ActiveX and IE using https and TLSv1

Question

I'm starting with the ActiveX - Internet Explorer example found in C:\Program Files\National Instruments\LabVIEW 2009\examples\comm\ axevent.llb.  In the Navigate2::URL parameter, I can type in the IP address of a device I'm trying to talk to and it will serve up some web pages.  In the WebBrowser in the front panel, I can see the pages, fill in form data and send it, and get stuff back (in the form of more html pages).  My problem is that I need to automate this interface, and in the final app, the server is going to send me XML, and not html.  So, what I want to do is to perform GET/POST commands, capture the data, parse/analyze the XML, repeat.  Also, this is all done in secure mode.  In the example above, I can type in https://xx.xx.xx.xx and after the login, I'm in.  I see the opening (secure) page in the webBrowser on the front panel.  What I need is
(1) to understand how to send in more specific commands.  Right now, after I type in the initial https link, my labview code really isn't doing anything.  It's all in the front panel's webBrowser.  I need to be able to send in specific GET commands, like GET https://xx.xx.xx.xx/setupRfData.asp HTTP/1.1.  And specific POST commands like POST https://xx.xx.xx.xx/setupRfData.cgi HTTP/1.1.  What I'm not sure about is how to pass in those parameters to the ActiveX Navigate(2) command.  I see the API at
http://msdn.microsoft.com/en-us/library/aa752093(VS.85).aspx
but it's not really clear how use it with LabVIEW.
(2) In the POST commands, I have to send in data, which I assume goes in the POSTdata field, but the format is unclear.
(3) And the Headers field also is where I'd like to specify certain things like "Connection::Keep-Alive"
(4) Finally, this may not have been clear above, but I need to send this device http commands, and it can send me either html or XML based on an input parameter I provide when I send it commands.  If I say "html" then I would just control the device via the web pages it sends back to me.  When it sends back XML, however, instead of a webBrowser, I could probably just have the XML displayed in an edit box?  My concern, though, is that the WebBrowser is what is taking care of all the secure handshaking for me.  How do I keep the link established and still receive/analyze XML?

Answer 1

BCho wrote:
Hey mrbean,
To address your first question; the web page continues to update even though the VI may not be running because
when you did run the VI you loaded the script for that page. This script continues to loop even though the VI
is not running. With an IE ActiveX container, it is a lot like you have a normal IE page open; it just happens
to also be a part of your front panel. If you negative, say, www.ni.com, you will notice that even though
the page is fully loaded and IE is not still attempting to load the page, it still 'updates'. You may also notice
that links still work, even though the VI is not still running. So, one way to think about and IE ActiveX
container is like a normal IE window that happens to also be in your VI.
In regards to how to send specific GET and POST commands; unfortunately, I am not very familiar with the ins and
outs of that particular ActiveX control. Judging by the msdn page that you pointed out, I would make the same
assumptions about GET and POST transactions. If you don't wire anything to the POSTData input you do a GET
transaction.
What happens if you send the entire URL https://10.4.17.1/setupRfData.axp HTTP/1.1? If you were
to do this same routine in a normal IE window, what URL would you use? What happens if you do include the HTTP/1.1? How about if you include it, do you get a different response? Here is a Thread that gives a little bit of relevant information.
When I type in the entire URL above, I get a message saying "The Web page you are viewing is trying to close the window (yes/no)".  I've read several forums about this, but most make mention of java script updates and IE7.  I'm using IE6 and I'm not doing any scripting.  I did see another post that said that it's possible I need to specify the "Referer", so I'm going to try that shortly by adding it to the Headers input of the Navigate2.
My underlying problem is that the in the current app, the user (me right now) only sees the html pages that the server is serving up (via https/TLSv1).  With that interface, you can enter info into various fields and send them to the server, or click a different tab which behind the scenes performs a GET, but no one has to type any of that.  In our app, we're trying to replace the IE with a labVIEW ActiveX IE and manually build/send these commands so that we can create automated sequences for a production test.  Therefore, I've used a sniffer (WireShark) and have determined what commands/fields are going across the ethernet to the server (and coming back). 
As I mentioned, I tried sending the fully qualified GET message ( URL https://10.4.17.1/setupRfData.axp HTTP/1.1) and saw the ...trying to close message.  Assuming I can get past this, my next issue is accessing the response information coming back from the server.  Right now it's html, but in the final product it's going to be XML.  (1) What is the LabVIEW Invoke Method that allows me access to the response information and (2) Is there a way to setup an Event that would notify me when a response came in.  (3) Can the same WebBrowser that currently displays html be used to display XML