home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!uknet!cam-orl!root
- From: thg@cam-orl.co.uk (Tim Glauert)
- Newsgroups: comp.windows.x
- Subject: Re: What would this window obscure?
- Message-ID: <1992Sep3.154113.10359@cam-orl.co.uk>
- Date: 3 Sep 92 15:41:13 GMT
- References: <RJC.92Aug27183542@daiches.cogsci.ed.ac.uk>
- Sender: root@cam-orl.co.uk (Operator)
- Distribution: comp
- Organization: Olivetti Research Ltd, Cambridge, England.
- Lines: 39
-
- In article <RJC.92Aug27183542@daiches.cogsci.ed.ac.uk>, rjc@cogsci.ed.ac.uk (Richard Caley) writes:
- |>
- |> Appologies if this is a simple question. Is there an efficiant way of
- |> finding out which parts of which windows show inside a given
- |> rectangle.
-
- No.
-
- |> I think this is equivalent to asking what expose events
- |> would be generated by creating a window coverring that rectangle and
- |> then removing it.
-
- Actually, it isn't, since the server has a lot of freedom about exacty which
- areas it returns in an expose event. It is allowed to generate expose events
- which cover obscured parts of a window and it is allowed to generate expose
- events on areas which are already visible. It is legal within the protocol
- specification for the server to send a single expose event covering the whole
- of a window whenever any part of it is uncovered.
-
- Of course, if the server implements backing store you have even more problems,
- since you might not get any expose events at all! And no, you can't turn
- backing store off, you can only say that it is "NotUseful".
-
- |> I could calculate this the hard way
-
- Getting the window geometry and calculating it yourself is the only reliable
- method. But it is *very* expensive if your windows are "shaped".
-
- |> but presumably the server can do
- |> it quickly and it would be nice to gbet hold of that information.
-
- Yes it would! I outlined a simple Obscure Extension at the last X technical
- conference, but it's not a high priority for me to get the implementation into
- a exportable state. I will probably do it some time next year,
-
- Tim.
-
- Tim Glauert, Olivetti Research Limited, | thg@cam-orl.co.uk
- 24a Trumpington Street, Cambridge, England. | +44 223 343232
-