Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa01923;
          26 Dec 96 23:10 PST
To: jay@JIMI.CS.UNLV.EDU
Subject: bug-chimera may 96
Date: Thu, 26 Dec 1996 23:10:47 -0800
From: Jay Nietling <jay@JIMI.CS.UNLV.EDU>


------- Forwarded Messages

Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa14264; 30 Apr 96 15:17 PDT
Received: from kroete2.freinet.de (arh166.ppp.uni-c.dk [130.228.13.166]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id AAA21151 for <bug-chimera@cs.unlv.edu>; Wed, 1 May 1996 00:17:00 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uENhb-0003mzC; Wed, 1 May 96 00:15 MET DST
Message-Id: <m0uENhb-0003mzC@kroete2.freinet.de>
Subject: Jumping to anchors
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Wed, 1 May 1996 00:15:22 +2600 (MET DST)
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1671      

The stuff that was supposed to make Chimera jump to anchors (those URLs
that end with #xxx) didn't work. So here are some patches. As the anchor
is rendered, Chimera now sets the scrollbars and jumps down to it. At the
same time the id=xxx feature from HTML3, which is the same as the
<a name=xxx>, but can be attached to anything, not just an <a ...> has
been added, because it wasn't much more work.

If the anchor you are jumping to is very near the end of the doc, Chimera
jumps twice. Once to place the anchor at the top of the page, and
once to put the bottom of the doc at the bottom of the page. I'm not sure
whether this is what people want. I think some beginners find it very
confusing if an application allows you to scroll the document so far that
the last line is off the top of the page.

Worse, is the situation where you press 'back' to a URL with an anchor.
Chimera will first scroll to the position of the anchor, and then scroll
to the place you were when you pressed back. Probably in this case the
anchor should be ignored, but it's a little module-crossing since it is
the widget and not the HTML code that does the back-scroll-to-the-right-
place trick.

You'll find the patch on https://inet.uni-c.dk/~ehcorry/chimera/
There you'll find:

option_patch.patch      Render popup-options correctly patch for 108
return_patch.patch      Let's you press return to submit a form.
108-3.patch             More fixes to forms (not announced to the list before)
108-4.patch             This one (#xxx patch).

These patches are cumulative, ie. I don't know whether they work out of
order.

- --
Erik Corry ehcorry@inet.uni-c.dk https://inet.uni-c.dk/~ehcorry/

------- Message 2

Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa18101; 30 Apr 96 17:03 PDT
Received: from kroete2.freinet.de (arh125.ppp.uni-c.dk [130.228.13.125]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id CAA29773 for <bug-chimera@cs.unlv.edu>; Wed, 1 May 1996 02:03:40 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uEPMp-0003kJC; Wed, 1 May 96 02:02 MET DST
Message-Id: <m0uEPMp-0003kJC@kroete2.freinet.de>
Subject: Oops
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Wed, 1 May 1996 02:02:02 +0200 (MET DST)
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 691       

Anyone who wants to try my anchor patch (108-4) would probably be well
advised to add this patch, too:

diff -u -r --exclude-from exclusions cfh108-2.0-erik4/html/layout.c cfh108-2.0-erik5/html/layout.c
- --- cfh108-2.0-erik4/html/layout.c      Wed May  1 01:59:47 1996
+++ cfh108-2.0-erik5/html/layout.c      Wed May  1 01:57:19 1996
@@ -557,7 +557,7 @@
   else if (tag == TAG_TD) HandleTD(li, p, end);
   else if (tag == TAG_NOTE) HandleNote(li, p, end);
 
- -  if ((value = MLFindAttribute(p, "id")) != NULL)
+  if (p && (value = MLFindAttribute(p, "id")) != NULL)
   {
     RenderPart *rp = HTMLCreateRP(li, p, SetupName, NULL, NULL, NULL, 0);
     rp->name = value;


Sorry,

- --
Erik Corry

------- Message 3

Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa10195; 1 May 96 14:44 PDT
Received: from kroete2.freinet.de (arh136.ppp.uni-c.dk [130.228.13.136]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id XAA15959; Wed, 1 May 1996 23:44:22 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uEjfW-0003l4C; Wed, 1 May 96 23:42 MET DST
Message-Id: <m0uEjfW-0003l4C@kroete2.freinet.de>
Subject: Plugins
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Wed, 1 May 1996 23:42:41 +0200 (MET DST)
Cc: Mark Podlipec <podlipec@BayNetworks.com>, 
    "Derek B. Noonburg" <derekn@vw.ece.cmu.edu>
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1265      

For things like inline video and inline Postscript/PDF/DVI previewing
it would be very cool to have a protocol for embedding other applications
in Chimera. This is something I know John has been thinking about for
some time.

Netscape have just started testing betas of Netscape 3 (there's a version
number inflation going on there!) which has support for plugins for
Unix (but not Linux :-(). The interface is described on
https://home.netscape.com/eng/mozilla/2.0/handbook/plugins/index.html
and there's a Unix SDK on the ftp site under
ftp://ftp20.netscape.com/sdk/unix/. I would suggest that we support the
same or almost the same interface on the grounds that:

1) It looks like about what we need
2) Someone already went to the trouble of specifying it for us.
   (Not that important, but still.)
3) It should be possible for people like Mark Podlipec (xanim) or the
   xpdf people to support both Chimera and Netscape with one source, and
   perhaps even one binary.
4) Netscape will be forced to keep the interface reasonable stable by
   other plugins.
5) It looks about 50 times simpler than CORBA!

Now all I want is an embedded version of vi, so I can edit the text
fields properly :-)

- --
Erik Corry echorry@inet.uni-c.dk https://inet.uni-c.dk/~ehcorry/

------- Message 4

Received: from iguana.reptiles.org by JIMI.CS.UNLV.EDU id aa19034;
          1 May 96 19:12 PDT
Received: by iguana.reptiles.org (/\##/\ Smail3.1.30.13 #30.5)
	id <m0uEnsS-000346C@iguana.reptiles.org>; Wed, 1 May 96 22:12:44 -0400 (EDT)
Message-Id: <m0uEnsS-000346C@iguana.reptiles.org>
Date: Wed, 1 May 96 22:12:20 -0400 (EDT)
From: Smarasderagd <smar@reptiles.org>
To: bug-chimera@cs.unlv.edu, ehcorry@inet.uni-c.dk
Subject: Re:  Plugins

>Now all I want is an embedded version of vi, so I can edit the text
>fields properly :-)

I rather miss Chimera's old behaviour of firing up an external viewer
for non-inline non-html objects.  I wouldn't mind being able to use an
arbitrary external program to edit text fields, either.

------- Message 5

Received: from halloran-eldar.lcs.mit.edu by JIMI.CS.UNLV.EDU id aa14123;
          4 May 96 2:35 PDT
Received: from freefall.FreeBSD.ORG by halloran-eldar.lcs.mit.edu; (5.65/1.1.8.2/19Aug95-0530PM)
	id AA16972; Sat, 4 May 1996 05:35:27 -0400
Received: from vector.jhs.no_domain (slip139-92-42-152.ut.nl.ibm.net [139.92.42.152])
          by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id CAA17177
          Sat, 4 May 1996 02:34:51 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by vector.jhs.no_domain (8.7.3/8.6.9) with SMTP id BAA05971; Fri, 3 May 1996 01:47:31 +0200 (MET DST)
Message-Id: <199605022347.BAA05971@vector.jhs.no_domain>
X-Authentication-Warning: vector.jhs.no_domain: Host localhost [127.0.0.1] didn't use HELO protocol
To: Erik Corry <ehcorry@inet.uni-c.dk>
Cc: Chimera Hacker List <bug-chimera@cs.unlv.edu>, 
    Mark Podlipec <podlipec@baynetworks.com>, 
    "Derek B. Noonburg" <derekn@vw.ece.cmu.edu>
Subject: Re: Plugins 
From: "Julian H. Stacey" <jhs@freebsd.org>
Reply-To: "Julian H. Stacey" <jhs@freebsd.org>
Organization: Vector Systems Ltd.
Address: Holz Strasse 27d, 80469 Munich, Germany
Phone: +49.89.268616
Fax: +49.89.2608126 (pending modem change)
Web: https://www.freebsd.org/~jhs/
Mailer: EXMH version 1.6.5  95 12 11,  PGP available
In-Reply-To: Your message of "Wed, 01 May 1996 23:42:41 +0200."
             <m0uEjfW-0003l4C@kroete2.freinet.de> 
Date: Fri, 03 May 1996 01:47:31 +0200
Sender: jhs@vector.jhs.no_domain

Hi, Reference:
> From: Erik Corry <ehcorry@inet.uni-c.dk> 
> 
> Now all I want is an embedded version of vi, so I can edit the text
> fields properly :-)

I have patches to vi & chimera so that I can run an xterm adjacent to
chimera, & each time I type a :w in vi, th adjacent chimera automatically 
does a redisplay.
In fact I also have patches for ghostview for same purpose.
The whole thing runs on FreeBSD
The patches are available to be incorporated by authors of chimera,
vi & ghostview, (& have been for maybe 6 months, but I guess they're all busy
people :-)

> 
> --
> Erik Corry echorry@inet.uni-c.dk https://inet.uni-c.dk/~ehcorry/
> 
Julian
- --
Julian H. Stacey	jhs@freebsd.org  	https://www.freebsd.org/~jhs/

------- Message 6

Received: from ttmath.ttu.edu by JIMI.CS.UNLV.EDU id aa28769; 5 May 96 9:59 PDT
Date: Sun, 5 May 1996 11:59:50 -0500 (CDT)
From: Jake Kesinger <kesinger@math.ttu.edu>
To: Bug-Chimera <bug-chimera@cs.unlv.edu>
Subject: Problem with large images (108)
Message-Id: <Pine.SOL.3.91.960504104234.16026C-100000@ttmath>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

I'm not sure  exactly how to explain this, take a look
at https://www.math.ttu.edu/~kesinger/chimera/xcol.html

If an image is loaded directly (i.e. not inlined) that is too large
for the display, 108 doesn't have any scrollbars or anything to 
view the part of the image outside the display.  

Also, is anyone having trouble looking at the PNG test suite off of
https://quest.nasa.jpl.gov/PNG/ ?
(In fact, I can hardly get any non-trivial PNG images to come up).
108 dumps with either a segfault, Bus error, or floating exception.
(I'm using libpng0.88 and zlib1.0).


  --Jake     _    Jake Kesinger (kesinger@math.ttu.edu), Outrageous Liar
LUBBOCK -> _|*~-  https://www.math.ttu.edu/~kesinger/
           \,  _} ``It's a damn poor mind that can only think of one way
             \(    to spell a word.'' (Andrew Jackson)



------- Message 7

Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa02185; 5 May 96 15:39 PDT
Received: from kroete2.freinet.de (arh164.ppp.uni-c.dk [130.228.13.164]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id AAA02001 for <bug-chimera@cs.unlv.edu>; Mon, 6 May 1996 00:39:05 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uGCQh-0003jjC; Mon, 6 May 96 00:37 MET DST
Message-Id: <m0uGCQh-0003jjC@kroete2.freinet.de>
Subject: Re: Problem with large images (108)
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Mon, 6 May 1996 00:37:25 +0200 (MET DST)
In-Reply-To: <Pine.SOL.3.91.960504104234.16026C-100000@ttmath> from "Jake Kesinger" at May 5, 96 11:59:50 am
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 181       

> Also, is anyone having trouble looking at the PNG test suite off of
> https://quest.nasa.jpl.gov/PNG/ ?

 https://quest.jpl.nasa.gov/PNG/

Yes.

- --
Erik Corry ehcorry@inet.uni-c.dk

------- Message 8

Received: from iguana.reptiles.org by JIMI.CS.UNLV.EDU id aa04184;
          5 May 96 19:52 PDT
Received: by iguana.reptiles.org (/\##/\ Smail3.1.30.13 #30.5)
	id <m0uGGPM-000345C@iguana.reptiles.org>; Sun, 5 May 96 22:52:28 -0400 (EDT)
Message-Id: <m0uGGPM-000345C@iguana.reptiles.org>
Date: Sun, 5 May 96 22:52:20 -0400 (EDT)
From: Smarasderagd <smar@reptiles.org>
To: bug-chimera@cs.unlv.edu, kesinger@math.ttu.edu
Subject: Re:  Problem with large images (108)

The PNG problems may be due to structure layout disagreements between
libpng and chimera.  You may want to check that the two have the same
size for "png_struct".  I've also run across a problem which seems to
hang chimera when loading a large image from a local file.

------- Message 9

Received: from ttmath.ttu.edu by JIMI.CS.UNLV.EDU id aa16949; 6 May 96 12:31 PDT
Date: Mon, 6 May 1996 14:31:13 -0500 (CDT)
From: Jake Kesinger <kesinger@math.ttu.edu>
To: Bug-Chimera <bug-chimera@cs.unlv.edu>
Subject: Sorry, my fault (re: PNG)
In-Reply-To: <Pine.SOL.3.91.960504104234.16026C-100000@ttmath>
Message-Id: <Pine.SOL.3.91.960506142446.16521A-100000@ttmath>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Sun, 5 May 1996, Jake Kesinger wrote:
> Also, is anyone having trouble looking at the PNG test suite off of
> https://quest.nasa.jpl.gov/PNG/ ?
> (In fact, I can hardly get any non-trivial PNG images to come up).
> 108 dumps with either a segfault, Bus error, or floating exception.
> (I'm using libpng0.88 and zlib1.0).

Sorry, sorry, sorry.  My fault all along.  One of my header files from
libpng was bad. 
It seems to work now.  Thanks for your patience.




------- Message 10

Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa14624; 7 May 96 17:55 PDT
Received: from citi.umich.edu by citi.umich.edu for bug-chimera@cs.unlv.edu with SMTP; Tue, 07 May 96 20:54:39 -0400
From: Jim Rees <rees@umich.edu>
To: Chimera Lovers <bug-chimera@cs.unlv.edu>
Date: Tue, 07 May 1996 20:54:39 -0400
Subject: Re: Jumping to anchors 
Sender: rees@citi.umich.edu
In-Reply-To: Erik Corry, Wed, 01 May 1996 00:15:22

  If the anchor you are jumping to is very near the end of the doc, Chimera
  jumps twice. Once to place the anchor at the top of the page, and
  once to put the bottom of the doc at the bottom of the page. I'm not sure
  whether this is what people want. I think some beginners find it very
  confusing if an application allows you to scroll the document so far that
  the last line is off the top of the page.

Since no one else has spoken out against this...

In this case it's harmless, in fact it's probably the right thing.  But I
really hate scroll implementations (like Motif) that scroll by different
amounts based on whether or not you are about to scroll past
end-of-document.

When I scroll a page, my eye jumps to where I expect the next line of text
will appear after the scroll.  If the scroll widget normally scrolls by a
half page (or full page or whatever), I find it extremely annoying when it
only scrolls by a few lines simply because I've reached the end of the
document.

If you really must put in a scroll limit so as not to confuse newbies, put
the limit at the last line of the document, not the last full page.  At
least this way it's clear what happened, and I won't waste time looking for
the line of text I just finished reading.

Also, the scrolling amount should be configurable.  I like to scroll by half
pages, but some people prefer scrolling by full pages, or full pages minus a
line.  Whatever.  Make it configurable.

------- Message 11

Received: from x400gate.bnr.ca by JIMI.CS.UNLV.EDU id aa21375; 8 May 96 5:51 PDT
X400-Received:  
 by mta bnr.ca in /PRMD=BNR/ADMD=TELECOM.CANADA/C=CA/; Relayed; Wed, 8 May 1996 08:50:57 -0400 
X400-Received:  
 by /PRMD=BNR/ADMD=TELECOM.CANADA/C=CA/; Relayed; Wed, 8 May 1996 08:50:19 -0400 
X400-Received:  
 by /PRMD=BNR/ADMD=TELECOM.CANADA/C=CA/; Relayed; Wed, 8 May 1996 08:50:00 -0400 
Date:  Wed, 8 May 1996 08:50:00 -0400 
X400-Originator:  /dd.id=0201342/g=raymond/i=re/s=marshall/@bnr.ca 
X400-MTS-Identifier:  
 [/PRMD=BNR/ADMD=TELECOM.CANADA/C=CA/;bcars520.b.162:08.04.96.12.50.19] 
X400-Content-Type:  P2-1984 (2) 
Content-Identifier:  Re: Jumping t... 
From: "raymond (r.e.) marshall" <raym@bnr.ca>
Sender: "raymond (r.e.) marshall" <raym@bnr.ca>
Message-ID:  <"1195 Wed May  8 08:50:25 1996"@bnr.ca> 
To: bug-chimera@cs.unlv.edu
Subject:  Re: Jumping to anchors 

Would it be so difficult to make the scrolling method a user selectable
option?  Like, by page, by half page, bottom line moved to top of page,
bottom two lines moved to top of page, full page (no line shown twice).

As for the double jump at the end of documents, that is only mildly
annoying to me.  My personal preference would be a single jump, that puts
the last line of the document at the bottom of the window.

Upon reflection, these seem like rather small concerns -- "back burner"
issues.  (Just my personal opinion...)

/  Ray
- --------------------------------------+---------------------------------
Raymond E. Marshall <raym@nortel.com> |  My opinions are not necessarily
NorTel, Customer Service              |  endorsed by my employer, etc.
RTP NC, USA    919-992-4731      Alternate access: raym@vnet.net


------- Message 12

Received: from convex.convex.com by JIMI.CS.UNLV.EDU id aa22629;
          8 May 96 8:04 PDT
Received: from mikey.convex.com by convex.convex.com (8.6.4.2/1.35)
	id KAA20863; Wed, 8 May 1996 10:04:54 -0500
Received: by mikey.convex.com (8.6.12/1.28)
	id KAA19786; Wed, 8 May 1996 10:04:22 -0500
From: David DeSimone <fox@mikey.convex.com>
Message-Id: <199605081504.KAA19786@mikey.convex.com>
Subject: Re: Jumping to anchors
To: Chimera Bug List <bug-chimera@cs.unlv.edu>
Date: Wed, 8 May 1996 10:04:21 -0500 (CDT)
Reply-To: David DeSimone <fox@rsn.hp.com>
X-Mailer: ELM [version 2.4 PL24 PGP6]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 909       

Jim Rees wrote:
> 
> I really hate scroll implementations (like Motif) that scroll by
> different amounts based on whether or not you are about to scroll past
> end-of-document.

As long as we're putting in our two cents, I'll raise my paw and say
that I prefer the opposite behavior.  I find it annoying when a document
scrolls its bottom window edge.

As my engineering buddy from long back would say, "Well, if we can't
decide which way it should be, then it'll have to be configurable." 
Then I would grumble because I had to write the code both ways AND allow
some sort of option setting between the two...  :)

- -- 
David DeSimone    | "The doctrine of human equality reposes on this:
fox@convex.hp.com |  that there is no man really clever who has not
Hewlett-Packard   |  found that he is stupid." -- Gilbert K. Chesterson
Convex Tech Ctr   |      PGP: 5B 47 34 9F 3B 9A B0 0D  AB A6 15 F1 BB BE 8C 44

------- Message 13

Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa02745; 8 May 96 16:51 PDT
Received: from kroete2.freinet.de (arh134.ppp.uni-c.dk [130.228.13.134]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id BAA23655; Thu, 9 May 1996 01:51:41 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uHIza-0003l4C; Thu, 9 May 96 01:50 MET DST
Message-Id: <m0uHIza-0003l4C@kroete2.freinet.de>
Subject: Text scrolling left and right
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Thu, 9 May 1996 01:50:02 +0200 (MET DST)
Cc: John Kilburg <john@isri.unlv.edu>
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 9175      

Here's a patch to do something I've been dying to see for some time.
The problem with the Athena text widget has been that it doesn't scroll
sideways, but just paints a black block at the right hand side to 
indicate that there wasn't any space for the text. This has really got
on my nerves, because it seems like a really useless 'feature'.

With this patch, one-line and multi-line text entry fields in forms
will scroll horizontally so as to ensure that the cursor is always
visible. In the case of multi-line text fields there is also a
scrollbar. There is no vertical scrollbar, because the Athena Text
Widget seems to handle vertical scrolling OK on it's own. You can
switch on a vertical scroll bar, but it doesn't work very well in
conjunction with an external horizontal one (when you scroll
horizontally, the vertical scrollbar get scrolled away!).

Problems with this feature (there always are :-)

* If you paste text, the cursor may not be visible afterwards. Press any
  key to scroll it back into view.
* You can't select more text in one go than is visible. Ie there's no
  automatic horizontal panning when selecting with the mouse.
* I use the Porthole widget. I'm not sure when that appeared, but I have
  an ifdef at the moment so it won't work with X11R4 (you get the old
  behaviour). Could someone with X11R5 or X11R4 check whether they have
  /usr/include/X11/Xaw/Porthole.h.
* The modification hasn't been added to the URL window yet, though in 
  principle there's no problem. It's a bit of an internal structure
  problem, as I'm not quite sure where to put a routine like this. Perhaps
  a new common/xutil.c file?

This goes in after the other 5 patches for 108. See
https://inet.uni-c.dk/~ehcorry/chimera/

diff -u -r --exclude-from exclusions cfh108-2.0-erik5/html/form.c cfh108-2.0-erik6/html/form.c
- --- cfh108-2.0-erik5/html/form.c	Thu May  9 01:29:32 1996
+++ cfh108-2.0-erik6/html/form.c	Thu May  9 01:33:12 1996
@@ -16,6 +16,11 @@
 #endif
 
 #include <X11/IntrinsicP.h>
+
+#if XtSpecificationRelease >= 5
+# define HAVE_PORTHOLE
+#endif
+
 #include <X11/StringDefs.h>
 #include <X11/Xaw/Scrollbar.h>
 #include <X11/Xaw/Command.h>
@@ -26,6 +31,12 @@
 #include <X11/Xaw/SimpleMenu.h>
 #include <X11/Xaw/SmeBSB.h>
 
+#ifdef HAVE_PORTHOLE
+# include <X11/Xaw/Porthole.h>
+#else
+# include <X11/Xaw/Paned.h>
+#endif
+
 #include "MxwText.h"
 
 #include "html.h"
@@ -48,6 +59,9 @@
   bool added;
   struct FormStateP *fs;
 
+  /* Text stuff */
+  Widget text;
+
   /* select stuff */
   bool multi;              /* multi-select? */
   struct OptionP *olist;   /* option list */
@@ -230,14 +244,14 @@
     if (xi->id == INPUT_TEXT || xi->id == INPUT_PASSWORD)
     {
       names[i] = n;
- -      values[i] = MxwGetText(xi->w);
+      values[i] = MxwGetText(xi->text);
       if (values[i] != NULL) values[i] = MPStrDup(mp, values[i]);
       i++;
     }
     else if (xi->id == INPUT_TEXTAREA)
     {
       names[i] = n;
- -      values[i] = MxwGetText(xi->w);
+      values[i] = MxwGetText(xi->text);
       if (values[i] != NULL) values[i] = MPStrDup(mp, values[i]);
       i++;
     }
@@ -602,6 +616,83 @@
   }
 }
 
+/*
+ * key_pressed_work_proc
+ *
+ * Does the stuff to make sure the cursor is visible
+ */
+static Boolean
+key_pressed_work_proc(
+  XtPointer client_data)
+{
+  Widget w = (Widget)client_data;
+  Widget sink = NULL;
+  Widget scroll = XtNameToWidget(XtParent(w), "horizontal");
+
+  XtVaGetValues(w, XtNtextSink, &sink, NULL);
+  if (sink != NULL)
+  {
+    Dimension viewport_width;
+    Dimension text_width;
+    Position text_widget_posn;
+    XRectangle rect;
+    float tp;
+    int t;
+
+    XawTextSinkGetCursorBounds(sink, &rect);
+    XtVaGetValues(XtParent(w), XtNwidth, &viewport_width, NULL);
+    XtVaGetValues(w, XtNx, &text_widget_posn, XtNwidth, &text_width, NULL);
+    if(text_widget_posn > (t = - rect.x + viewport_width - rect.width) ||
+       text_widget_posn < (t = - rect.x))
+    {
+      /*
+       * Cursor is not visible, t is the new window posn needed.
+       *
+       * Perhaps you think I should be using XtSetValues or
+       * XawScrollbarSetThumb or even XtMakeGeometryRequest to move the thumb
+       * Nothing works, except this, which seems totally backwards to me.
+       * There is XawViewportSetCoordinates, but that's new since X11R4.
+       */
+      tp = ((float)-t) / ((float)text_width);
+      if (scroll != NULL)
+        XtCallCallbacks(scroll, XtNjumpProc, (XtPointer)(&tp));
+      else
+      {
+        XtWidgetGeometry gr, gr2;
+        gr.x = t;
+        gr.request_mode = CWX;
+        if (XtMakeGeometryRequest(w, &gr, &gr2) == XtGeometryAlmost)
+          XtMakeGeometryRequest(w, &gr2, NULL);
+      }
+    }
+  }
+  /*
+   * Only call once!
+   */
+  return(True);
+}
+
+/*
+ * key_pressed_in_text
+ *
+ * Whenever a key is pressed, the cursor may be moved. So we
+ * have to check the position and scroll the window to make sure
+ * the cursor stays visible. But at the time of the Event, the cursor
+ * may well not yet have been moved. So we schedule a work procedure,
+ * which will be dealt with after the cursor movements have been done.
+ */
+static void
+key_pressed_in_text(
+  Widget w,
+  XtPointer client_data,
+  XEvent *event,
+  Boolean *coninue_to_displatch)
+{
+  XtAppAddWorkProc(XtWidgetToApplicationContext(w), key_pressed_work_proc,
+       (XtPointer)w);
+}
+
+
 static XtActionsRec textActionsList[] =
 {
   { "chimeraFormReturnAction", (XtActionProc)ChimeraFormReturnAction },
@@ -620,6 +711,7 @@
 int type;
 {
   Widget w;
+  Widget text;
   int width, height;
   XFontStruct *font;
   char *value;
@@ -640,12 +732,31 @@
     echo = True;
   }
 
- -  w = MxwCreateText(name, li->w, echo);
+  w = XtVaCreateWidget(
+      "viewport", 
+#ifdef HAVE_PORTHOLE
+      portholeWidgetClass,
+#else
+      panedWidgetClass,
+#endif
+      li->w,
+      XtNborderWidth, 0,
+      XtNwidth, 10,
+      XtNheight, 10,
+      NULL);
+
+  text = XtVaCreateManagedWidget(name, asciiTextWidgetClass, w,
+                                 XtNeditType, XtEtextEdit,
+                                 XtNecho, echo,
+                                 XtNresize, True,
+                                 NULL);
+
+  XtAddEventHandler(text, KeyPressMask, False, key_pressed_in_text, 0);
 
   if(!chimera_form_return_action_registered)
   {
     chimera_form_return_action_registered=true;
- -    XtAppAddActions(XtWidgetToApplicationContext(w),
+    XtAppAddActions(XtWidgetToApplicationContext(text),
               textActionsList, XtNumber(textActionsList));
   }
 
@@ -653,16 +764,16 @@
   {
     value = alloc_string(value);
     HTMLStringSpacify(value, strlen(value));
- -    MxwSetText(w, value);
+    MxwSetText(text, value);
     free_mem(value);
   }
   else
   {
- -    MxwSetText(w, "");
+    MxwSetText(text, "");
   }
 
- -  if ((font = MxwGetTextFont(w)) == NULL) height = 20;
- -  else height = font->ascent + font->descent + 5;
+  if ((font = MxwGetTextFont(text)) == NULL) height = 20;
+  else height = font->ascent + font->descent + 25;
   /* Addition by LRD.  tweaked to john's weird style */
   if ((value = MLFindAttribute(p, "size")) != NULL)
   {
@@ -678,13 +789,14 @@
   rp = CreateInputState(fs, p, width, height, type);
   ci = (InputState *)HTMLRPState(rp);
   ci->w = w;
+  ci->text = text;
 
- -  XtOverrideTranslations(w, XtParseTranslationTable
+  XtOverrideTranslations(text, XtParseTranslationTable
                                     ("<Key>Return: chimeraFormReturnAction()"));
- -  XtOverrideTranslations(w, XtParseTranslationTable
+  XtOverrideTranslations(text, XtParseTranslationTable
                                     ("<Key>Tab: chimeraFormTabAction()"));
 
- -  XtAddCallback(XawTextGetSource(w), XtNcallback, (XtCallbackProc)FormCallback,
+  XtAddCallback(XawTextGetSource(text), XtNcallback, (XtCallbackProc)FormCallback,
                       (XtPointer)rp);
 
   return(rp);
@@ -905,6 +1017,7 @@
 MLElement p;
 bool end;
 {
+  Widget text;
   FormState *fs;
   InputState *ci;
   int width, height;
@@ -933,14 +1046,30 @@
     {
       tatext = "";
     }
- -    w = XtVaCreateWidget("textarea",
- -			 asciiTextWidgetClass, li->w,
+
+    w = XtVaCreateWidget(
+      "viewport", 
+      viewportWidgetClass,
+      li->w,
+      XtNallowVert, false,
+      XtNallowHoriz, true,
+      XtNforceBars, true,
+      XtNuseBottom, true,
+      XtNwidth, 10,
+      XtNheight, 10,
+      NULL);
+
+    text = XtVaCreateManagedWidget("textarea",
+			 asciiTextWidgetClass, w,
 			 XtNeditType, XtEtextEdit,
 			 XtNstring, tatext,
 			 XtNecho, True,
+			 XtNresize, True,
 			 NULL);
- -    
- -    XtVaGetValues(w, XtNfont, &font, NULL);
+
+    XtAddEventHandler(text, KeyPressMask, False, key_pressed_in_text, 0);
+
+    XtVaGetValues(text, XtNfont, &font, NULL);
     if (font == NULL) height = 20;
     else height = font->ascent + font->descent + 2;
     width = XTextWidth(font, "X", 1);
@@ -964,6 +1093,7 @@
     rp = CreateInputState(fs, fs->tap, width, height, INPUT_TEXTAREA);
     ci = (InputState *)HTMLRPState(rp);
     ci->w = w;
+    ci->text = text;
     
     fs->tap = NULL;
   }

- --
Erik Corry ehcorry@inet.uni-c.dk

------- Message 14

Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa04069; 8 May 96 18:06 PDT
Received: from kroete2.freinet.de (arh129.ppp.uni-c.dk [130.228.13.129]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id DAA26900; Thu, 9 May 1996 03:06:11 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uHK9h-0003kJC; Thu, 9 May 96 03:04 MET DST
Message-Id: <m0uHK9h-0003kJC@kroete2.freinet.de>
Subject: Hold that patch
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Thu, 9 May 1996 03:04:33 +0200 (MET DST)
Cc: John Kilburg <john@isri.unlv.edu>
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 272       

Oops. There's a segmentation fault hiding in that text-field patch. I'm
a little too tired, so you'll get the fix tomorrow. This is turning into
a bad habit. (It's to do with the widget being deleted before the
work-proc gets called).

- --
Erik Corry ehcorry@inet.uni-c.dk

------- Message 15

Received: from hydra.ae.utexas.edu by JIMI.CS.UNLV.EDU id aa06202;
          8 May 96 21:11 PDT
Received: by hydra.ae.utexas.edu; id AA22379; Wed, 8 May 1996 23:11:06 -0500
Date: Wed, 8 May 1996 23:11:05 -0500 (CDT)
From: Rob McMullen <rwmcm@hydra.ae.utexas.edu>
To: bug-chimera@cs.unlv.edu
Subject: LessTif version of Chimera?
Message-Id: <Pine.OSF.3.91.960508225353.22392C@hydra.ae.utexas.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

In case you haven't been following the development of LessTif, the GNU 
project to develop a free Motif clone, lots of progress has been made 
recently.  Version 0.50 is available at:

ftp://ftp.hungry.com/pub/hungry/lesstif/lesstif-0.50.tar.gz

There are deficiencies in the Text widget that would prevent forms from 
working properly at the moment (soon to me remedied), but Erik Corry first 
brought this up and I thought that we'd bring it up for general 
discussion on the list:  What about a LessTif version of Chimera?

There are the obvious implications with respect to code size, but the 
more sophisticated toolkit would have possibilities...

Rob
- --
Rob McMullen <rwmcm@mail.ae.utexas.edu>  https://www.ae.utexas.edu/~rwmcm
Freeware Xt Widgets (available on home page):  SciPlot, ListTree, TextField
Member, LessTif core team (XmTextField & XmText) https://www.hungry.com:8000
Water is life!  Texas Aquatics Masters, Austin, TX



------- Message 16

Received: from xph029.physics.montana.edu by JIMI.CS.UNLV.EDU id aa07408;
          8 May 96 22:35 PDT
Received: from xph036.physics.montana.edu (michael@xph036.physics.montana.edu [153.90.240.200]) by xph029.physics.montana.edu (8.6.12/8.6.9) with ESMTP id XAA01434; Wed, 8 May 1996 23:34:13 -0600
Received: from localhost (michael@localhost) by xph036.physics.montana.edu (8.6.12/8.6.12) with SMTP id XAA00322; Wed, 8 May 1996 23:30:26 -0600
X-Authentication-Warning: xph036.physics.montana.edu: michael owned process doing -bs
Date: Wed, 8 May 1996 23:30:23 -0600 (MDT)
From: Michael Kellen <michael@xph029.physics.montana.edu>
To: Rob McMullen <rwmcm@hydra.ae.utexas.edu>
cc: bug-chimera@cs.unlv.edu
Subject: Re: LessTif version of Chimera?
In-Reply-To: <Pine.OSF.3.91.960508225353.22392C@hydra.ae.utexas.edu>
Message-ID: <Pine.LNX.3.92.960508232752.320A-100000@xph036.physics.montana.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Wed, 8 May 1996, Rob McMullen wrote:

> There are deficiencies in the Text widget that would prevent forms from
> working properly at the moment (soon to me remedied), but Erik Corry first
> brought this up and I thought that we'd bring it up for general
> discussion on the list:  What about a LessTif version of Chimera?

I'm against it.  The whole point is NOT to bloat up.  Or require
yet-another-gods-bedamned-dynamic-library-much-less-a-static-one.
And I don't want to even think about using an incomplete toolkit like
lesstif when we generate enough segfaults on our own.

Michael Kellen
{delay inline release RealSoonNow!!}


------- Message 17

Received: from iguana.reptiles.org by JIMI.CS.UNLV.EDU id aa07920;
          8 May 96 23:30 PDT
Received: by iguana.reptiles.org (/\##/\ Smail3.1.30.13 #30.5)
	id <m0uHPFL-00034AC@iguana.reptiles.org>; Thu, 9 May 96 02:30:51 -0400 (EDT)
Message-Id: <m0uHPFL-00034AC@iguana.reptiles.org>
Date: Thu, 9 May 96 02:30:43 -0400 (EDT)
From: Smarasderagd <smar@reptiles.org>
To: michael@xph029.physics.montana.edu, rwmcm@hydra.ae.utexas.edu
Subject: Re: LessTif version of Chimera?
Cc: bug-chimera@cs.unlv.edu

I'm not really against a new widget set.  Motif is nice if you want to
take your hand off the mouse for any length of time, and there are
still serious bugs in athena.  Chimera probably isn't a good choice for
a trial run, though.  NCSA Mosaic seems like a better candidate for
LessTif, if it hasn't been considered already.  Chimera has wandered
rather far from being even a typical athena application, much less a
Motif one.

------- Message 18

Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa09989; 9 May 96 2:24 PDT
Received: from kroete2.freinet.de (arh147.ppp.uni-c.dk [130.228.13.147]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id LAA29828 for <bug-chimera@cs.unlv.edu>; Thu, 9 May 1996 11:24:14 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uHRve-0003l5C; Thu, 9 May 96 11:22 MET DST
Message-Id: <m0uHRve-0003l5C@kroete2.freinet.de>
Subject: Re: LessTif version of Chimera?
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Thu, 9 May 1996 11:22:34 +0200 (MET DST)
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 8613      

Michael Kellen wrote:
> On Wed, 8 May 1996, Rob McMullen wrote:
> 
> > There are deficiencies in the Text widget that would prevent forms from
> > working properly at the moment (soon to me remedied), but Erik Corry first
> > brought this up and I thought that we'd bring it up for general
> > discussion on the list:  What about a LessTif version of Chimera?
> 
> I'm against it.  The whole point is NOT to bloat up.  Or require
> yet-another-gods-bedamned-dynamic-library-much-less-a-static-one.
> And I don't want to even think about using an incomplete toolkit like
> lesstif when we generate enough segfaults on our own.

There's no reason at all why Lesstif should be static.

I can believe that Lesstif will be larger than Athena or Athena 3d, but
I'm rapidly coming to the conclusion that Athena is a toy, a demo. It's
not really the sort of thing you want to base a serious product on. The
contorsions I had to go through to get the text widget to scroll
horizontally (take a look at the patch) are not nice at all, and as I
mentioned, text selection still won't work right. Similar contorsions
will probably be needed to implement mouse-less use, something that
is being seen as more and more important. The Swedish health-and-safety-
at-work organisation TCO (responsible for the strict monitor radiation
emmision standards) are talking about eliminating the mouse alltogether
due to irresolvable ergonomic problems. This is something that Windows
has taken more seriously up till now that X programmers. I know people
who use Windows every day without a mouse! By the way, here's a tip
I don't think most people know: Shift-Insert will insert the current
mouse selection into an X-term. This is _much_ faster than taking your
fingers off the keyboard to insert with the middle button of the mouse.
Doesn't work for Chimera yet...

The real solution would be to rewrite the text widget. Or adapt Rob's
single line text widget to do multiline, but that's plain ridiculous,
because then we'd be duplicating almost exactly the work Rob is doing
to integrate his text widget into Lesstif.

But perhaps we can go for a completely different solution that reduces
points of contact with the toolkit altogether. I'm thinking that the top
row of buttons plus the URL should be rendered as HTML. You would have
something like:

<form method=get action=chimera:newurl>
<input name="url" size=50>
</form>
<a href=chimera:back><img src=lib:back.gif width=32 height=32></a>
<a href=chimera:forward><img src=lib:forward.gif width=32 height=32></a>
<a href=chimera:bookmarks><img src=lib:bookmarks.gif width=32 height=32></a>

The lib: protocol would be a variation of the file: protocol that looks
up the file in Chimera's lib directory, and the chimera: protocol would
be a special pseudo-protocol that makes the correct action happen. This
would make the toolbar nicely extensible, (we could add some sort of
variation on multiple selects for drop-down command menus) with things
like the HTML3 link= stuff, etc. It's the way UdiWWW does it
(https://www.uni-ulm.de/~richter/udiwww/), and I think it makes sense. In
this scheme, popup windows would be small HTML windows too:

<title>Downloading cfh130-2.0.tar.gz</title>
<h1>Downloading /usr/users/erik/cfh130-2.0.tar.gz</h1>
<p align=center><a href=chimera:cancel?downloadid=0x12567854>
  <img source=lib:cancel.gif></a>

Perhaps we should save both ideas for after 2.0 is out. I see that Lesstif
are working on a GNATS-WWW interface so you can report and process bug
reports with a browser. This looks like a really good idea. Perhaps we
can have their code when it's working. I think we could do with a central
bugs-repository to make sure they all get splatted before the release.
Here's my current list of issues (you can add most of the things Michael
posted in his last test-results post):

XMP doesn't work (deprecated, but...)
   EG: https://home.netscape.com/eng/mozilla/2.0/handbook/plugins/ht1.html
Problem with lists not being closed properly?
   EG: https://developer.netscape.com/standards/index.html
POST method for forms
   EG: Add a new film to the Internet Movie Database.
Text mode still has display bugs.
   EG: See https://www.jclark.com/spam.txt
'No host name found'
   EG: https://www.mca.com/universal_pictures/12/phase1/prison_bed.html click on
      airplane
Page-reload-link-back goes back to original page, not the reloaded one
Implicit <p> and </p> not being handled
   EG: https://www.infoweb.dk/biografnet/index.cgi/aarhus_bifhtml2graphics.html
   only the top heading (Biografprogram i ...) should be centred, the rest
   should be left aligned.
Text selection in HTML window
   This is very important for usability.
Accellerators for keyboard scrolling, etc.
   This is very important for usability.
Various image problems
   I guess this is my dept.

There's also the problem of img's where the image comes back with the
wrong content-type. The problem is that 'other leading browsers' ignore
the content-type for img's and just look at the magic number. That means
it works for incorrectly set up web sites, and unfortunately (as a
consequence!) there are quite a lot of these. This one may not really be
soluble. An example is the large image at the top of the page on:
   https://www.mca.com/universal_pictures/12/index.html

By the way, I don't want to sound negative: I think Chimera is really
becoming very nice to use. It just still has a lot of rough edges.

Here as promised is the patch (108-7.patch) to make the horizontally
scrolling text widgets work without crashing. As usual you apply it
by placing yourself in the top directory (the one with the README in
it) and typing

patch -p1 < 108-7.patch

It goes in after the one I sent last night.

diff -u -r --exclude-from exclusions cfh108-2.0-erik6/html/form.c cfh108-2.0-erik7/html/form.c
- --- cfh108-2.0-erik6/html/form.c	Thu May  9 11:01:29 1996
+++ cfh108-2.0-erik7/html/form.c	Thu May  9 11:00:00 1996
@@ -61,6 +61,7 @@
 
   /* Text stuff */
   Widget text;
+  XtWorkProcId wp_id;
 
   /* select stuff */
   bool multi;              /* multi-select? */
@@ -617,6 +618,22 @@
 }
 
 /*
+ * text_destroy_callback
+ *
+ * The text widget has been destroyed - remove any pending work procs
+ */
+static void
+text_destroy_callback(
+  Widget w,
+  XtPointer client_data,
+  XtPointer widget_data)
+{
+  InputState *ci = (InputState *)client_data;
+  if (ci->wp_id != 0) XtRemoveWorkProc(ci->wp_id);
+  ci->wp_id = 0;
+}
+
+/*
  * key_pressed_work_proc
  *
  * Does the stuff to make sure the cursor is visible
@@ -625,10 +642,14 @@
 key_pressed_work_proc(
   XtPointer client_data)
 {
- -  Widget w = (Widget)client_data;
+  InputState *ci = (InputState *)client_data;
+  Widget w = ci->text;
   Widget sink = NULL;
   Widget scroll = XtNameToWidget(XtParent(w), "horizontal");
 
+  XtRemoveCallback(w, XtNdestroyCallback, text_destroy_callback, client_data);
+  ci->wp_id = 0;
+
   XtVaGetValues(w, XtNtextSink, &sink, NULL);
   if (sink != NULL)
   {
@@ -688,8 +709,13 @@
   XEvent *event,
   Boolean *coninue_to_displatch)
 {
- -  XtAppAddWorkProc(XtWidgetToApplicationContext(w), key_pressed_work_proc,
- -       (XtPointer)w);
+  InputState *ci = (InputState *)client_data;
+
+  if (ci->wp_id != 0) return;
+  ci->wp_id = XtAppAddWorkProc(
+       XtWidgetToApplicationContext(w), key_pressed_work_proc,
+       (XtPointer)ci);
+  XtAddCallback(w, XtNdestroyCallback, text_destroy_callback, (XtPointer)ci);
 }
 
 
@@ -751,8 +777,6 @@
                                  XtNresize, True,
                                  NULL);
 
- -  XtAddEventHandler(text, KeyPressMask, False, key_pressed_in_text, 0);
- -
   if(!chimera_form_return_action_registered)
   {
     chimera_form_return_action_registered=true;
@@ -791,6 +815,9 @@
   ci->w = w;
   ci->text = text;
 
+  XtAddEventHandler(text, KeyPressMask, False, key_pressed_in_text,
+                    (XtPointer)ci);
+
   XtOverrideTranslations(text, XtParseTranslationTable
                                     ("<Key>Return: chimeraFormReturnAction()"));
   XtOverrideTranslations(text, XtParseTranslationTable
@@ -1067,8 +1094,6 @@
 			 XtNresize, True,
 			 NULL);
 
- -    XtAddEventHandler(text, KeyPressMask, False, key_pressed_in_text, 0);
- -
     XtVaGetValues(text, XtNfont, &font, NULL);
     if (font == NULL) height = 20;
     else height = font->ascent + font->descent + 2;
@@ -1095,6 +1120,9 @@
     ci->w = w;
     ci->text = text;
     
+    XtAddEventHandler(text, KeyPressMask, False, key_pressed_in_text,
+                      (XtPointer)ci);
+
     fs->tap = NULL;
   }
 

- --
Erik Corry ehcorry@inet.uni-c.dk

------- Message 19

Received: from hydra.ae.utexas.edu by JIMI.CS.UNLV.EDU id aa17175;
          9 May 96 10:16 PDT
Received: by hydra.ae.utexas.edu; id AA00514; Thu, 9 May 1996 12:16:04 -0500
Date: Thu, 9 May 1996 12:16:04 -0500 (CDT)
From: Rob McMullen <rwmcm@hydra.ae.utexas.edu>
To: bug-chimera@cs.unlv.edu
Subject: Re: LessTif version of Chimera?
In-Reply-To: <m0uHRve-0003l5C@kroete2.freinet.de>
Message-Id: <Pine.OSF.3.91.960509121359.516A-100000@hydra.ae.utexas.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Thu, 9 May 1996, Erik Corry wrote:

> Michael Kellen wrote:
> > I'm against it.  The whole point is NOT to bloat up.  Or require
> > yet-another-gods-bedamned-dynamic-library-much-less-a-static-one.
> > And I don't want to even think about using an incomplete toolkit like
> > lesstif when we generate enough segfaults on our own.
>
> There's no reason at all why Lesstif should be static.

(LessTif is shared on most platforms, even Linux-aout and AIX, which are
tricky.)

I hate code bloat as much as probably every user on this list, but if a
Motif interface would be useful shouldn't we consider it?  I find Motif apps
much more convenient than Athena, and computers should be all about our
convenience.  How about something as simple as a menubar with a history list?

Mo/LessTif is of course larger than Athena, but if you use even one other
Motif app, you'll have most of the Motif code loaded as shared already.  Why
not use it?

Granted, LessTif is incomplete and does have problems yet to be sorted out.
An amazing amount does work, however.  I use it almost exclusively now to
develop demo programs for my freeware widgets.  Demo != major application,
but now is the time to start thinking about things if we are going to
consider a LessTif version.

> I can believe that Lesstif will be larger than Athena or Athena 3d, but
> I'm rapidly coming to the conclusion that Athena is a toy, a demo. It's
> not really the sort of thing you want to base a serious product on.

I agree completely.  There is no comparison between the usability and
consistency of the two user interfaces.  Don't laugh, but I have actually
seen people who couldn't figure out an Athena scrollbar without help!

Given two versions of the same app, the Motif version is virtually always
going to be preferred.  I realize that the goals of Chimera aren't to be
Netscape, but will the use of Athena prevent it from becoming accepted as
a browser for the average user?  I think that it might.

Rob
- --
Rob McMullen <rwmcm@mail.ae.utexas.edu>  https://www.ae.utexas.edu/~rwmcm
Freeware Xt Widgets (available on home page):  SciPlot, ListTree, TextField
Member, LessTif core team (XmTextField & XmText) https://www.hungry.com:8000
Water is life!  Texas Aquatics Masters, Austin, TX


------- Message 20

Received: from hydra.ae.utexas.edu by JIMI.CS.UNLV.EDU id aa17299;
          9 May 96 10:19 PDT
Received: by hydra.ae.utexas.edu; id AA00527; Thu, 9 May 1996 12:19:41 -0500
Date: Thu, 9 May 1996 12:19:40 -0500 (CDT)
From: Rob McMullen <rwmcm@hydra.ae.utexas.edu>
To: bug-chimera@cs.unlv.edu
Subject: Re: LessTif version of Chimera?
In-Reply-To: <m0uHPFL-00034AC@iguana.reptiles.org>
Message-Id: <Pine.OSF.3.91.960509100924.23191B-100000@hydra.ae.utexas.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Thu, 9 May 1996, Smarasderagd wrote:

> NCSA Mosaic seems like a better candidate for LessTif, if it hasn't 
> been considered already. 

Yep, FYI Mosaic and xmcd are the 2 big apps that probably get the most 
attention from the core team.  We have an additional big ol' list of 
complicated programs with which to test, but those two are the main test 
apps.

Rob
- --
Rob McMullen <rwmcm@mail.ae.utexas.edu>  https://www.ae.utexas.edu/~rwmcm
Freeware Xt Widgets (available on home page):  SciPlot, ListTree, TextField
Member, LessTif core team (XmTextField & XmText) https://www.hungry.com:8000
Water is life!  Texas Aquatics Masters, Austin, TX



------- Message 21

Received: from igw2.merck.com by JIMI.CS.UNLV.EDU id aa19066; 9 May 96 11:36 PDT
Received: from igw2.merck.com; Thu, 9 May 1996 14:33:55 -0400
Message-Id: <199605091833.OAA27390@igw2>
Received: from mailhost.merck.com(54.3.1.99) by ngatekeeper.merck.com via smap (g3.0.1)
	id sma027331; Thu, 9 May 96 14:33:47 -0400
From: Anthony Starks <anthony_starks@Merck.Com>
Subject: Chimera Java Port?
To: bug-chimera@cs.unlv.edu
Date: Thu, 9 May 1996 14:28:49 -0400 (EDT)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 331       

Has anyone considered porting the guts of
chimera to Java? We could let AWT handle
the UI (no more discussion of Mo/Lesstif vs Athena :)

A java chimera would also be instantly portable to
other platforms as well.

And finally, the totally modular browser that I've
always invisioned would be easier to realize.

- -- 
						Anthony

------- Message 22

Received: from cerebro.snm.com by JIMI.CS.UNLV.EDU id aa19560;
          9 May 96 11:59 PDT
Received: from mageworks.snm.com.snm.com (mageworks.snm.com [204.245.164.3]) by cerebro.snm.com (8.6.9/8.6.9) with ESMTP id PAA02340; Thu, 9 May 1996 15:00:12 -0400
Received: from localhost by mageworks.snm.com.snm.com (SMI-8.6/SMI-SVR4)
	id OAA02099; Thu, 9 May 1996 14:57:13 -0400
Date: Thu, 9 May 1996 14:57:13 -0400 (EDT)
From: Ben Taylor <bent@snm.com>
Reply-To: Ben Taylor <bent@snm.com>
To: Rob McMullen <rwmcm@hydra.ae.utexas.edu>
cc: bug-chimera@cs.unlv.edu
Subject: Re: LessTif version of Chimera?
In-Reply-To: <Pine.OSF.3.91.960509121359.516A-100000@hydra.ae.utexas.edu>
Message-ID: <Pine.GSO.3.93.960509143947.2002B-100000@mageworks.snm.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Thu, 9 May 1996, Rob McMullen wrote:

> I hate code bloat as much as probably every user on this list, but if a
> Motif interface would be useful shouldn't we consider it?  I find Motif apps
> much more convenient than Athena, and computers should be all about our
> convenience.  How about something as simple as a menubar with a history list?

How about something that runs fast.  I *tried* emacs with the motif
interface.  It was so incredibly slow, that I went back to the Athena
widgets.  We're talking Sparc 10/ multiple processors/ 80 Mb of memory.
We're talking about orders of magnitude in speed using Motif vs Athena.

I hate using Netscape for the same basic reason.  (I'll use it, but I
don't have to like it.)  It's slow.  The file browser is incredibly
misconfigured.  Not that Chimera has a file browser for saving or
loading files, but thats another subject.

> > I can believe that Lesstif will be larger than Athena or Athena 3d, but
> > I'm rapidly coming to the conclusion that Athena is a toy, a demo. It's
> > not really the sort of thing you want to base a serious product on.
> 
> I agree completely.  There is no comparison between the usability and
> consistency of the two user interfaces.  Don't laugh, but I have actually
> seen people who couldn't figure out an Athena scrollbar without help!

It's pretty obtuse.

> Given two versions of the same app, the Motif version is virtually always
> going to be preferred.  I realize that the goals of Chimera aren't to be
> Netscape, but will the use of Athena prevent it from becoming accepted as
> a browser for the average user?  I think that it might.

For the average user?  It's got a ways to go (No offense John or anyone
working on it - Thanks) before it's prime-time.  I doubt Athena has as
much to do with it, as the level of functionality does.

Now don't get me wrong.  There are several things I'd like to see really
improved.  But for the quick preview aspect and fast rendering, it really
can't be beat.  I don't want Chimera to lose that feature.

> Rob

Ben Taylor
bent@snm.com


------- Message 23

Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa19552; 9 May 96 12:00 PDT
Received: from citi.umich.edu by citi.umich.edu for bug-chimera@cs.unlv.edu with SMTP; Thu, 09 May 96 14:56:12 -0400
From: Jim Rees <rees@umich.edu>
To: Chimera Lovers <bug-chimera@cs.unlv.edu>
Date: Thu, 09 May 1996 14:56:11 -0400
Subject: Re: LessTif version of Chimera? 
Sender: rees@citi.umich.edu
In-Reply-To: Rob McMullen, Thu, 09 May 1996 12:16:04 CDT

  Given two versions of the same app, the Motif version is virtually always
  going to be preferred.

By you, but not by me.  I have many reasons for preferring athena apps, but
the biggest right now is that all the apps I use are athena or athena-like
(xmh, xterm, xrn, emacs) and I don't want two different interfaces on one
screen.

  I realize that the goals of Chimera aren't to be
  Netscape, but will the use of Athena prevent it from becoming accepted as
  a browser for the average user?  I think that it might.

One of the biggest advantages of chimera is that it's easy to build.  If it
depends on some other package like lesstif, and I can't get that package to
build easily, then I will search for another browser.

------- Message 24

Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa20694; 9 May 96 12:57 PDT
Received: from kroete2.freinet.de (arh139.ppp.uni-c.dk [130.228.13.139]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id VAA19871 for <bug-chimera@cs.unlv.edu>; Thu, 9 May 1996 21:56:59 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uHbnv-0003l5C; Thu, 9 May 96 21:55 MET DST
Message-Id: <m0uHbnv-0003l5C@kroete2.freinet.de>
Subject: Re: Chimera Java Port?
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Thu, 9 May 1996 21:55:12 +0200 (MET DST)
In-Reply-To: <199605091833.OAA27390@igw2> from "Anthony Starks" at May 9, 96 02:28:49 pm
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1354      

> Has anyone considered porting the guts of
> chimera to Java? We could let AWT handle
> the UI (no more discussion of Mo/Lesstif vs Athena :)

The impression I get is that AWT isn't really ready for big programs
yet.

> A java chimera would also be instantly portable to
> other platforms as well.

True, but basically it wouldn't be Chimera any more. I don't think you'd
be able to reuse much of the code at all.

> And finally, the totally modular browser that I've
> always invisioned would be easier to realize.

Chimera already is pretty modular - have you taken a look at the innards?

How about the other way around: Java as a plug-in, either using the Netscape
API or another if that proves inadequate. Ought to be doable, and it has
the great advantage that it takes no memory at all so long as you're not on
a page with any applets. I imagine something that uses dynamic loading, but
perhaps a two-process solution would be possible too.

Also remember that Java is interpreted at the moment on just about all
platforms. That doesn't smell like performance to me. There's a freeware
dynamic Java compiler project called Kaffee, but all I know is the name
and that it's somehow going to be part of the Jolt project. There's
a page on it at https://speedy.redhat.com/jolt/ that I never got time to
check out.

- --
Erik Corry ehcorry@inet.uni-c.dk

------- Message 25

Received: from xph029.physics.montana.edu by JIMI.CS.UNLV.EDU id aa23139;
          9 May 96 14:25 PDT
Received: from localhost (michael@localhost) by xph029.physics.montana.edu (8.6.12/8.6.9) with SMTP id PAA05049 for <bug-chimera@cs.unlv.edu>; Thu, 9 May 1996 15:24:06 -0600
Date: Thu, 9 May 1996 15:24:03 -0600 (MDT)
From: Michael Kellen <michael@xph029.physics.montana.edu>
Reply-To: Michael Kellen <michael@xph029.physics.montana.edu>
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Subject: Re: LessTif version of Chimera?
In-Reply-To: <Pine.OSF.3.91.960509121359.516A-100000@hydra.ae.utexas.edu>
Message-ID: <Pine.LNX.3.92.960509151317.5027A-100000@xph029.physics.montana.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII



I started with:

> > > yet-another-gods-bedamned-dynamic-library-much-less-a-static-one.

Erik Corry missed my point:

> > There's no reason at all why Lesstif should be static.

As did Rob McMullen:

> (LessTif is shared on most platforms, even Linux-aout and AIX, which are
> tricky.)


So obviously I should have been more clear.  :-)

LessTif would have to be available as a static lib for those who compile
the binaries for distribution.  Because not everyone has LessTif.  Not
everyone wants it, or has room for it.  Everyone has Xaw if they have X.

Nothing annoys me more than trying to compile up some app only to find
that to do so I have to go grab eighty Megs worth of other crap.  Hell,
the PNG and JPEG stuff would irritate me if I HAD to have it.  Since I
already *had* them, No Big Deal.

And LessTif JustAin'tStable.  It's cloning a widget set that has its own
bugs and segfaults, and incompletely at that.  Bad, bad, evil, evil.  One
set of alpha code at a time, please.

(Java is a worse idea until some SERIOUS security issues are resolved!).

Michael (opinionated) Kellen


PS, John -- could you update https://www.isri.unlv.edu/~john/chimera/src/ ?
            I'd like to be able to work on the latest code too.  :-)


------- Message 26

Received: from igw2.watson.ibm.com by JIMI.CS.UNLV.EDU id aa23832;
          9 May 96 15:04 PDT
Received: from hawpub.watson.ibm.com (hawpub.watson.ibm.com [9.2.90.19]) by igw2.watson.ibm.com (8.7.4/8.7.1) with SMTP id SAA04811; Thu, 9 May 1996 18:04:09 -0400
Received: by hawpub.watson.ibm.com (AIX 3.2/UCB 5.64/4/8/96)
          id AA34407; Thu, 9 May 1996 18:03:53 -0400
From: Uri Blumenthal <uri@watson.ibm.com>
Message-Id: <9605092203.AA34407@hawpub.watson.ibm.com>
Subject: Re: LessTif version of Chimera?
To: Jim Rees <rees@umich.edu>
Date: Thu, 9 May 1996 18:03:53 -0400 (EDT)
Cc: bug-chimera@cs.unlv.edu
In-Reply-To: <199605092017.QAA07704@igw2.watson.ibm.com> from "Jim Rees" at May 9, 96 02:56:11 pm
Reply-To: uri@watson.ibm.com
X-Mailer: ELM [version 2.4 PL25]
Content-Type: text

Jim Rees says:
>   Given two versions of the same app, the Motif version is virtually always
>   going to be preferred.
> 
> By you, but not by me.  I have many reasons for preferring athena apps, but
> the biggest right now is that all the apps I use are athena or athena-like....

I personally don't care - I can [and do] live with both Motif and Athena
applications on my machine, and they coexist quite peacefully with each
other. I did spend money and bought a "real" [i.e. - working] Motif-2.0
code...

>   I realize that the goals of Chimera aren't to be Netscape, but will
>   the use of Athena prevent it from becoming accepted as a browser......?
> 
> One of the biggest advantages of chimera is that it's easy to build.  If it
> depends on some other package like lesstif, and I can't get that package to
> build easily, then I will search for another browser.

I strongly second this. It is a great advantage of Chimera that it does not
require other packages. If it required Motif, I'd be able to live with it,
as I happen to have Motif installed. If it's LessTif or similar package
that's "almost-compatible" - I'd simply not bother...

Please, KISS. I.e. Keep It SIMPLE.
- -- 
Regards,
Uri		uri@watson.ibm.com
- -=-=-=-=-=-=-
<Disclaimer>

------- Message 27

Received: from hydra.ae.utexas.edu by JIMI.CS.UNLV.EDU id aa24133;
          9 May 96 15:19 PDT
Received: by hydra.ae.utexas.edu; id AA00665; Thu, 9 May 1996 17:19:56 -0500
Date: Thu, 9 May 1996 17:19:55 -0500 (CDT)
From: Rob McMullen <rwmcm@hydra.ae.utexas.edu>
To: bug-chimera@cs.unlv.edu
Subject: Re: LessTif version of Chimera? 
In-Reply-To: <9605092005.AA18866@orion.ae.utexas.edu>
Message-Id: <Pine.OSF.3.91.960509171750.663B-100000@hydra.ae.utexas.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Thu, 9 May 1996, Jim Rees wrote:

>   Given two versions of the same app, the Motif version is virtually always
>   going to be preferred.
> 
> By you, but not by me.  I have many reasons for preferring athena apps, but
> the biggest right now is that all the apps I use are athena or athena-like
> (xmh, xterm, xrn, emacs) and I don't want two different interfaces on one
> screen.

Obviously the same argument could be made for having all Motif apps.  But,
for you and others, a big drawing card of Chimera was its Athena
interface.  That is something we have to consider. 

> One of the biggest advantages of chimera is that it's easy to build.  If it
> depends on some other package like lesstif, and I can't get that package to
> build easily, then I will search for another browser.

I'm the same way; I always groan when a package requires me to download and
configure another library.  But, believe me, major effort has been put into
making LessTif easy to build.  2MB of library source compiles without 
even a single *warning* on most machines.

Noone is going to consider me unbiased here, but consistency of user
interface is important.  There are advantages to both, but for better or
worse the future is moving to Motif.

As long as nobody suggests XForms, we're safe. :)

Rob
- --
Rob McMullen <rwmcm@mail.ae.utexas.edu>  https://www.ae.utexas.edu/~rwmcm
Freeware Xt Widgets (available on home page):  SciPlot, ListTree, TextField
Member, LessTif core team (XmTextField & XmText) https://www.hungry.com:8000
Water is life!  Texas Aquatics Masters, Austin, TX


------- Message 28

Received: from hydra.ae.utexas.edu by JIMI.CS.UNLV.EDU id aa24434;
          9 May 96 15:24 PDT
Received: by hydra.ae.utexas.edu; id AA00673; Thu, 9 May 1996 17:24:12 -0500
Date: Thu, 9 May 1996 17:24:12 -0500 (CDT)
From: Rob McMullen <rwmcm@hydra.ae.utexas.edu>
To: bug-chimera@cs.unlv.edu
Subject: Re: LessTif version of Chimera?
In-Reply-To: <Pine.GSO.3.93.960509143947.2002B-100000@mageworks.snm.com>
Message-Id: <Pine.OSF.3.91.960509172128.663D-100000@hydra.ae.utexas.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Thu, 9 May 1996, Ben Taylor wrote:

> I hate using Netscape for the same basic reason.  (I'll use it, but I
> don't have to like it.)  It's slow.

It is a given that Motif is slower than Athena.  No doubt about that.  But
how much of the slowdown is due to Motif, and how much is Netscape itself?
Netscape is HUGE, and there just isn't that much UI interaction for Motif
to be the cause of the bottlenecks.

(Emacs is emacs.  In whatever UI incarnation, nobody would accuse it of
being fast.  That the Athena version is much faster than Motif...who 
knows?  Maybe the supporting code is slow.)

> For the average user?  It's got a ways to go (No offense John or anyone
> working on it - Thanks) before it's prime-time.  I doubt Athena has as
> much to do with it, as the level of functionality does.

Sure, it's still alpha code, but who knows how Athena has affected the
development path of Chimera.  (Well, I suppose John knows... :)

> Now don't get me wrong.  There are several things I'd like to see really
> improved.  But for the quick preview aspect and fast rendering, it really
> can't be beat.  I don't want Chimera to lose that feature.

Neither do I.  But, how much of that is speed due to Athena, and how much
is due to John's WWW widget?  Replacing the Athena Scrollbar with a Motif
one isn't going to noticably affect the rendering times.

So:  is it Athena that sets Chimera apart or is it speed?  I've seen plenty
of fast Motif apps, and as I've beaten to death in this post, I don't
believe the UI will have much to do at all with the future speed of a
Motif-ied Chimera.

Rob
- --
Rob McMullen <rwmcm@mail.ae.utexas.edu>  https://www.ae.utexas.edu/~rwmcm
Freeware Xt Widgets (available on home page):  SciPlot, ListTree, TextField
Member, LessTif core team (XmTextField & XmText) https://www.hungry.com:8000
Water is life!  Texas Aquatics Masters, Austin, TX


------- Message 29

Received: from hydra.ae.utexas.edu by JIMI.CS.UNLV.EDU id aa25644;
          9 May 96 16:21 PDT
Received: by hydra.ae.utexas.edu; id AA00714; Thu, 9 May 1996 18:21:55 -0500
Date: Thu, 9 May 1996 18:21:54 -0500 (CDT)
From: Rob McMullen <rwmcm@hydra.ae.utexas.edu>
To: bug-chimera@cs.unlv.edu
Subject: Re: LessTif version of Chimera?
In-Reply-To: <Pine.LNX.3.92.960509151317.5027A-100000@xph029.physics.montana.edu>
Message-Id: <Pine.OSF.3.91.960509182103.712B-100000@hydra.ae.utexas.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Thu, 9 May 1996, Michael Kellen wrote:

> LessTif would have to be available as a static lib for those who compile
> the binaries for distribution.  Because not everyone has LessTif.  Not
> everyone wants it, or has room for it.  Everyone has Xaw if they have X.

Not strictly true -- AIX, for instance.  It was distributed on my box here
at work as an optional package that I had to build myself.  Not that it
was a chore to compile, but neither is LessTif.  It is certainly true that
Xaw is freely available for all platforms that support X, though. 

> Nothing annoys me more than trying to compile up some app only to find
> that to do so I have to go grab eighty Megs worth of other crap.  Hell,
> the PNG and JPEG stuff would irritate me if I HAD to have it.  Since I
> already *had* them, No Big Deal.

I would imagine that when LessTif is complete, it would be distributed as
libXm.a by those package vendors that don't have a license for Motif.  (Most
Linux distributions, and I assume FreeBSD as well.)

If you needed Motif for something else, why wouldn't you grab LessTif and
compile it?  Just like PNG and JPEG, you would already have that around just
waiting for LessTif-Chimera. :)  I imagine that Chimera wouldn't be the first
thing that a user would need Motif for, nor would it be the last.

> And LessTif JustAin'tStable.

True enough, but it will be.  But, a LessTif Chimera isn't available today,
either.  Gotta plan before attempting.

> It's cloning a widget set that has its own bugs and segfaults, and
> incompletely at that.

It isn't "incompletely" cloning Motif, but the chimera list isn't the place
for than discussion.

Rob
- --
Rob McMullen <rwmcm@mail.ae.utexas.edu>  https://www.ae.utexas.edu/~rwmcm
Freeware Xt Widgets (available on home page):  SciPlot, ListTree, TextField
Member, LessTif core team (XmTextField & XmText) https://www.hungry.com:8000
Water is life!  Texas Aquatics Masters, Austin, TX


------- Message 30

Received: from homesick.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa28629;
          9 May 96 19:49 PDT
Received: from cygnus.com by HOMESICK.CS.UNLV.EDU id aa11746; 9 May 96 19:42 PDT
Received: from tweedledumb.cygnus.com (tweedledumb.cygnus.com [192.80.44.1]) by cygnus.com (8.6.12/8.6.9) with SMTP id TAA14689; Thu, 9 May 1996 19:41:36 -0700
Received: from maneki-neko.cygnus.com by tweedledumb.cygnus.com (4.1/4.7) id AA24612; Thu, 9 May 96 22:41:29 EDT
Received: by maneki-neko.cygnus.com; (8.7.5/1.1.8.2/20Sep95-0235PM)
	id WAA24817; Thu, 9 May 1996 22:41:26 -0400
Sender: eichin@cygnus.com
To: Rob McMullen <rwmcm@hydra.ae.utexas.edu>
Cc: bug-chimera@cs.unlv.edu
Subject: Re: LessTif version of Chimera?
References: <Pine.OSF.3.91.960509182103.712B-100000@hydra.ae.utexas.edu>
From: Mark Eichin <eichin@cygnus.com>
Date: 09 May 1996 22:41:25 -0400
In-Reply-To: Rob McMullen's message of Thu, 9 May 1996 18:21:54 -0500 (CDT)
Message-Id: <xe1n33huu5m.fsf@maneki-neko.cygnus.com>
Lines: 20
X-Mailer: September Gnus v0.78/Emacs 19.30


> Not strictly true -- AIX, for instance.  It was distributed on my box here
> at work as an optional package that I had to build myself.  Not that it

HP/UX likewise -- there are two ways to get Xaw for HP/UX:
	1) build it yourself from the net
	2) buy the "X terminal" package, for a couple of thousand
dollars, which supports using X terminals off an HP server (and
includes Xaw because xdm uses it, or something like that.)

Nonetheless, the issue is that Athena widgets are ugly, but (pretty
much) available. LessTif is ugly too (sorry, guys, that's what you
*get* when you clone Motif :-) but will probably end up shipping on
most free-unix systems once it works well enough to build Mosaic...

I guess I don't see what's the value of a lesstif-chimera; chimera is
*small* and no-frills; if you want a *tif browser, use mosaic... Of
course, if John wants to experiment seperating the html and http parts
of chimera from the UI, then lesstif could be a good choice for an
*alternate* UI, just as interviews or OI would be...

------- Message 31

Received: from hydra.ae.utexas.edu by JIMI.CS.UNLV.EDU id aa06334;
          10 May 96 7:27 PDT
Received: by hydra.ae.utexas.edu; id AA01072; Fri, 10 May 1996 09:27:47 -0500
Date: Fri, 10 May 1996 09:27:47 -0500 (CDT)
From: Rob McMullen <rwmcm@hydra.ae.utexas.edu>
To: bug-chimera@cs.unlv.edu
Subject: Re: LessTif version of Chimera?
In-Reply-To: <xe1n33huu5m.fsf@maneki-neko.cygnus.com>
Message-Id: <Pine.OSF.3.91.960510092550.1068A-100000@hydra.ae.utexas.edu>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On 9 May 1996, Mark Eichin wrote:

> if John wants to experiment seperating the html and http parts
> of chimera from the UI, then lesstif could be a good choice for an
> *alternate* UI, just as interviews or OI would be...

Obviously people are attached to Athena, and I'm not suggesting that John
devote his time to UI issues rather than the HTML parser.  My opinion is
that a Mo/LessTif interface is essential for Chimera to be taken seriously
in the future, but *Tif development by others as an alternate UI may be the 
only way to go for now.

Rob
- --
Rob McMullen <rwmcm@mail.ae.utexas.edu>  https://www.ae.utexas.edu/~rwmcm
Freeware Xt Widgets (available on home page):  SciPlot, ListTree, TextField
Member, LessTif core team (XmTextField & XmText) https://www.hungry.com:8000
Water is life!  Texas Aquatics Masters, Austin, TX


------- Message 32

Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa04031; 12 May 96 13:57 PDT
Received: from kroete2.freinet.de (arh135.ppp.uni-c.dk [130.228.13.135]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id WAA27550 for <bug-chimera@cs.unlv.edu>; Sun, 12 May 1996 22:57:27 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uIiBC-0003pNC; Sun, 12 May 96 22:55 MET DST
Message-Id: <m0uIiBC-0003pNC@kroete2.freinet.de>
Subject: A moving target?
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Sun, 12 May 1996 22:55:45 +0200 (MET DST)
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1050      

The WWW-consortium have announced a new HTML 'standard' called HTML 3.2
(codenamed Wilbur).  It's looks like HTML 2.0 plus tables, <img
align=left/right>, applets, scripts, and a few other bits of the HTML
3.0 proposal. A lot of the HTML 3.0 proposal (including maths) didn't
make it, and is being worked on, presumably for HTML 4.0. You can see
more on https://www.w3.org/pub/WWW/MarkUp/Activity and on
https://www.w3.org/pub/WWW/MarkUp/Wilbur/. There's a DTD, but there
doesn't seem to be a nice non-SGML-wizards- version of it, like there
was for HTML 3.0 (accesible through my homepage).

Of course, we still have to get HTML 2.0 working, though I don't think
I'm the only one who finds the temptation to start fiddling with post-2.0
stuff almost irresistable.

Is 3.2 a realistic goal for Chimera 2.0, or should we settle for HTML 2.0,
clean up the usability a little and get the damn thing released? The
only thing you need to be able to do with applets and scripts for HTML 3.2
is to ignore them by the way.

- --
Erik Corry ehcorry@inet.uni-c.dk

------- Message 33

Received: from homesick.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa12430;
          13 May 96 8:20 PDT
Received: from isy.liu.se by HOMESICK.CS.UNLV.EDU id aa18664; 13 May 96 8:18 PDT
Received: from dtr.isy.liu.se (dtr.isy.liu.se [130.236.20.22]) by isy.liu.se (8.7.5/8.7.3) with SMTP id RAA10163; Mon, 13 May 1996 17:16:56 +0200 (MET DST)
Received: from ladon.isy.liu.se by dtr.isy.liu.se (4.1/SMI-4.1)
	id AA03881; Mon, 13 May 96 17:16:53 +0200
Message-Id: <XFMail.960513171653.jj@isy.liu.se>
X-Mailer: XFMail 0.4-alpha [p0] on SunOS
Sender: jj@ladon.isy.liu.se
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Mime-Version: 1.0
In-Reply-To: <m0uDdC7-0003n8C@kroete2.freinet.de>
Date: Mon, 13 May 1996 17:07:18 +0200 (MET DST)
Organization: =?iso-8859-1?Q?Link=F6ping?= University
From: Jean-Jacques Moulis <jj@isy.liu.se>
To: Erik Corry <ehcorry@inet.uni-c.dk>
Subject: RE: More patches for forms
Cc: Chimera Hacker List <bug-chimera@cs.unlv.edu>

On 28-Apr-96 Erik Corry wrote:
>>Here are some more forms patches for handling forms in Chimera 108.
.....
>3) It's nice to be able to submit by pressing return in a text input
>   element. This now works if there's only one submit button and only
>   one (one-line) text input element in the form. This is quite a large
>   part of the patch (in html/form.c).
....
>Erik Corry ehcorry@inet.uni-c.dk

I applied the patch, it works for GET but generate an error on the server
with POST.

- ----------------------------------------------------------------------------
Jean-Jacques Moulis                                       Tel:  (013) 281684
ISY                                                       Fax:  (013) 281339
Linkvping Universitet                                  E-mail: jj@isy.liu.se
581 83 Linkvping                             www.dtr.isy.liu.se/dtr/staff/jj


------- Message 34

Received: from xph029.physics.montana.edu by JIMI.CS.UNLV.EDU id aa23279;
          13 May 96 16:24 PDT
Received: from localhost (michael@localhost) by xph029.physics.montana.edu (8.6.12/8.6.9) with SMTP id RAA10889 for <bug-chimera@cs.unlv.edu>; Mon, 13 May 1996 17:22:57 -0600
X-Authentication-Warning: xph029.physics.montana.edu: michael owned process doing -bs
Date: Mon, 13 May 1996 17:22:55 -0600 (MDT)
From: Michael Kellen <michael@xph036.physics.montana.edu>
X-Sender: michael@xph029.physics.montana.edu
Reply-To: Michael Kellen <michael@xph036.physics.montana.edu>
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Subject: [cfh108] Delayed Inlines Release
Message-ID: <Pine.LNX.3.92.960512194939.1223A-100000@xph036.physics.montana.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII


This release of the delayed inline option is much less kludgier than the
last.  The delay status information has been entirely moved into the info
structure for Inlines (html/inline.c) and is accessed by RenderHooks.

The cache checking was reworked to work with the new cache structure (as
of cfh106), which removed the need to stat a cache file.

For now, only text is available instead of the inlined image.  You will
see either the ALT text (if supplied) or "[IMAGE]" or "[LINK]" or "[MAP]"
in the place of the image when delay is on.

Cached images are not delayed.  If WIDTH and HEIGHT are supplied, the text
will be clipped by that size if it is larger.  This allows such images to
be downloaded later without re-rendering the whole page.

USAGE

Add the following to your .chimera/resources:

inline.delay:   1

Start up chimera.  All images should be delayed.  A value of 0 will
disable delays (default).

If you move your mouse over a simple image, the filename should appear in
parentheses in the message bar.  Left or Middle clicking will download and
display the image.  After the image is rendered, a MIDDLE MOUSE click will
pop up the Download dialogue to simplify grabbing images off of the web.

If you move over an Imagemap, you should see the same behaviour, with the
addition of the word "ImageMap" after the filename on the message bar.
After the image is rendered, this will function like a normal imagemap.
Again, a MIDDLE MOUSE click will allow you to grab the image.

If you move over a link, the file name will be followed by the usual URL
display showing the destination of the link.  A left click will FOLLOW THE
LINK.  Middle click will download the image.  After the image is rendered,
this will function like a normal image link (left to follow, middle to
download the linked document).


KNOWN BUGS

IMO, li->tmargin should NOT be applied to inline frames

Multiple instances of the same image on a page are still being annoying.
If you attempt to load any instance but the first, the browser will give a
"unable to render" type error.  It works fine if you click on the first
instance.

Unsized images require re-rendering of the page.  This must not occur
until after the image has been cached or it is not saved.  Currently
InlineMotionCallback checks to see if you are over an unrendered but
cached image, and if so re-renders the page.  This is a HACK.

file:/path URLs should really not be delayed.

inline.delay should be checked for sanity.  Ideally, it should be
possible to change on the fly via a toggle (a la "Source").  The toggle
would not have to be realized in the button bar if the user did not want
it.


This is patched against the distribution cfh108, which I grabbed from
Erik's site.

Please give it a whirl,

Michael




diff -c ../old/src/builtin.c src/builtin.c
*** ../old/src/builtin.c	Mon Apr 15 19:44:56 1996
- --- src/builtin.c	Sun May 12 19:36:29 1996
***************
*** 19,24 ****
- --- 19,28 ----
  extern void InitModule_Image _ArgProto((WWWContext));
  extern void InitModule_Download _ArgProto((WWWContext));
  extern void InitModule_View _ArgProto((WWWContext));
+ #ifdef DELAY_INLINE
+ extern void InitModule_Delayed _ArgProto((WWWContext));
+ #endif
+

  void
  InitChimeraBuiltins(wc)
***************
*** 32,36 ****
- --- 36,43 ----
    InitModule_FTP(wc);
    InitModule_File(wc);
    InitModule_Image(wc);
+ #ifdef DELAY_INLINE
+   InitModule_Delayed(wc);
+ #endif
    return;
  }
diff -c ../old/www/WWW.h www/WWW.h
*** ../old/www/WWW.h	Wed Apr 24 03:04:22 1996
- --- www/WWW.h	Sun May 12 19:36:29 1996
***************
*** 154,159 ****
- --- 154,162 ----
   * frame.c
   */
  int FrameLoad _ArgProto((Frame, URLParts *));
+ #ifdef DELAY_INLINE
+ int FrameCacheLoad _ArgProto((Frame, URLParts *));
+ #endif
  int FrameReload _ArgProto((Frame));
  int FrameBack _ArgProto((Frame));
  int FrameReplace _ArgProto((Frame, URLParts *));
***************
*** 180,185 ****
- --- 183,191 ----
  Document *DocumentCreate _ArgProto((bool));
  void DocumentDestroy _ArgProto((Document *));

+ #ifdef DELAY_INLINE
+ bool FrameCacheQuery _ArgProto((Frame, char *));
+ #endif
  /*
   * gui.c
   */
***************
*** 239,243 ****
- --- 245,252 ----
  WWWID WWWUniqueID _ArgProto((WWWContext));
  int WWWGetUserAuth _ArgProto((WWWContext, URLParts *, int,
                                WWWAuthCallback, void *));
+ #ifdef DELAY_INLINE
+ byte *WWWQueryInline _ArgProto((WWWContext, WWWProto, char *, size_t *));
+ #endif

  #endif
diff -c ../old/www/context.c www/context.c
*** ../old/www/context.c	Mon Apr 22 01:45:54 1996
- --- www/context.c	Sun May 12 19:36:29 1996
***************
*** 338,340 ****
- --- 338,369 ----
  {
    return(-1);
  }
+
+ #ifdef DELAY_INLINE
+ byte *
+ WWWQueryInline(wc, wp, key, size)
+ WWWContext wc;
+ WWWProto wp;
+ char *key;
+ size_t *size;
+ {
+   RenderHooks rh;
+   void *rc;
+
+   *size = 0;
+   if (wc == NULL || wp == NULL) return (NULL);
+   /* Frame `1' is the fake frame used by the cache */
+   if ((wc->flist == NULL) || wc->flist == (Frame )1) return(NULL);
+
+   rc = wc->flist->orc;
+   rh = wc->flist->orh;
+   if (rh.query == NULL) return(NULL);
+   if (rh.search != NULL)
+   {
+     rc = (byte *)(rh.search)(rc, wp->url, size);
+   }
+   if (rc == NULL) return(NULL);
+   return( (rh.query)(rc, key, size) );
+ }
+ #endif
+
diff -c ../old/www/frame.c www/frame.c
*** ../old/www/frame.c	Wed Apr 24 03:19:57 1996
- --- www/frame.c	Sun May 12 19:36:29 1996
***************
*** 463,468 ****
- --- 463,483 ----
    return(0);
  }

+ #ifdef DELAY_INLINE
+ /*
+  * FrameCacheLoad
+  */
+ int
+ FrameCacheLoad(frame, up)
+ struct _Frame *frame;
+ URLParts *up;
+ {
+   if (frame->wp->d != NULL) frame->wp->d->nocache=false;
+   RenderNew(frame, up, STACK_RELOAD);
+   return(0);
+ }
+ #endif
+
  /*
   * FrameHome
   */
diff -c ../old/www/gui.c www/gui.c
*** ../old/www/gui.c	Mon Apr 15 19:03:52 1996
- --- www/gui.c	Sun May 12 21:26:40 1996
***************
*** 92,99 ****
- --- 92,113 ----
  bool with_scroll;
  unsigned int width, height;
  {
+ #ifdef DELAY_INLINE
+   size_t isDelayed = 0;
+
+   if (with_scroll)
+   {
+     if (frame->wp != NULL)
+       WWWQueryInline(frame->wc,frame->wp,"delayed",&isDelayed);
+     else if (frame->orh.query != NULL)
+       (frame->orh.query)(frame->orc, "delayed",  &isDelayed);
+   }
+   WWWSetSize(frame->www, (with_scroll && !isDelayed), width, height);
+   WWWSetDrawSize(frame->www, (with_scroll && !isDelayed), width, height);
+ #else
    WWWSetSize(frame->www, with_scroll, width, height);
    WWWSetDrawSize(frame->www, with_scroll, width, height);
+ #endif
    frame->size_set = True;
    if (frame->size_callback != NULL)
    {
***************
*** 108,114 ****
- --- 122,141 ----
  bool with_scroll;
  unsigned int width, height;
  {
+ #ifdef DELAY_INLINE
+   size_t isDelayed = 0;
+
+   if (with_scroll)
+   {
+     if (frame->wp != NULL)
+       WWWQueryInline(frame->wc,frame->wp,"delayed",&isDelayed);
+     else if (frame->orh.query != NULL)
+       (frame->orh.query)(frame->orc, "delayed",  &isDelayed);
+   }
+   WWWSetDrawSize(frame->www, (with_scroll && !isDelayed), width, height);
+ #else
    WWWSetDrawSize(frame->www, with_scroll, width, height);
+ #endif
    return(0);
  }

diff -c ../old/www/proto.c www/proto.c
*** ../old/www/proto.c	Wed Apr 24 03:49:53 1996
- --- www/proto.c	Sun May 12 19:36:29 1996
***************
*** 117,122 ****
- --- 117,124 ----
  struct _WWWContext *wc;
  struct _WWWProto *wp;
  {
+   size_t isDelayed = 0;
+
    if (wc->iocnt >= wc->maxiocnt)
    {
      wp->waiting = true;
***************
*** 132,137 ****
- --- 134,150 ----
      }
    }

+ #ifdef DELAY_INLINE
+   WWWQueryInline(wp->wc, wp, "delayed", &isDelayed);
+   if (wp->pc == NULL && isDelayed)
+   {
+     if ((wp->ph = GetProtoHooks(wc, "delay-inline")) != NULL)
+     {
+       wp->pc = (wp->ph->init)(wp->up, NULL, wp->ph->class_context, wp);
+     }
+   }
+ #endif
+
    /* Check for a proxy. */
    if (wp->pc == NULL && (wp->pup = GetProxy(wp->mp, wp->up)) != NULL)
    {
***************
*** 411,413 ****
- --- 424,444 ----

    return;
  }
+
+ #ifdef DELAY_INLINE
+ /*
+  * FrameCacheQuery
+  */
+ bool
+ FrameCacheQuery(frame, url)
+ struct _Frame *frame;
+ char *url;
+ {
+   ProtoHooks *ph;
+
+   if ((ph = GetProtoHooks(frame->wc, "cache-read")) != NULL)
+        return( CacheQuery(url, ph->class_context) );
+
+   return(false);
+ }
+ #endif
diff -c ../old/Common.tmpl.dist Common.tmpl.dist
*** ../old/Common.tmpl.dist	Wed Apr 24 03:44:24 1996
- --- Common.tmpl.dist	Sun May 12 19:36:29 1996
***************
*** 78,84 ****
  XCOMM CC = gcc

  /* Add site stuff here */
! CEXTRA_DEFINES = $(OS_DEFINES)
  CEXTRA_LIBS = $(OS_LIBS)

  /*
- --- 78,84 ----
  XCOMM CC = gcc

  /* Add site stuff here */
! CEXTRA_DEFINES = $(OS_DEFINES) -DDELAY_INLINE
  CEXTRA_LIBS = $(OS_LIBS)

  /*
diff -c ../old/html/html.h html/html.h
*** ../old/html/html.h	Sat Apr 27 01:27:25 1996
- --- html/html.h	Sun May 12 19:39:30 1996
***************
*** 290,295 ****
- --- 290,300 ----
  /*
   * inline.c
   */
+ #ifdef DELAY_INLINE
+ void  InlineUndelay _ArgProto((void *));
+ Frame InlineGetFrame _ArgProto((void *));
+ bool  InlineSizeGiven _ArgProto((void *));
+ #endif
  void HandleImg _ArgProto((LayoutInfo *, MLElement, bool));

  /*
***************
*** 382,387 ****
- --- 387,399 ----
  int HTMLLoadAnchor _ArgProto((LayoutInfo *, RenderPart *, bool,int,int,int));
  void HTMLPrintAnchor _ArgProto((LayoutInfo *, RenderPart *, bool, int, int));
  URLParts *HTMLMakeURLParts _ArgProto((LayoutInfo *, MemPool, char *));
+ #ifdef DELAY_INLINE
+ int HTMLLoadInline _ArgProto((LayoutInfo *, RenderPart *, bool, int,int,
+                                             int,int,char *, void *));
+ void HTMLPrintInline _ArgProto((LayoutInfo *, RenderPart *, char *, int));
+ int HTMLSaveInline _ArgProto((LayoutInfo *, RenderPart *, bool,
+                                                   int, int, int, char *));
+ #endif

  /*
   * table.c
diff -c ../old/html/load.c html/load.c
*** ../old/html/load.c	Wed Apr 24 03:25:43 1996
- --- html/load.c	Sun May 12 19:36:30 1996
***************
*** 150,152 ****
- --- 150,310 ----

    return;
  }
+
+ #ifdef DELAY_INLINE
+ #define	InlineOther	0
+ #define InlineImage	1
+ #define InlineAnchor	2
+ #define InlineMap	3
+
+
+ int
+ HTMLLoadInline(li, rp, trans, x, y, button, type, src, callback)
+ LayoutInfo *li;
+ RenderPart *rp;
+ bool trans;
+ int x, y;
+ int button;
+ int type;
+ char *src;
+ void *callback;
+ {
+   if (rp == NULL) return(0);
+
+   if ( (button == 1) && (type == InlineAnchor) )
+   {
+     HTMLLoadAnchor(li, rp, trans, x, y, button);
+   }
+   else
+   {
+     URLParts *up;
+     MemPool mp;
+
+     mp = MPCreate();
+     InlineUndelay(li->over->state);
+     if ((up = HTMLMakeURLParts(li, mp, src)) == NULL)
+       FrameReload(InlineGetFrame(li->over->state));
+     else if (InlineSizeGiven(li->over->state))
+     {
+       InlineUndelay(li->over->state);
+       FrameReplace(InlineGetFrame(li->over->state),up);
+     }
+     else
+     {
+       Frame iframe = InlineGetFrame(li->over->state);
+
+       FrameForceType(iframe, "x-www/x-cache-write");
+       FrameCacheLoad(iframe,up);
+     }
+     MPDestroy(mp);
+   }
+
+
+
+   return(0);
+ }
+
+ void
+ HTMLPrintInline(li, rp, src, type)
+ LayoutInfo *li;
+ RenderPart *rp;
+ char *src;
+ int type;
+ {
+   char *buf,*target,*source,*url;
+   URLParts *up;
+   MemPool mp;
+
+   if (rp == NULL) return;
+   if (src == NULL) return;
+
+   mp = MPCreate();
+   if (type == InlineAnchor)
+   {
+     if(rp->anchor == NULL)
+     {
+       MPDestroy(mp);
+       return;
+     }
+
+     if ((url = MLFindAttribute(rp->anchor, "href")) == NULL)
+     {
+       MPDestroy(mp);
+       return;
+     }
+
+     if ((up = HTMLMakeURLParts(li, mp, url)) == NULL)
+     {
+       MPDestroy(mp);
+       return;
+     }
+
+     target = URLMakeString(mp, up, true);
+   }
+   else if (type == InlineMap)
+   {
+     target = MPStrDup(mp, "ImageMap");
+   }
+   else
+   {
+     target = MPStrDup(mp, "");
+   }
+   source = GetBaseFilename(src);
+
+   buf = (char *)MPGet(mp, strlen(source) + strlen(target) + 4);
+   sprintf (buf, "(%s) %s", source, target);
+
+   FrameMessage(li->frame, buf);
+
+   MPDestroy(mp);
+
+   return;
+ }
+
+
+ int
+ HTMLSaveInline(li, rp, trans, x, y, button, src)
+ LayoutInfo *li;
+ RenderPart *rp;
+ bool trans;
+ int x, y;
+ int button;
+ char *src;
+ {
+   Frame dframe;
+   char *ext;
+   WWWContext wc = WWWFrameToContext(li->frame);
+
+   if (rp == NULL) return(-1);
+
+   if (button == 2)
+   {
+     URLParts *up;
+     MemPool mp;
+
+     mp = MPCreate();
+     if ((up = HTMLMakeURLParts(li, mp, src)) == NULL)
+     {
+       MPDestroy(mp);
+       return(-1);
+     }
+     /*
+      * Load it up in a frame forced to a special type and hope the caller
+      * can deal with this.
+      */
+     if ((ext = WWWGetResource(wc, "www.lastResortType")) != NULL)
+     {
+       if ((dframe = FrameCreate(wc, NULL)) != NULL)
+       {
+         FrameForceType(dframe, ext);
+         FrameLoad(dframe, up);
+       }
+     }
+
+     MPDestroy(mp);
+   }
+   else
+     HTMLLoadAnchor(li, rp, false, x, y, button);
+   return(0);
+ }
+ #endif   /* DELAY_INLINE */
diff -c ../old/html/inline.c html/inline.c
*** ../old/html/inline.c	Sat Apr 27 01:29:26 1996
- --- html/inline.c	Mon May 13 17:09:31 1996
***************
*** 12,22 ****
- --- 12,47 ----

  #include "html.h"

+ #ifdef DELAY_INLINE
+ #include <X11/IntrinsicP.h>
+ #include <X11/StringDefs.h>
+
+ #ifndef ICONS
+ const char inline_icon[4][8] = { "", "[IMAGE]", "[LINK]", "[MAP]" };
+ const char inline_mime[4][12] = { "text/plain", "text/plain",
+                                 "text/plain", "text/plain"};
+ const int inline_length[4] = { 0, 7, 6, 5 };
+ #else
+ #include "icons.h"
+ #endif
+
+ #endif
+
  typedef struct inline_state
  {
    LayoutInfo *li;  /* need this always */
    bool added;      /* added yet */
    Frame frame;     /* embedded document ID */
+ #ifdef DELAY_INLINE
+   /* Inline delay information */
+   char        *src;                    /* SRC field info      */
+   enum        {available, requested, unavailable} status;
+   char        *alt_text;               /* either text or icon */
+   char        *alt_MIME;               /* MIME type for alt   */
+   int         alt_length;              /* Length of Alt data  */
+   enum        {Other, InlineImage, InlineAnchor, InlineMap} type;
+   bool        size_given;              /* HEIGHT & WIDTH specified? */
+ #endif
  } InlineState;

  /*
***************
*** 69,78 ****
    int offset;

    if (ci->added) return;
- -   ci->added = True;

    if (status == -1)
    {
      rp = HTMLCreateRP(ci->li, NULL, NULL, NULL, NULL, NULL, 0);
      HTMLAddRP(ci->li, rp, ADD_NOTHING, 0, 0, 0);
    }
- --- 94,109 ----
    int offset;

    if (ci->added) return;

    if (status == -1)
    {
+ #ifdef DELAY_INLINE
+     ci->status = unavailable;
+     /*
+      *  Should now (re)render the ALT frame if need be
+      */
+ #endif
+
      rp = HTMLCreateRP(ci->li, NULL, NULL, NULL, NULL, NULL, 0);
      HTMLAddRP(ci->li, rp, ADD_NOTHING, 0, 0, 0);
    }
***************
*** 88,96 ****
- --- 119,136 ----
      else if (av == ATTRIB_MIDDLE || av == ATTRIB_CENTER) offset = height / 2;
      else offset = 0;

+ #ifdef DELAY_INLINE
+     if (ci->status == requested)
+       ci->status = available;
+ #endif
      HTMLAddRP(ci->li, rp, ADD_SETBL, offset, width, height);
    }

+ #ifdef DELAY_INLINE
+   if (ci->status == available)
+ #endif
+   ci->added = True;
+
    return;
  }

***************
*** 107,113 ****
    InlineState *ci = HTMLRPState(rp);
    LayoutInfo *li = ci->li;

!   HTMLLoadAnchor(li, rp, false, x, y, button);

    return(0);
  }
- --- 147,164 ----
    InlineState *ci = HTMLRPState(rp);
    LayoutInfo *li = ci->li;

! #ifdef DELAY_INLINE
!   InlineState *ti;
!
!   ti = (InlineState *)(li->over->state);
!   if (ti->status == unavailable)
!     HTMLLoadInline(li, rp, false, x, y, button,
!                                     ti->type, ti->src, RealizeCallback);
!   else if (ti->type != InlineAnchor)
!     HTMLSaveInline(li, rp, false, x, y, button, ti->src);
!   else
! #endif
!     HTMLLoadAnchor(li, rp, false, x, y, button);

    return(0);
  }
***************
*** 122,139 ****
    InlineState *ci = HTMLRPState(rp);
    LayoutInfo *li = ci->li;

!   if (rp == NULL || rp->anchor == NULL || rp->p == NULL) return(0);
!   HTMLPrintAnchor(li, rp, false, x, y);
    li->over = rp;

    return(0);
  }

  /*
   *
   * Public functions
   *
   */

  /*
   * HandleImg
- --- 173,319 ----
    InlineState *ci = HTMLRPState(rp);
    LayoutInfo *li = ci->li;

! #ifdef DELAY_INLINE
!   /* This is a hack,  I should really be able to queue this request
!    * to occur AFTER the cache write is complete
!    */
!   if ((ci->status == requested) && FrameCacheQuery(li->frame, ci->src))
!   {
!     ci->status=available;
!     FrameReplace(li->frame,li->bup);
!   }
!   if (ci->status != available)
!     HTMLPrintInline(li, rp, ci->src, ci->type);
!   else
! #endif
!     HTMLPrintAnchor(li, rp, false, x, y);
!
    li->over = rp;

    return(0);
  }

+ #ifdef DELAY_INLINE
+ /*
+  *  InlineStateSearch
+  */
+ static byte *
+ InlineStateSearch(closure, url, size)
+ void *closure;
+ char *url;
+ size_t *size;
+ {
+   RenderPart *rp = (RenderPart *)closure;
+   InlineState *ci = (InlineState *)HTMLRPState(rp);
+
+   *size = 0;
+   if (rp->destroy != DestroyInline) return(NULL);
+
+   rp = ci->li->rphead;
+
+   while (rp != NULL)
+   {
+     if (rp->destroy == DestroyInline)
+     {
+       ci = (InlineState *)HTMLRPState(rp);
+       if ((strcmp(ci->src,url) == 0) && (ci->status != available))
+         return((byte *)(rp));
+     }
+     rp = rp->mnext;
+   }
+
+   return(NULL);
+ }
+
+
+ static byte *
+ InlineStateQuery(closure, key, size)
+ void *closure;
+ char *key;
+ size_t *size;
+ {
+   RenderPart *rp = (RenderPart *)closure;
+   InlineState *ci;
+   byte *retval = NULL;
+
+   *size = 0;
+   if ((rp == NULL) || (rp->destroy != DestroyInline)) return (retval);
+   ci = (InlineState *)HTMLRPState(rp);
+
+   if (strcmp(key,"type") == 0)
+   {
+     *size = ci->type;
+   }
+   if (strcmp(key,"delayed") == 0)
+   {
+     if (ci->status == unavailable)
+       *size = true;
+   }
+   if (strcmp(key,"alt_length") == 0)
+   {
+     *size = ci->alt_length;
+   }
+   if (strcmp(key,"text") == 0)
+   {
+     *size = ci->alt_length;
+     retval = (byte *) (ci->alt_text);
+   }
+   if (strcmp(key,"MIME") == 0)
+   {
+     *size = strlen(ci->alt_MIME);
+     retval = (byte *) (ci->alt_MIME);
+   }
+   if (strcmp(key,"source") == 0)
+   {
+     *size = strlen(ci->src);
+     retval = (byte *) (ci->src);
+   }
+   if (strcmp(key,"frame") == 0)
+   {
+     *size = sizeof(Frame);
+     retval = (byte *) (ci->frame);
+   }
+   if (strcmp(key,"presized") == 0)
+   {
+       *size = ci->size_given;
+   }
+
+   return(retval);
+ }
+
+ #endif
+
  /*
   *
   * Public functions
   *
   */
+ #ifdef DELAY_INLINE
+ void
+ InlineUndelay(state)
+ void *state;
+ {
+   InlineState *ci=(InlineState *)state;
+   if (ci->status > available)
+     --ci->status;
+ }
+
+ Frame
+ InlineGetFrame(state)
+ void *state;
+ {
+   InlineState *ci=(InlineState *)state;
+   return(ci->frame);
+ }
+
+ bool
+ InlineSizeGiven(state)
+ void *state;
+ {
+   InlineState *ci=(InlineState *)state;
+   return(ci->size_given);
+ }
+ #endif  /* DELAY_INLINE */

  /*
   * HandleImg
***************
*** 151,156 ****
- --- 331,341 ----
    char *ws, *hs;
    RenderHooks orh;
    MemPool tmp;
+ #ifdef DELAY_INLINE
+   URLParts *sup;
+   char *value;
+ #endif
+

    if (end) return;
    if ((url = MLFindAttribute(p, "src")) == NULL) return;
***************
*** 171,187 ****

    li->delay = True;

    if ((ws = MLFindAttribute(p, "width")) != NULL &&
        (hs = MLFindAttribute(p, "height")) != NULL)
    {
      unsigned int width, height;

      width = (unsigned int)atoi(ws);
      height = (unsigned int)atoi(hs);
      ci->frame = FrameCreateSized(li->frame, width, height);
      RealizeCallback(rp, 0, width, height);
    }
!   else
    {
      ci->frame = FrameCreateUnsized(li->frame, RealizeCallback, rp);
    }
- --- 356,450 ----

    li->delay = True;

+ #ifdef DELAY_INLINE
+   ci->type = InlineImage;
+   ci->size_given = false;
+ /*                                                                    */
+ /* FrameGetDelayState calls WWWGetDelayState which does an XtGetValue */
+ /* on the Inline Toggle Widget                                        */
+ /*                                                                    */
+
+
+   if ((value = FrameGetResource(li->frame, "inline.delay")) != NULL)
+   {
+     if ( atoi(value) != 0)
+       ci->status = unavailable;
+     else
+       ci->status = available;
+   }
+   else
+     ci->status = available;
+
+   if ((sup = HTMLMakeURLParts(li, tmp, MLFindAttribute(p, "src"))) != NULL)
+     ci->src = URLMakeString(li->mp, sup, false);
+
+   if (FrameCacheQuery(li->frame, ci->src)) ci->status = available;
+
+   /* Find out if this is an inline anchor */
+   if (rp->anchor != NULL) ci->type = InlineAnchor;
+
+   /* Find out if this is an imagemap */
+   if (MLFindAttribute(p, "ismap") != NULL) ci->type = InlineMap;
+
+   if ((ci->alt_text = MLFindAttribute(p, "alt")) == NULL)
+   {
+     ci->alt_text = (char *)inline_icon[ci->type];
+     ci->alt_MIME = (char *)inline_mime[ci->type];
+     ci->alt_length = inline_length[ci->type];
+   }
+   else
+   {
+     ci->alt_MIME = (char *)inline_mime[Other];
+     ci->alt_length = strlen(ci->alt_text);
+   }
+ #endif
+
    if ((ws = MLFindAttribute(p, "width")) != NULL &&
        (hs = MLFindAttribute(p, "height")) != NULL)
    {
      unsigned int width, height;

+ #ifdef DELAY_INLINE
+     ci->size_given = true;
+ #endif
+
      width = (unsigned int)atoi(ws);
      height = (unsigned int)atoi(hs);
      ci->frame = FrameCreateSized(li->frame, width, height);
      RealizeCallback(rp, 0, width, height);
    }
! #ifdef DELAY_INLINE
!   else if ( ci->status )
!   {
!     unsigned int width, height;
!
!     if (strcmp(ci->alt_MIME,inline_mime[Other]) == 0 )
!     {
!       HTMLSetupFonts(li);
!       if (li->lc->defaultFont == NULL)
!       {
!         height = 32;
!         width = 10 * strlen(ci->alt_text);
!       }
!       else
!       {
!         height = li->lc->defaultFont->ascent + li->lc->defaultFont->descent +
!                 + li->tmargin + li->textLineSpace;;
!         width = XTextWidth(li->lc->defaultFont, ci->alt_text, ci->alt_length)
!                 + li->rmargin;
!       }
!     }
!     else
!     {
!       width = 32;
!       height = 24;
!     }
!
!     ci->frame = FrameCreateSized(li->frame, width, height);
!     RealizeCallback(rp, 0, width, height);
!   }
! #endif  /* DELAY_INLINE */
!   else
    {
      ci->frame = FrameCreateUnsized(li->frame, RealizeCallback, rp);
    }
***************
*** 189,194 ****
- --- 452,461 ----
    memset(&orh, 0, sizeof(RenderHooks));
    orh.mouse = InlineMouseCallback;
    orh.motion = InlineMotionCallback;
+ #ifdef DELAY_INLINE
+   orh.search = InlineStateSearch;
+   orh.query = InlineStateQuery;
+ #endif
    FrameOverride(ci->frame, &orh, rp);
    FrameLoad(ci->frame, aup);

diff -c ../old/html/layout.c html/layout.c
*** ../old/html/layout.c	Sat Apr 27 04:47:18 1996
- --- html/layout.c	Sun May 12 19:36:30 1996
***************
*** 356,361 ****
- --- 356,364 ----

    if (rl->rphead != NULL &&
        rl->width + rp->width >= rl->maxwidth &&
+ #ifdef DELAY_INLINE
+       li->restack != NULL &&
+ #endif
        li->restack->wrap)
    {
      RenderPart *xp, *fxp;
diff -c ../old/proto/Imakefile proto/Imakefile
*** ../old/proto/Imakefile	Fri Apr 12 17:41:24 1996
- --- proto/Imakefile	Sun May 12 19:37:53 1996
***************
*** 1,8 ****
  #include <../Common.tmpl>

! SRCS = file.c http.c ftp.c cache.c

! OBJS = file.o http.o ftp.o cache.o

  INCLUDES = -I../ -I../common -I../www

- --- 1,8 ----
  #include <../Common.tmpl>

! SRCS = file.c http.c ftp.c cache.c delayed.c

! OBJS = file.o http.o ftp.o cache.o delayed.o

  INCLUDES = -I../ -I../common -I../www

diff -c ../old/proto/cache.c proto/cache.c
*** ../old/proto/cache.c	Sat Apr 27 03:39:22 1996
- --- proto/cache.c	Sun May 12 19:36:31 1996
***************
*** 574,576 ****
- --- 574,596 ----

    return;
  }
+
+ #ifdef DELAY_INLINE
+ /*
+  * CacheQuery
+  */
+ bool
+ CacheQuery(url,cc)
+ char *url;
+ CacheClass *cc;
+ {
+   CacheEntry *ce;
+
+   for (ce = cc->clist; ce != NULL; ce = ce->next)
+   {
+     if (strcmp(url, ce->url) == 0) return(true);
+   }
+
+   return(false);
+ }
+ #endif
diff -c ../old/proto/delayed.c proto/delayed.c
*** ../old/proto/delayed.c	Mon May 13 17:17:11 1996
- --- proto/delayed.c	Mon May 13 17:08:37 1996
***************
*** 0 ****
- --- 1,127 ----
+ /*
+  * delayed.c
+  *
+  * Copyright (c) 1996, Michael Kellen <kelle056@gold.tc.umn.edu>
+  * See the file COPYRIGHT for details.
+  */
+ #include <stdio.h>
+ #include <ctype.h>
+
+ #ifdef HAVE_STDLIB_H
+ #include <stdlib.h>
+ #endif
+
+ #ifdef HAVE_STRING_H
+ #include <string.h>
+ #endif
+
+ #ifdef HAVE_UNISTD_H
+ #include <unistd.h>
+ #endif
+
+ #include "WWW.h"
+
+ typedef struct _delay_info
+ {
+   MemPool mp;
+   WWWProto wp;
+   WWWContext wc;
+   WWWTimeOut iot;
+   URLParts *up;
+   Document *d;
+   bool init_done;
+   char *alt_text;
+   char *alt_MIME;
+   int  alt_length;
+ } DelayInfo;
+
+ /*
+  * DelayRead
+  */
+ static void
+ DelayRead(iot, closure)
+ WWWTimeOut iot;
+ void *closure;
+ {
+   DelayInfo *di = (DelayInfo *)closure;
+   char *b;
+
+   di->init_done = true;
+   if (WWWInitData(di->wp, di->up, NULL, di->d) == -1) return;
+   if (di->alt_length > 1)
+     di->d->len = di->alt_length - 1;
+   else
+     di->d->len = 0;
+   b = (char *)alloc_mem(di->alt_length);
+   memcpy(b,di->alt_text,di->alt_length);
+   di->d->data = (byte *)b;
+   WWWAddData(di->wp);
+   WWWEndData(di->wp);
+   return;
+ }
+
+ /*
+  * DelayDestroy
+  */
+ static int
+ DelayDestroy(di)
+ DelayInfo *di;
+ {
+   if (!di->init_done && di->d != NULL) DocumentDestroy(di->d);
+   if (di->iot != NULL) WWWDestroyTimeOut(di->iot);
+   MPDestroy(di->mp);
+
+   return(0);
+ }
+
+ /*
+  * DelayInit
+  */
+ static void *
+ DelayInit(up, pup, class_context, wp)
+ URLParts *up;
+ URLParts *pup;
+ void *class_context;
+ WWWProto wp;
+ {
+   DelayInfo *di;
+   MemPool mp;
+   size_t size;
+
+   mp = MPCreate();
+   di = (DelayInfo *)MPCGet(mp, sizeof(DelayInfo));
+   di->mp = mp;
+   di->wp = wp;
+   di->wc = WWWProtoToContext(wp);
+   di->iot = WWWCreateTimeOut(di->wc, di);
+   di->up = URLDup(di->mp, up);
+   di->d = DocumentCreate(true);
+
+   di->alt_MIME = MPStrDup(mp, WWWQueryInline(di->wc, wp, "MIME", &size));
+   di->alt_text= MPStrDup(mp, WWWQueryInline(di->wc, wp, "text", &size));
+   di->alt_length = size;
+
+   di->d->mflist = MIMECreateField(di->d->mp, "content-type", di->alt_MIME);
+   WWWSetTimeOut(di->iot, 10, DelayRead);
+
+   return(di);
+ }
+
+ /*
+  * InitModule_Delayed
+  */
+ void
+ InitModule_Delayed(wc)
+ WWWContext wc;
+ {
+   ProtoHooks ph;
+
+   memset(&ph, 0, sizeof(ph));
+   ph.protocol = "delay-inline";
+   ph.init = DelayInit;
+   ph.local = true;
+   ph.destroy = DelayDestroy;
+   WWWAddProtoHooks(wc, &ph);
+
+   return;
+ }


------- Message 35

Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id ab07100;
          14 May 96 11:49 PDT
Received: from oneworld.com by JIMI.CS.UNLV.EDU id aa07033; 14 May 96 11:45 PDT
Received: from redoak.otp.com by cvo.oneworld.com with smtp
	(Smail3.1.29.1 #16) id m0uJP5B-000SBmC; Tue, 14 May 96 11:44 PDT
Received: from 205.231.244.2 (205.231.244.2) by redoak.otp.com
 (EMWAC SMTPRS 0.70) with SMTP id <B0000005191@redoak.otp.com>;
 Tue, 14 May 1996 11:28:29 +0100
Message-ID: <B0000005191@redoak.otp.com>
Date: 14 May 96 11:28:29
From: Oak Tree Publications <nash@otp.com>
To: chimera-announce@cs.unlv.edu, nash@otp.com
Subject: Intel Internet/Intranet Directory2
X-Mailer: Allaire Cold Fusion 1.5

Dear Chimera, 

We would like to invite you to participate in the Intel Internet/Intranet  Directory.  
Please visit this site at:
https://scripts.intel.com/cgi-db/dbml.exe?template=iaweb/director/main.dbm.

As we work toward our goal of ensuring that business users can take full 
advantage of the Internet/Intranet, we created the Intel Internet/Intranet 
Directory. This resource gives users/developers access to the rapidly growing 
list of companies who are providing products and services that are compatible 
with Intel architecture, and can assist in all aspects of the 
development/usage of the Internet/Intranet. 

HOW THE DIRECTORY WORKS
=======================
The directory is connected to a database and HTML pages are served up as 
users choose what they want to see in the directory.  We take advantage of 
this process in the directory's product submission process.  When you submit 
your information, you will see your entry as it will appear in the directory 
and your information goes directly into the database.

HOW TO SUBMIT YOUR INFORMATION
=======================
To add yourself to the directory, please choose the "Add your company to the 
directory or modify information" graphic at the top of the main page of the 
directory.  For more information on how to update your information, please 
choose the "How to use the directory" graphic at the top of the main page of 
the directory.

Intel will review your listing and reserves the right to edit or reject the 
listing. The directory will be updated daily, but it may take up to 24 hours 
for your changes to appear in the directory. Once you submit your 
information, you are responsible for keeping it up to date.  You will be 
asked to review your information once a quarter.  If you do not check it 
quarterly, you will be dropped from the directory. 

PASSWORDS:
=======================
When submitting your company information, you will be asked to give a 
password, it must be at least eight characters and contain both letters and 
numbers.  Once you have submitted your information and Intel has approved 
it, you will then be considered an existing account.  When you need to 
update any information, you will use the password you created.  You may 
change it at any time.

SUBMIT YOUR COMPANY LOGO
========================
If you haven't done so already, you may submit your logo for display on your 
company page. For ease of use, your company logo should be 8K or less.  If 
you wish to submit your logo, please:

1.  Send us an URL pointing us to a graphic we can download from your site OR 
2.  Reply to this e-mail message with your logo attached in GIF or JPEG format 

RETURN LINK TO INTEL
====================
If you haven't done so already, please provide a return link to Intel at 
www.intel.com using the Intel linking button.  You can download this button 
once you've logged in as an existing account with your password given to you 
at the beginning of this message.

WE'D LIKE TO HEAR FROM YOU!
===========================
We'd like to know what you think about the directory today and how we can 
improve it in the future.  Please click on the "Questions and Comments" hot 
text at the bottom of the main page of the directory.  You will be given an 
opportunity to submit comments here.

We're looking forward to hearing from you!

Enterprise Server Group
Intel Corporation



------- Message 36

Received: from xph029.physics.montana.edu by JIMI.CS.UNLV.EDU id aa29820;
          15 May 96 14:22 PDT
Received: from localhost (michael@localhost) by xph029.physics.montana.edu (8.6.12/8.6.9) with SMTP id PAA16221 for <bug-chimera@cs.unlv.edu>; Wed, 15 May 1996 15:20:11 -0600
Date: Wed, 15 May 1996 15:20:07 -0600 (MDT)
From: Michael Kellen <michael@xph029.physics.montana.edu>
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Subject: [cfh108] Delayed Inlines Patch
In-Reply-To: <Pine.LNX.3.92.960512194939.1223A-100000@xph036.physics.montana.edu>
Message-ID: <Pine.LNX.3.92.960515151726.16218A-100000@xph029.physics.montana.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII


The following patch (mostly) fixes the problem with multiple instances of
an image in the same document.  It is cumulative with the first patch, or
you can grab the total patch from

    https://xph029.physics.montana.edu/chimera

Michael


diff -c ../old/html/inline.c html/inline.c
*** ../old/html/inline.c	Wed May 15 15:15:39 1996
- --- html/inline.c	Wed May 15 15:07:24 1996
***************
*** 205,210 ****
- --- 205,211 ----
  {
    RenderPart *rp = (RenderPart *)closure;
    InlineState *ci = (InlineState *)HTMLRPState(rp);
+   RenderPart *lastmatch = NULL;

    *size = 0;
    if (rp->destroy != DestroyInline) return(NULL);
***************
*** 216,228 ****
      if (rp->destroy == DestroyInline)
      {
        ci = (InlineState *)HTMLRPState(rp);
!       if ((strcmp(ci->src,url) == 0) && (ci->status != available))
!         return((byte *)(rp));
      }
      rp = rp->mnext;
    }

!   return(NULL);
  }


- --- 217,234 ----
      if (rp->destroy == DestroyInline)
      {
        ci = (InlineState *)HTMLRPState(rp);
!       if (strcmp(ci->src,url) == 0)
!       {
!          if (ci->status == requested)
!            return((byte *)(rp));
!          else if (ci->status == unavailable)
!            lastmatch = rp;
!       }
      }
      rp = rp->mnext;
    }

!   return((byte *)(lastmatch));
  }





------- Message 37

Received: from pf2.phil.uni-sb.de by JIMI.CS.UNLV.EDU id aa19221;
          16 May 96 14:17 PDT
Received: from schlunz.phil.uni-sb.de (benni@access177.telip.uni-sb.de [134.96.113.177]) by pf2.phil.uni-sb.de (8.7.5/8.7.5/chris) with ESMTP id XAA28606 for <bug-chimera@cs.unlv.edu>; Thu, 16 May 1996 23:18:41 +0200 (MET DST)
Received: (from benni@localhost) by schlunz.phil.uni-sb.de (8.7.5/8.7.3) id XAA00287 for benni; Thu, 16 May 1996 23:25:14 +0200
Received: from JIMI.CS.UNLV.EDU (jimi.CS.UNLV.EDU [131.216.22.2]) by pf2.phil.uni-sb.de (8.7.5/8.7.5/chris) with SMTP id XAA26054 for <benni@phil.uni-sb.de>; Sun, 12 May 1996 23:20:25 +0200 (MET DST)
Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa04079;
          12 May 96 14:01 PDT
Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa04031; 12 May 96 13:57 PDT
Received: from kroete2.freinet.de (arh135.ppp.uni-c.dk [130.228.13.135]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id WAA27550 for <bug-chimera@cs.unlv.edu>; Sun, 12 May 1996 22:57:27 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uIiBC-0003pNC; Sun, 12 May 96 22:55 MET DST
Message-Id: <m0uIiBC-0003pNC@kroete2.freinet.de>
Subject: A moving target?
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Sun, 12 May 1996 22:55:45 +0200 (MET DST)
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Status: O

The WWW-consortium have announced a new HTML 'standard' called HTML 3.2
(codenamed Wilbur).  It's looks like HTML 2.0 plus tables, <img
align=left/right>, applets, scripts, and a few other bits of the HTML
3.0 proposal. A lot of the HTML 3.0 proposal (including maths) didn't
make it, and is being worked on, presumably for HTML 4.0. You can see
more on https://www.w3.org/pub/WWW/MarkUp/Activity and on
https://www.w3.org/pub/WWW/MarkUp/Wilbur/. There's a DTD, but there
doesn't seem to be a nice non-SGML-wizards- version of it, like there
was for HTML 3.0 (accesible through my homepage).

Of course, we still have to get HTML 2.0 working, though I don't think
I'm the only one who finds the temptation to start fiddling with post-2.0
stuff almost irresistable.

Is 3.2 a realistic goal for Chimera 2.0, or should we settle for HTML 2.0,
clean up the usability a little and get the damn thing released? The
only thing you need to be able to do with applets and scripts for HTML 3.2
is to ignore them by the way.

- --
Erik Corry ehcorry@inet.uni-c.dk


------- Message 38

Received: from pf2.phil.uni-sb.de by JIMI.CS.UNLV.EDU id aa19282;
          16 May 96 14:22 PDT
Received: from schlunz.phil.uni-sb.de (benni@access177.telip.uni-sb.de [134.96.113.177]) by pf2.phil.uni-sb.de (8.7.5/8.7.5/chris) with ESMTP id XAA28736 for <bug-chimera@cs.unlv.edu>; Thu, 16 May 1996 23:23:11 +0200 (MET DST)
Received: (from benni@localhost) by schlunz.phil.uni-sb.de (8.7.5/8.7.3) id XAA00436 for benni; Thu, 16 May 1996 23:29:15 +0200
Received: from JIMI.CS.UNLV.EDU (jimi.CS.UNLV.EDU [131.216.22.2]) by pf2.phil.uni-sb.de (8.7.5/8.7.5/chris) with SMTP id CAA13322 for <benni@phil.uni-sb.de>; Tue, 14 May 1996 02:19:15 +0200 (MET DST)
Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa23330;
          13 May 96 16:28 PDT
Received: from xph029.physics.montana.edu by JIMI.CS.UNLV.EDU id aa23279;
          13 May 96 16:24 PDT
Received: from localhost (michael@localhost) by xph029.physics.montana.edu (8.6.12/8.6.9) with SMTP id RAA10889 for <bug-chimera@cs.unlv.edu>; Mon, 13 May 1996 17:22:57 -0600
X-Authentication-Warning: xph029.physics.montana.edu: michael owned process doing -bs
Date: Mon, 13 May 1996 17:22:55 -0600 (MDT)
From: Michael Kellen <michael@xph036.physics.montana.edu>
X-Sender: michael@xph029.physics.montana.edu
Reply-To: Michael Kellen <michael@xph036.physics.montana.edu>
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Subject: [cfh108] Delayed Inlines Release
Message-ID: <Pine.LNX.3.92.960512194939.1223A-100000@xph036.physics.montana.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Status: O


This release of the delayed inline option is much less kludgier than the
last.  The delay status information has been entirely moved into the info
structure for Inlines (html/inline.c) and is accessed by RenderHooks.

The cache checking was reworked to work with the new cache structure (as
of cfh106), which removed the need to stat a cache file.

For now, only text is available instead of the inlined image.  You will
see either the ALT text (if supplied) or "[IMAGE]" or "[LINK]" or "[MAP]"
in the place of the image when delay is on.

Cached images are not delayed.  If WIDTH and HEIGHT are supplied, the text
will be clipped by that size if it is larger.  This allows such images to
be downloaded later without re-rendering the whole page.

USAGE

Add the following to your .chimera/resources:

inline.delay:   1

Start up chimera.  All images should be delayed.  A value of 0 will
disable delays (default).

If you move your mouse over a simple image, the filename should appear in
parentheses in the message bar.  Left or Middle clicking will download and
display the image.  After the image is rendered, a MIDDLE MOUSE click will
pop up the Download dialogue to simplify grabbing images off of the web.

If you move over an Imagemap, you should see the same behaviour, with the
addition of the word "ImageMap" after the filename on the message bar.
After the image is rendered, this will function like a normal imagemap.
Again, a MIDDLE MOUSE click will allow you to grab the image.

If you move over a link, the file name will be followed by the usual URL
display showing the destination of the link.  A left click will FOLLOW THE
LINK.  Middle click will download the image.  After the image is rendered,
this will function like a normal image link (left to follow, middle to
download the linked document).


KNOWN BUGS

IMO, li->tmargin should NOT be applied to inline frames

Multiple instances of the same image on a page are still being annoying.
If you attempt to load any instance but the first, the browser will give a
"unable to render" type error.  It works fine if you click on the first
instance.

Unsized images require re-rendering of the page.  This must not occur
until after the image has been cached or it is not saved.  Currently
InlineMotionCallback checks to see if you are over an unrendered but
cached image, and if so re-renders the page.  This is a HACK.

file:/path URLs should really not be delayed.

inline.delay should be checked for sanity.  Ideally, it should be
possible to change on the fly via a toggle (a la "Source").  The toggle
would not have to be realized in the button bar if the user did not want
it.


This is patched against the distribution cfh108, which I grabbed from
Erik's site.

Please give it a whirl,

Michael




diff -c ../old/src/builtin.c src/builtin.c
*** ../old/src/builtin.c	Mon Apr 15 19:44:56 1996
- --- src/builtin.c	Sun May 12 19:36:29 1996
***************
*** 19,24 ****
- --- 19,28 ----
  extern void InitModule_Image _ArgProto((WWWContext));
  extern void InitModule_Download _ArgProto((WWWContext));
  extern void InitModule_View _ArgProto((WWWContext));
+ #ifdef DELAY_INLINE
+ extern void InitModule_Delayed _ArgProto((WWWContext));
+ #endif
+

  void
  InitChimeraBuiltins(wc)
***************
*** 32,36 ****
- --- 36,43 ----
    InitModule_FTP(wc);
    InitModule_File(wc);
    InitModule_Image(wc);
+ #ifdef DELAY_INLINE
+   InitModule_Delayed(wc);
+ #endif
    return;
  }
diff -c ../old/www/WWW.h www/WWW.h
*** ../old/www/WWW.h	Wed Apr 24 03:04:22 1996
- --- www/WWW.h	Sun May 12 19:36:29 1996
***************
*** 154,159 ****
- --- 154,162 ----
   * frame.c
   */
  int FrameLoad _ArgProto((Frame, URLParts *));
+ #ifdef DELAY_INLINE
+ int FrameCacheLoad _ArgProto((Frame, URLParts *));
+ #endif
  int FrameReload _ArgProto((Frame));
  int FrameBack _ArgProto((Frame));
  int FrameReplace _ArgProto((Frame, URLParts *));
***************
*** 180,185 ****
- --- 183,191 ----
  Document *DocumentCreate _ArgProto((bool));
  void DocumentDestroy _ArgProto((Document *));

+ #ifdef DELAY_INLINE
+ bool FrameCacheQuery _ArgProto((Frame, char *));
+ #endif
  /*
   * gui.c
   */
***************
*** 239,243 ****
- --- 245,252 ----
  WWWID WWWUniqueID _ArgProto((WWWContext));
  int WWWGetUserAuth _ArgProto((WWWContext, URLParts *, int,
                                WWWAuthCallback, void *));
+ #ifdef DELAY_INLINE
+ byte *WWWQueryInline _ArgProto((WWWContext, WWWProto, char *, size_t *));
+ #endif

  #endif
diff -c ../old/www/context.c www/context.c
*** ../old/www/context.c	Mon Apr 22 01:45:54 1996
- --- www/context.c	Sun May 12 19:36:29 1996
***************
*** 338,340 ****
- --- 338,369 ----
  {
    return(-1);
  }
+
+ #ifdef DELAY_INLINE
+ byte *
+ WWWQueryInline(wc, wp, key, size)
+ WWWContext wc;
+ WWWProto wp;
+ char *key;
+ size_t *size;
+ {
+   RenderHooks rh;
+   void *rc;
+
+   *size = 0;
+   if (wc == NULL || wp == NULL) return (NULL);
+   /* Frame `1' is the fake frame used by the cache */
+   if ((wc->flist == NULL) || wc->flist == (Frame )1) return(NULL);
+
+   rc = wc->flist->orc;
+   rh = wc->flist->orh;
+   if (rh.query == NULL) return(NULL);
+   if (rh.search != NULL)
+   {
+     rc = (byte *)(rh.search)(rc, wp->url, size);
+   }
+   if (rc == NULL) return(NULL);
+   return( (rh.query)(rc, key, size) );
+ }
+ #endif
+
diff -c ../old/www/frame.c www/frame.c
*** ../old/www/frame.c	Wed Apr 24 03:19:57 1996
- --- www/frame.c	Sun May 12 19:36:29 1996
***************
*** 463,468 ****
- --- 463,483 ----
    return(0);
  }

+ #ifdef DELAY_INLINE
+ /*
+  * FrameCacheLoad
+  */
+ int
+ FrameCacheLoad(frame, up)
+ struct _Frame *frame;
+ URLParts *up;
+ {
+   if (frame->wp->d != NULL) frame->wp->d->nocache=false;
+   RenderNew(frame, up, STACK_RELOAD);
+   return(0);
+ }
+ #endif
+
  /*
   * FrameHome
   */
diff -c ../old/www/gui.c www/gui.c
*** ../old/www/gui.c	Mon Apr 15 19:03:52 1996
- --- www/gui.c	Sun May 12 21:26:40 1996
***************
*** 92,99 ****
- --- 92,113 ----
  bool with_scroll;
  unsigned int width, height;
  {
+ #ifdef DELAY_INLINE
+   size_t isDelayed = 0;
+
+   if (with_scroll)
+   {
+     if (frame->wp != NULL)
+       WWWQueryInline(frame->wc,frame->wp,"delayed",&isDelayed);
+     else if (frame->orh.query != NULL)
+       (frame->orh.query)(frame->orc, "delayed",  &isDelayed);
+   }
+   WWWSetSize(frame->www, (with_scroll && !isDelayed), width, height);
+   WWWSetDrawSize(frame->www, (with_scroll && !isDelayed), width, height);
+ #else
    WWWSetSize(frame->www, with_scroll, width, height);
    WWWSetDrawSize(frame->www, with_scroll, width, height);
+ #endif
    frame->size_set = True;
    if (frame->size_callback != NULL)
    {
***************
*** 108,114 ****
- --- 122,141 ----
  bool with_scroll;
  unsigned int width, height;
  {
+ #ifdef DELAY_INLINE
+   size_t isDelayed = 0;
+
+   if (with_scroll)
+   {
+     if (frame->wp != NULL)
+       WWWQueryInline(frame->wc,frame->wp,"delayed",&isDelayed);
+     else if (frame->orh.query != NULL)
+       (frame->orh.query)(frame->orc, "delayed",  &isDelayed);
+   }
+   WWWSetDrawSize(frame->www, (with_scroll && !isDelayed), width, height);
+ #else
    WWWSetDrawSize(frame->www, with_scroll, width, height);
+ #endif
    return(0);
  }

diff -c ../old/www/proto.c www/proto.c
*** ../old/www/proto.c	Wed Apr 24 03:49:53 1996
- --- www/proto.c	Sun May 12 19:36:29 1996
***************
*** 117,122 ****
- --- 117,124 ----
  struct _WWWContext *wc;
  struct _WWWProto *wp;
  {
+   size_t isDelayed = 0;
+
    if (wc->iocnt >= wc->maxiocnt)
    {
      wp->waiting = true;
***************
*** 132,137 ****
- --- 134,150 ----
      }
    }

+ #ifdef DELAY_INLINE
+   WWWQueryInline(wp->wc, wp, "delayed", &isDelayed);
+   if (wp->pc == NULL && isDelayed)
+   {
+     if ((wp->ph = GetProtoHooks(wc, "delay-inline")) != NULL)
+     {
+       wp->pc = (wp->ph->init)(wp->up, NULL, wp->ph->class_context, wp);
+     }
+   }
+ #endif
+
    /* Check for a proxy. */
    if (wp->pc == NULL && (wp->pup = GetProxy(wp->mp, wp->up)) != NULL)
    {
***************
*** 411,413 ****
- --- 424,444 ----

    return;
  }
+
+ #ifdef DELAY_INLINE
+ /*
+  * FrameCacheQuery
+  */
+ bool
+ FrameCacheQuery(frame, url)
+ struct _Frame *frame;
+ char *url;
+ {
+   ProtoHooks *ph;
+
+   if ((ph = GetProtoHooks(frame->wc, "cache-read")) != NULL)
+        return( CacheQuery(url, ph->class_context) );
+
+   return(false);
+ }
+ #endif
diff -c ../old/Common.tmpl.dist Common.tmpl.dist
*** ../old/Common.tmpl.dist	Wed Apr 24 03:44:24 1996
- --- Common.tmpl.dist	Sun May 12 19:36:29 1996
***************
*** 78,84 ****
  XCOMM CC = gcc

  /* Add site stuff here */
! CEXTRA_DEFINES = $(OS_DEFINES)
  CEXTRA_LIBS = $(OS_LIBS)

  /*
- --- 78,84 ----
  XCOMM CC = gcc

  /* Add site stuff here */
! CEXTRA_DEFINES = $(OS_DEFINES) -DDELAY_INLINE
  CEXTRA_LIBS = $(OS_LIBS)

  /*
diff -c ../old/html/html.h html/html.h
*** ../old/html/html.h	Sat Apr 27 01:27:25 1996
- --- html/html.h	Sun May 12 19:39:30 1996
***************
*** 290,295 ****
- --- 290,300 ----
  /*
   * inline.c
   */
+ #ifdef DELAY_INLINE
+ void  InlineUndelay _ArgProto((void *));
+ Frame InlineGetFrame _ArgProto((void *));
+ bool  InlineSizeGiven _ArgProto((void *));
+ #endif
  void HandleImg _ArgProto((LayoutInfo *, MLElement, bool));

  /*
***************
*** 382,387 ****
- --- 387,399 ----
  int HTMLLoadAnchor _ArgProto((LayoutInfo *, RenderPart *, bool,int,int,int));
  void HTMLPrintAnchor _ArgProto((LayoutInfo *, RenderPart *, bool, int, int));
  URLParts *HTMLMakeURLParts _ArgProto((LayoutInfo *, MemPool, char *));
+ #ifdef DELAY_INLINE
+ int HTMLLoadInline _ArgProto((LayoutInfo *, RenderPart *, bool, int,int,
+                                             int,int,char *, void *));
+ void HTMLPrintInline _ArgProto((LayoutInfo *, RenderPart *, char *, int));
+ int HTMLSaveInline _ArgProto((LayoutInfo *, RenderPart *, bool,
+                                                   int, int, int, char *));
+ #endif

  /*
   * table.c
diff -c ../old/html/load.c html/load.c
*** ../old/html/load.c	Wed Apr 24 03:25:43 1996
- --- html/load.c	Sun May 12 19:36:30 1996
***************
*** 150,152 ****
- --- 150,310 ----

    return;
  }
+
+ #ifdef DELAY_INLINE
+ #define	InlineOther	0
+ #define InlineImage	1
+ #define InlineAnchor	2
+ #define InlineMap	3
+
+
+ int
+ HTMLLoadInline(li, rp, trans, x, y, button, type, src, callback)
+ LayoutInfo *li;
+ RenderPart *rp;
+ bool trans;
+ int x, y;
+ int button;
+ int type;
+ char *src;
+ void *callback;
+ {
+   if (rp == NULL) return(0);
+
+   if ( (button == 1) && (type == InlineAnchor) )
+   {
+     HTMLLoadAnchor(li, rp, trans, x, y, button);
+   }
+   else
+   {
+     URLParts *up;
+     MemPool mp;
+
+     mp = MPCreate();
+     InlineUndelay(li->over->state);
+     if ((up = HTMLMakeURLParts(li, mp, src)) == NULL)
+       FrameReload(InlineGetFrame(li->over->state));
+     else if (InlineSizeGiven(li->over->state))
+     {
+       InlineUndelay(li->over->state);
+       FrameReplace(InlineGetFrame(li->over->state),up);
+     }
+     else
+     {
+       Frame iframe = InlineGetFrame(li->over->state);
+
+       FrameForceType(iframe, "x-www/x-cache-write");
+       FrameCacheLoad(iframe,up);
+     }
+     MPDestroy(mp);
+   }
+
+
+
+   return(0);
+ }
+
+ void
+ HTMLPrintInline(li, rp, src, type)
+ LayoutInfo *li;
+ RenderPart *rp;
+ char *src;
+ int type;
+ {
+   char *buf,*target,*source,*url;
+   URLParts *up;
+   MemPool mp;
+
+   if (rp == NULL) return;
+   if (src == NULL) return;
+
+   mp = MPCreate();
+   if (type == InlineAnchor)
+   {
+     if(rp->anchor == NULL)
+     {
+       MPDestroy(mp);
+       return;
+     }
+
+     if ((url = MLFindAttribute(rp->anchor, "href")) == NULL)
+     {
+       MPDestroy(mp);
+       return;
+     }
+
+     if ((up = HTMLMakeURLParts(li, mp, url)) == NULL)
+     {
+       MPDestroy(mp);
+       return;
+     }
+
+     target = URLMakeString(mp, up, true);
+   }
+   else if (type == InlineMap)
+   {
+     target = MPStrDup(mp, "ImageMap");
+   }
+   else
+   {
+     target = MPStrDup(mp, "");
+   }
+   source = GetBaseFilename(src);
+
+   buf = (char *)MPGet(mp, strlen(source) + strlen(target) + 4);
+   sprintf (buf, "(%s) %s", source, target);
+
+   FrameMessage(li->frame, buf);
+
+   MPDestroy(mp);
+
+   return;
+ }
+
+
+ int
+ HTMLSaveInline(li, rp, trans, x, y, button, src)
+ LayoutInfo *li;
+ RenderPart *rp;
+ bool trans;
+ int x, y;
+ int button;
+ char *src;
+ {
+   Frame dframe;
+   char *ext;
+   WWWContext wc = WWWFrameToContext(li->frame);
+
+   if (rp == NULL) return(-1);
+
+   if (button == 2)
+   {
+     URLParts *up;
+     MemPool mp;
+
+     mp = MPCreate();
+     if ((up = HTMLMakeURLParts(li, mp, src)) == NULL)
+     {
+       MPDestroy(mp);
+       return(-1);
+     }
+     /*
+      * Load it up in a frame forced to a special type and hope the caller
+      * can deal with this.
+      */
+     if ((ext = WWWGetResource(wc, "www.lastResortType")) != NULL)
+     {
+       if ((dframe = FrameCreate(wc, NULL)) != NULL)
+       {
+         FrameForceType(dframe, ext);
+         FrameLoad(dframe, up);
+       }
+     }
+
+     MPDestroy(mp);
+   }
+   else
+     HTMLLoadAnchor(li, rp, false, x, y, button);
+   return(0);
+ }
+ #endif   /* DELAY_INLINE */
diff -c ../old/html/inline.c html/inline.c
*** ../old/html/inline.c	Sat Apr 27 01:29:26 1996
- --- html/inline.c	Mon May 13 17:09:31 1996
***************
*** 12,22 ****
- --- 12,47 ----

  #include "html.h"

+ #ifdef DELAY_INLINE
+ #include <X11/IntrinsicP.h>
+ #include <X11/StringDefs.h>
+
+ #ifndef ICONS
+ const char inline_icon[4][8] = { "", "[IMAGE]", "[LINK]", "[MAP]" };
+ const char inline_mime[4][12] = { "text/plain", "text/plain",
+                                 "text/plain", "text/plain"};
+ const int inline_length[4] = { 0, 7, 6, 5 };
+ #else
+ #include "icons.h"
+ #endif
+
+ #endif
+
  typedef struct inline_state
  {
    LayoutInfo *li;  /* need this always */
    bool added;      /* added yet */
    Frame frame;     /* embedded document ID */
+ #ifdef DELAY_INLINE
+   /* Inline delay information */
+   char        *src;                    /* SRC field info      */
+   enum        {available, requested, unavailable} status;
+   char        *alt_text;               /* either text or icon */
+   char        *alt_MIME;               /* MIME type for alt   */
+   int         alt_length;              /* Length of Alt data  */
+   enum        {Other, InlineImage, InlineAnchor, InlineMap} type;
+   bool        size_given;              /* HEIGHT & WIDTH specified? */
+ #endif
  } InlineState;

  /*
***************
*** 69,78 ****
    int offset;

    if (ci->added) return;
- -   ci->added = True;

    if (status == -1)
    {
      rp = HTMLCreateRP(ci->li, NULL, NULL, NULL, NULL, NULL, 0);
      HTMLAddRP(ci->li, rp, ADD_NOTHING, 0, 0, 0);
    }
- --- 94,109 ----
    int offset;

    if (ci->added) return;

    if (status == -1)
    {
+ #ifdef DELAY_INLINE
+     ci->status = unavailable;
+     /*
+      *  Should now (re)render the ALT frame if need be
+      */
+ #endif
+
      rp = HTMLCreateRP(ci->li, NULL, NULL, NULL, NULL, NULL, 0);
      HTMLAddRP(ci->li, rp, ADD_NOTHING, 0, 0, 0);
    }
***************
*** 88,96 ****
- --- 119,136 ----
      else if (av == ATTRIB_MIDDLE || av == ATTRIB_CENTER) offset = height / 2;
      else offset = 0;

+ #ifdef DELAY_INLINE
+     if (ci->status == requested)
+       ci->status = available;
+ #endif
      HTMLAddRP(ci->li, rp, ADD_SETBL, offset, width, height);
    }

+ #ifdef DELAY_INLINE
+   if (ci->status == available)
+ #endif
+   ci->added = True;
+
    return;
  }

***************
*** 107,113 ****
    InlineState *ci = HTMLRPState(rp);
    LayoutInfo *li = ci->li;

!   HTMLLoadAnchor(li, rp, false, x, y, button);

    return(0);
  }
- --- 147,164 ----
    InlineState *ci = HTMLRPState(rp);
    LayoutInfo *li = ci->li;

! #ifdef DELAY_INLINE
!   InlineState *ti;
!
!   ti = (InlineState *)(li->over->state);
!   if (ti->status == unavailable)
!     HTMLLoadInline(li, rp, false, x, y, button,
!                                     ti->type, ti->src, RealizeCallback);
!   else if (ti->type != InlineAnchor)
!     HTMLSaveInline(li, rp, false, x, y, button, ti->src);
!   else
! #endif
!     HTMLLoadAnchor(li, rp, false, x, y, button);

    return(0);
  }
***************
*** 122,139 ****
    InlineState *ci = HTMLRPState(rp);
    LayoutInfo *li = ci->li;

!   if (rp == NULL || rp->anchor == NULL || rp->p == NULL) return(0);
!   HTMLPrintAnchor(li, rp, false, x, y);
    li->over = rp;

    return(0);
  }

  /*
   *
   * Public functions
   *
   */

  /*
   * HandleImg
- --- 173,319 ----
    InlineState *ci = HTMLRPState(rp);
    LayoutInfo *li = ci->li;

! #ifdef DELAY_INLINE
!   /* This is a hack,  I should really be able to queue this request
!    * to occur AFTER the cache write is complete
!    */
!   if ((ci->status == requested) && FrameCacheQuery(li->frame, ci->src))
!   {
!     ci->status=available;
!     FrameReplace(li->frame,li->bup);
!   }
!   if (ci->status != available)
!     HTMLPrintInline(li, rp, ci->src, ci->type);
!   else
! #endif
!     HTMLPrintAnchor(li, rp, false, x, y);
!
    li->over = rp;

    return(0);
  }

+ #ifdef DELAY_INLINE
+ /*
+  *  InlineStateSearch
+  */
+ static byte *
+ InlineStateSearch(closure, url, size)
+ void *closure;
+ char *url;
+ size_t *size;
+ {
+   RenderPart *rp = (RenderPart *)closure;
+   InlineState *ci = (InlineState *)HTMLRPState(rp);
+
+   *size = 0;
+   if (rp->destroy != DestroyInline) return(NULL);
+
+   rp = ci->li->rphead;
+
+   while (rp != NULL)
+   {
+     if (rp->destroy == DestroyInline)
+     {
+       ci = (InlineState *)HTMLRPState(rp);
+       if ((strcmp(ci->src,url) == 0) && (ci->status != available))
+         return((byte *)(rp));
+     }
+     rp = rp->mnext;
+   }
+
+   return(NULL);
+ }
+
+
+ static byte *
+ InlineStateQuery(closure, key, size)
+ void *closure;
+ char *key;
+ size_t *size;
+ {
+   RenderPart *rp = (RenderPart *)closure;
+   InlineState *ci;
+   byte *retval = NULL;
+
+   *size = 0;
+   if ((rp == NULL) || (rp->destroy != DestroyInline)) return (retval);
+   ci = (InlineState *)HTMLRPState(rp);
+
+   if (strcmp(key,"type") == 0)
+   {
+     *size = ci->type;
+   }
+   if (strcmp(key,"delayed") == 0)
+   {
+     if (ci->status == unavailable)
+       *size = true;
+   }
+   if (strcmp(key,"alt_length") == 0)
+   {
+     *size = ci->alt_length;
+   }
+   if (strcmp(key,"text") == 0)
+   {
+     *size = ci->alt_length;
+     retval = (byte *) (ci->alt_text);
+   }
+   if (strcmp(key,"MIME") == 0)
+   {
+     *size = strlen(ci->alt_MIME);
+     retval = (byte *) (ci->alt_MIME);
+   }
+   if (strcmp(key,"source") == 0)
+   {
+     *size = strlen(ci->src);
+     retval = (byte *) (ci->src);
+   }
+   if (strcmp(key,"frame") == 0)
+   {
+     *size = sizeof(Frame);
+     retval = (byte *) (ci->frame);
+   }
+   if (strcmp(key,"presized") == 0)
+   {
+       *size = ci->size_given;
+   }
+
+   return(retval);
+ }
+
+ #endif
+
  /*
   *
   * Public functions
   *
   */
+ #ifdef DELAY_INLINE
+ void
+ InlineUndelay(state)
+ void *state;
+ {
+   InlineState *ci=(InlineState *)state;
+   if (ci->status > available)
+     --ci->status;
+ }
+
+ Frame
+ InlineGetFrame(state)
+ void *state;
+ {
+   InlineState *ci=(InlineState *)state;
+   return(ci->frame);
+ }
+
+ bool
+ InlineSizeGiven(state)
+ void *state;
+ {
+   InlineState *ci=(InlineState *)state;
+   return(ci->size_given);
+ }
+ #endif  /* DELAY_INLINE */

  /*
   * HandleImg
***************
*** 151,156 ****
- --- 331,341 ----
    char *ws, *hs;
    RenderHooks orh;
    MemPool tmp;
+ #ifdef DELAY_INLINE
+   URLParts *sup;
+   char *value;
+ #endif
+

    if (end) return;
    if ((url = MLFindAttribute(p, "src")) == NULL) return;
***************
*** 171,187 ****

    li->delay = True;

    if ((ws = MLFindAttribute(p, "width")) != NULL &&
        (hs = MLFindAttribute(p, "height")) != NULL)
    {
      unsigned int width, height;

      width = (unsigned int)atoi(ws);
      height = (unsigned int)atoi(hs);
      ci->frame = FrameCreateSized(li->frame, width, height);
      RealizeCallback(rp, 0, width, height);
    }
!   else
    {
      ci->frame = FrameCreateUnsized(li->frame, RealizeCallback, rp);
    }
- --- 356,450 ----

    li->delay = True;

+ #ifdef DELAY_INLINE
+   ci->type = InlineImage;
+   ci->size_given = false;
+ /*                                                                    */
+ /* FrameGetDelayState calls WWWGetDelayState which does an XtGetValue */
+ /* on the Inline Toggle Widget                                        */
+ /*                                                                    */
+
+
+   if ((value = FrameGetResource(li->frame, "inline.delay")) != NULL)
+   {
+     if ( atoi(value) != 0)
+       ci->status = unavailable;
+     else
+       ci->status = available;
+   }
+   else
+     ci->status = available;
+
+   if ((sup = HTMLMakeURLParts(li, tmp, MLFindAttribute(p, "src"))) != NULL)
+     ci->src = URLMakeString(li->mp, sup, false);
+
+   if (FrameCacheQuery(li->frame, ci->src)) ci->status = available;
+
+   /* Find out if this is an inline anchor */
+   if (rp->anchor != NULL) ci->type = InlineAnchor;
+
+   /* Find out if this is an imagemap */
+   if (MLFindAttribute(p, "ismap") != NULL) ci->type = InlineMap;
+
+   if ((ci->alt_text = MLFindAttribute(p, "alt")) == NULL)
+   {
+     ci->alt_text = (char *)inline_icon[ci->type];
+     ci->alt_MIME = (char *)inline_mime[ci->type];
+     ci->alt_length = inline_length[ci->type];
+   }
+   else
+   {
+     ci->alt_MIME = (char *)inline_mime[Other];
+     ci->alt_length = strlen(ci->alt_text);
+   }
+ #endif
+
    if ((ws = MLFindAttribute(p, "width")) != NULL &&
        (hs = MLFindAttribute(p, "height")) != NULL)
    {
      unsigned int width, height;

+ #ifdef DELAY_INLINE
+     ci->size_given = true;
+ #endif
+
      width = (unsigned int)atoi(ws);
      height = (unsigned int)atoi(hs);
      ci->frame = FrameCreateSized(li->frame, width, height);
      RealizeCallback(rp, 0, width, height);
    }
! #ifdef DELAY_INLINE
!   else if ( ci->status )
!   {
!     unsigned int width, height;
!
!     if (strcmp(ci->alt_MIME,inline_mime[Other]) == 0 )
!     {
!       HTMLSetupFonts(li);
!       if (li->lc->defaultFont == NULL)
!       {
!         height = 32;
!         width = 10 * strlen(ci->alt_text);
!       }
!       else
!       {
!         height = li->lc->defaultFont->ascent + li->lc->defaultFont->descent +
!                 + li->tmargin + li->textLineSpace;;
!         width = XTextWidth(li->lc->defaultFont, ci->alt_text, ci->alt_length)
!                 + li->rmargin;
!       }
!     }
!     else
!     {
!       width = 32;
!       height = 24;
!     }
!
!     ci->frame = FrameCreateSized(li->frame, width, height);
!     RealizeCallback(rp, 0, width, height);
!   }
! #endif  /* DELAY_INLINE */
!   else
    {
      ci->frame = FrameCreateUnsized(li->frame, RealizeCallback, rp);
    }
***************
*** 189,194 ****
- --- 452,461 ----
    memset(&orh, 0, sizeof(RenderHooks));
    orh.mouse = InlineMouseCallback;
    orh.motion = InlineMotionCallback;
+ #ifdef DELAY_INLINE
+   orh.search = InlineStateSearch;
+   orh.query = InlineStateQuery;
+ #endif
    FrameOverride(ci->frame, &orh, rp);
    FrameLoad(ci->frame, aup);

diff -c ../old/html/layout.c html/layout.c
*** ../old/html/layout.c	Sat Apr 27 04:47:18 1996
- --- html/layout.c	Sun May 12 19:36:30 1996
***************
*** 356,361 ****
- --- 356,364 ----

    if (rl->rphead != NULL &&
        rl->width + rp->width >= rl->maxwidth &&
+ #ifdef DELAY_INLINE
+       li->restack != NULL &&
+ #endif
        li->restack->wrap)
    {
      RenderPart *xp, *fxp;
diff -c ../old/proto/Imakefile proto/Imakefile
*** ../old/proto/Imakefile	Fri Apr 12 17:41:24 1996
- --- proto/Imakefile	Sun May 12 19:37:53 1996
***************
*** 1,8 ****
  #include <../Common.tmpl>

! SRCS = file.c http.c ftp.c cache.c

! OBJS = file.o http.o ftp.o cache.o

  INCLUDES = -I../ -I../common -I../www

- --- 1,8 ----
  #include <../Common.tmpl>

! SRCS = file.c http.c ftp.c cache.c delayed.c

! OBJS = file.o http.o ftp.o cache.o delayed.o

  INCLUDES = -I../ -I../common -I../www

diff -c ../old/proto/cache.c proto/cache.c
*** ../old/proto/cache.c	Sat Apr 27 03:39:22 1996
- --- proto/cache.c	Sun May 12 19:36:31 1996
***************
*** 574,576 ****
- --- 574,596 ----

    return;
  }
+
+ #ifdef DELAY_INLINE
+ /*
+  * CacheQuery
+  */
+ bool
+ CacheQuery(url,cc)
+ char *url;
+ CacheClass *cc;
+ {
+   CacheEntry *ce;
+
+   for (ce = cc->clist; ce != NULL; ce = ce->next)
+   {
+     if (strcmp(url, ce->url) == 0) return(true);
+   }
+
+   return(false);
+ }
+ #endif
diff -c ../old/proto/delayed.c proto/delayed.c
*** ../old/proto/delayed.c	Mon May 13 17:17:11 1996
- --- proto/delayed.c	Mon May 13 17:08:37 1996
***************
*** 0 ****
- --- 1,127 ----
+ /*
+  * delayed.c
+  *
+  * Copyright (c) 1996, Michael Kellen <kelle056@gold.tc.umn.edu>
+  * See the file COPYRIGHT for details.
+  */
+ #include <stdio.h>
+ #include <ctype.h>
+
+ #ifdef HAVE_STDLIB_H
+ #include <stdlib.h>
+ #endif
+
+ #ifdef HAVE_STRING_H
+ #include <string.h>
+ #endif
+
+ #ifdef HAVE_UNISTD_H
+ #include <unistd.h>
+ #endif
+
+ #include "WWW.h"
+
+ typedef struct _delay_info
+ {
+   MemPool mp;
+   WWWProto wp;
+   WWWContext wc;
+   WWWTimeOut iot;
+   URLParts *up;
+   Document *d;
+   bool init_done;
+   char *alt_text;
+   char *alt_MIME;
+   int  alt_length;
+ } DelayInfo;
+
+ /*
+  * DelayRead
+  */
+ static void
+ DelayRead(iot, closure)
+ WWWTimeOut iot;
+ void *closure;
+ {
+   DelayInfo *di = (DelayInfo *)closure;
+   char *b;
+
+   di->init_done = true;
+   if (WWWInitData(di->wp, di->up, NULL, di->d) == -1) return;
+   if (di->alt_length > 1)
+     di->d->len = di->alt_length - 1;
+   else
+     di->d->len = 0;
+   b = (char *)alloc_mem(di->alt_length);
+   memcpy(b,di->alt_text,di->alt_length);
+   di->d->data = (byte *)b;
+   WWWAddData(di->wp);
+   WWWEndData(di->wp);
+   return;
+ }
+
+ /*
+  * DelayDestroy
+  */
+ static int
+ DelayDestroy(di)
+ DelayInfo *di;
+ {
+   if (!di->init_done && di->d != NULL) DocumentDestroy(di->d);
+   if (di->iot != NULL) WWWDestroyTimeOut(di->iot);
+   MPDestroy(di->mp);
+
+   return(0);
+ }
+
+ /*
+  * DelayInit
+  */
+ static void *
+ DelayInit(up, pup, class_context, wp)
+ URLParts *up;
+ URLParts *pup;
+ void *class_context;
+ WWWProto wp;
+ {
+   DelayInfo *di;
+   MemPool mp;
+   size_t size;
+
+   mp = MPCreate();
+   di = (DelayInfo *)MPCGet(mp, sizeof(DelayInfo));
+   di->mp = mp;
+   di->wp = wp;
+   di->wc = WWWProtoToContext(wp);
+   di->iot = WWWCreateTimeOut(di->wc, di);
+   di->up = URLDup(di->mp, up);
+   di->d = DocumentCreate(true);
+
+   di->alt_MIME = MPStrDup(mp, WWWQueryInline(di->wc, wp, "MIME", &size));
+   di->alt_text= MPStrDup(mp, WWWQueryInline(di->wc, wp, "text", &size));
+   di->alt_length = size;
+
+   di->d->mflist = MIMECreateField(di->d->mp, "content-type", di->alt_MIME);
+   WWWSetTimeOut(di->iot, 10, DelayRead);
+
+   return(di);
+ }
+
+ /*
+  * InitModule_Delayed
+  */
+ void
+ InitModule_Delayed(wc)
+ WWWContext wc;
+ {
+   ProtoHooks ph;
+
+   memset(&ph, 0, sizeof(ph));
+   ph.protocol = "delay-inline";
+   ph.init = DelayInit;
+   ph.local = true;
+   ph.destroy = DelayDestroy;
+   WWWAddProtoHooks(wc, &ph);
+
+   return;
+ }



------- Message 39

Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id ab19441;
          16 May 96 14:29 PDT
Received: from pf2.phil.uni-sb.de by JIMI.CS.UNLV.EDU id aa19381;
          16 May 96 14:24 PDT
Received: from schlunz.phil.uni-sb.de (benni@access177.telip.uni-sb.de [134.96.113.177]) by pf2.phil.uni-sb.de (8.7.5/8.7.5/chris) with ESMTP id XAA28795; Thu, 16 May 1996 23:25:13 +0200 (MET DST)
Received: (from benni@localhost) by schlunz.phil.uni-sb.de (8.7.5/8.7.3) id XAA00510 for benni; Thu, 16 May 1996 23:31:43 +0200
Received: from JIMI.CS.UNLV.EDU (jimi.CS.UNLV.EDU [131.216.22.2]) by pf2.phil.uni-sb.de (8.7.5/8.7.5/chris) with SMTP id WAA05772 for <benni@phil.uni-sb.de>; Tue, 14 May 1996 22:36:59 +0200 (MET DST)
Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa07218;
          14 May 96 11:55 PDT
Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id ab07100;
          14 May 96 11:49 PDT
Received: from oneworld.com by JIMI.CS.UNLV.EDU id aa07033; 14 May 96 11:45 PDT
Received: from redoak.otp.com by cvo.oneworld.com with smtp
	(Smail3.1.29.1 #16) id m0uJP5B-000SBmC; Tue, 14 May 96 11:44 PDT
Received: from 205.231.244.2 (205.231.244.2) by redoak.otp.com
 (EMWAC SMTPRS 0.70) with SMTP id <B0000005191@redoak.otp.com>;
 Tue, 14 May 1996 11:28:29 +0100
Message-ID: <B0000005191@redoak.otp.com>
Date: 14 May 96 11:28:29
From: Oak Tree Publications <nash@otp.com>
To: chimera-announce@cs.unlv.edu, nash@otp.com
Subject: Intel Internet/Intranet Directory2
X-Mailer: Allaire Cold Fusion 1.5
Status: O

Dear Chimera, 

We would like to invite you to participate in the Intel Internet/Intranet  Directory.  
Please visit this site at:
https://scripts.intel.com/cgi-db/dbml.exe?template=iaweb/director/main.dbm.

As we work toward our goal of ensuring that business users can take full 
advantage of the Internet/Intranet, we created the Intel Internet/Intranet 
Directory. This resource gives users/developers access to the rapidly growing 
list of companies who are providing products and services that are compatible 
with Intel architecture, and can assist in all aspects of the 
development/usage of the Internet/Intranet. 

HOW THE DIRECTORY WORKS
=======================
The directory is connected to a database and HTML pages are served up as 
users choose what they want to see in the directory.  We take advantage of 
this process in the directory's product submission process.  When you submit 
your information, you will see your entry as it will appear in the directory 
and your information goes directly into the database.

HOW TO SUBMIT YOUR INFORMATION
=======================
To add yourself to the directory, please choose the "Add your company to the 
directory or modify information" graphic at the top of the main page of the 
directory.  For more information on how to update your information, please 
choose the "How to use the directory" graphic at the top of the main page of 
the directory.

Intel will review your listing and reserves the right to edit or reject the 
listing. The directory will be updated daily, but it may take up to 24 hours 
for your changes to appear in the directory. Once you submit your 
information, you are responsible for keeping it up to date.  You will be 
asked to review your information once a quarter.  If you do not check it 
quarterly, you will be dropped from the directory. 

PASSWORDS:
=======================
When submitting your company information, you will be asked to give a 
password, it must be at least eight characters and contain both letters and 
numbers.  Once you have submitted your information and Intel has approved 
it, you will then be considered an existing account.  When you need to 
update any information, you will use the password you created.  You may 
change it at any time.

SUBMIT YOUR COMPANY LOGO
========================
If you haven't done so already, you may submit your logo for display on your 
company page. For ease of use, your company logo should be 8K or less.  If 
you wish to submit your logo, please:

1.  Send us an URL pointing us to a graphic we can download from your site OR 
2.  Reply to this e-mail message with your logo attached in GIF or JPEG format 

RETURN LINK TO INTEL
====================
If you haven't done so already, please provide a return link to Intel at 
www.intel.com using the Intel linking button.  You can download this button 
once you've logged in as an existing account with your password given to you 
at the beginning of this message.

WE'D LIKE TO HEAR FROM YOU!
===========================
We'd like to know what you think about the directory today and how we can 
improve it in the future.  Please click on the "Questions and Comments" hot 
text at the bottom of the main page of the directory.  You will be given an 
opportunity to submit comments here.

We're looking forward to hearing from you!

Enterprise Server Group
Intel Corporation




------- Message 40

Received: from pf2.phil.uni-sb.de by JIMI.CS.UNLV.EDU id aa20440;
          16 May 96 15:10 PDT
Received: from schlunz.phil.uni-sb.de (benni@pfslip03.phil.uni-sb.de [134.96.82.203]) by pf2.phil.uni-sb.de (8.7.5/8.7.5/chris) with ESMTP id AAA29600 for <bug-chimera@cs.unlv.edu>; Fri, 17 May 1996 00:11:01 +0200 (MET DST)
Received: (from benni@localhost) by schlunz.phil.uni-sb.de (8.7.5/8.7.3) id AAA01053 for benni; Fri, 17 May 1996 00:17:39 +0200
Received: from JIMI.CS.UNLV.EDU (jimi.CS.UNLV.EDU [131.216.22.2]) by pf2.phil.uni-sb.de (8.7.5/8.7.5/chris) with SMTP id XAA28973 for <benni@phil.uni-sb.de>; Thu, 16 May 1996 23:40:42 +0200 (MET DST)
Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa19278;
          16 May 96 14:22 PDT
Received: from pf2.phil.uni-sb.de by JIMI.CS.UNLV.EDU id aa19221;
          16 May 96 14:17 PDT
Received: from schlunz.phil.uni-sb.de (benni@access177.telip.uni-sb.de [134.96.113.177]) by pf2.phil.uni-sb.de (8.7.5/8.7.5/chris) with ESMTP id XAA28606 for <bug-chimera@cs.unlv.edu>; Thu, 16 May 1996 23:18:41 +0200 (MET DST)
Received: (from benni@localhost) by schlunz.phil.uni-sb.de (8.7.5/8.7.3) id XAA00287 for benni; Thu, 16 May 1996 23:25:14 +0200
Received: from JIMI.CS.UNLV.EDU (jimi.CS.UNLV.EDU [131.216.22.2]) by pf2.phil.uni-sb.de (8.7.5/8.7.5/chris) with SMTP id XAA26054 for <benni@phil.uni-sb.de>; Sun, 12 May 1996 23:20:25 +0200 (MET DST)
Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa04079;
          12 May 96 14:01 PDT
Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa04031; 12 May 96 13:57 PDT
Received: from kroete2.freinet.de (arh135.ppp.uni-c.dk [130.228.13.135]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id WAA27550 for <bug-chimera@cs.unlv.edu>; Sun, 12 May 1996 22:57:27 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uIiBC-0003pNC; Sun, 12 May 96 22:55 MET DST
Message-Id: <m0uIiBC-0003pNC@kroete2.freinet.de>
Subject: A moving target?
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Sun, 12 May 1996 22:55:45 +0200 (MET DST)
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Status: O

The WWW-consortium have announced a new HTML 'standard' called HTML 3.2
(codenamed Wilbur).  It's looks like HTML 2.0 plus tables, <img
align=left/right>, applets, scripts, and a few other bits of the HTML
3.0 proposal. A lot of the HTML 3.0 proposal (including maths) didn't
make it, and is being worked on, presumably for HTML 4.0. You can see
more on https://www.w3.org/pub/WWW/MarkUp/Activity and on
https://www.w3.org/pub/WWW/MarkUp/Wilbur/. There's a DTD, but there
doesn't seem to be a nice non-SGML-wizards- version of it, like there
was for HTML 3.0 (accesible through my homepage).

Of course, we still have to get HTML 2.0 working, though I don't think
I'm the only one who finds the temptation to start fiddling with post-2.0
stuff almost irresistable.

Is 3.2 a realistic goal for Chimera 2.0, or should we settle for HTML 2.0,
clean up the usability a little and get the damn thing released? The
only thing you need to be able to do with applets and scripts for HTML 3.2
is to ignore them by the way.

- --
Erik Corry ehcorry@inet.uni-c.dk



------- Message 41

Received: from iguana.reptiles.org by JIMI.CS.UNLV.EDU id aa24132;
          19 May 96 11:04 PDT
Received: by iguana.reptiles.org (/\##/\ Smail3.1.30.13 #30.5)
	id <m0uLCq8-000345C@iguana.reptiles.org>; Sun, 19 May 96 14:04:31 -0400 (EDT)
Message-Id: <m0uLCq8-000345C@iguana.reptiles.org>
Date: Sun, 19 May 96 14:04:24 -0400 (EDT)
From: Smarasderagd <smar@reptiles.org>
To: bug-chimera@cs.unlv.edu
Subject: (redundant?) bug fix for cfh108

I ran into a problem related to interlaced GIF images with small heights.
I'm almost certain that this has been fixed already, but just in case, here's
a patch:

- --- 1.1	1996/05/19 18:03:48
+++ gif.c	1996/05/19 17:21:53
@@ -511,7 +511,7 @@
     if (image->x == image->width)
     {
       image->y += interlace_rate[image->pass];
- -      if (image->y >= image->height)
+      while (image->y >= image->height)
       {
 	image->pass++;
 	if (image->pass == 4) return(GS_DONE);
- --- 1.1	1996/05/19 18:04:09
+++ image.c	1996/05/19 16:48:22
@@ -10,6 +10,7 @@
 #include <stdlib.h>
 #endif
 
+#include <assert.h>
 #include <X11/StringDefs.h>
 #include <X11/Intrinsic.h>
 
@@ -871,6 +872,10 @@
 
     SetSize(is);
   }
+
+  /* "What time is it, boys and girls?"  "It's PARANOIA time!" */
+
+  assert(fline <= lline && fline >= 0 && lline < image->height);
 
   /*
    * Every time

------- Message 42

Received: from haymarket.ed.ac.uk by JIMI.CS.UNLV.EDU id aa08834;
          22 May 96 4:44 PDT
Received: from aiai.ed.ac.uk (skye-alter.aiai.ed.ac.uk [192.41.104.6]) by haymarket.ed.ac.uk (8.6.13/8.6.12) with SMTP id MAA08872; Wed, 22 May 1996 12:44:11 +0100
Received: from subnode.aiai.ed.ac.uk (scarp) by aiai.ed.ac.uk; Wed, 22 May 96 12:42:11 BST
Date: Wed, 22 May 96 12:42:11 BST
Message-Id: <7278.9605221142@subnode.aiai.ed.ac.uk>
From: Tim Bradshaw <tfb@aiai.ed.ac.uk>
To: Rob McMullen <rwmcm@hydra.ae.utexas.edu>
Cc: bug-chimera@cs.unlv.edu
Subject: Re: LessTif version of Chimera? 
In-Reply-To: <Pine.OSF.3.91.960509171750.663B-100000@hydra.ae.utexas.edu>
References: <9605092005.AA18866@orion.ae.utexas.edu>
	<Pine.OSF.3.91.960509171750.663B-100000@hydra.ae.utexas.edu>

* Rob McMullen wrote:
> On Thu, 9 May 1996, Jim Rees wrote:
>> Given two versions of the same app, the Motif version is virtually always
>> going to be preferred.
>> 
>> By you, but not by me.  I have many reasons for preferring athena apps, but
>> the biggest right now is that all the apps I use are athena or athena-like
>> (xmh, xterm, xrn, emacs) and I don't want two different interfaces on one
>> screen.

> Obviously the same argument could be made for having all Motif apps.  But,
> for you and others, a big drawing card of Chimera was its Athena
> interface.  That is something we have to consider. 

Just to add my tuppence worth: there really are people who prefer
athena to motif, primitive or no.  I'm one, and I would definitely not
be using chimera if it became motif or motif-alike.  Athena is
visually simple and has scrollbars that actually work properly...

- --tim

------- Message 43

Received: from igw2.watson.ibm.com by JIMI.CS.UNLV.EDU id aa10417;
          22 May 96 7:31 PDT
Received: from hawpub.watson.ibm.com (hawpub.watson.ibm.com [9.2.90.19]) by igw2.watson.ibm.com (8.7.4/8.7.1) with SMTP id KAA22097; Wed, 22 May 1996 10:30:19 -0400
Received: by hawpub.watson.ibm.com (AIX 3.2/UCB 5.64/5/18/96)
          id AA12076; Wed, 22 May 1996 10:27:58 -0400
From: Uri Blumenthal <uri@watson.ibm.com>
Message-Id: <9605221427.AA12076@hawpub.watson.ibm.com>
Subject: Re: LessTif version of Chimera?
To: Tim Bradshaw <tfb@aiai.ed.ac.uk>
Date: Wed, 22 May 1996 10:27:58 -0400 (EDT)
Cc: rwmcm@hydra.ae.utexas.edu, bug-chimera@cs.unlv.edu
In-Reply-To: <7278.9605221142@subnode.aiai.ed.ac.uk> from "Tim Bradshaw" at May 22, 96 12:42:11 pm
Reply-To: uri@watson.ibm.com
X-Mailer: ELM [version 2.4 PL25]
Content-Type: text

Tim Bradshaw says:
> Just to add my tuppence worth: there really are people who prefer
> athena to motif, primitive or no.  I'm one, and I would definitely not
> be using chimera if it became motif or motif-alike.  

I *might* use Chimera if it becomes Motif. I certainly will *not*,
if it becomes LessTif or other Motif look-alike.
- -- 
Regards,
Uri		uri@watson.ibm.com
- -=-=-=-=-=-=-
<Disclaimer>

------- Message 44

Received: from convex.convex.com by JIMI.CS.UNLV.EDU id aa13140;
          22 May 96 10:30 PDT
Received: from mikey.convex.com by convex.convex.com (8.6.4.2/1.35)
	id MAA21471; Wed, 22 May 1996 12:30:10 -0500
Received: by mikey.convex.com (8.6.12/1.28)
	id MAA22347; Wed, 22 May 1996 12:28:51 -0500
From: David DeSimone <fox@mikey.convex.com>
Message-Id: <199605221728.MAA22347@mikey.convex.com>
Subject: Re: LessTif version of Chimera?
To: Chimera Bug List <bug-chimera@cs.unlv.edu>
Date: Wed, 22 May 1996 12:28:50 -0500 (CDT)
Reply-To: David DeSimone <fox@rsn.hp.com>
X-Mailer: ELM [version 2.4 PL24 PGP6]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1054      

Uri Blumenthal wrote:
> 
> I *might* use Chimera if it becomes Motif. I certainly will *not*,
> if it becomes LessTif or other Motif look-alike.

I will use Chimera if/when it's a useful, stable browser.  I don't
particularly care what the buttons and menus look like; I'd much rather
that it be a nice browser that gives Netscape and Mosaic a run for their
money.

I *do* care if it's a free browser that I can compile from source on any
platform, even if I have to obtain other free libraries that I can also
compile.

I'm tired of being forced to use Netscape to browse the web because it's
the only fast, stable browser that supports tables, which all the web
authors seem to be enamored with.  I'd rather use something else, but
there isn't anything else.

- -- 
David DeSimone    | "The doctrine of human equality reposes on this:
fox@convex.hp.com |  that there is no man really clever who has not
Hewlett-Packard   |  found that he is stupid." -- Gilbert K. Chesterson
Convex Tech Ctr   |      PGP: 5B 47 34 9F 3B 9A B0 0D  AB A6 15 F1 BB BE 8C 44

------- Message 45

Received: from igw2.watson.ibm.com by JIMI.CS.UNLV.EDU id aa14040;
          22 May 96 11:26 PDT
Received: from hawpub.watson.ibm.com (hawpub.watson.ibm.com [9.2.90.19]) by igw2.watson.ibm.com (8.7.4/8.7.1) with SMTP id OAA26131; Wed, 22 May 1996 14:25:15 -0400
Received: by hawpub.watson.ibm.com (AIX 3.2/UCB 5.64/5/18/96)
          id AA33762; Wed, 22 May 1996 14:24:57 -0400
From: Uri Blumenthal <uri@watson.ibm.com>
Message-Id: <9605221824.AA33762@hawpub.watson.ibm.com>
Subject: Re: LessTif version of Chimera?
To: fox@rsn.hp.com
Date: Wed, 22 May 1996 14:24:57 -0400 (EDT)
Cc: bug-chimera@cs.unlv.edu
In-Reply-To: <199605221728.MAA22347@mikey.convex.com> from "David DeSimone" at May 22, 96 12:28:50 pm
Reply-To: uri@watson.ibm.com
X-Mailer: ELM [version 2.4 PL25]
Content-Type: text

David DeSimone says:
> > I *might* use Chimera if it becomes Motif. I certainly will *not*,
> > if it becomes LessTif or other Motif look-alike.
> 
> I will use Chimera if/when it's a useful, stable browser.  I don't
> particularly care what the buttons and menus look like; I'd much rather
> that it be a nice browser that gives Netscape and Mosaic a run for their
> money.

You're missing the point. If in order to compile/install that 
"useful, stable browser"  I have to indulge in crowding my
system with less-than-stable LessTif - I'm not interested.
- -- 
Regards,
Uri		uri@watson.ibm.com
- -=-=-=-=-=-=-
<Disclaimer>

------- Message 46

Received: from convex.convex.com by JIMI.CS.UNLV.EDU id aa15064;
          22 May 96 12:22 PDT
Received: from mikey.convex.com by convex.convex.com (8.6.4.2/1.35)
	id OAA27553; Wed, 22 May 1996 14:22:32 -0500
Received: by mikey.convex.com (8.6.12/1.28)
	id OAA29825; Wed, 22 May 1996 14:22:25 -0500
From: David DeSimone <fox@mikey.convex.com>
Message-Id: <199605221922.OAA29825@mikey.convex.com>
Subject: Re: LessTif version of Chimera?
To: uri@watson.ibm.com
Date: Wed, 22 May 1996 14:22:25 -0500 (CDT)
Cc: Chimera Bug List <bug-chimera@cs.unlv.edu>
In-Reply-To: <9605221824.AA33762@hawpub.watson.ibm.com> from "Uri Blumenthal" at May 22, 96 02:24:57 pm
Reply-To: David DeSimone <fox@rsn.hp.com>
X-Mailer: ELM [version 2.4 PL24 PGP6]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1467      

Uri Blumenthal wrote:
> 
> David DeSimone says:
> > 
> > I don't particularly care what the buttons and menus look like; I'd
> > much rather that it be a nice browser that gives Netscape and Mosaic a
> > run for their money.

> You're missing the point. If in order to compile/install that 
> "useful, stable browser"  I have to indulge in crowding my
> system with less-than-stable LessTif - I'm not interested.

I'm not missing the point.  I'm simply disagreeing with it.  :)

I don't consider downloading and compiling LessTif to be "crowding" my
system, any more than downloading libjpeg or libpng.  I guess it's just
a difference of considering LessTif to be useful software or not.

That being said, it wouldn't bother me a bit to have Chimera using
Athena widgets; there are so many X programs that use them, I can't
envision a user having trouble using Chimera with such an interface. 
The user would not be able to use other X tools without learning how to
use the widgets anyway.

Since Chimera is already using Athena, and switching to LessTif at this
juncture would only slow its development, I feel it wouldn't make any
sense to change things this drastically.

- -- 
David DeSimone    | "The doctrine of human equality reposes on this:
fox@convex.hp.com |  that there is no man really clever who has not
Hewlett-Packard   |  found that he is stupid." -- Gilbert K. Chesterson
Convex Tech Ctr   |      PGP: 5B 47 34 9F 3B 9A B0 0D  AB A6 15 F1 BB BE 8C 44

------- Message 47

Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa03794; 28 May 96 16:40 PDT
Received: from kroete2.freinet.de (arh166.ppp.uni-c.dk [130.228.13.166]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id BAA29875 for <bug-chimera@cs.unlv.edu>; Wed, 29 May 1996 01:39:35 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uOXwn-0003kLC; Wed, 29 May 96 01:13 MET DST
Message-Id: <m0uOXwn-0003kLC@kroete2.freinet.de>
Subject: Does everyone have Posix regular expressions
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Wed, 29 May 1996 01:13:04 +0200 (MET DST)
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 757       

I'd like to use the regular expression support that is supposed to
be in your C library accoring to Posix. But I'm not sure everyone has
it. I know Linux has, but if people with access to other OSs could
check that they have regcomp, regexec and regfree, I would be grateful.

They are supposed to look like this:

       #include <regex.h>

       int regcomp(regex_t *preg, const char *regex, int cflags);
       int regexec(const   regex_t  *preg,  const  char *string,
                   size_t  nmatch,   regmatch_t pmatch[],   int
                   eflags);
       size_t regerror(int  errcode,  const  regex_t  *preg, char
                       *errbuf, size_t errbuf_size);
       void regfree(regex_t *preg);

- --
Erik Corry ehcorry@inet.uni-c.dk

------- Message 48

Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa04553; 28 May 96 17:45 PDT
Received: from kroete2.freinet.de (arh139.ppp.uni-c.dk [130.228.13.139]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id CAA03009 for <bug-chimera@cs.unlv.edu>; Wed, 29 May 1996 02:45:20 +0200
Received: by kroete2.freinet.de (Smail3.1.28.1 #4)
	id m0uOZMU-0003jjC; Wed, 29 May 96 02:43 MET DST
Message-Id: <m0uOZMU-0003jjC@kroete2.freinet.de>
Subject: Re: Does everyone have Posix regular expressions
To: Chimera Hacker List <bug-chimera@cs.unlv.edu>
Date: Wed, 29 May 1996 02:43:42 +0200 (MET DST)
In-Reply-To: <199605290013.TAA23981@mikey.convex.com> from "David DeSimone" at May 28, 96 07:13:46 pm
Reply-To: Erik Corry <ehcorry@inet.uni-c.dk>
From: Erik Corry <ehcorry@inet.uni-c.dk>
X-Face:  5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS
  <jt,R~:$--|={s;V(G/wsw[@S(Fv4{7`b/@Z7k_QW-s]J-<8oe*5)dg+S.CeqQ$~m]}jY$c
X-Mailer: ELM [version 2.4 PL24 ME8]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1144      

> One of my target platforms, ConvexOS, does not have the library
> functions.  We're BSD-based, and we do have the re_comp and re_exec
> library functions; perhaps they will do?

No. They have static state, which means you can only have one
regular expression compiled at once. That basically means they're
just toy implementations, not suitable for real work. Also they're
not very powerful, they only tell you whether there was a regex
match, not where it was.

> I was under the impression that GNU autoconf will take care of
> determining this, but I'm not sure if that even applies to Chimera at
> this time.

Trouble is, I don't just want to determine the presence or not of the
routines, I also want to use them.

I'll look into the Henry Spencer stuff. I looked at Perl, but it's
not exactly easy getting code out of Perl because it's all so
interdependent. There's also the GNU rx library which has a Posix-
compatible interface, but that's GNU LGPL (library-public license).
That might be OK, but it means you can't ship binaries unless they
can be relinked to a newer version of the rx library.

- --
Erik Corry ehcorry@inet.uni-c.dk

------- Message 49

Received: from concorde.inria.fr by JIMI.CS.UNLV.EDU id aa09265;
          29 May 96 2:43 PDT
Received: from fantomas.inria.fr (fantomas.inria.fr [128.93.11.34]) by concorde.inria.fr (8.7.1/8.7.1) with ESMTP id LAA11512; Wed, 29 May 1996 11:42:54 +0200 (MET DST)
Received: (from lasgout@localhost) by fantomas.inria.fr (8.6.10/8.6.6) id LAA08896; Wed, 29 May 1996 11:42:53 +0200
Date: Wed, 29 May 1996 11:42:53 +0200
Message-Id: <199605290942.LAA08896@fantomas.inria.fr>
From: Jean-Marc Lasgouttes <Jean-Marc.Lasgouttes@inria.fr>
To: ehcorry@inet.uni-c.dk
CC: bug-chimera@cs.unlv.edu
In-reply-to: <m0uOXwn-0003kLC@kroete2.freinet.de> (message from Erik Corry on
	Wed, 29 May 1996 01:13:04 +0200 (MET DST))
Subject: Re: Does everyone have Posix regular expressions
Reply-to: Jean-Marc.Lasgouttes@inria.fr

>>>>> "Erik" == Erik Corry <ehcorry@inet.uni-c.dk> writes:

    Erik> I'd like to use the regular expression support that is
    Erik> supposed to be in your C library accoring to Posix. But I'm
    Erik> not sure everyone has it. I know Linux has, but if people
    Erik> with access to other OSs could check that they have regcomp,
    Erik> regexec and regfree, I would be grateful.

On SunOS4.1.3, what I found is:

NAME
     regex, re_comp, re_exec - regular expression handler

SYNOPSIS
     char *re_comp(s)
     char *s;

     re_exec(s)
     char *s;

DESCRIPTION
     re_comp() compiles a string into an internal  form  suitable
     for  pattern matching.  re_exec() checks the argument string
     against the last string passed to re_comp().

     re_comp() returns a NULL pointer if the string  s  was  com-
     piled  successfully;  otherwise a string containing an error
     message is returned.  If re_comp() is passed  0  or  a  null
     string,  it  returns without changing the currently compiled
     regular expression.

     re_exec() returns 1 if the string s matches  the  last  com-
     piled  regular expression, 0 if the string s failed to match
     the last compiled regular expression, and -1 if the compiled
     regular  expression  was  invalid  (indicating  an  internal
     error).

     The strings passed to both re_comp() and re_exec() may  have
     trailing or embedded NEWLINE characters; they are terminated
     by null characters.  The regular expressions recognized  are
     described  in  the  manual  entry for ed(1), given the above
     difference.


------- Message 50

Received: from cheviot.ncl.ac.uk by JIMI.CS.UNLV.EDU id aa09511;
          29 May 96 3:21 PDT
Received: from aire.ncl.ac.uk by cheviot.ncl.ac.uk id <LAA25928@cheviot.ncl.ac.uk>
  (8.6.12/ for ncl.ac.uk) with SMTP; Wed, 29 May 1996 11:20:21 +0100
Received: (njdc@localhost) by aire.ncl.ac.uk (8.6.12/8.6.x-cf revision 3 for SunOS 4.1.x) id LAA19913; Wed, 29 May 1996 11:18:02 +0100
From: "J.D.Coleman" <J.D.Coleman@newcastle.ac.uk>
Message-Id: <199605291018.LAA19913@aire.ncl.ac.uk>
Subject: Re: Does everyone have Posix regular expressions
To: Chimera Mailing List <bug-chimera@cs.unlv.edu>
Date: Wed, 29 May 1996 11:18:01 +0100 (BST)
In-Reply-To: <199605290942.LAA08896@fantomas.inria.fr> from "Jean-Marc Lasgouttes" at May 29, 96 11:42:53 am
X-Organisation: University of Newcastle Computing Service
X-Phone: +44-191-222-8068
X-Fax: +44-191-222-8765
X-Mailer: ELM [version 2.4 PL23]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 703       

Erik Corry wrote:
...
> They are supposed to look like this:
> 
>        #include <regex.h>
...

Jean-Marc Lasgoutt wrote:
...
> On SunOS4.1.3, what I found is:
> 
> NAME
>      regex, re_comp, re_exec - regular expression handler
...

Note that SunOS has /usr/include/regexp.h not /usr/include/regex.h.

J

PS.  Re: should Chimera 2.0 support HTML 2.0 or 3.2 - well I'd like to see
table support, but I think a release would be much preferable.  Just my 2
pennies worth.

PPS. Should we have a Chimera 3.2 for HTML 3.2?

- -- 
"The Ravenous Bugblatter Beast of Traal!  Is it safe?"
"Oh yes, it's perfectly safe.  It's just us who are in trouble!"
- --
Wargame (n): battle simulation using painted numbers.

------- Message 51

Received: from uucp-1.csn.net by JIMI.CS.UNLV.EDU id aa11339; 29 May 96 7:44 PDT
Received: from grendel.UUCP (uucp@localhost) by uucp-1.csn.net (8.7.5/8.7.3) with UUCP id IAA10961; Wed, 29 May 1996 08:42:57 -0600 (MDT)
Received: by grendel.objinc.com (5.0/SMI-SVR4)
	id AA05210; Wed, 29 May 1996 08:30:35 +0700
Date: Wed, 29 May 1996 08:30:35 +0700
From: Chip Atkinson <chip@grendel.objinc.com>
Message-Id: <9605291430.AA05210@grendel.objinc.com>
To: ehcorry@inet.uni-c.dk
Subject: Re: Does everyone have Posix regular expressions
Cc: bug-chimera@cs.unlv.edu
X-Sun-Charset: US-ASCII

Greetings,

My OS does not have regex.h, it has regexpr.h and regexp.h.
I grepped the /usr/include.... directories for 
regcomp
regexec
regerror 
regfree 
and found nothing.

I can mail these or other headers to you if desire.

My OS is Solaris 2.3

Chip
chip@objinc.com

------- End of Forwarded Messages