c ++ char数组unicode西里尔符号不正确

时间:2014-04-10 04:52:08

标签: c++ unicode cyrillic

我在c ++中有函数请求java servlet。它有参数char *,其中函数写入响应。除了西里尔字母之外,此参数中的符号都可以。示例:РвановРван(事实上,它必须是:ИвановИван)

但是当我将此参数写入文件时,西里尔符号也可以。示例代码:

ofstream outputFile;
outputFile.open("log.txt");
printf("Respond from servlet: %s", parameter);

这是我的功能代码

int sendPostRequest(char *pszPostData, LPCTSTR servletUrl, char* resultBuffer, ofstream &outputFile) {

    outputFile << "============== SENDING REEQUEST ====================" << endl;

    HINTERNET hSession = WinHttpOpen(
        userAgent,
        WINHTTP_ACCESS_TYPE_NO_PROXY,
        WINHTTP_NO_PROXY_NAME,
        WINHTTP_NO_PROXY_BYPASS,
        0);
    if (!hSession)
    {
        _tprintf(_TEXT("Failed to open WinHTTP session: %ld\n"), GetLastError());
        outputFile << "Failed to open WinHTTP session: %ld\n" << GetLastError() << endl;
        return NULL;
    }
    else {
        _tprintf(_TEXT("Oppening WinHTTP session successful: %ld\n"), GetLastError());
        outputFile << "Oppening WinHTTP session successful: %ld\n" << GetLastError() << endl;
    }

    HINTERNET hConnect = WinHttpConnect(
        hSession,
        serverIP,
        serverPort,
        0);
    if (!hConnect)
    {
        _tprintf(_TEXT("Failed to connect to server: %ld\n"), GetLastError());
        outputFile << "Failed to connect to server: %ld\n" << GetLastError() << endl;
        WinHttpCloseHandle(hSession);
        return NULL;
    }
    else {
        _tprintf(_TEXT("Connection to server successful: %ld\n"), GetLastError());
        outputFile << "Connection to server successful: %ld\n" << GetLastError() << endl;
    }

    _tprintf(_TEXT("Post data : %ld\n"), pszPostData);
    outputFile << "Post data : %ld\n" << pszPostData << endl;
    DWORD dwDataLen = strlen(pszPostData);

    HINTERNET hRequest = WinHttpOpenRequest(
        hConnect,
        _TEXT("POST"),
        servletUrl,
        NULL,
        WINHTTP_NO_REFERER,
        WINHTTP_DEFAULT_ACCEPT_TYPES,
        WINHTTP_FLAG_REFRESH);
    if (!hRequest)
    {
        _tprintf(_TEXT("Failed to open request: %ld\n"), GetLastError());
        outputFile << "Failed to open request: %ld\n" << GetLastError() << endl;
        WinHttpCloseHandle(hConnect);
        WinHttpCloseHandle(hSession);
        return -1;
    }
    else {
        _tprintf(_TEXT("Opening request successful: %ld\n"), GetLastError());
        outputFile << "Opening request successful: %ld\n" << GetLastError() << endl;
    }

    DWORD dwReqOpts = 0;
    DWORD dwSize = sizeof(DWORD);
    WinHttpSetOption(
        hRequest,
        WINHTTP_OPTION_SECURITY_FLAGS,
        &dwReqOpts,
        sizeof(DWORD));

    BOOL done = false;
    BOOL rc = WinHttpSendRequest(
        hRequest,
        contentTypeHeader,
        -1,
        (LPVOID)pszPostData,
        dwDataLen,
        dwDataLen,
        NULL);

    if (rc) {
        rc = WinHttpReceiveResponse(hRequest, NULL);
        _tprintf(_TEXT("Sending request successful: %ld\n"), GetLastError());
        outputFile << "Sending request successful: %ld\n" << GetLastError() << endl;
    }
    else
    {
        _tprintf(_TEXT("Send request failed: %ld\n"), GetLastError());
        outputFile << "Send request failed: %ld\n" << GetLastError() << endl;
        WinHttpCloseHandle(hRequest);
        WinHttpCloseHandle(hConnect);
        WinHttpCloseHandle(hSession);
        return -1;
    }

    // Get the status from the server
    DWORD dwCode = 0;
    if (rc)
    {
        rc = WinHttpQueryHeaders(
            hRequest,
            WINHTTP_QUERY_STATUS_CODE | WINHTTP_QUERY_FLAG_NUMBER,
            NULL,
            &dwCode,
            &dwSize,
            NULL);
    }

    if (dwCode != HTTP_STATUS_OK)
    {
        _tprintf(_TEXT("HTTP Request failed: %ld\n"), dwCode);
        outputFile << "HTTP Request failed: %ld\n" << dwCode << endl;
        WinHttpCloseHandle(hRequest);
        WinHttpCloseHandle(hConnect);
        WinHttpCloseHandle(hSession);
    }
    else
    {
        _tprintf(_TEXT("HTTP Request is ok: %ld\n"), dwCode);
        outputFile << "HTTP Request is ok : %ld\n" << dwCode << endl;

        // Keep reading from the remote server until there's nothing left to read
        DWORD dwBytesToBeRead = 0, dwBytesRead = 0;
        //char szBuffer[8192] = { 0 };
        //strcpy(resultBuffer, "");
        do
        {
            if (!WinHttpQueryDataAvailable(hRequest, &dwBytesToBeRead))
            {
                _tprintf(_TEXT("No data available from server? %ld"), GetLastError());
                outputFile << "No data available from server? %ld" << GetLastError() << endl;

                WinHttpCloseHandle(hRequest);
                WinHttpCloseHandle(hConnect);
                WinHttpCloseHandle(hSession);
                return -1;
            }

            if (!WinHttpReadData(
                hRequest,

                //szBuffer,
                resultBuffer,

                //sizeof(szBuffer),
                RESULT_BUFFER_SIZE,

                &dwBytesRead))
            {
                _tprintf(_TEXT("Failed to read data from server: %ld"), GetLastError());
                outputFile << "Failed to read data from server: %ld" << GetLastError() << endl;

                WinHttpCloseHandle(hRequest);
                WinHttpCloseHandle(hConnect);
                WinHttpCloseHandle(hSession);
                return -1;
            }
            if (dwBytesRead > 0)
            {
                //szBuffer[dwBytesRead] = 0;
                resultBuffer[dwBytesRead] = 0; // NULL-terminated returned buffer
            }
        } while (dwBytesRead > 0);
        WinHttpCloseHandle(hRequest);
        WinHttpCloseHandle(hConnect);
        WinHttpCloseHandle(hSession);
        return 0;
    }
    return -1;
}

其中: pszPostData - 发布请求正文 servletUrl - servlet的路径,不包括协议,IP地址和端口 resultBuffer必须由服务器响应写入 outputFile - 日志文件

2 个答案:

答案 0 :(得分:2)

当您将文本数据发布到网络服务器时(实际上,当您使用文本数据完全时),您需要了解正在使用的字符编码。这里,char *pszPostData只指向一些字节。我们知道它们是文本,但它可以用ASCII编码。或UTF-8。或者完全不同的东西。

在网络上,application/x-www-form-urlencoded的默认字符编码为ISO-8859-1或“latin1”,无法对Cyrillic进行编码。 (字符编码将字节映射到字符;在latin1的情况下,它不会将任何字节映射到Cryllic字符,但它会将所有字节映射到某些,这意味着您的数据将通过,但不会解码为正确的字符。)

这是默认值。您可以通过以下方式指定特定编码:

application/x-www-form-urlencoded; charset=<your data's character encoding>

然而,你需要弄明白这一点,因为我不清楚你的数据实际上是什么字符集。(我不确定“ИвановИван”如何解码为“РвановРРІР°” РЅ“)

答案 1 :(得分:0)

示例中的文本编码为UTF8并被视为CP1251。

请考虑阅读Character encoding