Hi,
> > 1) What is the "recommended" way to carry the > > GET/POST request identifier (inserted by the > > developer of the web page) from the client > > to Apache? > > You're not going to get the browser to send > a custom header by any changes you make to the > page, except possibly using Ajax. Javascript isn't > enabled on all browsers, so that's out for a > public web site. Cookies also are often turned > off, so that's out, if it's a public site. > That leaves the QUERY_STRING, mainly. (What > you're calling GET variables). We have another > way that's better for many sites, but there are > patent issues with that method. (My company > is seeking a patent.) Now I am curious, but I bet you are not going to tell me! > Using the query string > means rewriting the content of each page, so > one could do what PHP does and use cookies if > they are available, then failover to query string > if cookies are not sent by the browser. > > This is a bit awkward since the QUERY_STRING in POST can be anything, for example an XML-documents. I wonder why it is impossible to add a HTTP Request Header in Javascript. Does anybody know? With AJAX and ActionScript (or Flex 3) it can be done with a few lines of code. Anyway, I guess I could write an Apache module with a hook-method that attemps to get the identifier from 1) Custom HTTP request header 2) A Cookie 3) QUERY_STRING In this order and DECLINE if not found. Though, how should I deal with, for example, an XML-document in a POST requests? Where should I "hide" the transaction identifier? Best regards, Andrej
