Curl request with digest auth in PHP for download Bitbucket private repository

问题

I'm try to do this request on php, for download the last source from my Bitbucket private repository:

curl --digest --user user:pass https://bitbucket.org/user/repo/get/tip.zip -o test.zip

in command line its ok, the file download perfect, but in php dont work, this my php code:

$out = fopen('test.zip', 'w+');
$ch = curl_init();
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_VERBOSE, true);
curl_setopt($ch, CURLOPT_HTTPAUTH, CURLAUTH_DIGEST);
curl_setopt($ch, CURLOPT_USERPWD, 'user:pass');
curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false);
curl_setopt($ch, CURLOPT_FILE, $out);
curl_setopt($ch, CURLOPT_HEADER, false);
curl_setopt($ch, CURLOPT_NOBODY, false);
curl_setopt($ch, CURLOPT_URL, 'https://bitbucket.org/user/repo/get/tip.zip');
curl_exec($ch);

This is the response, the login its invalid and the server redirect to the login page:

Error CURL: '' 
Error number: 0
Array
(
    [url] => https://bitbucket.org/account/signin/?next=/user/repo/get/tip.tar.gz
    [content_type] => text/html; charset=utf-8
    [http_code] => 200
    [header_size] => 1099
    [request_size] => 194
    [filetime] => -1
    [ssl_verify_result] => 0
    [redirect_count] => 1
    [total_time] => 1.055465
    [namelookup_time] => 1.5E-5
    [connect_time] => 0.102969
    [pretransfer_time] => 0.216164
    [size_upload] => 0
    [size_download] => 10049
    [speed_download] => 9520
    [speed_upload] => 0
    [download_content_length] => 10049
    [upload_content_length] => 0
    [starttransfer_time] => 0.527512
    [redirect_time] => 0.527519
    [redirect_url] => 
)

Anyone know how I can solve my problem? Thank you very much!!!

回答1:

This code worked fine for me:

define('USERNAME','username');
define('PASSWORD','password');

function download($url, $destination) {
    try {
        $fp = fopen($destination, "w");
        $ch = curl_init();
        curl_setopt($ch, CURLOPT_URL, $url);
        curl_setopt($ch, CURLOPT_HTTPAUTH, CURLAUTH_BASIC);
        curl_setopt($ch, CURLOPT_USERPWD, USERNAME . ":" . PASSWORD);
        curl_setopt($ch, CURLOPT_FILE, $fp);
        curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false);
        curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
        curl_setopt($ch, CURLOPT_TIMEOUT, 30);
        $resp = curl_exec($ch);

        // validate CURL status
        if(curl_errno($ch))
            throw new Exception(curl_error($ch), 500);

        // validate HTTP status code (user/password credential issues)
        $status_code = curl_getinfo($ch, CURLINFO_HTTP_CODE);
        if ($status_code != 200)
            throw new Exception("Response with Status Code [" . $status_code . "].", 500);
    }
    catch(Exception $ex) {
        if ($ch != null) curl_close($ch);
        if ($fp != null) fclose($fp);
        throw new Exception('Unable to properly download file from url=[' + $url + '] to path [' + $destination + '].', 500, $ex);
    }
    if ($ch != null) curl_close($ch);
    if ($fp != null) fclose($fp);
}

download('https://bitbucket.org/brunobraga/playground/get/tip.tar.gz', '/tmp/test.tar.gz');

回答2:

I've given it the customary "20 minute tinker" and I'll be damned, working with CURL and my OSX HTTP Client this does not seem to want to budge, and I've worked successfully with HTTP Digest and Curl in the past.

One suggestion: if the command line is working then why not use the command? Does your production server support exec()?