Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa01667; 26 Dec 96 22:57 PST To: jay@JIMI.CS.UNLV.EDU Subject: bug-chimera aug 95 Date: Thu, 26 Dec 1996 22:56:59 -0800 From: Jay Nietling ------- Forwarded Messages Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa21174; 4 Aug 95 1:46 PDT Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa21063; 4 Aug 95 1:34 PDT To: chimera-announce@cephas.ISRI.UNLV.EDU Subject: new 2.0 alpha Date: Fri, 04 Aug 1995 01:34:51 -0700 From: John Kilburg https://www.isri.unlv.edu/~john/chimera/src/ Supports basic forms stuff (text, textarea, checkbox, submit, single select). Works on CUSI, for example. -john ------- Message 2 Received: from xph029.physics.montana.edu by JIMI.CS.UNLV.EDU id aa02520; 4 Aug 95 11:15 PDT Received: (from michael@localhost) by xph029.physics.montana.edu (8.6.12/8.6.9) id MAA06764; Fri, 4 Aug 1995 12:14:04 -0600 Date: Fri, 4 Aug 1995 12:13:57 -0600 (MDT) From: Michael Kellen To: Chimera Hacker List Subject: [cfh67-2.0] Buglist :-) Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Since some forms stuff now works, I tried out the HTML 2 bug checker at . There are some problems with the parser. Also, the default width for input fields is supposed to be 40 characters, isn't it? Scrollbars now correlate with text! Some inline GIFs are displayed side-by-side: (this was true in earlier alphas as well) You can input URL's into that line in the browser gopher request results are not displayed correctly. A trailing ?, if appended to the request, generates a response, but it is saved as a file rather than displayed. ftp requests no longer core dump. Not only that, they work great! Now if we can just get filesize and type in there. local & ftp lists start with "localheader" and "ftpheader" respectively, presumably these are stubs for later header strings. http indices still discard xbm's and choke on long filenames File popup still does not function. Search Popup does not function Cut-and-paste selections do not work in text viewing area. telnet and extension protocol requests are "Downloaded", but no document is actually requested or displayed Still screams along nicely. Keep up the good work! Michael ------- Message 3 Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa24122; 6 Aug 95 14:38 PDT Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa24100; 6 Aug 95 14:34 PDT To: chimera-announce@cephas.ISRI.UNLV.EDU Subject: another 2.0 alpha Date: Sun, 06 Aug 1995 14:34:25 -0700 From: John Kilburg Better forms support and various bug fixes. You might need to set Chimera*brokenHTML: True to relax the HTML parser. https://www.isri.unlv.edu/~john/chimera/src/ -john ------- Message 4 Received: from milkyway.com by JIMI.CS.UNLV.EDU id aa15118; 8 Aug 95 15:42 PDT Received: from metis.milkyway.com (rootmcr@metis.milkyway.com [192.168.77.21]) by jupiter.milkyway.com (8.6.12/8.6.12) with ESMTP id SAA01502 for ; Tue, 8 Aug 1995 18:40:03 -0400 Received: from metis.milkyway.com by metis.milkyway.com (8.6.9/BSDI-Client) id SAA09025; Tue, 8 Aug 1995 18:51:46 -0400 Message-Id: <199508082251.SAA09025@metis.milkyway.com> X-Mailer: exmh version 1.6.1 5/23/95 X-rating: You must be 18 years exactly to read this message X-Face: +o^+u7Z5}dB^gVlCgr.W/thrVG>63+@L&~6W3um$qzdHEf*o^b4g'.>AF*9jO,@sw.~gu*+ !Ld4U(yvY'QL7ZSB#r3zb[pTsR0K5ZHDs5.8'w.'$u(o;imk*Z-.g)V|2a-KM-waTKUvx'xM>xOlZL E=ghh49p2h$1`Vp&rOtYlnm{|ixN#45yL)*j$3>QbmWu-[)Nw;^P53@cMO[P#Q>k3Ut)?Vh^`IJYvB ZdB[z`5aM4Z"wW@l~~iWw0MY^%F$mP)~F\lBcgj`h^hOvIp< X-Uri: https://www.milkyway.com/People/Michael_Richardson/Bio.html To: bug-chimera@cs.unlv.edu Subject: compiling on X11R6 (BSD/OS v2.0) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 08 Aug 1995 18:51:43 -0400 From: Michael Richardson while compiling www/WWW.c www/WWWP.c, I get problems with prototypes. Boolean is a "char" --- which can not be used with with prototyped functions since it promotes to "int". cc on this system is gcc 1.4.2 (I still do not know why...) I haven't figured out the imake-friendly way to use "gcc2" ... ------- Message 5 Received: from tybalt.caltech.edu by JIMI.CS.UNLV.EDU id aa00969; 10 Aug 95 4:28 PDT Received: from alumni.caltech.edu by tybalt.caltech.edu with ESMTP (8.6.7/DEI:4.41) id EAA00292; Thu, 10 Aug 1995 04:28:38 -0700 Received: from localhost by alumni.caltech.edu (8.6.4/DEI:4.41) id EAA02339; Thu, 10 Aug 1995 04:28:37 -0700 Message-Id: <199508101128.EAA02339@alumni.caltech.edu> From: Lester Ingber Date: Thu, 10 Aug 1995 04:28:36 -0700 Reply-To: ingber@alumni.caltech.edu X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: bug-chimera@cs.unlv.edu Subject: problems with authentication window bug-chimera@cs.unlv.edu I'm using chimera-1.65 on my Sun SPARC 2/4.1.3_U1/OW-3.0 via alumni.caltech.edu. For example, this was true on https://www.cuc.com/ https://update.wsj.com/ On sites that require an "authentication" login, after I type my login, I can't seem to get to the next line to type in my password? I've tried placing the cursor, hitting return, tab, up-arrow, down-arrow, etc., but my typing just gets appended to the login space? Thanks. Lester /* RESEARCH ingber@alumni.caltech.edu * * INGBER ftp.alumni.caltech.edu:/pub/ingber * * LESTER https://www.alumni.caltech.edu/~ingber/ * * Prof. Lester Ingber _ P.O. Box 857 _ McLean, VA 22101 _ 1.800.L.INGBER */ ------- Message 6 Received: from mail.cs.tu-berlin.de by JIMI.CS.UNLV.EDU id aa09245; 10 Aug 95 10:40 PDT Received: from snake.cs.tu-berlin.de (czyborra@snake.cs.tu-berlin.de [130.149.29.116]) by mail.cs.tu-berlin.de (8.6.12/8.6.12) with ESMTP id TAA18813; Thu, 10 Aug 1995 19:38:00 +0200 Received: (czyborra@localhost) by snake.cs.tu-berlin.de (8.6.12/8.6.12) id TAA20179; Thu, 10 Aug 1995 19:37:52 +0200 Subject: Re: problems with authentication window Cc: bug-chimera@cs.unlv.edu X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII From: Roman Czyborra To: Lester Ingber Date: Thu, 10 Aug 1995 19:37:50 +0200 Message-ID: In-Reply-To: <199508101128.EAA02339@alumni.caltech.edu> > I can't seem to get to the next line to type in my password I think there is no keyboard shortcut to warp to the password field. But you can use the mouse to move the pointer into it. The password won't appear on the screen as you type it, but you can see the cursor move. ------- Message 7 Received: from crl14.crl.com by JIMI.CS.UNLV.EDU id aa10312; 11 Aug 95 8:26 PDT Received: (from root@localhost) by tailor.roman.org (8.6.11/8.6.9) id LAA00093; Fri, 11 Aug 1995 11:28:32 -0400 Date: Fri, 11 Aug 1995 11:28:32 -0400 Message-Id: <199508111528.LAA00093@tailor.roman.org> To: bug-chimera@cephas.ISRI.UNLV.EDU X-URL: mailto:bug-chimera@cephas.ISRI.UNLV.EDU X-Mailer: Lynx, Version 2-4-1 From: tailor@crl.com Subject: STOP! button Maybe the `Cancel' button is the `Stop' button on other browsers, but, once the URL is being fetched, you get a clock, and `Cancel' doesn't seem to respond. Could you enable the area on the menu bar during URL fetches, so that, e. g., `Cancel' works? Otherwise, chimera is a terrific idea, and really shows up Mosaic and Netscape by doing so much with so little code. ------- Message 8 Received: from convex.convex.com by JIMI.CS.UNLV.EDU id aa12225; 11 Aug 95 9:30 PDT Received: from mikey.convex.com by convex.convex.com (8.6.4.2/1.35) id LAA02918; Fri, 11 Aug 1995 11:30:06 -0500 Received: by mikey.convex.com (8.6.10/1.28) id LAA05642; Fri, 11 Aug 1995 11:30:06 -0500 Date: Fri, 11 Aug 1995 11:30:05 -0500 (CDT) From: David DeSimone To: tailor@crl.com cc: bug-chimera@cephas.ISRI.UNLV.EDU Subject: Re: STOP! button In-Reply-To: <199508111528.LAA00093@tailor.roman.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Fri, 11 Aug 1995 tailor@crl.com wrote: > Maybe the `Cancel' button is the `Stop' button on other browsers, > but, once the URL is being fetched, you get a clock, and `Cancel' > doesn't seem to respond. Which Chimera version are you using? I have not tried 2.0 yet, but 1.65 works properly in this regard. However, you do have to enable this option in options.h: /* * Define this if you want non-blocking connects. This causes problems * on some machines. */ #define NONBLOCKING_CONNECT 1 Furthermore, your OS must support non-blocking connects. If it does not, then the connect attempt will block until it either succeeds, or times out. That's the behavior you're seeing. - -- David DeSimone | "The doctrine of human equality reposes on this: fox@convex.com | that there is no man really clever who has not Convex Computers | found that he is stupid." -- Gilbert K. Chesterson ------- Message 9 Received: from mail.cs.tu-berlin.de by JIMI.CS.UNLV.EDU id aa13435; 11 Aug 95 10:05 PDT Received: from titanic.cs.tu-berlin.de (czyborra@titanic.cs.tu-berlin.de [130.149.18.9]) by mail.cs.tu-berlin.de (8.6.12/8.6.12) with ESMTP id TAA20210; Fri, 11 Aug 1995 19:05:05 +0200 Received: (czyborra@localhost) by titanic.cs.tu-berlin.de (8.6.12/8.6.6) id TAA01813; Fri, 11 Aug 1995 19:05:03 +0200 From: Roman Czyborra To: tailor@crl.com Cc: bug-chimera@cs.unlv.edu Subject: Re: STOP! button In-Reply-To: <199508111528.LAA00093@tailor.roman.org> by tailor@crl.com dated 1995-8-11 11:28:32 MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Date: Fri, 11 Aug 1995 19:05:02 +0200 Message-ID: > Maybe the `Cancel' button is the `Stop' button on other browsers, Yes, indeed. > `Cancel' doesn't seem to respond. Try hitting it twice or give it some time to wake up. ------- Message 10 Received: from mail.cs.tu-berlin.de by JIMI.CS.UNLV.EDU id aa15846; 11 Aug 95 11:08 PDT Received: from titanic.cs.tu-berlin.de (czyborra@titanic.cs.tu-berlin.de [130.149.18.9]) by mail.cs.tu-berlin.de (8.6.12/8.6.12) with ESMTP id TAA21222; Fri, 11 Aug 1995 19:43:34 +0200 Received: (czyborra@localhost) by titanic.cs.tu-berlin.de (8.6.12/8.6.6) id TAA02071; Fri, 11 Aug 1995 19:43:32 +0200 From: Roman Czyborra To: David DeSimone Cc: tailor@crl.com, bug-chimera@cs.unlv.edu Subject: Re: STOP! button In-Reply-To: by fox@convex.convex.com dated 1995-8-11 11:30:05 MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Date: Fri, 11 Aug 1995 19:43:31 +0200 Message-ID: > #define NONBLOCKING_CONNECT 1 I have defined this and still get the symptom on SunOS 4.1.3_U1. > Furthermore, your OS must support non-blocking connects. According to man 2 socket, it does. ------- Message 11 Received: from LINC.CIS.UPENN.EDU by JIMI.CS.UNLV.EDU id aa21046; 11 Aug 95 14:07 PDT Received: (anoop@localhost) by linc.cis.upenn.edu (8.6.12/UPenn 1.4) id RAA17700 for bug-chimera@cs.unlv.edu; Fri, 11 Aug 1995 17:07:05 -0400 Date: Fri, 11 Aug 1995 17:07:05 -0400 From: Anoop Sarkar Posted-Date: Fri, 11 Aug 1995 17:07:05 -0400 Message-Id: <199508112107.RAA17700@linc.cis.upenn.edu> To: bug-chimera@cs.unlv.edu Subject: subscribe Anoop Sarkar Content-Length: 1 ------- Message 12 Received: from little-charlie.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa24419; 11 Aug 95 16:04 PDT To: tailor@crl.com cc: bug-chimera@cephas.isri.unlv.edu Subject: Re: STOP! button In-reply-to: Your message of "Fri, 11 Aug 1995 11:28:32 EDT." <199508111528.LAA00093@tailor.roman.org> Date: Fri, 11 Aug 1995 16:04:27 -0700 From: John Kilburg >Maybe the `Cancel' button is the `Stop' button on other browsers, >but, once the URL is being fetched, you get a clock, and `Cancel' >doesn't seem to respond. Could you enable the area on the menu >bar during URL fetches, so that, e. g., `Cancel' works? Is this on 1.65? Cancel is hopeless on 1.65. The next release (2.0) should solve this problem. >Otherwise, chimera is a terrific idea, and really shows up Mosaic >and Netscape by doing so much with so little code. Thanks. Hopefully it will get much better with 2.0. -john ------- Message 13 Received: from convex.convex.com by JIMI.CS.UNLV.EDU id aa25793; 11 Aug 95 16:49 PDT Received: from mikey.convex.com by convex.convex.com (8.6.4.2/1.35) id SAA03214; Fri, 11 Aug 1995 18:49:16 -0500 Received: by mikey.convex.com (8.6.10/1.28) id SAA24027; Fri, 11 Aug 1995 18:49:14 -0500 Date: Fri, 11 Aug 1995 18:49:14 -0500 (CDT) From: David DeSimone To: Roman Czyborra cc: tailor@crl.com, bug-chimera@cs.unlv.edu Subject: Re: STOP! button In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Fri, 11 Aug 1995, Roman Czyborra wrote: > > #define NONBLOCKING_CONNECT 1 > > I have defined this and still get the symptom on SunOS 4.1.3_U1. I'm running it on the same OS, and it works for me. > > Furthermore, your OS must support non-blocking connects. > > According to man 2 socket, it does. I think it depends entirely on what operation you're trying to cancel. Perhaps chimera is trying to lookup the hostname via DNS, or maybe YP, and the interruptibility of these operations is questionable. - -- David DeSimone | "The doctrine of human equality reposes on this: fox@convex.com | that there is no man really clever who has not Convex Computers | found that he is stupid." -- Gilbert K. Chesterson ------- Message 14 Received: from billy-boy.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa25842; 11 Aug 95 16:51 PDT To: Anoop Sarkar cc: bug-chimera@cs.unlv.edu, slumos@billy-boy.CS.UNLV.EDU Subject: Re: subscribe Anoop Sarkar In-reply-to: Your message of "Fri, 11 Aug 1995 17:07:05 EDT." <199508112107.RAA17700@linc.cis.upenn.edu> Date: Fri, 11 Aug 1995 16:51:31 -0700 From: "Steve E. Lumos" > Perhaps big-chimera-request@cs.unlv.edu would be a better address for this next time.... TTYL - Steve ------- Message 15 Received: from msa.tte.vtt.fi by JIMI.CS.UNLV.EDU id aa04339; 12 Aug 95 0:30 PDT Received: (from msa@localhost) by msa.tte.vtt.fi (8.6.9/8.6.9) id KAA23743; Sat, 12 Aug 1995 10:29:32 +0300 Date: Sat, 12 Aug 1995 10:29:32 +0300 From: Markku Savela Message-Id: <199508120729.KAA23743@msa.tte.vtt.fi> To: bug-chimera@cs.unlv.edu In-reply-to: David DeSimone's message of Fri, 11 Aug 1995 18:49:14 -0500 (CDT) Subject: STOP! button Reply-to: Markku Savela >I think it depends entirely on what operation you're trying to cancel. >Perhaps chimera is trying to lookup the hostname via DNS, or maybe YP, >and the interruptibility of these operations is questionable. Nothing about interoperabitily. Just plain standard DNS libs are by default blocking. This became annoying enough in IRC server code, so that Darren Reed ripped the resolv lib open and modified an anynchoronous version of the DNS library for IRC server code. Of course, making a private non-blocking resolv library for a WWW client is probably an overkill (it became absolute necessity for ircd, which had to deal with tens or hundreds of connect attempts per minute...) - -- Markku Savela (msa@hemuli.tte.vtt.fi), Technical Research Centre of Finland Multimedia Systems, P.O.Box 1203,FIN-02044 VTT,https://www.vtt.fi/tte/staff/msa/ ------- Message 16 Received: from tybalt.caltech.edu by JIMI.CS.UNLV.EDU id aa22789; 13 Aug 95 6:12 PDT Received: from alumni.caltech.edu by tybalt.caltech.edu with ESMTP (8.6.7/DEI:4.41) id GAA27600; Sun, 13 Aug 1995 06:12:44 -0700 Received: from localhost by alumni.caltech.edu (8.6.4/DEI:4.41) id GAA01584; Sun, 13 Aug 1995 06:12:41 -0700 Message-Id: <199508131312.GAA01584@alumni.caltech.edu> From: Lester Ingber Date: Sun, 13 Aug 1995 06:12:41 -0700 Reply-To: ingber@alumni.caltech.edu X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: bug-chimera@cs.unlv.edu Subject: non-response in Authentication window under chimera-1.65 bug-chimera@cs.unlv.edu To make the situation more clear, below is a uuencoded compressed PostScript file of the Authentication window. Under chimera, placing the cursor in the Password box has no effect; continued typing just adds characters to the upper Username box? I have a "stock" Sun SPARC 2/4.1.3_U1/OW-3.0 and running chimera-1.65 under term-2.3.5. I have not had such problems on previous versions of chimera. Lester :I'm using chimera-1.65 on my Sun SPARC 2/4.1.3_U1/OW-3.0 via :alumni.caltech.edu. For example, this was true on : https://www.cuc.com/ : https://update.wsj.com/ : :On sites that require an "authentication" login, after I type my login, :I can't seem to get to the next line to type in my password? I've :tried placing the cursor, hitting return, tab, up-arrow, down-arrow, :etc., but my typing just gets appended to the login space? : :Thanks. : :Lester : :======================================================================== :Roman: : :No, I can't; that's the problem! As I mentioned: :: I've :: tried placing the cursor, :: ... :That of course was the first thing I tried (several times). Since this :problem has occurred on several URL sites, I thought it reasonable that :something is wrong with (my implementaiono f) chimera? I'm used to filling :out many forms, e.g., by placing the cursor, etc., but here it is failing? : :Thanks. : :Lester : :On Aug 10, 7:37pm, Roman Czyborra wrote: :: Subject: Re: problems with authentication window :: From czyborra@cs.tu-berlin.de Thu Aug 10 10:41:50 1995 :: Return-Path: :: Subject: Re: problems with authentication window :: Cc: bug-chimera@cs.unlv.edu :: From: Roman Czyborra :: To: Lester Ingber :: Date: Thu, 10 Aug 1995 19:37:50 +0200 :: In-Reply-To: <199508101128.EAA02339@alumni.caltech.edu> :: :: > I can't seem to get to the next line to type in my password :: :: I think there is no keyboard shortcut to warp to the password field. :: But you can use the mouse to move the pointer into it. The password won't :: appear on the screen as you type it, but you can see the cursor move. :: :: ::-- End of excerpt from Roman Czyborra /* --8<--------8<---- CUT HERE TO CUT-LINE BELOW ----8<--------8<-- */ begin 644 auth.ps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end /* -->8-------->8---- CUT HERE TO CUT-LINE ABOVE ---->8-------->8-- */ /* RESEARCH ingber@alumni.caltech.edu * * INGBER ftp.alumni.caltech.edu:/pub/ingber * * LESTER https://www.alumni.caltech.edu/~ingber/ * * Prof. Lester Ingber _ P.O. Box 857 _ McLean, VA 22101 _ 1.800.L.INGBER */ ------- Message 17 Received: from mail.crl.com by JIMI.CS.UNLV.EDU id aa24934; 13 Aug 95 7:24 PDT Received: from tailor.roman.org (crl14.crl.com) by mail.crl.com with SMTP id AA00998 (5.65c/IDA-1.5 for ); Sun, 13 Aug 1995 07:24:07 -0700 Date: Sun, 13 Aug 1995 10:26:17 -0400 (EDT) From: tailor X-Sender: root@tailor.roman.org To: bug-chimera@cs.unlv.edu Subject: news.pl and postto.pl on Linux (Chimera 1.65) Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII From : tailor@crl.com (Not root...) First of all, Chimera is really neat and clever. Only problem is that on Linux, it doesn't post properly to remote a news server. I found one place in postto.pl where adding a line made the posting of news go as far as the mock-pine Chimera window (where ^K erases a line, if you know it does), but the post still fails with a message of the form " can't post to news group /news.group " Herewith the minor change to postto.pl: while (<>) { chop; ($type,$value) = split ( '\s*:\s+' , $_, 2); if ($type eq "") { last; } if ( $type eq "X-hostname" ) { $server = $value; } if ( $type eq "X-filename" ) { $group = $value; ($type,$value) = split ( '/' , $value); } The correction is the last line. ------- Message 18 Received: from fornax.thenet.co.uk by JIMI.CS.UNLV.EDU id aa15847; 14 Aug 95 6:51 PDT Received: from debath.thenet.co.uk by mailhost.thenet.co.uk (8.6.12/FCI Sendmail Config v1.0) id WAA03512; Sat, 12 Aug 1995 22:12:58 GMT Received: from debath.thenet.co.uk by debath.thenet.co.uk with smtp (Smail3.1.29.1 #3) id m0shih8-0007dFC; Sun, 13 Aug 95 20:27 BST Date: Sun, 13 Aug 1995 20:27:37 +0100 (BST) From: Robert de Bath cc: bug-chimera@cs.unlv.edu Subject: Re: STOP! button In-Reply-To: <199508120729.KAA23743@msa.tte.vtt.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Sat, 12 Aug 1995, Markku Savela wrote: > > >I think it depends entirely on what operation you're trying to cancel. > >Perhaps chimera is trying to lookup the hostname via DNS, or maybe YP, > >and the interruptibility of these operations is questionable. ... > Of course, making a private non-blocking resolv library for a WWW > client is probably an overkill (it became absolute necessity for ircd, > which had to deal with tens or hundreds of connect attempts per > minute...) Of course, there is a way to avoid the entire problem for Chimera, it goes like this: You put the protocal engines for opening and collecting the data into tiny helper applications: For eg: Main program 'chimera' helpers: chimera_http chimera_ftp chimera_gopher chimera_nntp chimera_tftp ... These will be very small and you use these for _ALL_ URL accesses. (With the possible exception of file://... as it's practically guarenteed to be quick) The display program should start the fetcher (sending it the information it needs) and then return to the main wait loop, where it sits on a select statment. This has some important fallouts: 1) The program only blocks on the select statment so it's reasonably easy to make it respond to X events (redraws and aborts!) quickly even while it's fetching URLs (It could even do some animation while it's waiting :-) 2) It's not too difficult for the display program to start multiple requests at the same time (all the inline images in a document) 3) The display program should have little trouble with displaying partial documents. 4) Disconnected requests are reasonably easy ie fetch a URL into a file can be made so that the display program can go and look at other URLs while the fetch is still running. (It could be done either via the display program acting something like cat or simply redirecting the fetcher's standard out to a file) One point to note is that I've suggested fixed formats for the names of the fetcher programs, this is to avoid the need for another configuration file. This might seem like the current method of adding extended url types but the important difference is that the data is _not_ fetched from a tight loop. Oh and one last _very_ important advantage, it'd be a real pig to port to MSwindoze :-) :-) - -- Rob. (Robert de Bath ) ------- Message 19 Received: from mail.cs.tu-berlin.de by JIMI.CS.UNLV.EDU id aa16933; 15 Aug 95 20:52 PDT Received: from titanic.cs.tu-berlin.de (czyborra@titanic.cs.tu-berlin.de [130.149.18.9]) by mail.cs.tu-berlin.de (8.6.12/8.6.12) with ESMTP id FAA21388; Wed, 16 Aug 1995 05:36:59 +0200 Received: (czyborra@localhost) by titanic.cs.tu-berlin.de (8.6.12/8.6.6) id FAA10873; Wed, 16 Aug 1995 05:36:57 +0200 Subject: Re: non-response in Authentication window under chimera-1.65 Cc: bug-chimera@cs.unlv.edu, Kathrin Spiller X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII From: Roman Czyborra To: Lester Ingber Date: Wed, 16 Aug 1995 05:36:56 +0200 Message-ID: In-Reply-To: <199508131312.GAA01584@alumni.caltech.edu> Hello Lester, > To make the situation more clear, below is a uuencoded compressed > PostScript file of the Authentication window. For such purposes, I prefer MIME encoded mails containing an xv-launching Content-Type: image/gif like you can send with metamail, pine, zmail or netscape. > Under chimera, placing the cursor in the Password box has no effect; > continued typing just adds characters to the upper Username box? Sorry, I was wrong and you were right. Moving the mouse has no effect. Kathrin Spiller told me today that you have to type your username, then press Return and then move the mouse. ------- Message 20 Received: from mail.cs.tu-berlin.de by JIMI.CS.UNLV.EDU id aa12812; 16 Aug 95 16:19 PDT Received: from titanic.cs.tu-berlin.de (root@titanic.cs.tu-berlin.de [130.149.18.9]) by mail.cs.tu-berlin.de (8.6.12/8.6.12) with ESMTP id BAA24437; Thu, 17 Aug 1995 01:18:56 +0200 Received: (czyborra@localhost) by titanic.cs.tu-berlin.de (8.6.12/8.6.6) id FAA10873; Wed, 16 Aug 1995 05:36:57 +0200 Subject: Re: non-response in Authentication window under chimera-1.65 Cc: bug-chimera@cs.unlv.edu, Kathrin Spiller X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII From: Roman Czyborra To: Lester Ingber Date: Wed, 16 Aug 1995 05:36:56 +0200 Message-ID: In-Reply-To: <199508131312.GAA01584@alumni.caltech.edu> Hello Lester, > To make the situation more clear, below is a uuencoded compressed > PostScript file of the Authentication window. For such purposes, I prefer MIME encoded mails containing an xv-launching Content-Type: image/gif like you can send with metamail, pine, zmail or netscape. > Under chimera, placing the cursor in the Password box has no effect; > continued typing just adds characters to the upper Username box? Sorry, I was wrong and you were right. Moving the mouse has no effect. Kathrin Spiller told me today that you have to type your username, then press Return and then move the mouse. ------- Message 21 Received: from enhanced.texel.com by JIMI.CS.UNLV.EDU id aa21917; 18 Aug 95 22:54 PDT Received: from wisdom.enhanced.com (cmaguire@wisdom.enhanced.com [192.0.0.19]) by intech4.enhanced.com (8.6.12/8.6.9) with ESMTP id BAA25647 for ; Sat, 19 Aug 1995 01:58:09 -0400 Received: (from cmaguire@localhost) by wisdom.enhanced.com (8.6.12/8.6.9) id BAA17728 for bug-chimera@cs.unlv.edu; Sat, 19 Aug 1995 01:51:52 -0400 Date: Sat, 19 Aug 1995 01:51:52 -0400 From: Camm Maguire Message-Id: <199508190551.BAA17728@wisdom.enhanced.com> To: bug-chimera@cs.unlv.edu Subject: subscribe, please Subscribe, please! ------- Message 22 Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa23423; 22 Aug 95 23:51 PDT Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa23416; 22 Aug 95 23:47 PDT To: chimera-announce@cephas.ISRI.UNLV.EDU Subject: 2.0 alpha 72 Date: Tue, 22 Aug 1995 23:47:19 -0700 From: John Kilburg Hopefully more stable and "solid" looking. https://www.isri.unlv.edu/~john/chimera/src/ -john ------- Message 23 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa00966; 23 Aug 95 4:26 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id NAA19910 for cs.unlv.edu!bug-chimera; Wed, 23 Aug 1995 13:26:47 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0slEiB-0003mTC; Wed, 23 Aug 95 14:15 MET DST Message-Id: From: Erik Corry Subject: Gif colors To: bug-chimera@cs.unlv.edu Date: Wed, 23 Aug 1995 14:15:15 +0100 (MET DST) Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 209 It seems to me that the colors of inline GIFs are rather messed up in the new 2.0 alpha 72. Anybody else get this? Is this normal? This is on an 8-bit pseudocolor screen. - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 24 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa00970; 23 Aug 95 4:26 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id NAA19909 for cs.unlv.edu!bug-chimera; Wed, 23 Aug 1995 13:26:46 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0slEfl-0003mSC; Wed, 23 Aug 95 14:12 MET DST Message-Id: From: Erik Corry Subject: 2.0 alpha 72 To: bug-chimera@cs.unlv.edu Date: Wed, 23 Aug 1995 14:12:45 +0100 (MET DST) Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 607 I needed the following patch. I don't pretend to understand its implications. The problem seems to be that rw->www.pqhead has been zeroed during the call to CallTagHandler at line 935. This at least stops Chimera crashing when it encounters an inline JPEG. - --- layout.c.orig Wed Aug 23 05:43:37 1995 +++ layout.c Wed Aug 23 13:43:06 1995 @@ -899,7 +899,7 @@ } for (; !rw->www.delay && rw->www.pqhead != NULL; - - rw->www.pqhead = rw->www.pqhead->qnext) + rw->www.pqhead = rw->www.pqhead?rw->www.pqhead->qnext:0) { p = rw->www.pqhead; - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 25 Received: from cheviot.ncl.ac.uk by JIMI.CS.UNLV.EDU id aa02158; 23 Aug 95 5:26 PDT Received: from aire.ncl.ac.uk by cheviot.ncl.ac.uk id (8.6.12/ for ncl.ac.uk) with SMTP; Wed, 23 Aug 1995 13:26:28 +0100 Received: (njdc@localhost) by aire.ncl.ac.uk (8.6.7/8.6.x-cf revision 8 for SunOS 4.1.x) id NAA16596; Wed, 23 Aug 1995 13:25:37 +0100 From: "J.D.Coleman" Message-Id: <199508231225.NAA16596@aire.ncl.ac.uk> Subject: Re: Gif colors To: ehcorry@inet.uni-c.dk Date: Wed, 23 Aug 1995 13:25:37 +0100 (BST) Cc: bug-chimera@cs.unlv.edu In-Reply-To: from "Erik Corry" at Aug 23, 95 02:15:15 pm X-Organisation: University of Newcastle Computing Service X-Phone: +44-0191-222-8068 X-Fax: +44-0191-222-8765 X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 651 > It seems to me that the colors of inline GIFs are rather messed up in > the new 2.0 alpha 72. > > Anybody else get this? Is this normal? This is on an 8-bit pseudocolor > screen. Yes, I was going to remark on it. I see the problem where many inlines are displayed on a page and they all have 200+ colours. The later ones tend to get darker then eventually display as all black. Pages with many inlines with few colours (<20) seem to have no problems. I wondered if it is something to do with the colourmap allocation code, but I haven't had a chance to look. J PS. This happens with 2.0a69 too. - -- Core files are like Aero bars, you see. ------- Message 26 Received: from msa.tte.vtt.fi by JIMI.CS.UNLV.EDU id aa03885; 23 Aug 95 6:58 PDT Received: (from msa@localhost) by msa.tte.vtt.fi (8.6.9/8.6.9) id QAA20386; Wed, 23 Aug 1995 16:56:43 +0300 Date: Wed, 23 Aug 1995 16:56:43 +0300 From: Markku Savela Message-Id: <199508231356.QAA20386@msa.tte.vtt.fi> To: bug-chimera@cs.unlv.edu In-reply-to: Erik Corry's message of Wed, 23 Aug 1995 14:15:15 +0100 (MET DST) Subject: Gif colors Reply-to: Markku Savela >It seems to me that the colors of inline GIFs are rather messed up in >the new 2.0 alpha 72. I observed the same. I loaded a page with 4 different GIF images. The first was good, the following increasingly bad. The last one was shown totally black... - -- msa ------- Message 27 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa05430; 23 Aug 95 8:31 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id RAA02247 for cs.unlv.edu!bug-chimera; Wed, 23 Aug 1995 17:35:03 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0slH5g-0003mbC; Wed, 23 Aug 95 16:47 MET DST Message-Id: From: Erik Corry Subject: 2.0 alpha 72 To: bug-chimera@cs.unlv.edu Date: Wed, 23 Aug 1995 16:47:39 +0100 (MET DST) Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 1384 The below may only apply to file: URLs. I seem to be getting local gif's loaded twice under some circumstances (ie my own home page ;-). What is happening is that at line 553 in inline.c in the function HandleImg, the width and height are not available, so rw->www.delay is set and AddImageRenderPart is not called. Then the call sequence goes LoadDocument => LocalProtocol => file => HandleInlineDocument => gifAddData => gs_decode_data => ImageToXImage Here (in inline.c at line 284) AddImageRenderPart is called without resetting rw->www.delay (is that the problem?). Then the call sequence is AddImageRenderPart => AddRenderPart. In AddRenderPart (line 339 in layout.c) the delay flag is checked and found to be set. Therefore => WWWLayout => CallTagHandler => and then HandleImg. The Image gets rendered again. I may of course be misunderstanding it all. But there are certainly two (somewhat William-Gibson-esque renderings) of me on my home page. Of course it's pretty clueless HTML (it worked in N******e :-) so that may be the reason. Here's some cut down code that demonstrates it: Erik's Home Page

Erik Corry

Send me mail mug2.gif is about 55k, which I will send you if necessary. - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 28 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa05434; 23 Aug 95 8:31 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id RAA02249 for cs.unlv.edu!bug-chimera; Wed, 23 Aug 1995 17:35:05 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0slIgD-0003mdC; Wed, 23 Aug 95 18:29 MET DST Message-Id: From: Erik Corry Subject: Re: Gif colors To: bug-chimera@cs.unlv.edu Date: Wed, 23 Aug 1995 18:29:28 +0100 (MET DST) Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2563 > > It seems to me that the colors of inline GIFs are rather messed up in > the new 2.0 alpha 72. > > Anybody else get this? Is this normal? This is on an 8-bit pseudocolor > screen. OK, following up to my own mail (and having J.D.Colemans post and the sources). I think the problem is that the GIFs are always displayed by allocating the colors in the GIF, and then simply mapping the GIF color indeces to the colors reserved on the X server. The problem with this is that it doesn't work when you have many GIFs using different colors, or (as in my case) one GIF that uses more colors than can be allocated from the X server. One solution is to allocate a color cube of 4x4x4 or 5x5x5 colors and then dither to this. This can be easily exended to the case for 15/16/24 bit displays, where you have a 32x32x32 color cube, or more. Another advantage to this is that we could read 24-bit images like jpegs and ppms without first having to convert them to GIFs. The problems with this solution are: * It doesn't look quite so nice for the simple case where Chimera would have been able to allocate all colors with the present system. In this case Chimera presently allows the images to be displayed without any dithering at all. This looks super, especially if the image was not dithered to make it into a GIF in the first place. * We will still need more colors for colored text, buttons etc. * We need to think about handling transparent bits of GIFs. The nicest results on an 8-bit screen are those of xv. When displaying a JPEG, xv will allocate a 6x6x6 cube. This rarely succeeds, but xv can cope with a few holes in the cube, substituting similar colors. xv also has a good algorithm for picking similar colors for a GIF when the ideal colors cannot be allocated. Unfortunately xv is a color hog (as is Chimera at the moment) and it is difficult to display 2 GIFs simultaneously with xv, without the second one suffering. This is not really good for a WWW-browser, since displaying several images simultaneously is the norm. One solution might be to start by allocating a 4x4x4 cube and then use it as a fallback if the colors needed by a GIF cannot be allocated. This might give the good results we are used to for GIFs with low numbers of colors, while offering a safe fallback if it fails. But 4x4x4 looks quite a bit worse than 6x6x6 so maybe it is simpler to always dither. What do you think? I am about to add 15/16/24 bit screen support, but if I do, I might as well do it right. - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 29 Received: from rosebud.turing.toronto.edu by JIMI.CS.UNLV.EDU id aa09674; 23 Aug 95 11:08 PDT Received: by turing.toronto.edu id <5381>; Wed, 23 Aug 1995 14:08:13 -0400 From: Smarasderagd To: bug-chimera@cs.unlv.edu, ehcorry@inet.uni-c.dk Subject: Re: Gif colors Message-Id: <95Aug23.140813edt.5381@turing.toronto.edu> Date: Wed, 23 Aug 1995 14:08:10 -0400 Dithering to a fixed color cube (or color brick -- you get more for your trouble if you use more shades for green than red or blue) is about the simplest way of coping with all the different possible inline images out there. A considerate web page designer will attempt to ensure that all the images on a page fit within a reasonable sized colormap, but JPEG images are a bit of a problem. Providing a resource/option to allocate a private colormap would help to some extent. Short of having chimera do all the dithering / quantizing / squooshing internally (at a probably unacceptable cost in memory/time) I don't have a lot of ideas. Maybe a menu/keyboard action to make chimera try for a better color allotment? ------- Message 30 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa14685; 23 Aug 95 13:56 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id WAA16565 for cs.unlv.edu!bug-chimera; Wed, 23 Aug 1995 22:59:30 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0slNPl-0003mlC; Wed, 23 Aug 95 23:32 MET DST Message-Id: From: Erik Corry Subject: Re: Gif colors To: bug-chimera@cs.unlv.edu Date: Wed, 23 Aug 1995 23:32:49 +0100 (MET DST) In-Reply-To: <95Aug23.140813edt.5381@turing.toronto.edu> from "Smarasderagd" at Aug 23, 95 02:08:10 pm Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2005 > > Dithering to a fixed color cube (or color brick -- you get more for your > trouble if you use more shades for green than red or blue) is about the > simplest way of coping with all the different possible inline images > out there. A considerate web page designer will attempt to ensure that Yes. 4x8x4 (128) looks OK, but has the disadvantage that it contains no grays (apart from white and black). Perhaps 5x7x5 (175) though that is a bit too many colors. I think 5x7x3 (105) is going to be too awful for blue pictures. Actually it's probably worth having 16 gray levels separate from the cube for those rare gray images. Maybe the really sneaky way is to do 4x7x4, (112) but select the blues and reds as 0% 50% 75% 100%. You don't notice the dark bits so much, I think. > all the images on a page fit within a reasonable sized colormap, but JPEG > images are a bit of a problem. Not just JPEG. Actually it's unreasonable to assume you are the only image-displaying program on the screen. > Providing a resource/option to allocate a private colormap would help to > some extent. Short of having chimera do all the dithering / quantizing / > squooshing internally (at a probably unacceptable cost in memory/time) I > don't have a lot of ideas. Maybe a menu/keyboard action to make chimera > try for a better color allotment? I'm not a great fan of private colormaps. Have you tried the visual schnauzer in 'xv'? It is a very disturbing effect, and looks more like a serious bug than anything else. You can't really optimise for all images simultaneously unless you wait for all images to arrive before displaying the first one. I'm still not clear to what extent you have to do that anyway at the moment with Chimera, but generally it is to be avoided, I should have thought. A separate button to optimise? First impressions count. Also, processing time goes unnoticed if you can interleave it with network delays, but is very obvious at other times. - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 31 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa14690; 23 Aug 95 13:56 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id WAA16566 for cs.unlv.edu!bug-chimera; Wed, 23 Aug 1995 22:59:31 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0slNi1-0003mlC; Wed, 23 Aug 95 23:51 MET DST Message-Id: From: Erik Corry Subject: Re: Gif colors To: bug-chimera@cs.unlv.edu Date: Wed, 23 Aug 1995 23:51:40 +0100 (MET DST) In-Reply-To: from "Robert de Bath" at Aug 23, 95 08:44:47 pm Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2308 > > > How about this: > > 1) If the display `truecolour' or greyscale then display to available > colour cube. Yes. A line rather than a cube in the case of the grayscale screen. > 2) If the display is 2-32 pallette use a greyscale. This will probably give very nice pictures, but people like color. Witness the death of the high-quality monochrome monitor. Note that palette screens with less than 256 colors are very rare. And when they appear, they are always 16-color. Of course you may not get all of those colors. > 3) If the display is > 32 colours palletted do this: > > a) Initially display the images using the pallette in the image. > b) If you seriously (rule of thumb :-) run out of colours switch to a fixed > colour cube. > c) When you get some spare cpu time use the `heckbert median cut' to choose > a better pallette (see the pnm utilitys - ppmquant) and possibly redither. Ha! You don't want much, do you. Switching to a color cube means redithering every image currently displayed. That's not going to be easy. Unless you start by allocating the cube, and then ignore it until the crisis appears. This latter option means you get a smaller cube and less palette colors, so you are compromising on both fronts. Probably suggestion (b) gives the best-looking results, but there will be a display glitch. When the crisis appears you start by freeing all the colors you had allocated and then allocate the cube. Now all images will be in false colors. Then you redither, and the images recover one by one. > Obviously (c) is optional :-) Thanks :-) See my other mail. I don't think a global optimisation over all images is feasible. I also think we should try to be nice to the colormap. Remember, it is easy to launch Ghostscript/view, a TeX dvi viewer, xv, a video viewer etc from a WWW browser. If we have a cube allocated there is a large chance that the other programs can share colors. Ghostscript, for example tries to allocate a 5x5x5 cube, then a 4x4x4, 3x3x3 or 2x2x2 cube if that fails. I think xv is similar for JPEGs etc. Certainly it starts by going for 6x6x6. If we use a cuboid like 4x7x4 (or a custom palette) they will not be able to share colors with Chimera. Of course they look bad, not Chimera, but still... - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 32 Received: from fornax.thenet.co.uk by JIMI.CS.UNLV.EDU id aa18414; 23 Aug 95 15:56 PDT Received: from debath.thenet.co.uk by mailhost.thenet.co.uk (8.6.12/FCI Sendmail Config v1.0) id VAA08820; Wed, 23 Aug 1995 21:09:41 GMT Received: from debath.thenet.co.uk by debath.thenet.co.uk with smtp (Smail3.1.29.1 #3) id m0slLjE-0007THC; Wed, 23 Aug 95 20:44 BST Date: Wed, 23 Aug 1995 20:44:47 +0100 (BST) From: Robert de Bath To: Smarasderagd cc: bug-chimera@cs.unlv.edu, ehcorry@inet.uni-c.dk Subject: Re: Gif colors In-Reply-To: <95Aug23.140813edt.5381@turing.toronto.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII How about this: 1) If the display `truecolour' or greyscale then display to available colour cube. 2) If the display is 2-32 pallette use a greyscale. 3) If the display is > 32 colours palletted do this: a) Initially display the images using the pallette in the image. b) If you seriously (rule of thumb :-) run out of colours switch to a fixed colour cube. c) When you get some spare cpu time use the `heckbert median cut' to choose a better pallette (see the pnm utilitys - ppmquant) and possibly redither. Obviously (c) is optional :-) Oh and remember when you choose colours the eye is _far_ more sensitive to variations in green than blue (RGB ratio is 3:6:1 IIRC) - -- Rob. (Robert de Bath ) ------- Message 33 Received: from fornax.thenet.co.uk by JIMI.CS.UNLV.EDU id aa19485; 23 Aug 95 16:38 PDT Received: from debath.thenet.co.uk by mailhost.thenet.co.uk (8.6.12/FCI Sendmail Config v1.0) id AAA10558; Thu, 24 Aug 1995 00:40:27 GMT Received: from debath.thenet.co.uk by debath.thenet.co.uk with smtp (Smail3.1.29.1 #3) id m0slPKw-0007THC; Thu, 24 Aug 95 00:35 BST Date: Thu, 24 Aug 1995 00:35:57 +0100 (BST) From: Robert de Bath To: Erik Corry cc: bug-chimera@cs.unlv.edu Subject: Re: Gif colors In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Wed, 23 Aug 1995, Erik Corry wrote: > This will probably give very nice pictures, but people like color. Auwl right, they can have RGB,CMY too then but that makes it look rather dotty! :-) > Note that palette screens with less than 256 colors are very > rare. And when they appear, they are always 16-color. Of course > you may not get all of those colors. They are rare but I have come across mono and 4 colour before now, not recently tho it's true. > Ha! You don't want much, do you. Switching to a color cube means > redithering every image currently displayed. That's not going to be > easy. Me! All I want is the 64k truecolour to work :-) > Unless you start by allocating the cube, and then ignore it until > the crisis appears. This latter option means you get a smaller cube and > less palette colors, so you are compromising on both fronts. No way I definitly didn't mean that. Humm, thinking about this tho I tend to start a few copies of chimera at the same time. If they didn't have the same CLUT they'd make each other look poor. xv will do that if you tell it to use it's own (or a perfect) colour map, in it's other modes xv uses a 4x8x4 pallette and `normal' colours and there is little `glitching' It looks like it would be possible to, like xv, pick a `perfect' pallette but it would produce a lot a display `glitching' (like xv can!). I suppose you could arrange to have 2 copies of each image (by the time the document is complete) one to (say) 4x4x4 and the other to a calculated pallette _including_ the 4x4x4. When you have the focus you use the images for the generated pallette otherwise you use the 4x4x4 images. It _should_ be possible to get good images (somewhat poorer when you loose the focus) without clobbering everyone else too badly and be free of glitches. You would still have to redither the image at some point tho. BUT like I said, as long a truecolour 32x64x32 (and 256x256x256) works _I'll_ be happy :-) - -- Rob. (Robert de Bath ) ------- Message 34 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa03642; 24 Aug 95 3:50 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id MAA25440 for cs.unlv.edu!bug-chimera; Thu, 24 Aug 1995 12:53:40 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0slZow-0003mJC; Thu, 24 Aug 95 12:47 MET DST Message-Id: From: Erik Corry Subject: Blockquote To: bug-chimera@cs.unlv.edu Date: Thu, 24 Aug 1995 12:47:37 +0100 (MET DST) Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 97 Presumably you know that blockquote doesn't work right yet. - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 35 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa03650; 24 Aug 95 3:51 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id MAA25441 for cs.unlv.edu!bug-chimera; Thu, 24 Aug 1995 12:53:40 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0slakF-0003mJC; Thu, 24 Aug 95 13:46 MET DST Message-Id: From: Erik Corry Subject: Proxies in 2.0 alpha 72 To: bug-chimera@cs.unlv.edu Date: Thu, 24 Aug 1995 13:46:50 +0100 (MET DST) Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 2772 I changed a few things in the no_proxy handling. I think the patches are self-explanatory. Problems solved are: 1) file: URLs should not be proxied 2) no_proxy defaulted to nothing instead of 'localhost', which is more useful 3) no_proxy matched too eagerly (see comment) 4) a null no_proxy string resulted in the proxy never being used Problems that remain are: 5) Error messages from (at least) the CERN proxy server are not displayed. 6) A value for xxx_proxy of hostname.domain is not automatically expanded to https://hostname.domain/ which would be nice. A question: 7) Is it a good idea to match all_proxy for all protocols? Are there any proxy servers that can handle nntp/news requests? Perhaps all_proxy should stick to the http/wais/gopher/ftp protocols. - --- cfh72-2.0/wwwio/document.c Sun Aug 20 09:38:10 1995 +++ cfh72-2.0-erik/wwwio/document.c Thu Aug 24 13:27:38 1995 @@ -58,6 +58,14 @@ char *pname; /* + * file: URLs are never put through a proxy. + */ + if(strcasecmp(up->protocol, "file") == 0) + { + return(0); + } + + /* * Hee, allow any weird protocol to be checked for proxy. */ pname = (char *)alloc_mem(strlen(up->protocol) + strlen("_proxy") + 1); @@ -84,21 +92,36 @@ /* * Check for domains that are not allowed for proxy + * + * Default to not proxying localhost. A null no_proxy string + * overrides this. TODO: get canonical hostname of current + * host and exclude this by default, too. */ - - if ((no_proxy = getenv("no_proxy")) != NULL - - || (no_proxy = NGetFromStringDB("no_proxy")) != NULL) + if ((no_proxy = getenv("no_proxy")) == NULL + && (no_proxy = NGetFromStringDB("no_proxy")) == NULL) + { + no_proxy = "localhost"; + } + + hlen = strlen(up->hostname); + cp = no_proxy; + while ((ph = mystrtok(cp, ',', &cp)) != NULL) { - - hlen = strlen(up->hostname); - - cp = no_proxy; - - while ((ph = mystrtok(cp, ',', &cp)) != NULL) + /* + * Check whether the no_proxy variable either matches the + * hostname exactly, or has the same ending. The ending must + * be a domain name, not part of a domain name ie. no_proxy= + * era.org does not match hostname=www.chimera.org. + */ + plen = strlen(ph); + if (plen != 0 && + plen <= hlen && + strcasecmp(ph, up->hostname + (hlen - plen)) == 0 && + (plen == hlen || up->hostname[hlen - plen - 1] == '.' + || up->hostname[hlen - plen] == '.')) { - - plen = strlen(ph); - - if (plen <= hlen && - - strncasecmp(ph, up->hostname + (hlen - plen), plen) == 0) - - { - - free_mem(pproxy); - - return(0); - - } + free_mem(pproxy); + return(0); } } - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 36 Received: from eikon.regent.e-technik.tu-muenchen.de by JIMI.CS.UNLV.EDU id aa15040; 24 Aug 95 12:46 PDT Received: from vector.eikon.e-technik.tu-muenchen.de (vector.eikon.e-technik.tu-muenchen.de [129.187.142.36]) by eikon.e-technik.tu-muenchen.de (8.6.12/8.6.9) with ESMTP id VAA03325 for ; Thu, 24 Aug 1995 21:45:39 +0200 Received: (from jhs@localhost) by vector.eikon.e-technik.tu-muenchen.de (8.6.11/8.6.9) id QAA03441 for bug-chimera@cs.unlv.edu; Thu, 24 Aug 1995 16:56:10 +0200 Date: Thu, 24 Aug 1995 16:56:10 +0200 From: Julian Howard Stacey Message-Id: <199508241456.QAA03441@vector.eikon.e-technik.tu-muenchen.de> To: bug-chimera@cs.unlv.edu Subject: Hint solicited on browser detection. I crave your indulgence for this non Chimera specific Q: I'd like to write some cgi code to detect browser type, & present different pages accordingly. I would appreciate `RTFM' statements with URLs for where I can fetch appropriate docs or examples to read. Thanks for any hints :-) Julian S jhs@freebsd.org ------- Message 37 Received: from rosebud.turing.toronto.edu by JIMI.CS.UNLV.EDU id aa19549; 24 Aug 95 15:10 PDT Received: by turing.toronto.edu id <5380>; Thu, 24 Aug 1995 18:10:44 -0400 From: Smarasderagd To: bug-chimera@cs.unlv.edu Subject: dumb question re image gamma Message-Id: <95Aug24.181044edt.5380@turing.toronto.edu> Date: Thu, 24 Aug 1995 18:10:29 -0400 Do html 2.0 or 3.0 support some sort of gamma hint for inline images? It seems to me this would be super helpful in cases where the image format doesn't provide a gamma value, or it does but the gamma is wrong. ------- Message 38 Received: from mail.cs.tu-berlin.de by JIMI.CS.UNLV.EDU id aa24339; 24 Aug 95 17:58 PDT Received: from titanic.cs.tu-berlin.de (czyborra@titanic.cs.tu-berlin.de [130.149.18.9]) by mail.cs.tu-berlin.de (8.6.12/8.6.12) with ESMTP id CAA10984; Fri, 25 Aug 1995 02:58:27 +0200 Received: (czyborra@localhost) by titanic.cs.tu-berlin.de (8.6.12/8.6.6) id CAA28963; Fri, 25 Aug 1995 02:58:25 +0200 From: Roman Czyborra To: Julian Stacey Cc: bug-chimera@cs.unlv.edu Subject: Re: Hint solicited on browser detection. In-Reply-To: <199508241456.QAA03441@vector.eikon.e-technik.tu-muenchen.de> by jhs@vector.eikon.e-technik.tu-muenchen.de dated 1995-8-24 16:56:10 MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Date: Fri, 25 Aug 1995 02:58:24 +0200 Message-ID: Hello Julian, > I'd like to write some cgi code to detect browser type, HTTP_USER_AGENT holds the identification sent by the browser. > & present different pages accordingly. Note that your output may get cached and picked up by another browser. > I would appreciate `RTFM' statements with URLs for where I can fetch > appropriate docs or examples to read. Sample scripts that change behavior for certain HTTP_USER_AGENTs are https://www.cs.tu-berlin.de/cgi/launch https://www.cs.tu-berlin.de/cgi/gnuinfo https://www.cs.tu-berlin.de/cgi/sendmail ------- Message 39 Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa27762; 24 Aug 95 21:19 PDT Received: from fsb1.aist-nara.ac.jp by JIMI.CS.UNLV.EDU id aa27692; 24 Aug 95 21:10 PDT Received: from bls6.aist-nara.ac.jp by mailgate.aist-nara.ac.jp (8.6.10+2.5Wb1/2.8Wb/NAIST-1.6[gate]) id NAA19713; Fri, 25 Aug 1995 13:10:30 +0900 Return-Path: Received: from localhost by bls6.aist-nara.ac.jp (5.67+1.6W[kuis-17]/2.8Wb/NAIST-1.3[is]) id AA27762; Fri, 25 Aug 95 13:10:23 GMT+0900 Message-Id: <9508250410.AA27762@bls6.aist-nara.ac.jp> To: chimera-announce@cephas.ISRI.UNLV.EDU Subject: Re: 2.0 alpha 72 Reply-To: k-chinen@is.aist-nara.ac.jp In-Reply-To: Your message of "Tue, 22 Aug 1995 23:47:19 -0700" References: <199508230723.QAA03155@mailgate.aist-nara.ac.jp> X-Mailer: Mew beta version 0.98 on Emacs 18.59.1, Mule 1.1 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Fri, 25 Aug 1995 13:10:21 +0900 From: Kenichi Chinen cfh72-2.0 is very nice. But I meet little problem when compile cfh72-2.0 on ULTRIX. I cannot compile wwwio/local.h . Because struct-stat is not defined. Please insert 3 lines in wwwio/local.c . #ifdef HAVE_SYS_TYPES_H #include #endif So, If you have sys/types.h, define HAVE_SYS_TYPES_H in Common.tmpl . ------- Message 40 Received: from little-charlie.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa28043; 24 Aug 95 21:33 PDT To: Smarasderagd cc: bug-chimera@cs.unlv.edu Subject: Re: dumb question re image gamma In-reply-to: Your message of "Thu, 24 Aug 1995 18:10:29 EDT." <95Aug24.181044edt.5380@turing.toronto.edu> Date: Thu, 24 Aug 1995 21:33:12 -0700 From: John Kilburg >Do html 2.0 or 3.0 support some sort of gamma hint for inline images? >It seems to me this would be super helpful in cases where the image format >doesn't provide a gamma value, or it does but the gamma is wrong. The last time I checked there wasn't a hint for this in 2 or 3. -john ------- Message 41 Received: from little-charlie.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa28163; 24 Aug 95 21:44 PDT To: Erik Corry cc: bug-chimera@cs.unlv.edu Subject: Re: Blockquote In-reply-to: Your message of "Thu, 24 Aug 1995 12:47:37 BST." Date: Thu, 24 Aug 1995 21:44:43 -0700 From: John Kilburg >Presumably you know that blockquote doesn't work right yet. What's wrong with them? I wouldn't be surprised if there is a bug but I just tried out my copy and it worked...I can't recall making changes to the blockquote stuff since 72. Send me a URL when you get a chance. -john ------- Message 42 Received: from little-charlie.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa29137; 24 Aug 95 22:45 PDT To: Erik Corry cc: bug-chimera@cs.unlv.edu Subject: Re: Proxies in 2.0 alpha 72 In-reply-to: Your message of "Thu, 24 Aug 1995 13:46:50 BST." Date: Thu, 24 Aug 1995 22:45:42 -0700 From: John Kilburg Thanks for the patch. >5) Error messages from (at least) the CERN proxy server are not displayed. There is a general lack of error reporting...I'll hopefully start working on this soon. >7) Is it a good idea to match all_proxy for all protocols? Are there > any proxy servers that can handle nntp/news requests? Perhaps all_proxy > should stick to the http/wais/gopher/ftp protocols. I think this is OK because if the exceptions are few you can override all_proxy with a specific xxx_proxy. If the exceptions are many then all_proxy probably shouldn't be used anyways. Leaving it as is ensures that all_proxy will work for new-fangled protocols in the future without having to change the code to allow more than http/wais/gopher/ftp. -john ------- Message 43 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa08248; 25 Aug 95 6:34 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id PAA04749; Fri, 25 Aug 1995 15:36:14 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0slxpC-0003mHC; Fri, 25 Aug 95 14:25 MET DST Message-Id: From: Erik Corry Subject: Re: Proxies in 2.0 alpha 72 To: John Kilburg Date: Fri, 25 Aug 1995 14:25:30 +0100 (MET DST) Cc: bug-chimera@cs.unlv.edu In-Reply-To: <199508251134.NAA07398@inet.uni-c.dk> from "John Kilburg" at Aug 24, 95 10:45:42 pm Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 878 John: > > >7) Is it a good idea to match all_proxy for all protocols? Are there > > any proxy servers that can handle nntp/news requests? Perhaps all_proxy > > should stick to the http/wais/gopher/ftp protocols. > > I think this is OK because if the exceptions are few you can override > all_proxy with a specific xxx_proxy. If the exceptions are many then > all_proxy probably shouldn't be used anyways. Leaving it as is ensures > that all_proxy will work for new-fangled protocols in the future without > having to change the code to allow more than http/wais/gopher/ftp. Is there a way to switch off all proxying for a certain protocol? I may not want to set the proxy to something else, just switch it off. If you remove the eager all_proxy you could still set blobp_proxy for the new blob protocol without source code changes. - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 44 Received: from LINC.CIS.UPENN.EDU by JIMI.CS.UNLV.EDU id aa10891; 25 Aug 95 8:47 PDT Received: (anoop@localhost) by linc.cis.upenn.edu (8.6.12/UPenn 1.4) id LAA28129; Fri, 25 Aug 1995 11:47:09 -0400 Date: Fri, 25 Aug 1995 11:47:09 -0400 From: Anoop Sarkar Posted-Date: Fri, 25 Aug 1995 11:47:09 -0400 Message-Id: <199508251547.LAA28129@linc.cis.upenn.edu> To: bug-chimera@cs.unlv.edu Subject: Strange href behaviour Reply-to: anoop@linc.cis.upenn.edu Chimera-2.0 seems to show erratic behaviour when faced with a href inside a header. Have a look at https://www.cis.upenn.edu/~anoop/ where although each section has identical code in the third section the href spills over and links/underlines the entire text after it. The page works just fine with Chimera-1.65. I can't seem to figure out why. - -Anoop ------- Message 45 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa23543; 25 Aug 95 16:46 PDT To: Erik Corry cc: bug-chimera@cs.unlv.edu Subject: Re: Proxies in 2.0 alpha 72 In-reply-to: Your message of "Fri, 25 Aug 1995 14:25:30 BST." Date: Fri, 25 Aug 1995 16:46:45 -0700 From: John Kilburg >> >7) Is it a good idea to match all_proxy for all protocols? Are there >> > any proxy servers that can handle nntp/news requests? Perhaps all_proxy >> > should stick to the http/wais/gopher/ftp protocols. >> >> I think this is OK because if the exceptions are few you can override >> all_proxy with a specific xxx_proxy. If the exceptions are many then >> all_proxy probably shouldn't be used anyways. Leaving it as is ensures >> that all_proxy will work for new-fangled protocols in the future without >> having to change the code to allow more than http/wais/gopher/ftp. > >Is there a way to switch off all proxying for a certain protocol? No, unfortunately. >I may not want to set the proxy to something else, just switch it >off. If you remove the eager all_proxy you could still set blobp_proxy >for the new blob protocol without source code changes. Someone on the list asked for all_proxy a while back and I would rather not change it without hearing from them. Maybe something like export nntp_proxy="no" would be a good idea for disabling all proxying for a particular protocol? -john ------- Message 46 Received: from ucsub.Colorado.EDU by JIMI.CS.UNLV.EDU id aa24094; 25 Aug 95 17:08 PDT Received: (from johnsoma@localhost) by ucsub.colorado.edu (8.6.12/8.6.12/CNS-3.6) id SAA00872; Fri, 25 Aug 1995 18:08:23 -0600 Date: Fri, 25 Aug 1995 18:08:23 -0600 (MDT) From: JOHNSON MICHAEL ALLEN To: bug-chimera@cs.unlv.edu Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII unsubscribe ------- Message 47 Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa29902; 25 Aug 95 23:18 PDT Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa29872; 25 Aug 95 22:51 PDT To: chimera-announce@cephas.ISRI.UNLV.EDU Subject: alpha 73 Date: Fri, 25 Aug 1995 22:50:39 -0700 From: John Kilburg Just released another one with minor fixes. Includes a change list so you can decide whether or not you should grab it. https://www.isri.unlv.edu/~john/chimera/src/ -john ------- Message 48 Received: from aohakobe.ipc.chiba-u.ac.jp by JIMI.CS.UNLV.EDU id aa04787; 26 Aug 95 2:37 PDT Received: from aohakobe (localhost [127.0.0.1]) by aohakobe.ipc.chiba-u.ac.jp (8.6.12/3.4Wbeta3 (-: 95041617) with ESMTP id SAA08602 for ; Sat, 26 Aug 1995 18:37:58 +0900 Message-Id: <199508260937.SAA08602@aohakobe.ipc.chiba-u.ac.jp> To: bug-chimera@cs.unlv.edu Subject: Re: alpha 73 In-reply-to: Your message of "Fri, 25 Aug 1995 22:50:39 MST." <9508260756.AA15703@cuipc.ipc.chiba-u.ac.jp> Date: Sat, 26 Aug 1995 18:37:58 +0900 From: Yozo Toda (TELEPHONE +81-43-290-3539) here is my note when compiling cfh73 on Solaris2.4. wwwio/pipe.c: I faced the following error. %%%% %%%% %%%% %%%% gcc -fpcc-struct-return -g -Wall -I../ -I../common -I/opt/X11R6/include -Dsun -DSVR4 -DX_LOCALE -DHAVE_MKTIME -DHAVE_STDLIB_H -DHAVE_STRING_H -DHAVE_UNISTD_H -DHAVE_MALLOC_H -DUSER_AGENT=\"Chimera/2.0alpha\" -c pipe.c pipe.c: In function `PipeCommand': pipe.c:103: `FNDELAY' undeclared (first use this function) pipe.c:103: (Each undeclared identifier is reported only once pipe.c:103: for each function it appears in.) *** Error code 1 make: Fatal error: Command failed for target `pipe.o' Current working directory /export1/local/src/chimera/cfh73-2.0/wwwio %%%% %%%% %%%% %%%% wwwio/net.c also contains FNDELAY, but no errors. I think the following three lines should be inserted to wwwio/pipe.c (under "#include "?) %%%% %%%% %%%% %%%% #if ((defined(SYSV) || defined(SVR4)) && defined(sun)) #include #endif %%%% %%%% %%%% %%%% src/Imakefile: I defined CEXTRA_LIBS in Common.tmpl (for socket library), but linking failed. the definition of EXTRA_LIBRARIES in src/Imakefile must contain $(CEXTRA_LIBS). Common.tmpl: seeing "SHM_STUFF = -DHAVE_SHM", I suppose $(HAVE_SHM) in the definition of CEXTRA_DEFINES should be $(SHM_STUFF)? but I find there is no place checking if HAVE_SHM is defined!! - -- yozo. ------- Message 49 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa14941; 26 Aug 95 17:21 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id CAA11736; Sun, 27 Aug 1995 02:26:28 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0smLgP-0003mJC; Sat, 26 Aug 95 15:54 MET DST Message-Id: From: Erik Corry Subject: The protocols file To: John Kilburg Date: Sat, 26 Aug 1995 15:54:00 +0100 (MET DST) Cc: bug-chimera@cs.unlv.edu In-Reply-To: <199508230926.LAA09947@inet.uni-c.dk> from "John Kilburg" at Aug 22, 95 11:47:19 pm Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 136 Is it just me setting it up wrongly, or has support for the protocols file been removed in 2.0 72? - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 50 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa04426; 27 Aug 95 14:50 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id XAA06232; Sun, 27 Aug 1995 23:52:28 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0smpSm-0003mHC; Sun, 27 Aug 95 23:41 MET DST Message-Id: From: Erik Corry Subject: Re: alpha 73 To: John Kilburg Date: Sun, 27 Aug 1995 23:41:56 +0100 (MET DST) Cc: bug-chimera@cs.unlv.edu In-Reply-To: <199508260712.JAA18919@inet.uni-c.dk> from "John Kilburg" at Aug 25, 95 10:50:39 pm Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 200 The error messages from the proxy daemon are displayed very nicely now. But invalid protocol names still don't produce an error message. Even without all_proxy. - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 51 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa12730; 27 Aug 95 23:22 PDT To: yozo@aohakobe.ipc.chiba-u.ac.jp cc: bug-chimera@cs.unlv.edu Subject: Re: alpha 73 In-reply-to: Your message of "Sat, 26 Aug 1995 18:37:58 +0900." <199508260937.SAA08602@aohakobe.ipc.chiba-u.ac.jp> Date: Sun, 27 Aug 1995 23:22:33 -0700 From: John Kilburg Thanks for your fixes. I'll include them. >Common.tmpl: >seeing "SHM_STUFF = -DHAVE_SHM", I suppose >$(HAVE_SHM) in the definition of CEXTRA_DEFINES should be $(SHM_STUFF)? >but I find there is no place checking if HAVE_SHM is defined!! Well, I was messing around with using the shm extension for XImage but I never actually got around to doing it. I will remove HAVE_SHM until it is actually needed. Thanks. -john ------- Message 52 Received: from LINC.CIS.UPENN.EDU by JIMI.CS.UNLV.EDU id aa24301; 28 Aug 95 8:39 PDT Received: (anoop@localhost) by linc.cis.upenn.edu (8.6.12/UPenn 1.4) id LAA21274; Mon, 28 Aug 1995 11:39:39 -0400 Date: Mon, 28 Aug 1995 11:39:39 -0400 From: Anoop Sarkar Posted-Date: Mon, 28 Aug 1995 11:39:39 -0400 Message-Id: <199508281539.LAA21274@linc.cis.upenn.edu> To: john@cephas.ISRI.UNLV.EDU CC: bug-chimera@cs.unlv.edu In-reply-to: <199508280650.CAA04571@linc.cis.upenn.edu> (message from John Kilburg on Sun, 27 Aug 1995 23:22:33 -0700) Subject: Re: alpha 73 Reply-to: anoop@linc.cis.upenn.edu The new release is easier to handle. This might be useful to document in the hints file: Under Solaris you need to add -lsocket -lnsl for getsockname, etc. to work. Also under my Solaris setup FNDELAY exists in sys/file.h I can't get X11 bitmaps to be rendered as inline images in 2.0. I notice that you've changed inline.c considerably and it no longer calls seperate routines like "Image *xbitmapLoad" anymore as in ver 1.65. I'm not sure if the problem is in the inline.c code however. Any ideas? - -Anoop ------- Message 53 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa04711; 28 Aug 95 14:03 PDT To: anoop@linc.cis.upenn.edu cc: bug-chimera@cs.unlv.edu Subject: Re: alpha 73 In-reply-to: Your message of "Mon, 28 Aug 1995 11:39:39 EDT." <199508281539.LAA21274@linc.cis.upenn.edu> Date: Mon, 28 Aug 1995 14:03:38 -0700 From: John Kilburg >The new release is easier to handle. This might be useful to document >in the hints file: > >Under Solaris you need to add -lsocket -lnsl for getsockname, etc. to >work. Also under my Solaris setup FNDELAY exists in sys/file.h > >I can't get X11 bitmaps to be rendered as inline images in 2.0. I >notice that you've changed inline.c considerably and it no longer >calls seperate routines like "Image *xbitmapLoad" anymore as in ver >1.65. I'm not sure if the problem is in the inline.c code however. Any >ideas? XBM isn't supported yet. Only GIFs are supported in 2.0 right now. Other formats will be added as soon as the basics are reasonably stable. Thanks for the Solaris tips. -john ------- Message 54 Received: from aohakobe.ipc.chiba-u.ac.jp by JIMI.CS.UNLV.EDU id aa16314; 28 Aug 95 19:55 PDT Received: from aohakobe (localhost [127.0.0.1]) by aohakobe.ipc.chiba-u.ac.jp (8.6.12/3.4Wbeta3 (-: 95041617) with ESMTP id LAA13283 for ; Tue, 29 Aug 1995 11:55:53 +0900 Message-Id: <199508290255.LAA13283@aohakobe.ipc.chiba-u.ac.jp> To: bug-chimera@cs.unlv.edu Subject: Re: alpha 73 In-reply-to: Your message of "Mon, 28 Aug 1995 11:39:39 -0400." <199508281539.LAA21274@linc.cis.upenn.edu> Date: Tue, 29 Aug 1995 11:55:53 +0900 From: Yozo Toda (TELEPHONE +81-43-290-3539) > Under Solaris you need to add -lsocket -lnsl for getsockname, etc. to > work. Also under my Solaris setup FNDELAY exists in sys/file.h ^^^^^^^^^^ Now I check Solaris2.1, 2.3, 2.4. FNDELAY is defined in sys/file.h of all three version. maybe O_NDELAY is available on SunOS4 too, but I don't know which is preferrable... - -- yozo. ------- Message 55 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa23014; 29 Aug 95 16:59 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id CAA12851; Wed, 30 Aug 1995 02:03:09 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0snW2n-0003mHC; Tue, 29 Aug 95 21:09 MET DST Message-Id: From: Erik Corry Subject: ANSI C To: john@cephas.ISRI.UNLV.EDU Date: Tue, 29 Aug 1995 21:09:57 +0100 (MET DST) Cc: bug-chimera@cs.unlv.edu Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 322 How important is it to remain K&R C compatible? I'm wondering whether it is a good idea to write new stuff with K&R-style parameters: void WWWHome(w) Widget w; { ... } Surely there are more compilers that are going to complain about that than that really need that. Who needs K&R? - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 56 Received: from rosebud.turing.toronto.edu by JIMI.CS.UNLV.EDU id aa06123; 29 Aug 95 23:12 PDT Received: by turing.toronto.edu id <5379>; Wed, 30 Aug 1995 02:12:39 -0400 From: Smarasderagd To: bug-chimera@cs.unlv.edu Subject: odd problem in 2.0 Message-Id: <95Aug30.021239edt.5379@turing.toronto.edu> Date: Wed, 30 Aug 1995 02:12:30 -0400 I'm experiencing a very odd problem with chimera 2.0. For my window manager I use 9wm, which attempts to duplicate some of the behaviour of 8-1/2, the plan 9 windowing system. When I run chimera 2.0 (any version), mouse clicks in the html window don't do anything. If I kill 9wm, it works. If I run twm instead, it works. I didn't experience this problem with chimera 1.65. I have the same problem under Linux. I'm well and truly mystified. The current 9wm code can be found at ftp://ftp.cs.su.oz.au/dhog/9wm. ------- Message 57 Received: from recycle.cebaf.gov by JIMI.CS.UNLV.EDU id aa04336; 31 Aug 95 9:50 PDT From doolitt Thu Aug 31 12:49:51 1995 remote from recycle Received: by recycle.cebaf.gov.cebaf.gov (Linux Smail3.1.28.1 #14) id m0soCo4-000ASjC; Thu, 31 Aug 95 12:49 EDT Message-Id: From: Larry Doolittle Subject: weird stuff To: bug-chimera@cs.unlv.edu Date: Thu, 31 Aug 1995 12:49:35 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 457 Hello - cfh69, 72, and 73 all crash for me (SIGSEGV dereferencing p->id in WWWLayout.c) when I load a page with an inline image, but only if they come from file://localhost/. The same page will display fine if it comes from https://myhost.domain/. I have started looking into it, but I don't have an answer yet. If anybody finds a fix, let me know - and please use direct e-mail, I don't know if I am correctly set up on this list. - Larry Doolittle ------- Message 58 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa11492; 31 Aug 95 13:21 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id WAA28067 for cs.unlv.edu!bug-chimera; Thu, 31 Aug 1995 22:29:28 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0soBoG-0003mBC; Thu, 31 Aug 95 17:45 MET DST Message-Id: From: Erik Corry Subject: Home page To: bug-chimera@cs.unlv.edu Date: Thu, 31 Aug 1995 17:45:44 +0100 (MET DST) Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 606 Doesn't someone want to design a really cool home page for Chimera, to be distributed in the lib directory. I think a really natty initial screen would really help Chimera look good. I think you could justify up to about 100k, since it's all local. It should probably include all the info that is currently on the home page. I'd like to do it, but I don't know what a chimera even looks like. :-) Maybe we should organise a competition. Could you put up a free copy of Chimera 2.0 version 74 as a prize for the best page, John? With a cybersurf to Las Vegas for two? - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 59 Received: from bock.freinet.de by JIMI.CS.UNLV.EDU id aa11515; 31 Aug 95 13:21 PDT Received: from kroete2.freinet.de (kroete2@localhost) by bock.freinet.de (8.6.12/8.6.12) with UUCP id WAA28065; Thu, 31 Aug 1995 22:29:25 +0200 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0soAdZ-0003mAC; Thu, 31 Aug 95 16:30 MET DST Message-Id: From: Erik Corry Subject: HTML glitch To: John Kilburg Date: Thu, 31 Aug 1995 16:30:34 +0100 (MET DST) Cc: bug-chimera@cs.unlv.edu In-Reply-To: <199508310317.FAA17502@inet.uni-c.dk> from "John Kilburg" at Aug 30, 95 08:09:35 pm Reply-To: Erik Corry X-Mailer: ELM [version 2.4 PL22] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 231 The URL: https://www.w3.org/hypertext/WWW/MarkUp/html3/CoverPage.html is supposed to have a hyperlink at the end, which doesn't show up in Chimera 2.0 73. It was still there in Chimera 2.0 72. - -- Erik Corry ehcorry@inet.uni-c.dk ------- End of Forwarded Messages