Recently I needed to get the content of a web page into a project and I didn't want to have WinHttp APIs mixed in my source code so I created a simple WinHttp wrapper named
WinHttpClient takes a URL as a parameter, sends the HTTP request and gets the response. The method that sends the HTTP request is a synchronous method so it should NOT be called in the UI thread.
WinHttpClient is thread safe which means it can be used in many threads at one time.
Using the Code
WinHttpClient is pretty simple. The class diagram is as follows.
For example, the code snap to get the content of http://www.codeproject.com/ is as follows.
wstring httpResponseHeader = client.GetHttpResponseHeader();
wstring httpResponse = client.GetHttpResponse();
There are several things that may interest you.
- The method
SendHttpRequest is a synchronous method and it will retry several times until it succeeds which causes this method to take a long time to finish. So do not use this in a UI thread. It is recommended to create some worker threads and use
WinHttpClient in them since
WinHttpClient is a thread safe class.
- You can parse and enumerate the URLs in the string returned by the method
GetHttpResponse and get the contents of the them and do it over and over again. Then you will get the contents of a whole website.
- URLs with default index pages can be handled well. Fox example:
I don’t really know how the HTTP verb affects the result. So a user can select use the GET or POST verb to send the HTTP request through the parameter of
SendHttpRequest. The default verb is GET.
Points of Interest
MultiByteToWideChar to convert to the wide characters, deleting the buffer I created to hold the output wide leads to a crash if I set the last character to 0. It is strange and I didn't figure it out.
This is my first time writing a technical article. I hope I explained everything clearly.