Oh, but Matt, MACs are perfect, so what could you possibly mean?
(Tongue planted firmly in cheek.) ;-)> John T From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Matt Sent: Friday, June 01, 2007 1:59 PM To: [email protected] Subject: Re: [IMail Forum] Alternative sign on page with 2006.2 Mike, I can certainly see the utility here. Like I said, we were both very tired at that point and incapable of doing any reverse engineering of code. The instructions that Mark gave worked great...except that this won't work with Safari. I was told that it barfs somewhere late in some of the AJAX stuff. "What that does is make an ajax request to the login.aspx, saving the source in a string, updating the form data in the string with query param info, writing this updated login form to the page, then submitting that form. Safari seems to be tripping up in the second to last bit. I clearly is making the ajax request, I think its updating the code, and I think that's where it goes no further. This would take some time (with a Mac) to figure out but I did confirm that the login works with Firefox (2.0.0.4 but prob earlier versions too) on the mac." I am thinking about using your code in order to avoid that. Before we dig into it, do you have any idea if you code works with Safari browsers? Why the hell did Apple decide to create it's own browser based on code other than Mozilla anyway??? Thanks, Matt Mike N wrote: The other file is not designed to accept arguments. It presents an alternative by accepting a POSTed login username and password. The "<enteredUser>" and "<enteredPassword>" are substituted by the VBScript. It does require a tiny bit of VBScript understanding to adapt to your environment. The advantage is that the URL is not cached by the browser or proxies; particularly if not using SSL. ----- Original Message ----- From: Marc Rosamond <mailto:[EMAIL PROTECTED]> To: [email protected] Sent: Thursday, May 31, 2007 9:41 AM Subject: RE: [IMail Forum] Alternative sign on page with 2006.2 The other file which was provided by a user appears to work differently from what the article mentions. If you use that file, your username ends up being "<enteredUser>" and "<enteredPassword>" which of course will result in an invalid login message. It doesn't seem to support being able to pass the username and password via the querystring as the original 2006 file did so I am guessing the user who made it intended for you to hardcode your username and password in the "<enteredUser>" and "<enteredPassword>" locations.
