// Copyright (C) 2020 The Qt Company Ltd. // SPDX-License-Identifier: LicenseRef-Qt-Commercial OR GFDL-1.3-no-invariants-only /*! \example http \examplecategory {Networking} \meta tags {http,network,https,proxy} \title HTTP Client \ingroup examples-network \brief Demonstrates a simple HTTP client. This example demonstrates how a simple HTTP client can fetch files from remote hosts. \image http-example.webp The main work of this example is done in the HttpWindow class. Thus we will focus on that. \snippet http/httpwindow.cpp qnam-download Using QNetworkAccessManager, we begin the download of a resource as pointed to by the \c url. If you are unfamiliar with it or the function used, QNetworkAccessManager::get(), or simply want to look into it in more detail, take a look at its documentation and the documentation for QNetworkReply and QNetworkRequest. \snippet http/httpwindow.cpp connecting-reply-to-slots Above, we connect some of the reply's signals to slots in the class. These slots will take care of both incoming data and finalizing the download/handling errors. \snippet http/httpwindow.cpp networkreply-readyread-1 As for handling the incoming data, since we don't know the maximum download size of any potential input and we don't want to exhaust the memory of any computer which might run the example program, we handle incoming data in QNetworkReply::readyRead() instead of in QNetworkReply::finished(). \snippet http/httpwindow.cpp networkreply-readyread-2 Then we write the data to file as it arrives. It is less convenient, but the application will consume less memory at its peak! \snippet http/httpwindow.cpp sslerrors-1 With the QNetworkReply::sslErrors() signal we can also handle errors that may occur during the TLS handshake when connecting to secure websites (i.e. HTTPS). \snippet http/httpwindow.cpp sslerrors-2 In this example, we show a dialog to the user so that they can choose whether or not to ignore the errors. \snippet http/httpwindow.cpp networkreply-error-handling-1 \snippet http/httpwindow.cpp networkreply-error-handling-2 If an error occurs then QNetworkReply will emit the QNetworkReply::errorOccurred() signal, followed by the QNetworkReply::finished() signal. In this example, we only connect to the latter. We handle any potential error(s) in the respective slot by deleting the file we were writing to, and display the error with our status label. \snippet http/httpwindow.cpp qnam-auth-required-1 If you connect to a website that uses \l{https://developer.mozilla.org/en-US/docs/Web/HTTP/Authentication}{HTTP authentication}, assuming you didn't supply the credentials that should be used ahead of time, you can handle missing credentials when the website requests it. With QNetworkAccessManager, we do this in a slot connected to the signal QNetworkAccessManager::authenticationRequired(). We make this connection once, in the constructor. \snippet http/httpwindow.cpp qnam-auth-required-2 In this example, we show a dialog where the user can either insert a username and password, or cancel. Canceling causes the request to fail. */