Looking at the code for creating sub requests (ap_sub_req_lookup_* and ap_sub_req_method_uri) I get the impression sub requests always have to remain "inside" the path of their original request's URI. That is, the path specified as the new URI for the before mentioned function calls is appended to the original request's URI.
Given an arbitrary request, what would I need to do to create a "sub" request based off the first request which I can use to inspect the server pointed at by the original request ? For example, think of using some sort of authentication detection in a reverse proxy. Upon receiving a request the reverse proxy should inspect the target server by using a configured "authentication URL" to check if the target server requires authentication and then display some nice form/query basic/do whatever. In this scenario the configured "authentication URL" might be outside the path queried by the original request, e.g. original request : uri == "/foo/bar/show_pretty.html" inspection request : uri == "/foo/auth.html" which does not seem possible with the ap_sub_req_* functions, as they will construct something like "/foo/bar/foo/auth.html" or "/foo/bar/auth.html". Is there a way to do this ? Cheers, Thomas
