From mboxrd@z Thu Jan  1 00:00:00 1970
Message-ID: <3e1162e60701021434x2b4645lbadc2eb581264c47@mail.gmail.com>
Date: Tue,  2 Jan 2007 14:34:28 -0800
From: "David Leimbach" <leimy2k@gmail.com>
To: "Fans of the OS Plan 9 from Bell Labs" <9fans@cse.psu.edu>
Subject: Re: [9fans] winwatch
In-Reply-To: <5ae4abb1c9261dbdfe9e4828d5620e27@proxima.alt.za>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
References: <5ae4abb1c9261dbdfe9e4828d5620e27@proxima.alt.za>
Topicbox-Message-UUID: fcd99172-ead1-11e9-9d60-3106f5b1d025

I implemented hide once, then decided I didn't need it :-).  So did a
few other people.

And given your comment about communicating with mice is probably well
warranted :-)

See: http://research.microsoft.com/Users/luca/Papers/Squeak.pdf



On 1/1/07, lucio@proxima.alt.za <lucio@proxima.alt.za> wrote:
> I replaced the code in click() with the following to add a half-baked
> facility to hide a window with winwatch.  It uses the B3-B1 chord, but
> somehow loses the ability to pick a window with B3 when the winwatch
> window is not active.
>
> void
> click(Mouse m)
> {
>         int fd, w, b = m.buttons;
>         Point xy = m.xy;
>         char buf[128];
>
>         do
>                 m = emouse();
>         while(m.buttons == b);
>
>         while (b) {
>                 for(w = 0; w < nwin; w++)
>                         if(ptinrect(xy, win[w].r))
>                                 break;
>                 if(w < nwin) {
>                         sprint(buf, "/dev/wsys/%d/wctl", win[w].n);
>                         switch (b & 7) {
>                                 case 1:
>                                         break;
>                                 case 2:
>                                 case 3:
>                                         break;
>                                 case 4:
>                                 case 6:
>                                 case 7:
>                                         if((fd = open(buf, OWRITE)) < 0)
>                                                 return;
>                                         write(fd, "unhide\n", 7);
>                                         write(fd, "top\n", 4);
>                                         write(fd, "current\n", 8);
>                                         close(fd);
>                                         break;
>                                 case 5:
>                                         if((fd = open(buf, OWRITE)) < 0)
>                                                 return;
>                                         write(fd, "top\n", 4);
>                                         write(fd, "hide\n", 7);
>                                         close(fd);
>                                         break;
>                         }
>                 }
>                 b = m.buttons;
>                 xy = m.xy;
>                 do {
>                         m = emouse();
>                 } while (m.buttons);
>         }
> }
>
> I am a little disappointed as to how complicated it is to implement
> mouse functionality.  It seems a lot easier to expect the interactive
> human to recover from unexpected conditions rather than try to design
> the interaction so no condition is unexpected.  This is not a
> criticism of Plan 9, far from it.  It is the realisation that by
> putting a computer on every desk we (computer programmer, glorified or
> otherwise) have turned computers from convenient slaves to eliminate
> tedium to instruments requiring constant human attention.
>
> ++L
>
>