如何扩展Boost Beast异步HTTP服务器示例代码以支持HTTP Range字段？

Great question! The good news is that Boost.Beast (version 1.70 included) has built-in tools to handle HTTP Range requests without requiring you to write custom POSIX file handling logic. You can extend your existing file body code with just a few extra steps—let me walk you through it.

Step 1: Parse the Range Request Header #

First, you'll need to extract and validate the Range header from the incoming HTTP request. Boost.Beast provides the http::parse_range function that does all the heavy lifting here: it checks if the Range header is valid, ensures the requested range fits within the file's bounds, and returns a structured result you can work with.

Step 2: Get the File Size #

Before processing the range, retrieve the total size of the opened file using the body.size() method. This helps validate the range and construct the correct response headers later.

Step 3: Adjust the File Body for the Requested Range #

Since http::file_body supports random access, you can use body.seek() to jump directly to the start of the requested range. No need to read the entire file or handle low-level file operations yourself.

Step 4: Construct the Partial Content Response #

For valid range requests, you'll return a 206 Partial Content status code, set the Content-Range header (using Boost.Beast's http::make_content_range to format it correctly), and set the response's content length to the size of the requested range.

Updated Code Example #

Here's how to integrate this into your existing code snippet:

// Attempt to open the file
beast::error_code ec;
http::file_body::value_type body;
body.open(path.c_str(), beast::file_mode::scan, ec);
if(ec)
    return fail(ec, "open");

// Get total file size
std::uint64_t file_size = body.size(ec);
if(ec)
    return fail(ec, "get file size");

// Check for Range request in headers
auto range_result = http::parse_range(request[http::field::range], file_size);

if(range_result)
{
    // Valid range request: process partial content
    auto const& requested_range = range_result.value();
    
    // Seek to the start of the requested range
    body.seek(requested_range.offset, beast::file_seek::begin, ec);
    if(ec)
        return fail(ec, "seek to range start");
    
    // Build partial content response
    http::response<http::file_body> res{http::status::partial_content, request.version()};
    res.set(http::field::server, BOOST_BEAST_VERSION_STRING);
    res.set(http::field::content_type, mime_type(path));
    // Use built-in helper to format Content-Range header
    res.set(http::field::content_range, http::make_content_range(requested_range, file_size));
    res.content_length(requested_range.length);
    res.body() = std::move(body);
    res.prepare_payload();
    
    // Send the response (use your existing async send logic here)
}
else
{
    // No valid Range request: send full file as before
    http::response<http::file_body> res{http::status::ok, request.version()};
    res.set(http::field::server, BOOST_BEAST_VERSION_STRING);
    res.set(http::field::content_type, mime_type(path));
    res.content_length(file_size);
    res.body() = std::move(body);
    res.prepare_payload();
    
    // Send the response (use your existing async send logic here)
}

Key Notes #

• Error Handling: Don't forget to handle cases where http::parse_range returns an error (e.g., invalid range format, range beyond file size). In those cases, you should return a 416 Requested Range Not Satisfiable status code.
• Efficiency: Using body.seek() is efficient because it leverages the OS's file system capabilities—you're not reading unnecessary parts of the file.
• No Custom File Logic: All file operations and HTTP header formatting are handled by Boost.Beast, so you avoid reinventing the wheel.

内容的提问来源于stack exchange，提问作者user5685755