home *** CD-ROM | disk | FTP | other *** search
- Date: Fri, 8 Jul 1994 03:11:30 -0400 (EDT)
- Date: Thu, 7 Jul 94 22:51:56 EDT
- Subject: Gem Listing (fwd)
- Subject: Gem Listing
- Subject: Re: Use of the Alternate key
- Subject: Re: available keys
- Subject: Re: available keys
- Subject: Extended object types
- Subject: ^A or ^a or ...
- Date: Fri, 8 Jul 1994 03:11:30 -0400 (EDT)
- Mime-Version: 1.0
- Precedence: bulk
-
- Forwarded message:
- >From goemon@venice.mps.ohio-state.edu Thu Jul 7 22:52:07 1994
- From: Goemon <goemon@venice.mps.ohio-state.edu>
- Message-Id: <9407080251.AA06175@venice.mps.ohio-state.edu>
- Subject: Gem Listing
- To: gem-list-approval@world.std.com
- Date: Thu, 7 Jul 94 22:51:56 EDT
-
-
- To: gem-list-approval@world.std.com
- Subject: Re: Use of the Alternate key
-
- Ofir:
- >I think it makes sense to leave the Alt key for dialogue shortcuts if your
- >dialogues are non-modal. Another possibility is to give the top window
- >priority. Alt+[A-Z] should be OK on all keyboards.
-
- I totally agree. That's why most dialog boxes require that you use the
- ALT-key combination no matter what... As for modal dialogs, I believe the
- best way to handle that is to disable the menubar if the modal dialog
- requires that it be displayed and something be done to it before anything
- else can happen.
- -----
- Subject: Re: available keys
-
- (Whoever),
- >I KNOW it's hard, which is why I suggested using a 1-pixel rectangle.
- >But others are correct... if you track a 1-pixel rectangle, then you'd
- >have to call objc_find every time the mouse moved, which WOULD cause a
- >lot of overhead.
-
- WHAT is so INCREDIBLY DIFFICULT about using objc_find to locate the editable
- object under the mouse?? HERE IS SOME EXAMPLE CODE!!!!!!
-
- do {
- /* Do your event thingie here and get mouse coords in mx and my */
- object = objc_find(tree, mx, my);
- if (tree[object].ob_flags & EDITABLE)
- graf_mouse(TEXT_CRSR, 0L);
- else
- graf_mouse(ARROW, 0L);
- } while(!(exitobject));
-
- Total elapsed time : 0.003 microseconds
-
- WHAT IS SO DIFFICULT ABOUT THIS?!?!?!?!?!?!?!?!?!?!?!?!?!?!?! THIS CREATES
- THE *SAME* AMOUNT OF TIME THAT THE 1-PIXEL RECTANGLE CREATES! WHY MUST YOU
- KEEP INSISTING ON THE HARD WAY OUT?? Are you *REALLY* a coder?
- -----
- Subject: Re: available keys
-
- >>> * Selectric - German
- >> yep - the best fileselector around :)
- >Yes, although I've started using BoxKite now which is nearly as good and
- >supports long filenames with minixfs.
-
- BoxKite and Selectric the best? <laughs> You've obviously not used UIS. :)
-
- And how many >>GEM<< programs support long filenames? :-)
- -----
- Subject: Extended object types
-
- Annius:
- > I don't understand the fuss about having to define extended object
- > types. There are only 256 ext types and they're there for the
- > programmer. Why try and give world-wide meaning to them and make it
- > completely impossible for a programmer to add anything to his/her
- > resource files that deviates from the general proposal?
-
- I think you COMPLETELY missed what the entire idea of a GEM Master Listing
- was for. Take, for example, the Cookie Master Listing. GEM Master Listing
- provides exactly that same purpose: It tells programmers which Extended
- Object Types (and such) have already been used, and if they wish to make
- their programs compatible with those types, they refer to this list.
-
- It is extremely practical and extremely useful for the programmers to use
- this listing so they know whether or not they are clashing with other
- indices. Remember, some programs write global messages, and if another
- program is installed and uses a different GUI, they will clash, and the
- other program may even end up crashing.
-
- > What would be the advantage of a standard? That people can exchange
- > their resource files? Why would they want to do that?
-
- What if you used some GEM library, say E_GEM. Now say you wrote a program
- around it, and then decided to dump E_GEM in favor of some other (newer)
- library? If E_GEM didn't follow the listing and decided to assign messages
- and indices hodge-podge, your task would be damn near impossible.
-
- Also, the indices are there not just to keep programmers from clashing, but
- also as a central reference point for this information, instead of having
- to refer to 60 separate documents which may or may not be in some consistent
- format. By having all the related information in one place in the same format,
- writing programs becomes that much simpler.
-
- Say for example, you wanted to write your program so that it was compatible
- with MultiTOS, Mag!x, Geneva and GEM. Normally you would have to refer to
- *AT LEAST* 4 different documents for that information, and they would likely
- be in different formats.
-
- See the problem?
-
- The GEM Master listing is the solution.
-
- Also, for programmers wanting to exchange messages and events with other
- programs, the GEM Master listing is a handy reference.
-
- If you like to live in your own proprietary world and never ever intend on
- exchanging messages or events with any other programs, then maybe you don't
- need the GEM Master listing.
-
- > I also strongly reject the idea that there should be a list where
- > everyone can propose their brand new WF/WM codes. If GEM needs
- > new functionality, make a proposal based on that need. But it is
- > ridiculous to start a discussion by immediately proposing new
- > WF/WM codes.
-
- Making a proposal would take way too long, and would have to be voted on.
- This way your indices can be checked before you allocate them, to make sure
- you don't stomp on some other program's codes, and that they remain
- compatible. Who knows, maybe looking at the GEM List you will see someone
- else's message that does essentially what you need, and by implementing it
- you also add compatiblity for your program with that message type.
-
- This is analagous to the problem with vector sharing, and is essentially
- the reason why XBRA (and the XBRA registration list) was created.
-
- If you are so against such standards, then why are you even on this
- mailing list?
-
- Of course, some people enjoy re-inventing the wheel. Are you one of those
- people? :-)
-
- > This was of course all MHO.
-
- Same here. :-)
- --------
-
- Subject: ^A or ^a or ...
-
- Ofir Gal:
-
- > > * TrakCom - Dutch/German
- > -> German
- Ja!
-
- > > * Mag!X - German
- > Now it's MagiC
-
- Still unavailable in USA, still doesn't work on anything but german TOS, and
- doesn't even work on Falcon030 ?!?! I don't see much of a future in this
- one :-)
-
- -- Ken Hollis (Bitgate Software)
-
-