du4822 2013-08-13 15:22
浏览 55
已采纳

PHP HTTP 1.1无法执行部分PUT请求

I have a file upload form on my website and when the user uploads its file to my server, I want to transport this uploaded file from my server to the cloud server. File can be large, so I decided to send the file to the cloud with 100KB portions.

Here is my PHP code sample which tries to send the file from my server to the cloud:

<?php
$cloud_server = '%my_server%';
$uploading_file_name = '%folders%/file.mp3';
$auth_token = '%token%';

$file_path = $_SERVER["DOCUMENT_ROOT"] . '/file.mp3';
$file_handler = fopen($file_path, 'rb');
$file_size = filesize($file_path);
$chunk_size = 100 * 1024;

$start_byte_position = 0;

while (!feof($file_handler)) {
    $chunk_content = fread($file_handler, $chunk_size);
    $chunk_content_length = strlen($chunk_content);
    $final_byte_position = $start_byte_position + $chunk_content_length;

    $socket_handler = fsockopen($cloud_server, 80, $errno, $errstr);
    if (!$socket_handler)
        die ($errstr);

    //Headers
    $request = "PUT /{$uploading_file_name} HTTP/1.1
";
    $request .= "Connection: Close
";
    $request .= "Host: {$cloud_server}
";
    $request .= "X-Auth-Token: {$auth_token}
";
    $request .= "Content-Range: {$start_byte_position}-{$final_byte_position}/{$file_size}
";
    $request .= "Content-Length: {$chunk_content_length}
";
    $request .= "
";

    //Body
    $request .=  $chunk_content;
    $request .= "
";

    fwrite($socket_handler, $request);

    $server_response = '';
    while (!feof($socket_handler))
        $server_response .= fgets($socket_handler, 128);
    echo $server_response, '<br><br>';

    $start_byte_position = $final_byte_position;
    fclose($socket_handler);
}

That code practically works, except one thing - each portion overwrites previous content. On the first loop's iteration 100KB file is created on the cloud, on the second iteration these 100KB are replaced with another 100KB. On the last iteration approximately 28KB are uploading, so finally I have 28KB file on the cloud's server instead of whole file. What am I doing wrong? I want these parts to follow each other, not to replace.

Here are server responses if that would help:

HTTP/1.1 201 Created Date: Tue, 13 Aug 2013 15:11:09 GMT Connection: close Server: Selectel_Storage/1.0 last-modified: Tue, 13 Aug 2013 15:11:09 GMT content-length: 0 etag: ea64231f21c952cdb57a5d3109415d09 content-type: text/html; charset=UTF-8

HTTP/1.1 201 Created Date: Tue, 13 Aug 2013 15:11:10 GMT Connection: close Server: Selectel_Storage/1.0 last-modified: Tue, 13 Aug 2013 15:11:09 GMT content-length: 0 etag: 01ab42f6ad58b401284540d6631dae9d content-type: text/html; charset=UTF-8

HTTP/1.1 201 Created Date: Tue, 13 Aug 2013 15:11:10 GMT Connection: close Server: Selectel_Storage/1.0 last-modified: Tue, 13 Aug 2013 15:11:10 GMT content-length: 0 etag: 99bd6f0d3cefd75abc140b9359464d6d content-type: text/html; charset=UTF-8

HTTP/1.1 201 Created Date: Tue, 13 Aug 2013 15:11:10 GMT Connection: close Server: Selectel_Storage/1.0 last-modified: Tue, 13 Aug 2013 15:11:10 GMT content-length: 0 etag: 07cf64ef10e96710b8cb48022ee5dd16 content-type: text/html; charset=UTF-8

................................

HTTP/1.1 201 Created Date: Tue, 13 Aug 2013 15:11:33 GMT Connection: close Server: Selectel_Storage/1.0 last-modified: Tue, 13 Aug 2013 15:11:33 GMT content-length: 0 etag: f01e66c96b21665bbbd19ee5e283a4e1 content-type: text/html; charset=UTF-8

Thanks in advance!

  • 写回答

1条回答 默认 最新

  • dongxuan1660 2013-08-15 21:03
    关注

    Well, I finally found an appropriate solution. As Marc B responded, it's impossible to perform partial PUT request (because this type of request must be idempotent by definition). However, if we speak about uploading files to the huge cloud server, it is possible to use appropriate server's API. As far as I can judge, OpenStack (its Swift component to be precise) is popular enough among file storages.

    So here is brief and clear official documentation.

    Cloud server, which I use, allows uploading large files only with DLO (Dynamic Large Objects) method, so all you need is:

    1) Split your large file into segments (or simply use fseek and fread).

    2) Send these segments to one specified directory, which name will further be considered as whole file's name.

    3) Send special request (create so-called manifest file), imploding all downloaded segments in one file.

    Quite easy!

    Here is a good example from the documentation page using CURL :

    # First, upload the segments
    curl -X PUT -H 'X-Auth-Token: <token>' \
        http://<storage_url>/container/myobject/1 --data-binary '1'
    curl -X PUT -H 'X-Auth-Token: <token>' \
        http://<storage_url>/container/myobject/2 --data-binary '2'
    curl -X PUT -H 'X-Auth-Token: <token>' \
        http://<storage_url>/container/myobject/3 --data-binary '3'
    
    # Next, create the manifest file
    curl -X PUT -H 'X-Auth-Token: <token>' \
        -H 'X-Object-Manifest: container/myobject/' \
        http://<storage_url>/container/myobject --data-binary ''
    
    # And now we can download the segments as a single object
    curl -H 'X-Auth-Token: <token>' \
        http://<storage_url>/container/myobject
    

    Hope this would help somebody.

    本回答被题主选为最佳回答 , 对您是否有帮助呢?
    评论

报告相同问题?

悬赏问题

  • ¥15 如何让企业微信机器人实现消息汇总整合
  • ¥50 关于#ui#的问题:做yolov8的ui界面出现的问题
  • ¥15 如何用Python爬取各高校教师公开的教育和工作经历
  • ¥15 TLE9879QXA40 电机驱动
  • ¥20 对于工程问题的非线性数学模型进行线性化
  • ¥15 Mirare PLUS 进行密钥认证?(详解)
  • ¥15 物体双站RCS和其组成阵列后的双站RCS关系验证
  • ¥20 想用ollama做一个自己的AI数据库
  • ¥15 关于qualoth编辑及缝合服装领子的问题解决方案探寻
  • ¥15 请问怎么才能复现这样的图呀