

Request header information Original header information
Accept text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
Accept-Encoding gzip, deflate
Accept-Language zh-cn,zh;q=0.8,en-us;q=0.5,en;q=0.3
Connection keep-alive
Cookie __utma=225240837.787252530.1317310581.1335406161.1335411401.1537; __utmz=225240837.1326850415.887.=google|utmccn=(organic)|utmcmd=organic|utmctr=%E4%BB%BB%E4%BD%95%E9%A1%B9%E7%9B%AE%E9%83%BD%E4%B8%8D%E4%BC%9A%E9%82%A3%E4%B9%88%E7%AE%80%E5%8D%95%20site%; PHPSESSID=888mj4425p8s0m7s0frre3ovc7; __utmc=225240837; __utmb=225240837.1.10.1335411401
Host 
User-Agent Mozilla/5.0 (Windows NT 5.1; rv:12.0) Gecko/20100101 Firefox/12.0

Here are some solutions:

1. Use the built-in zlib library
If the server has already installed the zlib library, the following code can easily solve the garbled code problem.

Copy the codeThe code is as follows:

$data = file_get_contents("://".$url);

2. Use CURL instead of file_get_contents

Copy the codeThe code is as follows:

function curl_get($url, $gzip=false){
 $curl = curl_init($url);
 curl_setopt($curl, CURLOPT_RETURNTRANSFER, 1);
 curl_setopt($curl, CURLOPT_CONNECTTIMEOUT, 10);
if($gzip) curl_setopt($curl, CURLOPT_ENCODING, "gzip"); // The key is here
 $content = curl_exec($curl);
 curl_close($curl);
 return $content;
}

3. Use gzip decompression function

Copy the codeThe code is as follows:

function gzdecode($data) {
  $len = strlen($data);
  if ($len < 18 || strcmp(substr($data,0,2),"\x1f\x8b")) {
    return null;  // Not GZIP format (See RFC 1952)
  }
  $method = ord(substr($data,2,1));  // Compression method
  $flags  = ord(substr($data,3,1));  // Flags
  if ($flags & 31 != $flags) {
    // Reserved bits are set -- NOT ALLOWED by RFC 1952
    return null;
  }
  // NOTE: $mtime may be negative (PHP integer limitations)
  $mtime = unpack("V", substr($data,4,4));
  $mtime = $mtime[1];
  $xfl   = substr($data,8,1);
  $os    = substr($data,8,1);
  $headerlen = 10;
  $extralen  = 0;
  $extra     = "";
  if ($flags & 4) {
    // 2-byte length prefixed EXTRA data in header
    if ($len - $headerlen - 2 < 8) {
      return false;    // Invalid format
    }
    $extralen = unpack("v",substr($data,8,2));
    $extralen = $extralen[1];
    if ($len - $headerlen - 2 - $extralen < 8) {
      return false;    // Invalid format
    }
    $extra = substr($data,10,$extralen);
    $headerlen += 2 + $extralen;
  }

  $filenamelen = 0;
  $filename = "";
  if ($flags & 8) {
    // C-style string file NAME data in header
    if ($len - $headerlen - 1 < 8) {
      return false;    // Invalid format
    }
    $filenamelen = strpos(substr($data,8+$extralen),chr(0));
    if ($filenamelen === false || $len - $headerlen - $filenamelen - 1 < 8) {
      return false;    // Invalid format
    }
    $filename = substr($data,$headerlen,$filenamelen);
    $headerlen += $filenamelen + 1;
  }

  $commentlen = 0;
  $comment = "";
  if ($flags & 16) {
    // C-style string COMMENT data in header
    if ($len - $headerlen - 1 < 8) {
      return false;    // Invalid format
    }
    $commentlen = strpos(substr($data,8+$extralen+$filenamelen),chr(0));
    if ($commentlen === false || $len - $headerlen - $commentlen - 1 < 8) {
      return false;    // Invalid header format
    }
    $comment = substr($data,$headerlen,$commentlen);
    $headerlen += $commentlen + 1;
  }

  $headercrc = "";
  if ($flags & 1) {
    // 2-bytes (lowest order) of CRC32 on header present
    if ($len - $headerlen - 2 < 8) {
      return false;    // Invalid format
    }
    $calccrc = crc32(substr($data,0,$headerlen)) & 0xffff;
    $headercrc = unpack("v", substr($data,$headerlen,2));
    $headercrc = $headercrc[1];
    if ($headercrc != $calccrc) {
      return false;    // Bad header CRC
    }
    $headerlen += 2;
  }

  // GZIP FOOTER - These be negative due to PHP's limitations
  $datacrc = unpack("V",substr($data,-8,4));
  $datacrc = $datacrc[1];
  $isize = unpack("V",substr($data,-4));
  $isize = $isize[1];

  // Perform the decompression:
  $bodylen = $len-$headerlen-8;
  if ($bodylen < 1) {
    // This should never happen - IMPLEMENTATION BUG!
    return null;
  }
  $body = substr($data,$headerlen,$bodylen);
  $data = "";
  if ($bodylen > 0) {
    switch ($method) {
      case 8:
        // Currently the only supported compression method:
        $data = gzinflate($body);
        break;
      default:
        // Unknown compression method
        return false;
    }
  } else {
    // I'm not sure if zero-byte body content is allowed.
    // Allow it for now...  Do nothing...
  }

  // Verifiy decompressed size and CRC32:
  // NOTE: This may fail with large data sizes depending on how
  //       PHP's integer limitations affect strlen() since $isize
  //       may be negative for large sizes.
  if ($isize != strlen($data) || crc32($data) != $datacrc) {
    // Bad format!  Length or CRC doesn't match!
    return false;
  }
  return $data;
}


use:
Copy the codeThe code is as follows:

$html=file_get_contents('https:///');
$html=gzdecode($html);

I will introduce these three methods, which should solve the problem of garbled crawling caused by gzip.