From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-0.4 required=5.0 tests=AC_FROM_MANY_DOTS,BAYES_00 autolearn=no autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 103376,24a59fbc07128ff8 X-Google-Attributes: gid103376,public X-Google-ArrivalTime: 2001-05-09 07:57:31 PST Path: newsfeed.google.com!newsfeed.stanford.edu!news.tele.dk!193.251.151.101!opentransit.net!wanadoo.fr!isdnet!psinet-france!psiuk-f4!psiuk-p4!uknet!psiuk-n!news.pace.co.uk!nh.pace.co.uk!not-for-mail From: "Marin David Condic" Newsgroups: comp.lang.ada Subject: Re: rename missing in Text_IO Date: Wed, 9 May 2001 10:44:51 -0400 Organization: Posted on a server owned by Pace Micro Technology plc Message-ID: <9dbl52$hd$1@nh.pace.co.uk> References: <9d6rbl$6vs$1@nh.pace.co.uk> <9d8rk2$sgg$1@nh.pace.co.uk> NNTP-Posting-Host: 136.170.200.133 X-Trace: nh.pace.co.uk 989419490 557 136.170.200.133 (9 May 2001 14:44:50 GMT) X-Complaints-To: newsmaster@pace.co.uk NNTP-Posting-Date: 9 May 2001 14:44:50 GMT X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.50.4522.1200 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4522.1200 Xref: newsfeed.google.com comp.lang.ada:7378 Date: 2001-05-09T14:44:50+00:00 List-Id: "Stephen Leake" wrote in message news:uheyuzn0y.fsf@gsfc.nasa.gov... > It would be interesting to hear which parts of Claw you think are "too > thick". My nomination for that category is its use of a hidden task to > serialize access to the GUI. On the other hand, I have not yet > succeeded in building a multi-tasking Windex app, so I can't really > complain! > Never said Claw was "too thick". What I said was that it didn't meet the specification of being simply an Ada equivalent to the Win32api. I think we'd both agree that Claw provides a bit more than what you would see from Win32ada. It does a bit more than just wrap Ada-isms around the Win32api. I don't object to that at all - far from it. I just don't think it is the answer to having a slightly more than "thin" binding to the Win32api. (Also, it isn't likely to end up distributed with all of the Ada/PC targeted compilers.) I wish I had a correct term for this kind of binding. "Thick" kind of implies you've packaged something up to provide similar {rather than identical} services, but in a more friendly way. "Thin" kind of implies that it is just a set of hooks to the routines to provide absolutely identical services from another language - right down to matching data types, etc. There ought to be a word for a binding that provides identical services, but smooths over the things that don't translate well from C to Ada. (A thin binding with "software spackle"? :-) Arguably, that ought to be a "thick" binding with something like Claw becoming a "subsystem" - it provides similar and *extended* services. > Windex is _not_ built on top of the Win32Ada binding, and neither is > Claw. Part of the reason is licensing; the Win32Ada binding is > actually copyright by Microsoft, since it was built by automatic > translation from their C source. It is _not_ "freely distributable"; > the license terms state that you must own a Microsoft development > environment to use Win32Ada. Windex is GMGPL. > Fair enough. My casual attitude towards precise writing may once again have me in trouble! :-) Let me say this: The Ada compilers that I've seen targeted to Windows/PCs have all come with the Win32ada bindings. I don't know of any that don't - maybe you know of one? Hence (even if it isn't a "standard") I could go off and develop a Spackle Binding (tm) that "withed" the Win32ada stuff and pretty much be asured it would work any place I needed it to. One could develop a Spackle Binding (tm) that had in the package body all of the same sorts of pragmas, etc. one finds in the Win32ada binding and sort of reverse engineer the Win32api. That would probably not infringe on anyone's copyright or license. But if Win32ada exists everywhere I'd want to compile my stuff, is that really necessary? > The other part of the reason is that there is lots of stuff in Win32 > that should not be used; either because it is superceded by other > parts of Win32, or there is a better Ada way. > Yes - many aspects of Win32api are Morally Evil and should be avoided like cigarettes, whisky & wild women. Still, if one wants to make a binding to it, one ought to be thorough so that Joe Programmer who is familiar with Win32api can find whatever he might be used to using. Tough call. At some point you're no longer making a binding - but a subsystem. Its not a *bad* thing to make a subsystem, but I'd probably do it in layers -the binding is the binding and the simplification of services rides on top of that. > I haven't worked on Windex in quite a while. I'll probably make > another release when GNAT 3.14p comes out, but I don't anticipate much > beyond that. I just don't seem to have the urge to write Win32 apps > any more; I may do something for my latest toy, a Palm handheld. > Perhaps Windex fits the bill. I've not looked it over. The rest of it comes down to how freely available it is for use. That starts moving from technical questions to business & legal questions. MDC -- Marin David Condic Senior Software Engineer Pace Micro Technology Americas www.pacemicro.com Enabling the digital revolution e-Mail: marin.condic@pacemicro.com Web: http://www.mcondic.com/