Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa01860; 26 Dec 96 23:05 PST To: jay@JIMI.CS.UNLV.EDU Subject: bug-chimera feb 96 Date: Thu, 26 Dec 1996 23:05:36 -0800 From: Jay Nietling ------- Forwarded Messages Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa19367; 31 Jan 96 16:43 PST Received: from kroete2.freinet.de (arhppp40.uni-c.dk [130.225.243.202]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id BAA01128 for ; Thu, 1 Feb 1996 01:40:33 +0100 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0thmpp-0003mBC; Thu, 1 Feb 96 01:25 MET Message-Id: From: Erik Corry Subject: Re: bgcolor To: Chimera Hacker List Date: Thu, 1 Feb 1996 01:25:08 +0100 (MET) In-Reply-To: <4492.9601311434@subnode.aiai.ed.ac.uk> from "Tim Bradshaw" at Jan 31, 96 02:34:38 pm Reply-To: Erik Corry X-Face: 5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS Can I vote for `netscapisms if there is an SGML DTD that could support > them' and preferably for HTML3isms? For people like me who use an > SGML-based editor (psgml for emacs) to write html, the awful random > stuff that netscape puts in are really a pain. Not to mention most of > them are just spurious prettiness. Chimera supporting Netscapisms doesn't stop you using an SGML DTD to write your own pages. If you don't want to use them, then don't. Much as I dislike the things Netscape have done to HTML, we do need to be a little pragmatic about it: Some of these extensions have become all but ubiquitous, and Chimera wins little by being pedantic about it. The simple pleasure of having discovered yet another page that doesn't stick to the standard is soon overshadowed by the irritation when yet another page looks completely different to the way it was intended. - -- 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 aa19373; 31 Jan 96 16:43 PST Received: from kroete2.freinet.de (arhppp40.uni-c.dk [130.225.243.202]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id BAA01137 for ; Thu, 1 Feb 1996 01:40:36 +0100 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0thn4r-0003maC; Thu, 1 Feb 96 01:40 MET Message-Id: From: Erik Corry Subject: Re: Chimera and accents. To: Chimera Hacker List Date: Thu, 1 Feb 1996 01:40:41 +0100 (MET) In-Reply-To: <199601311657.RAA26886@fantomas.inria.fr> from "Jean-Marc Lasgouttes" at Jan 31, 96 05:57:04 pm Reply-To: Erik Corry X-Face: 5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS A note about the latest chimera alpha: some accents fail to be > rendered correctly, in particular è, which is *very* useful in > french. Others like é work correctly. I had a quick look in > html/html.c, but I do not see why it does not work. How are they failing? If the accented letters are just rendered as spaces, then it is probably a bug in the isspace() function, which is supposed to tell whether a given character is a space character. This function often has bugs when used with 8-bit characters. In particular some Linux libcs have this problem. Actually it is probably a mistake to use this function at all. The function is localised, which means it works according to the character set associated with the current locale. The locale is usually set with the environment variable LANG. The problem with this approach is that what Chimera needs is globalisation, not localisation. That means that the document should not be rendered using the local character set, but rather using the character set that the document was written for. The charset of the document is determined by the charset= parameter to the Content-Type returned by the server. For example: Content-Type: text/html; charset=ISO-8859-1 means it is an HTML doc written with Latin-1 (Western European) letters. If no charset is specified, then ISO-8859-1 is assumed. If a character is explicitly specified with é or similar, then this can provide a way to specify characters not in the document charset. If you use numeric references, like ) (I'm not sure of the syntax here) then a character from ISO-8859-1 is always meant, regardless of the document charset from the header. There is a suggested amendment to HTML3 which allows 4-digit characters from the Unicode character set. This is upwards compatible because the first 256 Unicode characters are the same as the 256 ISO-8859-1 characters. The upshot of all this is that Chimera should be using the document charset to decide what is a space and what isn't, and not the locale-dependent functions like isspace. In the long run, we should convert all input characters to an internal representation, and then write our own isspace that can handle the characters in that representation. As far as I can see, Unicode is the only reasonable choice for that internal representation. The short term solution, since several people seem to be being hit with this problem, is to write our own isspace routine that can cope with ISO 8859-1 characters. If anyone does this they should most likely use a table-based solution for speed reasons and make sure machines where 'char' is a signed type are handled correctly. Other routines than isspace may be needed, too. - -- Erik Corry ehcorry@inet.uni-c.dk ------- Message 3 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa21896; 31 Jan 96 18:44 PST Received: from citi.umich.edu by citi.umich.edu for bug-chimera@cs.unlv.edu with SMTP; Wed, 31 Jan 96 21:43:58 -0500 From: Jim Rees To: Chimera Lovers Date: Wed, 31 Jan 1996 21:43:57 -0500 Subject: Re: Chimera and accents. Sender: rees@citi.umich.edu In-Reply-To: Erik Corry, Thu, 01 Feb 1996 01:40:41 +0100 The charset of the document is determined by the charset= parameter to the Content-Type returned by the server. That's what the spec says, but I have yet to see any web pages that actually do this. There doesn't seem to be any standard for non-8859-1 pages. Usually they just use some other character set, with no indication as to which set is in use. If you know of counter-examples, I'd like to see them. But chimera certainly should be able to render 8859-1 correctly, since that's the default. Chimera only knows about a very small number of the '&' escapes. The complete set is found in ISO 8879. A useful reference is the ISO Mapping Tables, section 6.1 of Unicode 1.0 volume 1 (yes I know it's obsolete, it was on sale, $20 for both volumes). I've considered adding the rest, but since the escapes are just a crutch for people hobbled by inadequate text editors, it didn't seem worth the effort. But it would be a useful thing for someone to do. ------- Message 4 Received: from nag.cs.Colorado.EDU by JIMI.CS.UNLV.EDU id aa22801; 31 Jan 96 19:38 PST Received: from nag.cs.colorado.edu (localhost.cs.colorado.edu [127.0.0.1]) by nag.cs.colorado.edu (8.6.10/8.6.10) with ESMTP id UAA13823; Wed, 31 Jan 1996 20:35:21 -0700 Message-Id: <199602010335.UAA13823@nag.cs.colorado.edu> To: Jim Rees cc: Chimera Lovers , popiel@nag.cs.colorado.edu Subject: Re: Chimera and accents. In-reply-to: Your message of "Wed, 31 Jan 1996 21:43:57 MST." <199602010309.UAA12156@nag.cs.colorado.edu> Date: Wed, 31 Jan 1996 20:35:13 -0700 From: "T. Alexander Popiel" In message: <199602010309.UAA12156@nag.cs.colorado.edu> Jim Rees writes: > >Chimera only knows about a very small number of the '&' escapes. The >complete set is found in ISO 8879. A useful reference is the ISO Mapping >Tables, section 6.1 of Unicode 1.0 volume 1 (yes I know it's obsolete, it >was on sale, $20 for both volumes). I've considered adding the rest, but >since the escapes are just a crutch for people hobbled by inadequate text >editors, it didn't seem worth the effort. But it would be a useful thing >for someone to do. I don't think the escapes are just a crutch. Some of us, for various reasons of our own, prefer to keep our documents 7-bit clean ASCII (Codes 0x1b to 0x7e (ESC to '~'), 0x0a (LF), 0x0d (CR), and 0x09 (TAB); 0x0a and 0x0d are always used as end-of-line, and 0x09 is always used as whitespace). Some of the reasons include: In source form, - - they tend to print better, when looked at with generic tools. - - they tend to look the same, regardless of which tools are in use. In transmission, - - they don't suffer as much from translation to and from arbitrary encodings like EBCDIC. - - they don't get messed up by LF/CRLF translation automagically done by many MSLOSS machines. - - they don't trip over people's XON/XOFF codes (or their equivalents). In interpretation, - - they tend to remain comprehensible even if they are not fully recognized. Please don't discount users of the old tech, merely because they haven't upgraded to the boated nightmares which attempt to interpret every misbegotten encoding available... - - Alex ------- Message 5 Received: from cygnus.com by JIMI.CS.UNLV.EDU id aa23011; 31 Jan 96 19:49 PST Received: from tweedledumb.cygnus.com (tweedledumb.cygnus.com [192.80.44.1]) by cygnus.com (8.6.12/8.6.9) with SMTP id TAA27279; Wed, 31 Jan 1996 19:49:17 -0800 Received: by tweedledumb.cygnus.com (4.1/4.7) id AA12453; Wed, 31 Jan 96 22:49:10 EST Received: by perdiem.cygnus.com (8.6.12/4.7) id WAA04501; Wed, 31 Jan 1996 22:48:53 -0500 Date: Wed, 31 Jan 1996 22:48:53 -0500 From: "Mark W. Eichin" Message-Id: <199602010348.WAA04501@perdiem.cygnus.com> To: rees@umich.edu Cc: bug-chimera@cs.unlv.edu In-Reply-To: <199602010326.TAA26867@cygnus.com> (message from Jim Rees on Wed, 31 Jan 1996 21:43:57 -0500) Subject: Re: Chimera and accents. >> Chimera only knows about a very small number of the '&' escapes. The I'm sure I saw patches posted here to fill out the list (from some HTML reference on the web) for 1.x, maybe a year ago. I can dig my archives if they've been lost... ------- Message 6 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa23414; 31 Jan 96 20:04 PST Received: from citi.umich.edu by citi.umich.edu for bug-chimera@cs.unlv.edu with SMTP; Wed, 31 Jan 96 23:03:24 -0500 From: Jim Rees To: Chimera Lovers Date: Wed, 31 Jan 1996 23:03:23 -0500 Subject: Re: Chimera and accents. Sender: rees@citi.umich.edu In-Reply-To: "Mark W. Eichin", Wed, 31 Jan 1996 22:48:53 EST I'm sure I saw patches posted here to fill out the list (from some HTML reference on the web) for 1.x, maybe a year ago. I can dig my archives if they've been lost... Actually, I'm the one who sent those diffs. I'd forgotten about it. Guess I'm getting senile. I just checked the chimera 2.0 code, and it seems to be missing the same set of escapes. The table is in a different format and order, so adding them back in isn't trivial, but it isn't hard either. ------- Message 7 Received: from mail.cs.tu-berlin.de by JIMI.CS.UNLV.EDU id ab28857; 31 Jan 96 23:30 PST Received: from titanic.cs.tu-berlin.de (czyborra@titanic.cs.tu-berlin.de [130.149.17.141]) by mail.cs.tu-berlin.de (8.6.12/8.6.12) with ESMTP id IAA07331; Thu, 1 Feb 1996 08:30:17 +0100 Received: (czyborra@localhost) by titanic.cs.tu-berlin.de (8.6.12/8.6.6) id IAA17043; Thu, 1 Feb 1996 08:30:16 +0100 Subject: Re: Chimera and accents. Cc: bug-chimera@cs.unlv.edu X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII From: Roman Czyborra To: Jim Rees Date: Thu, 01 Feb 1996 08:30:15 +0100 Message-ID: In-Reply-To: <199602010253.DAA00847@mail.cs.tu-berlin.de> > The charset of the document is determined by the charset= > parameter to the Content-Type returned by the server. Actually, sending a real MIME Content-Type with a charset parameter by HTTP screws up both current server and browser software. That's why you should only be using an unparameterized type of text/html and stuff the charset into the contents: > That's what the spec says, but I have yet to see any web pages that actually > do this. There doesn't seem to be any standard for non-8859-1 pages. > Usually they just use some other character set, with no indication as to > which set is in use. If you know of counter-examples, I'd like to see them. If you have a Latin 2 font in your font path and are using Netscape 2 with a disk cache, https://www.cs.tu-berlin.de/~czyborra/foreign.html will come up with Polish letters, at least upon Reload, Redisplay, Document Info or something like that. But you're right that lots of charset tags still have to be added, but Netscape 2 ain't close to being stable yet, either. > But chimera certainly should be able to render 8859-1 correctly, since > that's the default. Chimera works like a charm with other charsets, too. For instance, my https://www.cs.tu-berlin.de/~czyborra/~/bin/koimera lets you read koimera https://nuweb.jinr.dubna.su/~kovaler/TEXTS/ ------- Message 8 Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa08704; 1 Feb 96 6:50 PST Received: from kroete2.freinet.de (arhppp20.uni-c.dk [130.225.243.180]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id PAA00693; Thu, 1 Feb 1996 15:47:13 +0100 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0thzLR-0003kOC; Thu, 1 Feb 96 14:46 MET Message-Id: From: Erik Corry Subject: Re: Chimera and accents. To: Roman Czyborra Date: Thu, 1 Feb 1996 14:46:35 +0100 (MET) Cc: Chimera Hacker List In-Reply-To: from "Roman Czyborra" at Feb 1, 96 08:30:15 am Reply-To: Erik Corry X-Face: 5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS Erik wrote: > > The charset of the document is determined by the charset= > > parameter to the Content-Type returned by the server. > > Actually, sending a real MIME Content-Type with a charset parameter by > HTTP screws up both current server and browser software. That's why you > should only be using an unparameterized type of text/html and stuff the > charset into the contents: > > We don't, as far as I know, parse META yet (John?), and though it's really only a hack to get around inadequate http servers. It's probably necessary to support this. How does setting a charset in the content-type mess up browsers? Perhaps people could take a look at https://www.infoweb.dk/sprogtest.cgi which is a page with the header: Content-Type: text/html; charset=ISO-8859-1 and some Danish characters. It seems to work for me. > > That's what the spec says, but I have yet to see any web pages that actually > > do this. There doesn't seem to be any standard for non-8859-1 pages. > > Usually they just use some other character set, with no indication as to > > which set is in use. If you know of counter-examples, I'd like to see them. This is very sad. How are we supposed to render them correctly? I suppose you could have an override for the user to set the document character set if it's not ISO-8859-1. If we supported META then at least the author could correct the problem without help from the http server administrator. The nasty thing about META is that it changes the charset after we already started to read the file. This is conceptually very messy. > If you have a Latin 2 font in your font path and are using Netscape 2 with > a disk cache, https://www.cs.tu-berlin.de/~czyborra/foreign.html will come > up with Polish letters, at least upon Reload, Redisplay, Document Info or > something like that. But you're right that lots of charset tags still > have to be added, but Netscape 2 ain't close to being stable yet, either. So it half works. Are you using META or the server to set the charset? > > But chimera certainly should be able to render 8859-1 correctly, since > > that's the default. > > Chimera works like a charm with other charsets, too. For instance, my > https://www.cs.tu-berlin.de/~czyborra/~/bin/koimera lets you read > koimera https://nuweb.jinr.dubna.su/~kovaler/TEXTS/ Do you use a specially modified version of Chimera for this? It looks like it. - -- Erik Corry ehcorry@inet.uni-c.dk https://inet.uni-c.dk/~ehcorry/ ------- Message 9 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa10432; 1 Feb 96 8:19 PST Received: from citi.umich.edu by citi.umich.edu for bug-chimera@cs.unlv.edu with SMTP; Thu, 01 Feb 96 11:18:13 -0500 From: Jim Rees To: Chimera Lovers Date: Thu, 01 Feb 1996 11:18:12 -0500 Subject: Re: Chimera and accents. Sender: rees@citi.umich.edu In-Reply-To: Erik Corry, Thu, 01 Feb 1996 14:46:35 +0100 How does setting a charset in the content-type mess up browsers? ...It seems to work for me. Setting the chartype to 8859-1 isn't a good test, since that's the default. Try some other set and see what happens. This is very sad. How are we supposed to render them correctly? I suppose you could have an override for the user to set the document character set if it's not ISO-8859-1. That's exactly how the national versions of Mosaic work. This is an especially annoying situation since http uses mime. The information is there, but the people who have made national versions of browsers have chosen to ignore it in favor of manual intervention. I hope chimera will do better. Actually, the situation is more complicated than this. The Japanese come close to getting it right. They contend, and I agree, that having a single character set for an entire doocument is not correct. What you want is the ability to switch languages, and possibly character sets, in the middle of the document. The Japanese use ISO 2022 for this purpose. The other way to do it, widely advocated outside of Japan but seldom implemented, is Unicode with html language tags. ------- Message 10 Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa25776; 1 Feb 96 16:23 PST Received: from kroete2.freinet.de (arhppp49.uni-c.dk [130.225.243.211]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id BAA03715 for ; Fri, 2 Feb 1996 01:20:51 +0100 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0ti6P9-0003mDC; Thu, 1 Feb 96 22:18 MET Message-Id: From: Erik Corry Subject: Re: Chimera and accents. To: Jim Rees Date: Thu, 1 Feb 1996 22:18:54 +0100 (MET) Cc: Chimera Hacker List In-Reply-To: <199602011633.RAA20475@inet.uni-c.dk> from "Jim Rees" at Feb 1, 96 11:18:12 am Reply-To: Erik Corry X-Face: 5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS That's exactly how the national versions of Mosaic work. This is an > especially annoying situation since http uses mime. The information is > there, but the people who have made national versions of browsers have > chosen to ignore it in favor of manual intervention. I hope chimera will do > better. How? > Actually, the situation is more complicated than this. The Japanese come > close to getting it right. They contend, and I agree, that having a single > character set for an entire doocument is not correct. What you want is the > ability to switch languages, and possibly character sets, in the middle of > the document. The Japanese use ISO 2022 for this purpose. The other way to > do it, widely advocated outside of Japan but seldom implemented, is Unicode > with html language tags. I'd very much like to implement Unicode as an internal format for Chimera. This would make it simple to use it as an input character set for Chimera, and it wouldn't be difficult to convert ISO 2022 to Unicode on input. The problem I see with ISO 2022 is that the way HTML is going we have to have support for Unicode anyway, so why should we implement ISO 2022 as well. A question: Does ISO 2022 allow ISO-8859-2 and other non-ascii 8-bit character sets? - -- Erik Corry ehcorry@inet.uni-c.dk https://inet.uni-c.dk/~ehcorry/ ------- Message 11 Received: from fsb1.aist-nara.ac.jp by JIMI.CS.UNLV.EDU id aa06542; 2 Feb 96 0:17 PST Received: from decpc56.aist-nara.ac.jp by mailgate.aist-nara.ac.jp (8.6.10+2.5Wb1/2.8Wb/NAIST-1.6[gate]) id RAA27680; Fri, 2 Feb 1996 17:16:41 +0900 Return-Path: Received: from decpc56.aist-nara.ac.jp by decpc56.aist-nara.ac.jp (8.6.12/2.7W-AIST/1.3) id RAA13879; Fri, 2 Feb 1996 17:16:39 +0900 Message-Id: <199602020816.RAA13879@decpc56.aist-nara.ac.jp> To: bug-chimera@cs.unlv.edu Subject: Re: Chimera and accents. In-Reply-To: Your message of "Thu, 1 Feb 1996 22:18:54 +0100 (MET)" References: X-Mailer: Mew version 1.01 on Emacs 19.28.451, Mule 2.3 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Fri, 02 Feb 1996 17:16:37 +0900 From: =?ISO-2022-JP?B?GyRCOzNLXBsoQg==?= =?ISO-2022-JP?B?GyRCTFAbKEI=?= For Unicode lovers, please read following papers.
  • Yutaka KATAOKA, Masato MORISAKI, Hiroshi KURIBAYASHI and Hiroyoshi OHARA, "A Model for Input and Output of Multilingual Text in Windowing Environment", Transactions on Information Systems, Vol. 10 No. 4, pp438-451, ACM, October 1992
  • Yutaka KATAOKA and et. al., "The Worldwide Multilingual Computing (1)--(8), Proceedings of the 51th General Meeting of IPSJ, Vol. 3, pp245-260, IPSJ, September 1995
  • Yutaka KATAOKA and et. al., "Codeset Independent Full Multilingual Operationg System: Principles, Model and Optimal Architecture, IPSJ SIG System Software & Operating System 68-4, pp25-32, IPSJ, March 1995
  • Masataka OHTA, "CHARACTER ENCODING METHOD FOR INTERNATIONALIZED PLAIN TEXT PROCESSING", 8th JWCC, 12-14 December 1993, B2-1
After reading those papers, do you believe that Unicode is better code for I18N? Erik> A question: Does ISO 2022 allow ISO-8859-2 and other non-ascii 8-bit Erik> character sets? ISO 8859-2 is a 96 character set. A designation escape sequence is 02/13 04/03, 02/14 04/03 or 02/15 04/03. ISO 2022 defines an escape sequence, which are 02/05 and 02/05 , to designate and invoke non-ISO 2022 based character sets. Escape sequences to designate and invoke ISO 10646 are, 02/05 02/15 04/00 (UCS-2, Level 1) 02/05 02/15 04/01 (UCS-4, Level 1) 02/05 02/15 04/02 (UCS-2, Level 2) 02/05 02/15 04/03 (UCS-4, Level 2) 02/05 02/15 04/04 (UCS-2, Level 3) 02/05 02/15 04/05 (UCS-4, Level 3) # = 1/11, = intermidiate character, = final character. The details are writen in ISO 2022 and ISO 10646. - ------------------------------------------------------------------------------- YAMAMOTO Shigeru Information Networks, Department of Information Systems Nara Advanced Institute of Science and Technology Nara, Japan ------- Message 12 Received: from aohakobe.ipc.chiba-u.ac.jp by JIMI.CS.UNLV.EDU id aa09157; 2 Feb 96 0:45 PST Received: from aohakobe (localhost [127.0.0.1]) by aohakobe.ipc.chiba-u.ac.jp (8.6.12/3.4Wbeta3 (-: 95041617) with ESMTP id RAA12795 for ; Fri, 2 Feb 1996 17:46:11 +0900 Message-Id: <199602020846.RAA12795@aohakobe.ipc.chiba-u.ac.jp> To: Chimera Hacker List Subject: Re: Chimera and accents. In-reply-to: Your message of "Thu, 01 Feb 1996 22:18:54 +0100." Date: Fri, 02 Feb 1996 17:46:11 +0900 From: Yozo Toda (TELEPHONE +81-43-290-3539) > so why should we implement ISO 2022 as well. A question: Does ISO > 2022 allow ISO-8859-2 and other non-ascii 8-bit character sets? as I understand, those character sets are "designated" as follows in ISO2022 encoding: ISO-8859-1 ESC - A ISO-8859-2 ESC - B ISO-8859-3 ESC - C ISO-8859-4 ESC - D ISO-8859-5 ESC - L ISO-8859-6 ESC - G ISO-8859-7 ESC - F ISO-8859-8 ESC - H ISO-8859-9 ESC - M for example, with mterm (multilingual variation of xterm) you can display alef character by putting 4 byte: 0x1b 0x2d 0x48 0xe0 ESC - H "alef" - -- yozo. ------- Message 13 Received: from msa.tte.vtt.fi by JIMI.CS.UNLV.EDU id aa09607; 2 Feb 96 1:14 PST Received: (from msa@localhost) by msa.tte.vtt.fi (8.7/8.7) id LAA22632; Fri, 2 Feb 1996 11:12:07 +0200 (EET) Date: Fri, 2 Feb 1996 11:12:07 +0200 (EET) From: Markku Savela Message-Id: <199602020912.LAA22632@msa.tte.vtt.fi> To: bug-chimera@cs.unlv.edu In-reply-to: Erik Corry's message of Thu, 1 Feb 1996 22:18:54 +0100 (MET) Subject: Chimera and accents. Reply-to: Markku Savela >A question: Does ISO 2022 allow ISO-8859-2 and other non-ascii 8-bit >character sets? Just to clarify some concepts. ISO 2022 is not a character set, it is a specificication of ESC sequences that are used to inform about the meaning of the octets that follow (e.g. how the following octets map into character sets, or looking at other way, how the character sets are encoded into binary octet values) ISO 8859-2 and any other ISO 8859-x, EUC, are just different encodings of character sets that can be invoked/specified with the ISO 2022 escape sequences. (All ISO 8859-x sets are actually compositions of two character sets: one "us-ascii", the other some supplementary set of characters that are mapped to the 8bit characters of the encoding). For "simple user" the issue is often clouded, because due to their localisation, they implicitly assume by default some encoding (such as ISO 8859-1) for the whole input stream. HTML/WEB is specified so that the default starting (ISO 2022) state is ISO 8859-1. It is also defined, that the state cannot be changed within the document. ISO 2022 aware application would allow changing the encoding on the fly inside the content, even if the default state is ISO 8859-1. One difference of ISO 2022 scheme and Unicode/ISO 10646 is if your default starting state is some ISO 2022 defined encoding, and you allow code switching, there is a standard method to switch into unicode, if your default starting state is unicode, there is NO WAY to switch the character set or encoding. In this sense unicode is a "dead end" :-) - -- 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 14 Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa13789; 2 Feb 96 6:11 PST Received: from kroete2.freinet.de (arhppp10.uni-c.dk [130.225.243.170]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id PAA09300 for ; Fri, 2 Feb 1996 15:08:13 +0100 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0tiJuc-0003mBC; Fri, 2 Feb 96 12:44 MET Message-Id: From: Erik Corry Subject: Re: Chimera and accents. To: Chimera Hacker List Date: Fri, 2 Feb 1996 12:44:17 +0100 (MET) In-Reply-To: <199602020912.LAA22632@msa.tte.vtt.fi> from "Markku Savela" at Feb 2, 96 11:12:07 am Reply-To: Erik Corry X-Face: 5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS > > >A question: Does ISO 2022 allow ISO-8859-2 and other non-ascii 8-bit > >character sets? > > Just to clarify some concepts. ISO 2022 is not a character set, it is > a specificication of ESC sequences that are used to inform about the > meaning of the octets that follow (e.g. how the following octets map > into character sets, or looking at other way, how the character sets > are encoded into binary octet values) OK, so with ISO 2022, the input stream is split up into parts, and there is a character set associated with each part. This seems to me to be an inconvenient way to handle the strings internally. In particular, part of the text might use the code points associated with the HTML special characters for something else. I have nothing against ISO 2022, but I doubt that it is a suitable model for internal representation of text in a browser. > ISO 8859-2 and any other ISO 8859-x, EUC, are just different encodings > of character sets that can be invoked/specified with the ISO 2022 > escape sequences. (All ISO 8859-x sets are actually compositions of > two character sets: one "us-ascii", the other some supplementary set > of characters that are mapped to the 8bit characters of the encoding). > > For "simple user" the issue is often clouded, because due to their > localisation, they implicitly assume by default some encoding (such as > ISO 8859-1) for the whole input stream. This is why I don't like localisation for our purposes. The default encoding for HTML should be ISO-8859-1, and nothing else. > HTML/WEB is specified so that the default starting (ISO 2022) state is > ISO 8859-1. It is also defined, that the state cannot be changed > within the document. ISO 2022 aware application would allow changing > the encoding on the fly inside the content, even if the default state > is ISO 8859-1. The solution to the problem of reading in data with several character sets is to support ISO 2022 and Unicode for input. My question is, what do we use as an internal format? > One difference of ISO 2022 scheme and Unicode/ISO 10646 is > > if your default starting state is some ISO 2022 defined > encoding, and you allow code switching, there is a standard > method to switch into unicode, > > if your default starting state is unicode, there is NO WAY to > switch the character set or encoding. In this sense unicode is > a "dead end" :-) But do you need to switch the character set if you start with Unicode? Anyway, if people want to make their documents in ISO-2022, then that's fine, but that's an entirely different question to the question of whether Chimera should use some sort of ISO-2022-based system internally. - -- Erik Corry ehcorry@inet.uni-c.dk https://inet.uni-c.dk ------- Message 15 Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa13791; 2 Feb 96 6:11 PST Received: from kroete2.freinet.de (arhppp10.uni-c.dk [130.225.243.170]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id PAA09314 for ; Fri, 2 Feb 1996 15:08:27 +0100 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0tiKUz-0003maC; Fri, 2 Feb 96 13:21 MET Message-Id: From: Erik Corry Subject: Re: Chimera and accents. To: Chimera Hacker List Date: Fri, 2 Feb 1996 13:21:51 +0100 (MET) In-Reply-To: <199602020816.RAA13879@decpc56.aist-nara.ac.jp> from "=?ISO-2022-JP?B?GyRCOzNLXBsoQg==?= =?ISO-2022-JP?B?GyRCTFAbKEI=?=" at Feb 2, 96 05:16:37 pm Reply-To: Erik Corry X-Face: 5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS > For Unicode lovers, please read following papers. >
    >
  • Yutaka KATAOKA, Masato MORISAKI, Hiroshi KURIBAYASHI and Hiroyoshi > OHARA, "A Model for Input and Output of Multilingual Text in > Windowing Environment", Transactions on Information Systems, Vol. 10 > No. 4, pp438-451, ACM, October 1992 >
  • Yutaka KATAOKA and et. al., "The Worldwide Multilingual Computing > (1)--(8), Proceedings of the 51th General Meeting of IPSJ, Vol. 3, > pp245-260, IPSJ, September 1995 >
  • Yutaka KATAOKA and et. al., "Codeset Independent Full Multilingual > Operationg System: Principles, Model and Optimal Architecture, IPSJ > SIG System Software & Operating System 68-4, pp25-32, IPSJ, March > 1995 >
  • Masataka OHTA, "CHARACTER ENCODING METHOD FOR INTERNATIONALIZED > PLAIN TEXT PROCESSING", 8th JWCC, 12-14 December 1993, B2-1 >
I shall see if I can get hold of these. Are any of them online? > Erik> A question: Does ISO 2022 allow ISO-8859-2 and other non-ascii 8-bit > Erik> character sets? > > ISO 8859-2 is a 96 character set. > A designation escape sequence is 02/13 04/03, 02/14 04/03 or > 02/15 04/03. I'm sorry, I don't understand this. Does this mean I can switch into ISO-8859-1 with the sequence \023\002\015\004\003 (octal character specifications in 'C' string notation)? Are the alternatives equivalent? Is this the same as what Yozo Toda tells me: # ISO-8859-2 ESC - B # #for example, with mterm (multilingual variation of xterm) #you can display alef character by putting 4 byte: # 0x1b 0x2d 0x48 0xe0 # ESC - H "alef" which looks completely different. > The details are writen in ISO 2022 and ISO 10646. Unfortunately, ISO documents are very expensive. I have Ken Lunde's O'Reilly book, "Understanding Japanese Information Processing", but it doesn't explain ISO 2022 :-( ISO-8859-2 is a charset with about 190 characters, including the 95 ASCII characters as a subset. - -- Erik Corry ehcorry@inet.uni-c.dk https://inet.uni-c.dk/~ehcorry/ ------- Message 16 Received: from fsb1.aist-nara.ac.jp by JIMI.CS.UNLV.EDU id aa17944; 2 Feb 96 9:40 PST Received: from decpc56.aist-nara.ac.jp by mailgate.aist-nara.ac.jp (8.6.10+2.5Wb1/2.8Wb/NAIST-1.6[gate]) id CAA23072; Sat, 3 Feb 1996 02:40:15 +0900 Return-Path: Received: from decpc56.aist-nara.ac.jp by decpc56.aist-nara.ac.jp (8.6.12/2.7W-AIST/1.3) id CAA15718; Sat, 3 Feb 1996 02:40:15 +0900 Message-Id: <199602021740.CAA15718@decpc56.aist-nara.ac.jp> To: bug-chimera@cs.unlv.edu Subject: Re: Chimera and accents. In-Reply-To: Your message of "Fri, 2 Feb 1996 13:21:51 +0100 (MET)" References: X-Mailer: Mew version 1.01 on Emacs 19.28.451, Mule 2.3 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Sat, 03 Feb 1996 02:40:14 +0900 From: =?ISO-2022-JP?B?GyRCOzNLXBsoQg==?= =?ISO-2022-JP?B?GyRCTFAbKEI=?= Erik> I shall see if I can get hold of these. Are any of them online? Unfortunately, those papars are not on-line. Erik> I'm sorry, I don't understand this. Does this mean I can switch into Erik> ISO-8859-1 with the sequence \023\002\015\004\003 (octal character Erik> specifications in 'C' string notation)? Are the alternatives Erik> equivalent? No. 02/13 04/03 means \033 \055 \103 (octal), 0x1b 0x2d 0x43 (hex). Erik> Unfortunately, ISO documents are very expensive. I have Ken Lunde's Erik> O'Reilly book, "Understanding Japanese Information Processing", but it Erik> doesn't explain ISO 2022 :-( I don't read original ISO 2022. I'm reading a JIS X 0202 which is a translation of ISO 2022. #JIS = Japanese Industrial Standard So, I supporse there is a translation of ISO 2022 in your country industrial standard. # I bought a JIS hand book, which is about 9,000 yen. ($1 = about 100 yen) Erik> ISO-8859-2 is a charset with about 190 characters, including the 95 Erik> ASCII characters as a subset. I think ISO-8859-2, you write, means ISO-8859-2 condig. ISO-8859-2 coding use ASCII character set and Latin 2 (ISO 8859-2) character set. ASCII is a 94 character set and Latin 2 is 96 character set. So, ISO-8859-2 is a charset with 94 + 96 = 190 characters. About ISO-8859-2 coding, following four escape sequences are omitted. 02/00 04/11 (0x1b 0x20 0x4b) # this coding system is only 8bit. 02/00 04/03 (0x1b 0x20 0x43) # use G0 and G1 character buffer. # not use locking shift. # when a character set is designated to G0 # or G1, it's automaticaly invoked from G0 # to GL or from G1 to GR. # GL = from 02/01 to 07/14 (0x21 -- 0x7E) # GR = from 10/00 to 15/15 (0xa0 -- 0xff) 02/08 04/02 (0x1b 0x28 0x42) # designate ASCII character set to G0 02/13 04/02 (0x1b 0x2d 0x42) # designate ISO 8859-2 character set to G1 - ------------------------------------------------------------------------------- YAMAMOTO Shigeru Information Networks, Department of Information Systems Nara Advanced Institute of Science and Technology Nara, Japan ------- Message 17 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa18664; 2 Feb 96 10:19 PST Received: from citi.umich.edu by citi.umich.edu for bug-chimera@cs.unlv.edu with SMTP; Fri, 02 Feb 96 13:18:30 -0500 From: Jim Rees To: Chimera Lovers Date: Fri, 02 Feb 1996 13:18:29 -0500 Subject: unicode Sender: rees@citi.umich.edu Unicode seems to be the best choice for internal representation. It has the advantages of fixed width characters and statelessness. It's not sufficient by itself, since it carries no language tags. I think language tagging can be handled in a way similar to how font and face changes are handled now. In my opinion, ISO 2022 for data interchange by itself is a bad idea, since it confuses character set with language. Unicode with language tags seems like a better choice to me, since you can put any tag on any character, and since html is well suited to the addition of language tags. But, we're getting off the subject. Chimera has to deal with the situation as it exists today, while also making it possible to do the Right Thing. The people who publish non-8859-1 text have decided to make it impossible to read their pages automatically. So chimera should do what Netscape does, which is to allow the user to select the character set manually. Ugh. It would probably also be useful to automatically detect 2022 escapes, so at least Japanese can be read without manual intervention. Using mime headers to select the character set doesn't seem right to me, since you can't change character sets within a page. I think that's why the Japanese went with 2022 instead of (for instance) a JIS mime type. ------- Message 18 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa25499; 2 Feb 96 13:58 PST Received: from citi.umich.edu by citi.umich.edu for bug-chimera@cs.unlv.edu with SMTP; Fri, 02 Feb 96 16:57:07 -0500 From: Jim Rees To: Chimera Lovers Date: Fri, 02 Feb 1996 16:57:07 -0500 Subject: startup time with big caches Sender: rees@citi.umich.edu 1.65 takes a long time to start up if you have a big cache. I haven't looked to see if 2.0 does it differently, but maybe there should be a cache index to speed startup time. ------- Message 19 Received: from aohakobe.ipc.chiba-u.ac.jp by JIMI.CS.UNLV.EDU id aa06879; 3 Feb 96 0:27 PST Received: from aohakobe (localhost [127.0.0.1]) by aohakobe.ipc.chiba-u.ac.jp (8.6.12/3.4Wbeta3 (-: 95041617) with ESMTP id RAA15585 for ; Sat, 3 Feb 1996 17:28:51 +0900 Message-Id: <199602030828.RAA15585@aohakobe.ipc.chiba-u.ac.jp> To: Chimera Lovers Subject: Re: unicode In-reply-to: Your message of "Fri, 02 Feb 1996 13:18:29 EST." <9602021936.AA05912@cuipc.ipc.chiba-u.ac.jp> Date: Sat, 03 Feb 1996 17:28:51 +0900 From: Yozo Toda (TELEPHONE +81-43-290-3539) there is a discussion for internal representation for multilingual texts, but here only data exchange issue... > In my opinion, ISO 2022 for data interchange by itself is a bad idea, since > it confuses character set with language. Unicode with language tags seems > like a better choice to me, since you can put any tag on any character, and > since html is well suited to the addition of language tags. hmm. how about using ISO2022 for (only) character set specification? ISO2022 with language tags is an alternative choice. here I assume tags like "C'est la vie", which is an example shown in RFC1766. > Using mime headers to select the character set doesn't seem right to me, > since you can't change character sets within a page. I think that's why the > Japanese went with 2022 instead of (for instance) a JIS mime type. as I understand, MIME headers Content-language: or Content-type: only give hints which languages the page contain. I think it is a similar semantics as PostScript Document Structuring Convention describes. - -- yozo. ------- Message 20 Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa11688; 3 Feb 96 6:11 PST Received: from kroete2.freinet.de (arhppp31.uni-c.dk [130.225.243.193]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id PAA09627 for ; Sat, 3 Feb 1996 15:08:38 +0100 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0tihRu-0003mSC; Sat, 3 Feb 96 13:52 MET Message-Id: From: Erik Corry Subject: Re: unicode To: Chimera Hacker List Date: Sat, 3 Feb 1996 13:52:13 +0100 (MET) In-Reply-To: <199602030828.RAA15585@aohakobe.ipc.chiba-u.ac.jp> from "Yozo Toda" at Feb 3, 96 05:28:51 pm Reply-To: Erik Corry X-Face: 5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS hmm. how about using ISO2022 for (only) character set specification? > ISO2022 with language tags is an alternative choice. > here I assume tags like "C'est la vie", > which is an example shown in RFC1766. But we could also use ISO2022 for language hints, especially in documents that don't contain any LANG tags. That's not quite correct, but I think it will be much more useful, given that most people won't bother with LANG tags. > > Using mime headers to select the character set doesn't seem right to me, > > since you can't change character sets within a page. I think that's why the > > Japanese went with 2022 instead of (for instance) a JIS mime type. > > as I understand, MIME headers Content-language: or Content-type: > only give hints which languages the page contain. I think it is > a similar semantics as PostScript Document Structuring Convention describes. I don't think the language can be indicated at all in the Content-type of the document. But "The Content-Language entity-header field describes the natural language(s) of the intended audience for the enclosed entity. Note that this may not be equivalent to all the languages used within the entity." according to draft-ietf-http-v11-spec-01.txt. There doesn't seem to be any differentiation between 'and' and 'or' in the case where several languages are specified. So if the header contains: Content-Language: dk, en you don't know whether you need to understand both Danish and English to understand the text, or whether one is enough. But Content-Language is only intended for content negotiation, which noone does anyway. We could use it for a language hint in the absence of tags, I suppose. - -- Erik Corry ehcorry@inet.uni-c.dk https://inet.uni-c.dk/~ehcorry/ ------- Message 21 Received: from ai.mit.edu by JIMI.CS.UNLV.EDU id aa20817; 8 Feb 96 15:09 PST Received: from jacal.bertronics (camelot.ai.mit.edu) by life.ai.mit.edu (4.1/AI-4.10) for bug-chimera@cs.unlv.edu id AA27710; Thu, 8 Feb 96 18:09:10 EST Return-Path: Message-Id: Date: Thu, 8 Feb 96 18:06 EST To: bug-chimera@cs.unlv.edu Subject: X resources in chimera-1.65 Reply-To: jaffer@ai.mit.edu From: Aubrey Jaffer In README.hints, "Term Notes" it says: C. Uncomment the TERM line in options.h. There was no such line in chimera-1.65/options.h -=-=-=- Flickering light gives me migraine headaches. The option "-rv", when supported by X programs makes them usable. But chimera doesn't do anything with "-rv". So, as with many X programs, I have to set all the colors. The files "Chimera.ad.test" and "src/Chimera.ad" from the chimera-1.65 distribution have some colors set in them, but not a complete set: *background: moccasin *Scrollbar.background: burlywood2 *Command.background: burlywood2 *Toggle.background: burlywood2 I grepped through the source and found more: *background: Dark Slate Gray *foreground: White *borderColor: gray *html.visitedAnchorColor: light blue *html.anchorColor: light green *html.activeAnchorBG: Dark Gray *html.activeAnchorFG: White This gets all the colors except for the lines between the panes. I can't find that one. I think it would be helpful to others to include the complete list somewhere in your documentation. -=-=-=- The next problem was that the Chimera's default fonts are a bit small on a 1504x1128 screen. Again, "Chimera.ad.test" and "src/Chimera.ad" have some, but not all the fonts: *Label.font: -*-lucidatypewriter-medium-r-normal-*-*-120-*-*-*-*-iso8859-1 *Command.font: -*-lucida-bold-r-normal-sans-*-120-*-*-*-*-iso8859-1 *Toggle.font: -*-lucida-bold-r-normal-sans-*-120-*-*-*-*-iso8859-1 *urldisplay.font: -*-lucida-bold-r-normal-sans-*-120-*-*-*-*-iso8859-1 *titledisplay.font: -*-lucida-bold-r-normal-sans-*-120-*-*-*-*-iso8859-1 *header1Font: -*-helvetica-bold-r-normal-*-*-240-*-*-*-*-iso8859-1 *header2Font: -*-helvetica-bold-r-normal-*-*-180-*-*-*-*-iso8859-1 *header3Font: -*-helvetica-bold-r-normal-*-*-140-*-*-*-*-iso8859-1 *header4Font: -*-helvetica-bold-r-normal-*-*-120-*-*-*-*-iso8859-1 *header5Font: -*-helvetica-bold-r-normal-*-*-100-*-*-*-*-iso8859-1 *header6Font: -*-helvetica-bold-r-normal-*-*-80-*-*-*-*-iso8859-1 *font: -*-helvetica-medium-r-normal-*-*-120-*-*-*-*-iso8859-1 *boldFont: -*-helvetica-bold-r-normal-*-*-120-*-*-*-*-iso8859-1 *italicFont: -*-helvetica-medium-o-normal-*-*-120-*-*-*-*-iso8859-1 *addressFont: -*-helvetica-medium-o-normal-*-*-140-*-*-*-*-iso8859-1 Study of the source turned up these additional fonts: *fixedFont: -adobe-courier-medium-r-normal-*-21-*-*-*-*-*-*-* *fixedboldFont: -adobe-courier-bold-r-normal-*-21-*-*-*-*-*-*-* *fixeditalicFont: -adobe-courier-medium-o-normal-*-21-*-*-*-*-*-*-* *plainFont: -adobe-courier-medium-r-normal-*-21-*-*-*-*-*-*-* *plainboldFont: -adobe-courier-bold-r-normal-*-21-*-*-*-*-*-*-* *plainitalicFont: -adobe-courier-medium-o-normal-*-21-*-*-*-*-*-*-* *listingFont: -adobe-courier-medium-r-normal-*-18-*-*-*-*-*-*-* I don't know if X will let you, but it would make sense to have these scale automatically. -=-=-=- Lastly, I tried to redefine some keys to do the obvious things: Page-Up, Page-Dn, Escape (cancel). In this I was unsuccessful. I spent the whole afternoon reading X man pages and trying out .translations in .Xdefaults. But I can find no examples of key translations in the documentation and the ones I tried did not effect Chimera. Do I need to modify the source for this? -=-=-=- My system's parameters are: Linux jacal 1.2.13 #4 Wed Jan 31 18:56:03 EST 1996 i486 XFree86 Version 3.1 / X Window System (protocol Version 11, revision 0, vendor release 6000) Operating System: Linux Configured drivers: Mach32: accelerated server for ATI Mach32 graphics adaptors (Patchlevel 0) (using VT number 7) - -- -=-=-=-=-=- I am a guest and *not* a member of the MIT Artificial Intelligence Lab. My actions and comments do not reflect in any way on MIT. ------- Message 22 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa28705; 8 Feb 96 20:06 PST Received: from citi.umich.edu by citi.umich.edu for bug-chimera@cs.unlv.edu with SMTP; Thu, 08 Feb 96 23:05:22 -0500 From: Jim Rees To: Chimera Lovers Date: Thu, 08 Feb 1996 23:05:22 -0500 Subject: Re: X resources in chimera-1.65 Sender: rees@citi.umich.edu In-Reply-To: Aubrey Jaffer, Thu, 08 Feb 1996 18:06:00 EST Someone sent a list of all font resources to this list a while back. I can try to locate the message and resend it if you're interested. I'm not sure, but I would guess the lines between the panes would be something like *Paned*background. Look in the X manual for resources associated with the Paned widget. To get the page keys to do the right thing: Chimera*keyTrans: #override\n\ Prior: scroll-down("Halfscreen") \n\ Next: scroll-up("Halfscreen") \n ------- Message 23 Received: from mail.cs.tu-berlin.de by JIMI.CS.UNLV.EDU id aa16614; 9 Feb 96 8:41 PST Received: from titanic.cs.tu-berlin.de (czyborra@titanic.cs.tu-berlin.de [130.149.17.141]) by mail.cs.tu-berlin.de (8.6.12/8.6.12) with ESMTP id RAA12192; Fri, 9 Feb 1996 17:40:10 +0100 Received: (czyborra@localhost) by titanic.cs.tu-berlin.de (8.6.12/8.6.6) id RAA29403; Fri, 9 Feb 1996 17:40:08 +0100 From: Roman Czyborra To: Aubrey Jaffer Cc: bug-chimera@cs.unlv.edu Subject: Re: X resources in chimera-1.65 In-Reply-To: by jaffer@martigny.ai.mit.edu dated 1996-2- 8 18:06 MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Date: Fri, 09 Feb 1996 17:40:08 +0100 Message-ID: > chimera doesn't do anything with "-rv". Yes, it does turn the foreground color from black to white for me. This isn't very readable on yellow ground but that's the usual problem with color defaults. chimera -rv -bg white -fg white -xrm *anchorColor:white seems quite usable. Also, to prevent the color lookup failures on older X11R4 servers I'd prefer black and white defaults, too. > The next problem was that the Chimera's default fonts are a bit > small on a 1504x1128 screen. You sound like you have a problem with your font path. Does Netscape have small default fonts on your display, too? Check out your font path with xset q and if it lists 75dpi prior to 100dpi, put something like xset fp- /usr/lib/X11/fonts/75dpi/ fp+ /usr/lib/X11/fonts/75dpi/ to push back the 75dpi fonts in your ~/.xsession or ~/.xinitrc. > Lastly, I tried to redefine some keys to do the obvious things: > Page-Up, Page-Dn, Escape (cancel). Jim has taught you this one already. ------- Message 24 Received: from ai.mit.edu by JIMI.CS.UNLV.EDU id aa07477; 10 Feb 96 19:20 PST Received: from jacal.bertronics (berne.ai.mit.edu) by life.ai.mit.edu (4.1/AI-4.10) for bug-chimera@cs.unlv.edu id AA25485; Sat, 10 Feb 96 22:19:17 EST Return-Path: Message-Id: Date: Sat, 10 Feb 96 22:16 EST To: czyborra@cs.tu-berlin.de Cc: bug-chimera@cs.unlv.edu In-Reply-To: (message from Roman Czyborra on Fri, 09 Feb 1996 17:40:08 +0100) Subject: Re: X resources in chimera-1.65 Reply-To: jaffer@ai.mit.edu From: Aubrey Jaffer From: Roman Czyborra Cc: bug-chimera@cs.unlv.edu Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Date: Fri, 09 Feb 1996 17:40:08 +0100 > chimera doesn't do anything with "-rv". Yes, it does turn the foreground color from black to white for me. This isn't very readable on yellow ground but that's the usual problem with color defaults. chimera -rv -bg white -fg white -xrm *anchorColor:white On my system this gives me white text on a white background. Impossible to read! seems quite usable. Also, to prevent the color lookup failures on older X11R4 servers I'd prefer black and white defaults, too. I am assuming you meant: chimera -rv -bg white -fg black -xrm *anchorColor:white chimera -bg white -fg black -xrm *anchorColor:white Both of these give black characters on a white background. Removing the "-rv" makes *no difference* to chimera on my system. > The next problem was that the Chimera's default fonts are a bit > small on a 1504x1128 screen. You sound like you have a problem with your font path. Does Netscape have small default fonts on your display, too? Check out your font path with xset q and if it lists 75dpi prior to 100dpi, put something like xset fp- /usr/lib/X11/fonts/75dpi/ fp+ /usr/lib/X11/fonts/75dpi/ to push back the 75dpi fonts in your ~/.xsession or ~/.xinitrc. Thank you very much. I removed the 75dpi fonts entirely, and now all my displays look much nicer. Perchance, do you know a source of (or X command to create) 150dpi fonts? > Lastly, I tried to redefine some keys to do the obvious things: > Page-Up, Page-Dn, Escape (cancel). Jim has taught you this one already. Jim? Yours is the only message I have received about chimera. Could you please forward his message to me? -=-=- XFree86 Version 3.1 / X Window System (protocol Version 11, revision 0, vendor release 6000) Operating System: Linux Configured drivers: Mach32: accelerated server for ATI Mach32 graphics adaptors (Patchlevel 0) (using VT number 7) ------- Message 25 Received: from iguana.reptiles.org by JIMI.CS.UNLV.EDU id aa16742; 11 Feb 96 0:56 PST Received: by iguana.reptiles.org (/\##/\ Smail3.1.30.13 #30.5) id ; Sun, 11 Feb 96 03:55:58 -0500 (EST) Message-Id: Date: Sun, 11 Feb 96 03:55:31 -0500 (EST) From: Smarasderagd To: czyborra@cs.tu-berlin.de, jaffer@ai.mit.edu Subject: Re: X resources in chimera-1.65 Cc: bug-chimera@cs.unlv.edu If I recall _X_Window_System_Toolkit_ (Paul Asente and Ralph Swick, Digital Press) correctly, the "-rv" option simply reverses Xt's notion of the default white and black pixel conversions. If an application says anything about the colour of a widget, these aren't used. ------- Message 26 Received: from mail.cs.tu-berlin.de by JIMI.CS.UNLV.EDU id aa28061; 12 Feb 96 11:35 PST Received: from titanic.cs.tu-berlin.de (czyborra@titanic.cs.tu-berlin.de [130.149.17.141]) by mail.cs.tu-berlin.de (8.6.12/8.6.12) with ESMTP id UAA05479; Mon, 12 Feb 1996 20:35:17 +0100 Received: (czyborra@localhost) by titanic.cs.tu-berlin.de (8.6.12/8.6.6) id UAA14538; Mon, 12 Feb 1996 20:35:16 +0100 From: Roman Czyborra To: Aubrey Jaffer Cc: bug-chimera@cs.unlv.edu Subject: Re: X resources in chimera-1.65 In-Reply-To: by jaffer@martigny.ai.mit.edu dated 1996-2-10 22:16 MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Date: Mon, 12 Feb 1996 20:35:16 +0100 Message-ID: Hi Aubrey! > chimera -rv -bg white -fg white -xrm *anchorColor:white > On my system this gives me white text on a white background. Oh, sorry, I meant to write chimera -rv -bg black -fg white -xrm *anchorColor:white > I removed the 75dpi fonts entirely, and now all my displays look > much nicer. Yeah, don't they? No, I don't know of any 150 dpi fonts. > Jim? Yours is the only message I have received about chimera. > Could you please forward his message to me? Jim answered to the bug-chimera address only. Probably because he thought you have subscribed to the list and he wanted to spare you from receiving his answer twice. I've resent his answer to you. Cheers ------- Message 27 Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa09957; 13 Feb 96 11:50 PST Received: from mail.tenon.com by JIMI.CS.UNLV.EDU id aa09916; 13 Feb 96 11:47 PST Received: from [192.83.246.26] (cheetah.tenon.com [192.83.246.26]) by mail.tenon.com (8.7.3/8.7.3-MT2.2) with SMTP id LAA24651 for ; Tue, 13 Feb 1996 11:57:04 -0800 (PST) Date: Tue, 13 Feb 1996 11:57:04 -0800 (PST) Message-Id: <199602131957.LAA24651@mail.tenon.com> X-Sender: neil@pop.tenon.com Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: chimera-announce@cs.unlv.edu From: Neil Cutcliffe Subject: subscribe subscribe  ------- Message 28 Received: from mail.tenon.com by JIMI.CS.UNLV.EDU id aa10181; 13 Feb 96 11:55 PST Received: from [192.83.246.26] (cheetah.tenon.com [192.83.246.26]) by mail.tenon.com (8.7.3/8.7.3-MT2.2) with SMTP id MAA24930 for ; Tue, 13 Feb 1996 12:05:11 -0800 (PST) Date: Tue, 13 Feb 1996 12:05:11 -0800 (PST) Message-Id: <199602132005.MAA24930@mail.tenon.com> X-Sender: neil@pop.tenon.com Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: bug-chimera@cs.unlv.edu From: Neil Cutcliffe Subject: URL encoding bug in chimera 1.65 When Chimera "posts" forms containing the plus character ('+'), the + is not escaped, and thus the cgi will incorrectly unencoded it as a space. The file url.c should be modified as follows: /* * EscapeURL * * Puts escape codes in URLs. NOT complete. */ char * EscapeURL(url, s2p) unsigned char *url; int s2p; { unsigned char *cp; char *n, *s; static char *hex = "0123456789ABCDEF"; /* * use a bit of memory so i don't have to mess around here */ s = n = alloc_mem(strlen(url) * 3 + 2); for (cp = url; *cp; cp++, n++) { if (*cp == ' ' && s2p != 0) { *n = '+'; } #ifdef __NOTDEF__ else if (isalnum(*cp) || strchr("$-_.'(),+!*", *cp) || (*cp=='+' && !s2p)) #else /* '+' should not be included in strchr() lookup. */ else if (isalnum(*cp) || strchr("$-_.'(),!*", *cp) || (*cp=='+' && !s2p)) #endif { *n = *cp; } else { *n = '%'; n++; *n = hex[*cp / 16]; n++; *n = hex[*cp % 16]; } } *n = '\0'; return(s); }  ------- Message 29 Received: from ny.frontiercomm.net by JIMI.CS.UNLV.EDU id aa17220; 13 Feb 96 14:54 PST Received: from monroe-80.dialup.frontiercomm.net by ny.frontiercomm.net (NX5.67e/NX3.0M) id AA02208; Tue, 13 Feb 96 17:54:40 -0500 Received: (from uri@localhost) by angmar.watson.ibm.com (8.6.12/8.6.11) id RAA02887; Tue, 13 Feb 1996 17:54:37 -0500 Date: Tue, 13 Feb 1996 17:54:37 -0500 From: Uri Blumenthal Message-Id: <199602132254.RAA02887@angmar.watson.ibm.com> To: Jim Rees Cc: Chimera Lovers Subject: cache bug fix for 1.65 In-Reply-To: <9511301955.AA32299@hawpub.watson.ibm.com> References: <9511301955.AA32299@hawpub.watson.ibm.com> Reply-To: uri@watson.ibm.com Mime-Version: 1.0 (generated by tm-edit 7.38) Content-Type: text/plain; charset=US-ASCII >>>>> "Jim" == Jim Rees writes: Jim> There's a bug in cache.c that causes the cache to be disabled Jim> on machines where off_t is bigger than an int (NetBSD, eg). Jim> Here's the fix. This diff is against the original source, so Jim> it contains the other patches too. I think this fix will cause memory leak, as "char *path" is allocated over and over again, and never freed. Thus, one has to either change the declaration to char path[MAXPATHLEN + 1]; or add free(path); at the end of while-loop in "ReadCacheInfoFile()"... Regards, Uri. - -=-=-==-=-=- uri@watson.ibm.com I'm not sure which upsets me more: that people are so unwilling to accept responsibility for their own actions, or that they are so eager to regulate everyone else's. ------- Message 30 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa23610; 13 Feb 96 18:23 PST Received: from [141.211.170.99] by citi.umich.edu for bug-chimera@cs.unlv.edu with SMTP; Tue, 13 Feb 96 21:22:06 -0500 From: Jim Rees To: Chimera Lovers Date: Tue, 13 Feb 1996 21:22:03 -0500 Subject: Re: cache bug fix for 1.65 Sender: rees@citi.umich.edu In-Reply-To: Uri Blumenthal, Tue, 13 Feb 1996 17:54:37 EST I think this fix will cause memory leak, as "char *path" is allocated over and over again, and never freed. I think I noticed this when that patch was posted (the leak is not in my part of the patch, which was just the single "off_t" line). But it hardly seems worth fixing since you only leak at startup time, and only once per cache info file. Obviously just changing the declaration wouldn't work, but adding a free() would be easy. ------- Message 31 Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa18289; 14 Feb 96 10:31 PST Received: from cpmx.saic.com by JIMI.CS.UNLV.EDU id aa18144; 14 Feb 96 10:26 PST Received: from cpqm.saic.com by cpmx.saic.com; Wed, 14 Feb 96 10:26:15 -0800 Message-ID: Date: 14 Feb 1996 13:23:28 -0700 From: Wendy Holtzman Subject: chimera-1.65.tar To: chimera-announce X-Mailer: Mail*Link SMTP-QM 3.0.2 Hello, I am trying to compile chimera-1.65.tar on a concurrent machine. It uses MAXION/OS operating system. So far I have successfully made it to the final stages. I had to remove the ranlib in the Makefiles for it to run. The question I have is ... while running make all the src compilation for local.c and cache.c continued to get the following errors: "local.c", line 171: undefined symbol: header "local.c", line 171: warning: improper pointer/integer combination: arg #1 "local.c", line 172: undefined symbol: f "local.c", line 177: cannot dereference non-pointer type etc do you have any ideas as to how to fix this ??? Thank you Wendy Holtzman wendy_holtzman@cpqm.saic.com ------- Message 32 Received: from ki1.Chemie.FU-Berlin.DE by JIMI.CS.UNLV.EDU id aa02259; 23 Feb 96 9:58 PST Received: by ki1.chemie.fu-berlin.de (Smail3.1.28.1) from uranus.central.de (194.45.71.1) with smtp id ; Fri, 23 Feb 96 18:58 MET Received: from pertron.central.de by uranus.central.de with uucp (Smail3.1.28.1 #5) id m0tq1oL-000X5mC; Fri, 23 Feb 96 19:01 MET Received: from luna.ke.central.de by pertron.ke.central.de with smtp (Smail3.1.28.1 p1) id m0tq1Go-00047SC; Fri, 23 Feb 96 18:27 MET Received: by luna.ke.central.de id (Debian /\oo/\ Smail3.1.29.1 #29.33); Fri, 23 Feb 96 18:31 MET Message-Id: Date: Fri, 23 Feb 96 18:31 MET From: Karl Eichwalder To: bug-chimera@cs.unlv.edu Subject: inPort problem My system: Linux 1.2.13, XFree 3.1.2, gcc 2.6.3; two machines, not on the internet. Chimera-1.65 does not answer incoming connections as I wish ;-) As recommended I enabled the inPort feature in my .Xdefaults: Chimera.inPort: 8082 Fire up Chimera. Now I can telnet to Chimera and display plain text: luna:/home/karl$ telnet luna 8082 Trying 192.168.1.1... Connected to luna.ke.central.de. Escape character is '^]'. PUT / HTTP/1.0 Content-Type: text/plain; charset=iso-8859-1 Hello world! ^] telnet > quit Connection closed. luna:/home/karl$ Now Chimera schows the message (Hello world!). But it refuses to show an HTML document. luna:/home/karl$ telnet luna 8082 Trying 192.168.1.1... Connected to luna.ke.central.de. Escape character is '^]'. PUT / HTTP/1.0 URI: https://pertron.central.de/Welcome.html ^] telnet> quit Then Chimera pops up a message box "Enter filename" - that all. This box only appears, if I enter an empty line after the URI:-request. If you need additional information, please ask. Thanks for the time, Karl. ------- Message 33 Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id ac18270; 25 Feb 96 18:44 PST Received: from ry.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa17700; 25 Feb 96 18:40 PST To: chimera-announce@ry.ISRI.UNLV.EDU Subject: 2.0 alpha 101 Date: Sun, 25 Feb 1996 18:40:05 -0800 From: John Kilburg I just put a new one out there (100 was an internal release :). https://www.unlv.edu/~john/chimera/src/ https://www.isri.unlv.edu/~john/chimera/src/ I tested it on DEC alpha OSF/3.0 Sun sparc SunOS 4.1.3 Some testing was done on Linux 1.2.13 gcc 2.6.3 SGI IRIX 6.1 Sorry about the delay. -john ------- Message 34 Received: from cleanhead.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa00361; 26 Feb 96 3:17 PST To: bug-chimera@cleanhead.CS.UNLV.EDU Subject: app-defaults and bookmarks Date: Mon, 26 Feb 1996 03:16:56 -0800 From: John Kilburg I forgot to do something about the app-defaults problems so you may want to watch out for that. If you are running 1.x then remove the app-defaults file from the X app-defaults directory and use a wrapper that sets XFILESEARCHPATH. This is documented in README.hints... I think I mentioned this before...it is safe to set the bookmark filename to ~/.netscape/bookmarks.html or ~/.netscape-bookmarks.html -john ------- Message 35 Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id ad02438; 26 Feb 96 8:57 PST Received: from hal.Physics.Unlv.Edu by JIMI.CS.UNLV.EDU id aa02369; 26 Feb 96 8:54 PST Received: from localhost (lagrange.Physics.Unlv.Edu [131.216.64.42]) by hal.physics.unlv.edu (8.7.4/8.7.3) with ESMTP id CAA13948 for ; Mon, 26 Feb 1996 02:13:29 -0800 (PST) Message-Id: <199602261013.CAA13948@hal.physics.unlv.edu> To: chimera-announce@cs.unlv.edu Subject: alternate site Date: Mon, 26 Feb 1996 02:13:38 -0800 From: John Kilburg You can also grab it from https://www.physics.unlv.edu/~john/chimera/src/ -john ------- Message 36 Received: from ki1.Chemie.FU-Berlin.DE by JIMI.CS.UNLV.EDU id aa03011; 26 Feb 96 9:01 PST Received: by ki1.chemie.fu-berlin.de (Smail3.1.28.1) from uranus.central.de (194.45.71.1) with smtp id ; Mon, 26 Feb 96 18:00 MET Received: from pertron.central.de by uranus.central.de with uucp (Smail3.1.28.1 #5) id m0tr6La-000Vj4C; Mon, 26 Feb 96 18:04 MET Received: from luna.ke.central.de by pertron.ke.central.de with smtp (Smail3.1.28.1 p1) id m0tr3Pj-000472C; Mon, 26 Feb 96 14:56 MET Received: by luna.ke.central.de id (Debian /\oo/\ Smail3.1.29.1 #29.33); Mon, 26 Feb 96 15:00 MET Message-Id: Date: Mon, 26 Feb 96 15:00 MET From: Karl Eichwalder To: bug-chimera@cs.unlv.edu In-reply-to: (message from John Kilburg on Sun, 25 Feb 1996 18:40:05 -0800) Subject: Re: 2.0 alpha 101 Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit | Some testing was done on | Linux 1.2.13 gcc 2.6.3 I finished compilation on Linux 1.3.68 (libc-4.6.27, gcc-2.6.3, XFree-3.1.2, 8514-server, fvwm-2.0.41); to link it the following patch was needed (I don't know why the linker fails to find the libc; I think my setup is okay: at least, the compilation of the new window manager finished successfully): *** cfh101-2.0/src/Imakefile.~1~ Mon Feb 26 01:38:57 1996 - --- cfh101-2.0/src/Imakefile Mon Feb 26 10:15:45 1996 *************** *** 24,30 **** -L../mxw -lmxw \ -L../common -lcommon \ XawClientLibs \ ! -lm $(CDEBUGLIB) LOCAL_DEPLIBS = ../common/libcommon.a $(HTMLDEPLIB) \ ../mxw/libmxw.a \ - --- 24,30 ---- -L../mxw -lmxw \ -L../common -lcommon \ XawClientLibs \ ! -lm -lc $(CDEBUGLIB) LOCAL_DEPLIBS = ../common/libcommon.a $(HTMLDEPLIB) \ ../mxw/libmxw.a \ - -=-=-=-=-=-=-=-=-=-=-=-=-=- cut here -=-=-=-=-=-=-=-=-=-=-=-=-=- There's a resizing problem. If you make make the chimera window smaller the text wraps not automatically -- you have to reload the document then scrolling works again. Thanks for the new release. Is there already some documentation for chimera-2.0? - -- | keichwa@gwdg.de | Ch ora ? - Karl Eichwalder | ke@ke.central.de | gi domani. ------- Message 37 Received: from inet.uni-c.dk by JIMI.CS.UNLV.EDU id aa14269; 26 Feb 96 13:52 PST Received: from kroete2.freinet.de (arhppp19.uni-c.dk [130.225.243.179]) by inet.uni-c.dk (8.6.12/8.6.9) with SMTP id WAA15575 for ; Mon, 26 Feb 1996 22:49:11 +0100 Received: by kroete2.freinet.de (Smail3.1.28.1 #4) id m0trAof-0003mUC; Mon, 26 Feb 96 22:50 MET Message-Id: From: Erik Corry Subject: Re: alternate site To: Chimera Hacker List Date: Mon, 26 Feb 1996 22:50:43 +0100 (MET) In-Reply-To: <199602261013.CAA13948@hal.physics.unlv.edu> from "John Kilburg" at Feb 26, 96 02:13:38 am Reply-To: Erik Corry X-Face: 5am4$s<`Jfx-,*w6$)uX,dDt3Z2w?4ZNGS-i@_w(Fzv%(<[(XtN*t\,OmV_a=lS > You can also grab it from > > https://www.physics.unlv.edu/~john/chimera/src/ Or https://inet.uni-c.dk/~ehcorry/cfh101-2.0.tar.gz - -- Erik Corry ehcorry@inet.uni-c.dk https://inet.uni-c.dk/~ehcorry/ ------- Message 38 Received: from xph029.physics.montana.edu by JIMI.CS.UNLV.EDU id aa16547; 26 Feb 96 14:44 PST Received: (from michael@localhost) by xph029.physics.montana.edu (8.6.12/8.6.9) id PAA10079; Mon, 26 Feb 1996 15:40:48 -0700 Date: Mon, 26 Feb 1996 15:40:42 -0700 (MST) From: Michael Kellen To: Chimera Hacker List Subject: [cfh101-2.0] Shakedown Run In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII The reported linux problem sonds an AWFULLY LOT like a misconfigured gcc, most likely in the debugging support. I had no problems compiling the latest alpha straight out of the box on my linux system. The '#ANCHOR' support works nicely. (Yay!) The previously reported shtml error still generates new connections until the system locks up: https://www.vni.net/~kwelch/penguin.html/blue_ice.shtml I was disappointed with the bookmark "facility". IMO it has taken a big step downward in functionality. The lack of groups, inability to edit, and replacement of the current page with the list are a worse interface than the original. The speed of chimera is still quite good. When the user jumps to an ANCHOR, the "Back" button does not jump back to the previous position in the document. Resize events are not handled. Changing the browser size allows you to clip text and images. No scrollbars are generated to allow viewing such text if they were unneeded before the resize. io.done is not defined in the resources file. IMO, the resources file should not segfault the client by its absence. Michael Kellen Linux version 1.3.32 gcc version 2.7.0 XFree 3.1.2 ------- Message 39 Received: from iguana.reptiles.org by JIMI.CS.UNLV.EDU id aa19768; 26 Feb 96 16:21 PST Received: by iguana.reptiles.org (/\##/\ Smail3.1.30.13 #30.5) id ; Mon, 26 Feb 96 19:21:34 -0500 (EST) Message-Id: Date: Mon, 26 Feb 96 19:21:22 -0500 (EST) From: Smarasderagd To: bug-chimera@cs.unlv.edu, ke@ke.Central.DE Subject: Re: 2.0 alpha 101 Karl Eichwalder writes: >I finished compilation on Linux 1.3.68 (libc-4.6.27, gcc-2.6.3, >XFree-3.1.2, 8514-server, fvwm-2.0.41); to link it the following patch >was needed (I don't know why the linker fails to find the libc; I think >my setup is okay: at least, the compilation of the new window manager >finished successfully): This sounds like a fairly well-known problem with older (pre gcc-2.7.0) library distributions. When linking with -g, gcc attempts to link with - -lg instead of -lc. When linking statically with -g, ld looks for libg.a, which is (on my system, using gcc-2.6.2 and libc-4.6.27) symlinked to libc.a. By default, though, linking is shared, so it looks for libg.sa, the shared library stub. On some Linux systems no such file exists. One solution is to symlink libg.sa to libc.sa. Alternatively, you can link statically, using CEXTRA_LOAD_FLAGS in Common.tmpl. ------- Message 40 Received: from ttmath.ttu.edu by JIMI.CS.UNLV.EDU id aa22109; 26 Feb 96 17:40 PST Date: Mon, 26 Feb 1996 19:40:13 -0600 (CST) From: Jake Kesinger To: Chimera Hacker List Subject: Re: [cfh101-2.0] Shakedown Run In-Reply-To: Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Mon, 26 Feb 1996, Michael Kellen wrote: > I was disappointed with the bookmark "facility". IMO it has taken a big > step downward in functionality. The lack of groups, inability to edit, > and replacement of the current page with the list are a worse interface > than the original. Not to sound like a l4M3R, but I agree. To be honest, though, I liked the ond style bookmark window. In fact, I consider it my favorite BM interface. What are the long-range plans for BM functionality? Something else I've noticed with nearly all cfh\d+-2.0 chimeras is that the scrollbars don't have a border. That is, on many progs using what I assume to be the same scrollbar widget as chim (xterm, for one) the scrollbar looks kind of like: | | | | | | | | where the left column is the edge of the window, the XXs are |XX| the shaded part of the scrollbar, and the right column |XX| is between the scrollbar and the main window. | | | | On the 2.\d+ series the right column is not visible. Is this a resource that I can change? Because it's kind of annoying. Compiles (and seems to work) fine on Solaris2.mumble... --Jake _ Jake Kesinger (kesinger@math.ttu.edu) <*> 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 41 Received: from ki1.Chemie.FU-Berlin.DE by JIMI.CS.UNLV.EDU id aa17362; 27 Feb 96 10:11 PST Received: by ki1.chemie.fu-berlin.de (Smail3.1.28.1) from uranus.central.de (194.45.71.1) with smtp id ; Tue, 27 Feb 96 19:10 MET Received: from pertron.central.de by uranus.central.de with uucp (Smail3.1.28.1 #5) id m0trTvS-000VF2C; Tue, 27 Feb 96 19:15 MET Received: from luna.ke.central.de by pertron.ke.central.de with smtp (Smail3.1.28.1 p1) id m0trSsy-0004AHC; Tue, 27 Feb 96 18:08 MET Received: by luna.ke.central.de id (Debian /\oo/\ Smail3.1.29.1 #29.33); Tue, 27 Feb 96 17:11 MET Message-Id: Date: Tue, 27 Feb 96 17:11 MET From: Karl Eichwalder To: bug-chimera@cs.unlv.edu Subject: Display bug if picture is not available Organization: The Moon On Earth Linux 1.2.13, XFree 3.1.2, gcc 2.6.3. chimera101-2.0 fails to display a page, if an XBM file is not found (1.65 does it anyway). Since I use to fetch HTML documents by email (w3mail@gmd.de), I normally do not care about pictures. Please make a local copy from https://www.cs.pdx.edu/~trent/gnu/gnu.html and then try to load it as e.g. file:/tmp/gnu.html GNU's Not Unix!

[GNU icon]GNU's Not Unix!

Karl ------- Message 42 Received: from ilex.fernuni-hagen.de by JIMI.CS.UNLV.EDU id aa25006; 27 Feb 96 13:33 PST Received: from ES-sun1.fernuni-hagen.de by ilex.FernUni-Hagen.de with SMTP (PP); Tue, 27 Feb 1996 22:33:07 +0100 Received: from ES-sun2 by ES-sun1.fernuni-hagen.de (4.1/SMI-4.1) id AA17840; Tue, 27 Feb 96 22:33:06 +0100 Message-Id: <9602272133.AA17840@ES-sun1.fernuni-hagen.de> To: Karl Eichwalder Cc: bug-chimera@cs.unlv.edu Subject: Re: Display bug if picture is not available In-Reply-To: Your message of "Tue, 27 Feb 1996 17:11:00 +0700." Date: Tue, 27 Feb 1996 22:33:04 +0100 From: Johann Friedrich Heinrichmeyer simple question: where is chimera101-2.0.tar.gz > 152k (archive is not complete after download, maybe an error of our http-proxy-cache?) Viele Gruesse, Fritz Heinrichmeyer FernUniversitaet Hagen FAX: +49 2371/566236 LG Elektronische Schaltungen EMAIL: fritz.heinrichmeyer@fernuni-hagen.de Frauenstuhlweg 31 PHONE: +49 02371/566-243 58644 Iserlohn (Germany) WWW: https://ES-sun2.fernuni-hagen.de ------- Message 43 Received: from ry.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa17862; 28 Feb 96 1:37 PST To: Jake Kesinger cc: Chimera Hacker List Subject: Re: [cfh101-2.0] Shakedown Run In-reply-to: Your message of "Mon, 26 Feb 1996 19:40:13 CST." Date: Wed, 28 Feb 1996 01:37:18 -0800 From: John Kilburg >Not to sound like a l4M3R, but I agree. To be honest, though, I liked >the ond style bookmark window. In fact, I consider it my favorite >BM interface. What are the long-range plans for BM functionality? Well, I kind of like the current setup because it is easy to use and I can select a bookmark with very few mouse clicks. Also, the code is small and simple. I suppose that popular demand could bring the old bookmark interface back. >On the 2.\d+ series the right column is not visible. Is this a resource >that I can change? Because it's kind of annoying. I'll try to make sure this can be done in the next release. -john ------- Message 44 Received: from ry.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa18081; 28 Feb 96 1:50 PST To: Karl Eichwalder cc: bug-chimera@cs.unlv.edu Subject: Re: 2.0 alpha 101 In-reply-to: Your message of "Mon, 26 Feb 1996 15:00:00 +0700." Date: Wed, 28 Feb 1996 01:50:50 -0800 From: John Kilburg >Thanks for the new release. Is there already some documentation for >chimera-2.0? No documentation yet. One of these days I hope to get some written... -john ------- Message 45 Received: from ry.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa18566; 28 Feb 96 2:31 PST To: Chimera Hacker List Subject: Re: [cfh101-2.0] Shakedown Run In-reply-to: Your message of "Mon, 26 Feb 1996 15:40:42 MST." Date: Wed, 28 Feb 1996 02:31:38 -0800 From: John Kilburg >The previously reported shtml error still generates new connections until >the system locks up: > > https://www.vni.net/~kwelch/penguin.html/blue_ice.shtml This one is really weird. You can try https://www.vni.net/~kwelch/penguin.shtml until I fix the problem. It kind of looks like the server is broken, too. >Resize events are not handled. Changing the browser size allows you to >clip text and images. No scrollbars are generated to allow viewing such >text if they were unneeded before the resize. I'll try to work on this for the next release. >io.done is not defined in the resources file. Grab the resources file from https://www.isri.unlv.edu/~john/chimera/src/ (it wasn't on the physics page...). -john ------- Message 46 Received: from nova.gmi.edu by JIMI.CS.UNLV.EDU id aa21873; 28 Feb 96 5:46 PST Received: by gmi.edu (5.x/SMI-SVR4) id AA08451; Wed, 28 Feb 1996 08:46:10 -0500 Date: Wed, 28 Feb 1996 08:46:09 -0500 (EST) From: "R. Stewart Ellis" To: John Kilburg Cc: Chimera Hacker List Subject: Re: [cfh101-2.0] Shakedown Run In-Reply-To: <9602281219.AA04053@gmi.edu> Message-Id: Organization: GMI Engineering&Management Institute - Flint MI Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Wed, 28 Feb 1996, John Kilburg wrote: > Date: Wed, 28 Feb 1996 02:31:38 -0800 > From: John Kilburg > To: Chimera Hacker List > Subject: Re: [cfh101-2.0] Shakedown Run > > >The previously reported shtml error still generates new connections until > >the system locks up: > > > > https://www.vni.net/~kwelch/penguin.html/blue_ice.shtml The graphics on this were not rendered by Netscape 2.0 on SPARC Solaris 2.5. > > This one is really weird. You can try > > https://www.vni.net/~kwelch/penguin.shtml This one worked. > > until I fix the problem. > > It kind of looks like the server is broken, too. > > >Resize events are not handled. Changing the browser size allows you to > >clip text and images. No scrollbars are generated to allow viewing such > >text if they were unneeded before the resize. > > I'll try to work on this for the next release. > > >io.done is not defined in the resources file. > > Grab the resources file from https://www.isri.unlv.edu/~john/chimera/src/ > (it wasn't on the physics page...). > > -john > R.Stewart(Stew) Ellis, Assoc.Prof., (Off)810-762-9765 ___________________ Humanities & Social Science, GMI Eng.& Mgmt.Inst. () / _____ ______ Flint, MI 48504 Free speech and free software! /\ / / / / / / Web admin: https://www.gmi.edu/ ellis@gmi.edu /________/ / / / / ------- Message 47 Received: from ki1.Chemie.FU-Berlin.DE by JIMI.CS.UNLV.EDU id aa28302; 28 Feb 96 10:30 PST Received: by ki1.chemie.fu-berlin.de (Smail3.1.28.1) from uranus.central.de (194.45.71.1) with smtp id ; Wed, 28 Feb 96 19:29 MET Received: from pertron.central.de by uranus.central.de with uucp (Smail3.1.28.1 #5) id m0trqgj-000WokC; Wed, 28 Feb 96 19:33 MET Received: from luna.ke.central.de by pertron.ke.central.de with smtp (Smail3.1.28.1 p1) id m0trpol-0004AlC; Wed, 28 Feb 96 18:37 MET Received: by luna.ke.central.de id (Debian /\oo/\ Smail3.1.29.1 #29.33); Wed, 28 Feb 96 18:42 MET Message-Id: Date: Wed, 28 Feb 96 18:42 MET From: Karl Eichwalder To: john@ry.ISRI.UNLV.EDU CC: bug-chimera@cs.unlv.edu In-reply-to: (message from John Kilburg on Wed, 28 Feb 1996 01:20:14 -0800) Subject: Paragraf and align bug (Re: Display bug) Organization: The Moon On Earth This has been fixed in the next release. At least you probably won't see the problem again. :) Thanks ;-) Next one -- I hope it's clear, what I want to say: Paragraf and align bug

Chimera displays things right

but only stops to do this, if there is an explicit end-tag </p>.

This should go left.

As far as I know the standard html-2.0 says nothing about attributes of the

element... Seems to be an extension. Karl. ------- Message 48 Received: from ry.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa07900; 28 Feb 96 15:33 PST To: Karl Eichwalder cc: bug-chimera@cs.unlv.edu Subject: Re: Paragraf and align bug (Re: Display bug) In-reply-to: Your message of "Wed, 28 Feb 1996 18:42:00 +0700." Date: Wed, 28 Feb 1996 15:33:25 -0800 From: John Kilburg >Next one -- I hope it's clear, what I want to say: > >[...] > >As far as I know the standard html-2.0 says nothing about attributes of >the

element... Seems to be an extension. Karl. They are HTML3 attributes. Chimera supports a subset of HTML2 (almost all of it) and a very small number of HTML3 attributes. -john ------- Message 49 Received: from xph029.physics.montana.edu by JIMI.CS.UNLV.EDU id aa10740; 28 Feb 96 16:42 PST Received: (from michael@localhost) by xph029.physics.montana.edu (8.6.12/8.6.9) id RAA15027; Wed, 28 Feb 1996 17:38:03 -0700 Date: Wed, 28 Feb 1996 17:37:58 -0700 (MST) From: Michael Kellen To: Chimera Hacker List Subject: [cfh101-2.0] Bookmarks Widget "Patch" Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII For those of us pining for the "good old days" of the old bookmark widget, I have kludged together a hideous mishmash of cfh92 and cfh101 that seems to work. Please do not report any bugs, this is merely a stopgap. The "patch" is a tar file of the altered code, as it also includes the mxw/Bookmark widget code (modified from cfh92). To use it, you must (1) retrieve the modified source code https://xph029.physics.montana.edu/chimera/oldmarks.tgz or ftp://xph029.physics.montana.edu/pub/html/chimera/oldmarks.tgz (2) untar it in cfh101-2.0 (3) Edit the Common.tmpl provided BE SURE TO DEFINE OLDMARK (4) xmkmf -a (If you have already compiled cfh101 and kept the .o files, only cfh101-2.0/mxw and cfh101-2.0/src have changed, so you can save a few steps in the 'make's: cd mxw;make;cd ../src;make Or you can just remake the whole thing) (5) Edit your ~/.chimera/resources file to point at a chimera_bookmarkv2 file. (6) Away you go. It seems to work ... Michael ------- Message 50 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa10618; 29 Feb 96 12:05 PST Received: from citi.umich.edu by citi.umich.edu for bug-chimera@cs.unlv.edu with SMTP; Thu, 29 Feb 96 15:04:15 -0500 From: Jim Rees To: Chimera Lovers Date: Thu, 29 Feb 1996 15:04:14 -0500 Subject: open shortcut Sender: rees@citi.umich.edu Here's a shortcut for opening urls. If you give a url with no '/' in it, chimera will add "https://" to it. For example, "foo.com" becomes "https://foo.com/". The only side effect is that if you want to open a local file in the current directory, you must give its name as "./foo". I wish I'd thought of this, but I got the idea from Netscape. This patch includes all the other patches to main.c. The last hunk is the only one needed for the url shortcut. This is for chimera 1.65. diff -cb main.c- main.c *** main.c- Fri Mar 31 20:45:19 1995 - --- main.c Thu Feb 29 14:03:02 1996 *************** *** 121,126 **** - --- 121,128 ---- offset(showURL), XtRImmediate, (XtPointer)True }, { "showTitle", XtCBoolean, XtRBoolean, sizeof(Boolean), offset(showTitle), XtRImmediate, (XtPointer)True }, + { "anchorDisplay", XtCBoolean, XtRBoolean, sizeof(Boolean), + offset(anchorDisplay), XtRImmediate, (XtPointer)False }, { "button1Box", "BoxList", XtRString, sizeof(char *), offset(button1Box), XtRString, (XtPointer)BUTTON_LIST }, { "button2Box", "BoxList", XtRString, sizeof(char *), *************** *** 918,927 **** } /* ! * Convert needs to be called in case there is a transfer or content ! * encoding. */ ! if (d->ptext == NULL) { /* * This frightens me. I know it is bad but I can't think of anything - --- 920,929 ---- } /* ! * Convert needs to be called in case there is a content transfer or content ! * encoding. But not for downloads, says Roman. */ ! if (d->ptext == NULL && ! download) { /* * This frightens me. I know it is bad but I can't think of anything *************** *** 949,955 **** if (d->status == DS_OK && (download || (d->content == NULL && d->pcontent == NULL))) { ! CreateStrReq("filename", root.savestr, False, SaveOCallback, SaveDCallback, (XtPointer)d); } else if ((d->content != NULL && strcasecmp(d->content, "text/html") == 0) || - --- 951,976 ---- if (d->status == DS_OK && (download || (d->content == NULL && d->pcontent == NULL))) { ! ! /* Let's offer the save the downloaded file under the same name. ! Roman Czyborra, 1995-04-13 */ ! ! char *filename=NULL, *basename; int length; ! ! if (d->up) if (filename= d->up->filename) if (length = strlen(filename)) ! { ! if (filename[length-1]=='/') ! { ! filename[length-1]='\0'; ! } ! if (basename = strrchr (filename, '/')) ! { ! filename = basename + 1; ! } ! } ! if (! filename) filename=""; ! ! CreateStrReq("filename", filename, False, SaveOCallback, SaveDCallback, (XtPointer)d); } else if ((d->content != NULL && strcasecmp(d->content, "text/html") == 0) || *************** *** 979,985 **** else { DestroyDocument(d); ! d = NULL; } } else - --- 1000,1006 ---- else { DestroyDocument(d); ! return; } } else *************** *** 1066,1071 **** - --- 1087,1109 ---- return; } + /* rwmcm */ + /* + * AnchorURLDisplay + * + * Called when the mouse moves over an anchor + */ + void + AnchorURLDisplay(w, c) + Widget w; + char *c; + { + Arg args[5]; + + XtSetArg(args[0], XtNlabel, c); + XtSetValues(root.anchordisplay,args,1); + } + /* * OpenDocument * *************** *** 1089,1097 **** else { if (root.loadstr) free_mem(root.loadstr); root.loadstr = alloc_string(url); ! up = ParseURL(url); if (up != NULL) { d = LoadDoc(up, 0, 0); - --- 1127,1143 ---- else { if (root.loadstr) free_mem(root.loadstr); + /* + * short cut -- if user types in "foo.com" turn that into "https://foo.com/" + * this means you must type "./foo.com" if opening a local file by that name + */ + if (!strchr(url, '/')) { + root.loadstr = alloc_mem(strlen(url) + 9); + sprintf(root.loadstr, "https://%s/", url); + } else root.loadstr = alloc_string(url); ! up = ParseURL(root.loadstr); if (up != NULL) { d = LoadDoc(up, 0, 0); ------- End of Forwarded Messages