The direct use of socket to realize the HTTP protocol (download only) Author: querw download the source code downloaded from the HTTP server a file There are many ways, Microsoft is "enthusiastic" provides WinInet classes, using them is also very convenient. Of course, we can also implement these functions themselves. It is easy to achieve breakpoint resume and check updates by formatting the request. There is a support for the HTTP1.1 protocol that provides the HTTP1.1 protocol, directly using the Socket to implement the DLL of the download function, and implement the following functions:
Connecting the host formatting the request head setting, send timeout reception and analyze the response head connection, send, set timeout, receive data, etc. I will not say, Windows Socket is ready, call the corresponding function is OK. To download the file from the server, you must first send a request to the server. The HTTP request head consists of several row strings. The format of the HTTP request header is explained below. Suppose you want to download http://www.sina.com.cn/index.html this page, then the request head is as follows: Chapter 1: Method, request content, HTTP protocol version download can generally use GET methods, request The content is "/index.html" The version of the HTTP protocol refers to the version supported by the browser. It doesn't matter for downloading software, so use the 1.1 version "http / 1.1"; "get /index.html http / 1.1" 2 lines: Host name, format is "Host: Host" in this example: "Host: www.sina.com.cn" line 3: Accepted data type, the download software must be received all data types, so : "Accept: * / *" Chapter 4: Specify the type of browser in some servers will increase or decrease in some content according to the different types of customer server, in this example, "User-Agent: Mozilla / 4.0 ( Compatible; Msie 5.00; Windows 98) "Chapter 5: Connection settings Set to keep the connection:" Connection: Keep-alive "Chapter 6: To achieve the breakpoint, you want to specify where to receive data, The format is as follows: "Range: Bytes = Start position - Terminal location" For example, 500 bytes to read can be written this: "Range: Bytes = 0 - 499"; starting from the 1000th byte: "Range: Bytes = 999 - "Finally, don't forget to add a row of space, indicating the end of the request head. The whole request head is as follows: get /index.html http / 1.1
Host: www.sina.com.cn
Accept: * / *
User-agent: mozilla / 4.0 (compatible; msie 5.00; windows 98)
Connection: Keep-Alive Chttpsocket provides the FormatRequestHeader () function to format the output HTTP request header. The code is as follows: // / Output HTTP request head according to the relative URL of the request
Const Char * Chttpsocket :: FormatRequestHeader (Char * PServer, Char * POBJECT, LONG & LENGTH,
Char * Pcookie, Char * Preferer, Long NFROM,
Long NTO, INT NSERVERTYPE
{
Char szport [10];
Char sztemp [20];
Sprintf (Szport, "% D", M_Port); MEMSET (M_RequestHeader, '' / 0 '', 1024);
/// Chapter 1: Method, the path, version, version
STRCAT (M_RequestHeader, "Get");
STRCAT (M_RequestHeader, POBJECT);
STRCAT (M_RequestHeader, "http / 1.1");
STRCAT (M_RequestHeader, "/ R / N");
/// 2nd line: host
STRCAT (M_RequestHeader, "Host:");
STRCAT (M_RequestHeader, PServer);
STRCAT (M_RequestHeader, "/ R / N");
/// Chapter 3:
IF (prefere! = null)
{
STRCAT (M_RequestHeader, "Referr:");
STRCAT (M_RequestHeader, Preferer);
STRCAT (M_RequestHeader, "/ R / N");
}
/// Chapter 4: Received data type
STRCAT (M_RequestHeader, "Accept: * / *");
STRCAT (M_RequestHeader, "/ R / N");
/// Chapter 5: Browser type
STRCAT (M_RequestHeader, "User-Agent: Mozilla / 4.0 (Compatible; Msie 5.00; Windows 98));
STRCAT (M_RequestHeader, "/ R / N");
/// Chapter 6: Connection settings, keep
STRCAT (M_RequestHeader, "Connection: Keep-Alive");
STRCAT (M_RequestHeader, "/ R / N");
/// Chapter 7: cookie.
IF (Pcookie! = NULL)
{
STRCAT (M_RequestHeader, "SET Cookie: 0");
STRCAT (M_RequestHeader, Pcookie);
STRCAT (M_RequestHeader, "/ R / N");
}
/// Chapter 8: Request data start byte position (key to resume breakpoint)
IF (nfrom> 0)
{
STRCAT (M_RequestHeader, "Range: Bytes =");
_LTOA (NFROM, SZTEMP, 10);
STRCAT (M_RequestHeader, SzTemp);
STRCAT (M_RequestHeader, "-");
IF (NTO> NFROM)
{
_LTOA (NTO, SZTEMP, 10);
STRCAT (M_RequestHeader, SzTemp);
}
STRCAT (M_RequestHeader, "/ R / N");
}
/// Last line: blank line
STRCAT (M_RequestHeader, "/ R / N");
/// Return the result
Length = strlen (M_RequestHeader);
Return M_RequestHead;
} When the request is sent to the server, you can receive the response from the server. The response head is also composed of several row strings, except for the first row and the last space, each line consists of one domain and a value. The first line includes the server's response state, from 2xx to 5xx, each status code has different meaning, details can be viewed to the RFC document download, you can care, you can continue to read data; 3xx indicate the target It has been transferred, the new address is in the "location" domain; 4xx indicates that the client is fault, may be the download address is not right, wait; 5xx indicates server end fault. The field in the head is "Content-Length", "Accept-Ranges", "Content-Type", "Date", "Last-Modified", "Location", etc., download more concerned "Content- Length "Domain and" Location "domain. "Content-Length" means the size of the download file, "location" indicates the actual storage location of the target, and when the response code is 3XX, use the value in this domain to reconnect. The CHTTPSocket class in the attached source code provides the following methods to read the server status code, the value of a domain, the response head and the entire response head: int GetServerstate (); // Return to the server status code - 1 means unsuccessful int getfield (const char * szsession, char * szvalue, int nmaxlength); // Returns a domain value, -1 means unsuccessful
Int getResponseline (Char * pline, int nmaxlength); // Get a line of returning
Const Char * GetResponseHeader (INT & Length);
After obtaining the response, if the response code is 2xx and "content-length" does not equal 0, it means that the download file data can be received, the next job is very simple, call chttpsocket :: removeVie () until the received data The length is equal to the value of "content-length". A complete use process consists of the following steps:
Call the AFXPARSEURL () Analysis URL gets the server and download path; call chttpsocket :: socket () Create a socket; call chttpsocket :: connection () connection server; call chttpsocket :: formatRequestHeader () formatted request head; call chttpsocket :: SendRequest () Send a request header to the server; call chttpsocket :: getServersTate () get the response status code; call chttpsocket :: getfield ("content-length") get the size of the download file; call chttpsocket :: remove () Receive data until data Receive completion; this article comes with source code also includes an example engineering using the ChttpSocket implementation download function. Note that all calls are blocked, so it is best to create a thread for a download task, otherwise it will cause the interface to respond to user input. The program running interface is shown below: This figure shows the request head, the response head, and the download progress. Of course, there are many jobs to do with multi-task multithreaded downloads. This article only discusses one possibility of downloading yourself, I hope to help readers. Welcome to Mail advice.
