Received: from jimi.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa25993; 22 Dec 94 22:57 PST To: jay@JIMI.CS.UNLV.EDU Subject: bug-chimera nov 94 Date: Thu, 22 Dec 1994 22:57:43 -0800 From: Jay Nietling ------- Forwarded Messages Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa09494; 1 Nov 94 0:57 PST To: bug-chimera@cephas.ISRI.UNLV.EDU Subject: chimera 1.62 notes Date: Tue, 01 Nov 1994 00:57:19 -0800 From: John Kilburg I will try to release 1.62 within a couple of days. I am trying to fix a problem with the authentication code and then I will release it. Here is the change list for 1.62: The bookmarks can now be grouped. In fact, there has to be at least one group. A new bookmark file format was created. 1.62 can read the old format but not write it. Changed test-chimera so that it works a little better. Added to util.c Create lang.c and stringdb.c to deal with alternate language stuff and to provide a way of passing environment strings and X resources around. Added X resources for httpProxy, ftpProxy, gopherProxy, nntpProxy, newsProxy, urnProxy, waisProxy. The authentication code changed again. Now if a document has the same protocol and hostname as its parent the authentication information is passed from the parent to the child so that the user doesn't have to type the auth info again. This may not be exactly right but it should keep the auth info from being sent to the wrong site. Changed help.html to explain delayImageLoads. Removed DELAY_IMAGE_LOADS from options.h because it wasn't used. The cursor gets warped to the middle of the text widget on string popups. This sucks but I really didn't feel like messing around with making the text widget appear under the cursor. Fixed a problem with the delayed image load code. I am making radical changes to a newer version of chimera so 1.62 will probably be the last stable release for a long time. Hint: the new version (2.0) currently does multiple downloads at one time and incorporates the GIF reader into chimera. -john ------- Message 2 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa17634; 1 Nov 94 7:00 PST Received: from citi.umich.edu by citi.umich.edu for john@cephas.ISRI.UNLV.EDU bug-chimera@cephas.ISRI.UNLV.EDU with SMTP; Tue, 01 Nov 94 09:59:17 -0500 From: Jim.Rees@umich.edu To: John Kilburg Cc: bug-chimera@cephas.ISRI.UNLV.EDU Date: Tue, 01 Nov 94 09:59:09 EST Subject: Re: chimera 1.62 notes Sender: rees@citi.umich.edu In-Reply-To: John Kilburg, Tue, 01 Nov 94 00:57:19 PST Added to util.c Hooray! Create lang.c and stringdb.c to deal with alternate language stuff and to provide a way of passing environment strings and X resources around. When you say "alternate language" do you mean alternate character sets? How is this going to work? ------- Message 3 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa10042; 1 Nov 94 15:59 PST To: bug-chimera@cephas.ISRI.UNLV.EDU Subject: Re: chimera 1.62 notes In-reply-to: Your message of "Tue, 01 Nov 1994 09:59:09 EST." Date: Tue, 01 Nov 1994 15:59:25 -0800 From: John Kilburg > Added to util.c > >Hooray! > > Create lang.c and stringdb.c to deal with alternate language stuff and > to provide a way of passing environment strings and X resources around. > >When you say "alternate language" do you mean alternate character sets? >How is this going to work? I mean you can replace all of the english/american strings (error and informational messages) with whatever you want. The original strings are still hardcoded but are in one file (lang.c) and a file can be specified with a resource that holds alternatives. A function that needs a message gives a string (say "noload") to a function and gets back the appropriate message to use. This is kind of bogus but I did not want to use the X resources for messages (mainly because I didn't want to mess with having a big global structure running around). If someone has a better idea, please let me know. -john ------- Message 4 Received: from concorde.inria.fr by JIMI.CS.UNLV.EDU id aa11518; 3 Nov 94 5:18 PST Received: from fantomas.inria.fr (fantomas.inria.fr [128.93.11.34]) by concorde.inria.fr (8.6.9/8.6.9) with ESMTP id OAA16865 for ; Thu, 3 Nov 1994 14:18:14 +0100 Received: (from lasgout@localhost) by fantomas.inria.fr (8.6.8/8.6.6) id OAA24533; Thu, 3 Nov 1994 14:18:08 +0100 Date: Thu, 3 Nov 1994 14:18:08 +0100 From: Jean-Marc Lasgouttes Message-Id: <199411031318.OAA24533@fantomas.inria.fr> To: bug-chimera@cs.unlv.edu Subject: Problem with test-chimera. Reply-to: Jean-Marc.Lasgouttes@inria.fr I ran into a problem with test-chimera with version 1.62. test-chimera adds ressources with xrdb -merge, but does not restores the resources to their original state after execution. This implies that I could not run a normal chimera session after running test-chimera. I guess that the fix is simple, but I do not understand xrdb enough to propose a fix. Concerning the cursor warp on popups, on my machine the cursor does not warp in the text widget, so the popup does not accept input unless I move the cursor at the right place. Would it be possible to make the text widget active as soon as the popup has the focus (as it is done in xrn)? Jean-Marc. ------- Message 5 Received: from concorde.inria.fr by JIMI.CS.UNLV.EDU id aa12748; 3 Nov 94 6:13 PST Received: from fantomas.inria.fr (fantomas.inria.fr [128.93.11.34]) by concorde.inria.fr (8.6.9/8.6.9) with ESMTP id PAA19134 for ; Thu, 3 Nov 1994 15:12:58 +0100 Received: (from lasgout@localhost) by fantomas.inria.fr (8.6.8/8.6.6) id PAA24628; Thu, 3 Nov 1994 15:12:57 +0100 Date: Thu, 3 Nov 1994 15:12:57 +0100 From: Jean-Marc Lasgouttes Message-Id: <199411031412.PAA24628@fantomas.inria.fr> To: bug-chimera@cs.unlv.edu Subject: Bookmarks in Chimera 1.62 Reply-to: Jean-Marc.Lasgouttes@inria.fr I played a bit with the new bookmark format of Chimera and had found the following: - - the bookmark file is not saved is you only add a group, but no bookmarks. - - it would be nice to have the first group automatically opened when you invoke Bookmarks for the first time. Also, I'd like to use the second mouse button to open a group (in addition to double click with first button). - - for symmetry with the 'group' buttons, I'd rather see buttons like 'open item', 'add item' and 'delete item' below the item list, and a separate 'Dismiss' button. The current layout could be misleading. - - I'd like to suppress the internal borders in all windows except Bookmarks window. What shall I do? - - On a related note, I'd like to turn the background color of the HTML widget to grey (for mosaic compatibility :-(), without turning all windows to grey. How can I do that? Jean-Marc. ------- Message 6 Received: from cheviot.ncl.ac.uk by JIMI.CS.UNLV.EDU id aa13737; 3 Nov 94 6:55 PST Received: from blagdon.ncl.ac.uk by cheviot.ncl.ac.uk id (8.6.9/ for ncl.ac.uk) with SMTP; Thu, 3 Nov 1994 14:55:48 GMT Received: (njkw@localhost) by blagdon.ncl.ac.uk (8.6.7/8.6.x-cf revision 8 for SunOS 4.1.x) id OAA20180; Thu, 3 Nov 1994 14:55:46 GMT Date: Thu, 3 Nov 1994 14:55:46 GMT From: Jim Wight Message-Id: <199411031455.OAA20180@blagdon.ncl.ac.uk> To: bug-chimera@cs.unlv.edu In-reply-to: <199411031318.OAA24533@fantomas.inria.fr> Subject: Re: Problem with test-chimera. Reply-to: J.K.Wight@newcastle.ac.uk Jean-Marc Lasgouttes writes: > > > I ran into a problem with test-chimera with version 1.62. test-chimera > adds ressources with xrdb -merge, but does not restores the resources > to their original state after execution. This implies that I could not > run a normal chimera session after running test-chimera. > > I guess that the fix is simple, but I do not understand xrdb enough to > propose a fix. It would be much better if the resources that are currently merged into the server were held in a file, e.g. src/Chimera.ad.test, and the environment variable XUSERFILESEARCHPATH set to point to it. Thus, XFILESEARCHPATH=$PWD/src/Chimera.ad XUSERFILESEARCHPATH=$PWD/src/Chimera.ad.test export XFILESEARCHPATH XUSERFILESEARCHPATH if [ "$1" = "d" ] then cd src if [ -f ../core ] then gdb ./chimera ../core else gdb ./chimera fi else src/chimera $@ fi Alternatively test-chimera could combine Chimera.ad and the additional resources in a temporary file and point XFILESEARCHPATH at that. It is definitely a bad idea to mess with the server resources. Jim ------- Message 7 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa28763; 3 Nov 94 15:11 PST To: Jean-Marc.Lasgouttes@inria.fr cc: bug-chimera@cs.unlv.edu Subject: Re: Problem with test-chimera. In-reply-to: Your message of "Thu, 03 Nov 1994 14:18:08 +0100." <199411031318.OAA24533@fantomas.inria.fr> Date: Thu, 03 Nov 1994 15:11:14 -0800 From: John Kilburg >I ran into a problem with test-chimera with version 1.62. test-chimera >adds ressources with xrdb -merge, but does not restores the resources >to their original state after execution. This implies that I could not >run a normal chimera session after running test-chimera. > >I guess that the fix is simple, but I do not understand xrdb enough to >propose a fix. Type: xrdb -remove this removes all of your resources...I don't know how to do it selectively. I don't have any custom resources so I don't worry about it. >Concerning the cursor warp on popups, on my machine the cursor does >not warp in the text widget, so the popup does not accept input unless >I move the cursor at the right place. Would it be possible to make the >text widget active as soon as the popup has the focus (as it is done >in xrn)? Ah! OK. Warping is a bad idea. -john ------- Message 8 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa11812; 3 Nov 94 22:51 PST Received: from citi.umich.edu by citi.umich.edu for bug-chimera@cephas.ISRI.UNLV.EDU with SMTP; Fri, 04 Nov 94 01:50:36 -0500 From: Jim.Rees@umich.edu To: bug-chimera@cephas.ISRI.UNLV.EDU Date: Fri, 04 Nov 94 01:50:28 EST Subject: gopher types Sender: rees@citi.umich.edu In chimera 1.61, gopher images sometimes get loaded as text rather than images. Here's an example: gopher://gopher.archive.merit.edu:7055/II/apollo/dn330-sm.gif I don't know enough about gopher to say whether this is a bug in chimera or in the gopher server. ------- Message 9 Received: from nova.gmi.edu by JIMI.CS.UNLV.EDU id aa19572; 4 Nov 94 4:17 PST Received: by nova.gmi.edu (4.1/SMI-4.1-DNI) id AA25669; Fri, 4 Nov 94 07:22:03 EST Date: Fri, 4 Nov 1994 07:22:02 -0500 (EST) From: "R. Stewart Ellis" To: Jim.Rees@umich.edu Cc: bug-chimera@cephas.ISRI.UNLV.EDU Subject: Re: gopher types In-Reply-To: <9411040711.AA11940@nova.gmi.edu> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Fri, 4 Nov 1994 Jim.Rees@umich.edu wrote: > In chimera 1.61, gopher images sometimes get loaded as text rather than > images. Here's an example: > > gopher://gopher.archive.merit.edu:7055/II/apollo/dn330-sm.gif > > I don't know enough about gopher to say whether this is a bug in chimera or > in the gopher server. > I just checked this with the UMN gopher client. I did not check it with any web browser other than chimera 1.60, which opened it as a GIF. From serving graphic URL's from a gopher myself, I believe the type designator should be I9 rather than II. R.Stewart(Stew) Ellis, Assoc.Prof., (Off)313-762-9765 ___________________ Humanities & Social Science, GMI Eng.& Mgmt. Inst. / _____ ______ Flint, MI 48504 ellis@nova.gmi.edu / / / / / / Gopher,chimera,nn,tin,jove,modems, free code is best!/________/ / / / / ------- Message 10 Received: from fsb1.aist-nara.ac.jp by JIMI.CS.UNLV.EDU id aa20870; 4 Nov 94 5:24 PST Received: from alpha234.aist-nara.ac.jp by mailgate.aist-nara.ac.jp (8.6.9+2.4W/2.8Wb/NAIST-1.5[gate]) id WAA20735; Fri, 4 Nov 1994 22:23:59 +0900 Return-Path: Received: from localhost by alpha234.aist-nara.ac.jp (5.67+1.6W[kuis-17]/2.8Wb/NAIST-1.3[is]) id AA19020; Fri, 4 Nov 94 22:23:57 GMT+0900 Message-Id: <9411041323.AA19020@alpha234.aist-nara.ac.jp> To: john@cephas.isri.unlv.edu Cc: bug-chimera@cephas.isri.unlv.edu Subject: Re: chimera 1.62 notes Reply-To: k-chinen@is.aist-nara.ac.jp In-Reply-To: Your message of "Tue, 01 Nov 1994 00:57:19 -0800" References: <199411010915.SAA13684@mailgate.aist-nara.ac.jp> X-Mailer: Mew beta version 0.73 on Emacs 18.59.1, Mule 1.1 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 04 Nov 1994 22:23:57 +0900 From: Kenichi Chinen Hi john. It is not bug-report. It is request of New feature. I want reload promotion for proxy-server when chimera reloading anything via proxy-server. Thanks. - --- k-chinen@is.aist-nara.ac.jp ------- Message 11 Received: from masschaos.de.convex.com by JIMI.CS.UNLV.EDU id aa26600; 4 Nov 94 8:45 PST Received: from localhost.de.convex.com by masschaos.de.convex.com (4.1/1.35r) id AA14413; Fri, 4 Nov 94 17:45:00 +0100 Message-Id: <9411041645.AA14413@masschaos.de.convex.com> To: bug-chimera@cs.unlv.edu Subject: Chimera 1.61 under SunOS 4.1.3 Date: Fri, 04 Nov 1994 17:44:59 +0100 From: Ray Davis I thought I would check out chimera, so I built it under SunOS 4.1.3. It likes to memory fault core dump quite often. For example, I open a local html file, then click on a pointer to a local regular file, and boom. Would anyone already have a clue what's wrong so I don't have to debug the thing? In fact I'd rather deinstall it than debug it. No time. Here are the things I changed while following the INSTALL procedure: $ diff Common.tmpl Common.tmpl.dist 12,15c12,15 < CBINDIR = /usr/local/bin < CUTILDIR = /usr/local/chimera/util < CLIBDIR = /usr/local/chimera < CMANDIR = /usr/local/man --- > CBINDIR = $(BINDIR) > CUTILDIR = $(BINDIR) > CLIBDIR = $(LIBDIR)/chimera > CMANDIR = $(MANDIR) $ diff options.h options.h.dist (no output) Thanks, Ray ------- Message 12 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa02868; 4 Nov 94 11:42 PST To: Ray Davis cc: bug-chimera@cs.unlv.edu Subject: Re: Chimera 1.61 under SunOS 4.1.3 In-reply-to: Your message of "Fri, 04 Nov 1994 17:44:59 +0100." <9411041645.AA14413@masschaos.de.convex.com> Date: Fri, 04 Nov 1994 11:42:46 -0800 From: John Kilburg >I thought I would check out chimera, so I built it under SunOS 4.1.3. >It likes to memory fault core dump quite often. For example, I open >a local html file, then click on a pointer to a local regular file, >and boom. > >Would anyone already have a clue what's wrong so I don't have to >debug the thing? In fact I'd rather deinstall it than debug it. >No time. Make sure the utilities are in your path. Chimera needs one of them to convert ASCII text to HTML. -john ------- Message 13 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa05926; 7 Nov 94 0:03 PST To: bug-chimera@cephas.ISRI.UNLV.EDU Subject: chimera 1.62 beta 3 Date: Mon, 07 Nov 1994 00:03:39 -0800 From: John Kilburg ftp://ftp.cs.unlv.edu/pub/chimera-misc/xc3-1.62.tar.gz This version contains mostly fixes for authentication problems. It also features a decent test-chimera (I think). The focus is now set correctly for the string popup. There were some other changes that I don't remember. If you access Hotwired then I highly recommend that you pick this one up. I just updated the CHANGES file (after making the tar.gz) and I have not extensively tested it but please grab it and look for any nasty problems...I would like to go into bug fixing mode (instead of significant coding mode) on 1.62 soon. -john ------- Message 14 Received: from tel13.tel.vtt.fi by JIMI.CS.UNLV.EDU id aa09930; 7 Nov 94 2:12 PST Received: by tel13.tel.vtt.fi id AA16971 (5.65c/IDA-1.4.4 for bug-chimera@cephas.ISRI.UNLV.EDU); Mon, 7 Nov 1994 12:11:50 +0200 Date: Mon, 7 Nov 1994 12:11:50 +0200 From: Markku Savela Message-Id: <199411071011.AA16971@tel13.tel.vtt.fi> To: bug-chimera@cephas.ISRI.UNLV.EDU In-Reply-To: John Kilburg's message of Mon, 07 Nov 1994 00:03:39 -0800 <199411070835.AA00141@tel.vtt.fi> Subject: chimera 1.62 beta 3 Reply-To: Markku Savela Trying this on HPUX 9.x in non-ANSI mode (with MIT X11R5 libs) >ftp://ftp.cs.unlv.edu/pub/chimera-misc/xc3-1.62.tar.gz 'xmkmf -a' went okay, but in make I get making all in ./src... cc -I../libhtmlw -I../mxw -I../ -I/usr/X11R5/include -DSYSV -DXRELEASE=5 -DCONVERT_FILES=\"~/.chimera_convert:/usr/local/lib/chimera/convert\" -DMIME_TYPE_FILES=\"~/.chimera_mime.types:/usr/local/lib/chimera/mime.types\" -DMAIL_CAPS=\"~/.mailcap:/usr/local/lib/chimera/mailcap\" -DPROTOCOL_FILES=\"~/.chimera_protocol:/usr/local/lib/chimera/protocol\" -DHELP_URL=\"file:/usr/local/lib/chimera/help.html\" -DHOME_URL=\"file:/usr/local/lib/chimera/home.html\" -c main.c cc: "main.c", line 95: error 1705: Function prototypes are an ANSI feature. ... this is easy to fix, you left int DisplayTransferStatus(int, int); into wrong place. The next thing is the old . SYSV is defined for HPUX xmkmf. No idea what the correct fix is (I just took off the 'defined(SYSV) ||' for now.. ) cc -I../libhtmlw -I../mxw -I../ -I/usr/X11R5/include -DSYSV -DXRELEASE=5 -DCONVERT_FILES=\"~/.chimera_convert:/usr/local/lib/chimera/convert\" -DMIME_TYPE_FILES=\"~/.chimera_mime.types:/usr/local/lib/chimera/mime.types\" -DMAIL_CAPS=\"~/.mailcap:/usr/local/lib/chimera/mailcap\" -DPROTOCOL_FILES=\"~/.chimera_protocol:/usr/local/lib/chimera/protocol\" -DHELP_URL=\"file:/usr/local/lib/chimera/help.html\" -DHOME_URL=\"file:/usr/local/lib/chimera/home.html\" -c input.c cpp: "input.c", line 17: error 4036: Can't open include file 'sys/select.h'. Running './test-chimera' works now nicely. I still have the strange problem with inline images appearing rather random, they appear to come and go. (For example, I have several blueball.gif's in a doc, and sometimes they all appear as X's and sometimes some are shown and rest is X's, when repeatingly doing reload). Is this some delayed image loading problem? Parallel stuff messing up something? I notice that images appear more deterministic when I am using a real HTTP (get documents via server). In that case some images first appear as X's, but repeated loads get more of them and they stick. The randomness happens when I access the local pages directly, with references like file://localhost/usr/local/etc/httpd/htdocs/home.html Hmm.. perhaps the caching gets confused when exactly same file/document/image is accessed via http: and file: ? - -- msa ------- Message 15 Received: from igw.merck.com by JIMI.CS.UNLV.EDU id aa22533; 7 Nov 94 11:50 PST Received: by igw.merck.com (5.65/fma-120691); id AA24518; Mon, 7 Nov 94 14:56:59 -0500 Message-Id: <9411071956.AA24518@igw.merck.com> Date: Mon, 7 Nov 1994 14:46:20 -0500 From: Anthony Starks To: bug-chimera@cs.unlv.edu Subject: 1.62 beta == molasses in January the 1.62 beta is dog slow. could this be a combination of the disk-based cache loading and processing and the fact that each image has fork/exec overhead? It's gotten to the point of un-usability! ------- Message 16 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa25918; 7 Nov 94 13:48 PST To: Anthony Starks cc: bug-chimera@cs.unlv.edu Subject: Re: 1.62 beta == molasses in January In-reply-to: Your message of "Mon, 07 Nov 1994 14:46:20 EST." <9411071956.AA24518@igw.merck.com> Date: Mon, 07 Nov 1994 13:48:16 -0800 From: John Kilburg >the 1.62 beta is dog slow. >could this be a combination of the disk-based cache loading >and processing and the fact that each image has fork/exec overhead? > >It's gotten to the point of un-usability! Do you have timing information from 1.61 and the 1.62 beta? If you send me a URL that really slows 1.62 down but not earlier releases then I will try to improve it a bit. 1.62 doesn't fork/exec more than any other version as far as I know. At least it shouldn't. If you are using a DECstation then maybe its not chimera's fault. :) :) -john ------- Message 17 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa28423; 7 Nov 94 14:48 PST Received: from citi.umich.edu by citi.umich.edu for savela@tel.vtt.fi bug-chimera@cephas.ISRI.UNLV.EDU with SMTP; Mon, 07 Nov 94 17:47:06 -0500 From: Jim.Rees@umich.edu To: Markku Savela Cc: bug-chimera@cephas.ISRI.UNLV.EDU Date: Mon, 07 Nov 1994 17:47:05 -0500 Subject: Re: chimera 1.62 beta 3 Sender: rees@citi.umich.edu In-Reply-To: Markku Savela, Mon, 07 Nov 1994 12:11:50 +0200 Running './test-chimera' works now nicely. I still have the strange problem with inline images appearing rather random, they appear to come and go... I notice that images appear more deterministic when I am using a real HTTP (get documents via server). In that case some images first appear as X's, but repeated loads get more of them and they stick. I can confirm all of this, running chimera 1.62 on aix3.2.5. It's similar to the select/stdio bug, but is non-deterministic. I get about half the inline images on the first try, and more on each successive try. I get a different set of inlines every time I load. ------- Message 18 Received: from fsb1.aist-nara.ac.jp by JIMI.CS.UNLV.EDU id aa01006; 9 Nov 94 3:52 PST Received: from alpha234.aist-nara.ac.jp by mailgate.aist-nara.ac.jp (8.6.9+2.4W/2.8Wb/NAIST-1.5[gate]) id UAA28057; Wed, 9 Nov 1994 20:50:51 +0900 Return-Path: Received: from localhost by alpha234.aist-nara.ac.jp (5.67+1.6W[kuis-17]/2.8Wb/NAIST-1.3[is]) id AA21750; Wed, 9 Nov 94 20:50:49 GMT+0900 Message-Id: <9411091150.AA21750@alpha234.aist-nara.ac.jp> To: john@cephas.isri.unlv.edu Cc: bug-chimera@cephas.isri.unlv.edu Subject: Re: chimera 1.62 notes Reply-To: k-chinen@is.aist-nara.ac.jp In-Reply-To: Your message of "Fri, 04 Nov 1994 22:23:57 +0900" References: <9411041323.AA19020@alpha234.aist-nara.ac.jp> X-Mailer: Mew beta version 0.73 on Emacs 18.59.1, Mule 1.1 Mime-Version: 1.0 Content-Type: Multipart/Mixed; boundary="--Next_Part(Wed_Nov__9_20:50:45_1994)--" Date: Wed, 09 Nov 1994 20:50:49 +0900 From: Kenichi Chinen - ----Next_Part(Wed_Nov__9_20:50:45_1994)-- Content-Type: Text/plain; charset=us-ascii Hi john. It is not bug-report. It is sample of patch for 'Pragma: no-cache'. From: Kenichi Chinen Subject: Re: chimera 1.62 notes Date: Fri, 04 Nov 1994 22:23:57 +0900 > I want reload promotion for proxy-server when chimera reloading > anything via proxy-server. If send 'Progma: no-cache' to proxy server over HTTP/1.0 . Proxy server try to GET and reply such answer. Is it true ? This patch only support HTTP. Thanks. - --- k-chinen@is.aist-nara.ac.jp - ----Next_Part(Wed_Nov__9_20:50:45_1994)-- Content-Type: text/plain; charset=us-ascii diff -c ../orig-chimera-1.61/src/cache.c src/cache.c *** ../orig-chimera-1.61/src/cache.c Tue Oct 25 09:06:57 1994 - --- src/cache.c Wed Nov 9 13:39:31 1994 *************** *** 221,228 **** MD5Final(digest, &context); free(url); ! cp = hex; ! for (ep = (dp = digest) + (sizeof(digest) / sizeof(digest[0])); dp < ep; cp += 2) { (void)sprintf(cp, "%02x", *dp++ & 0xff); } - --- 221,228 ---- MD5Final(digest, &context); free(url); ! cp = (char*) hex; ! for (ep = (dp = (char*) digest) + (sizeof(digest) / sizeof(digest[0])); dp < ep; cp += 2) { (void)sprintf(cp, "%02x", *dp++ & 0xff); } diff -c ../orig-chimera-1.61/src/document.c src/document.c *** ../orig-chimera-1.61/src/document.c Thu Oct 27 08:31:20 1994 - --- src/document.c Wed Nov 9 20:34:29 1994 *************** *** 106,114 **** * to get information. Allows new protocols without changing chimera. */ Document * ! ExternalProtocol(up, plist) URLParts *up; Protocol *plist; { Document *d; Protocol *p; - --- 106,115 ---- * to get information. Allows new protocols without changing chimera. */ Document * ! ExternalProtocol(up, plist, reload) URLParts *up; Protocol *plist; + int reload; { Document *d; Protocol *p; *************** *** 167,175 **** * Handles proxy protocols. */ Document * ! ProxyProtocol(up, mtlist) URLParts *up; MIMEType *mtlist; { int i, len; Document *d = NULL; - --- 168,177 ---- * Handles proxy protocols. */ Document * ! ProxyProtocol(up, mtlist, reload) URLParts *up; MIMEType *mtlist; + int reload; { int i, len; Document *d = NULL; *************** *** 219,225 **** { char *msg = "

Error

Invalid proxy URL. The proxy URL must be an absolute URL."; ! if (IsAbsoluteURL(rup)) d = http_proxy(rup, up, mtlist); else d = BuildDocument(msg, strlen(msg), "text/html", 1, 0); DestroyURLParts(rup); - --- 221,227 ---- { char *msg = "

Error

Invalid proxy URL. The proxy URL must be an absolute URL."; ! if (IsAbsoluteURL(rup)) d = http_proxy(rup, up, mtlist, reload); else d = BuildDocument(msg, strlen(msg), "text/html", 1, 0); DestroyURLParts(rup); *************** *** 239,247 **** * Handle protocols that have been implemented in the chimera source. */ static Document * ! LocalProtocol(up, mtlist) URLParts *up; MIMEType *mtlist; { int i; Document *d = NULL; - --- 241,250 ---- * Handle protocols that have been implemented in the chimera source. */ static Document * ! LocalProtocol(up, mtlist, reload) URLParts *up; MIMEType *mtlist; + int reload; { int i; Document *d = NULL; *************** *** 258,264 **** if (protocols[i].func != NULL && strcmp(protocols[i].name, up->protocol) == 0) { ! d = (protocols[i].func)(up, mtlist); break; } } - --- 261,267 ---- if (protocols[i].func != NULL && strcmp(protocols[i].name, up->protocol) == 0) { ! d = (protocols[i].func)(up, mtlist, reload); break; } } *************** *** 366,384 **** * LoadDocument */ Document * ! LoadDocument(up, plist, mtlist) URLParts *up; Protocol *plist; MIMEType *mtlist; { Document *d; /* * Call a function to grab the URL. */ ! if ((d = ExternalProtocol(up, plist)) != NULL) ; ! else if ((d = ProxyProtocol(up)) != NULL) ; ! else d = LocalProtocol(up, mtlist); /* * Now figure out what we've got and do something with it. - --- 369,388 ---- * LoadDocument */ Document * ! LoadDocument(up, plist, mtlist, reload) URLParts *up; Protocol *plist; MIMEType *mtlist; + int reload; { Document *d; /* * Call a function to grab the URL. */ ! if ((d = ExternalProtocol(up, plist, reload)) != NULL) ; ! else if ((d = ProxyProtocol(up, mtlist, reload)) != NULL) ; ! else d = LocalProtocol(up, mtlist, reload); /* * Now figure out what we've got and do something with it. *************** *** 410,416 **** if (tup != NULL) { DestroyDocument(d); ! return(LoadDocument(tup, plist, mtlist)); } else /* bad relocation URL given */ { - --- 414,420 ---- if (tup != NULL) { DestroyDocument(d); ! return(LoadDocument(tup, plist, mtlist, reload)); } else /* bad relocation URL given */ { diff -c ../orig-chimera-1.61/src/document.h src/document.h *** ../orig-chimera-1.61/src/document.h Tue Oct 4 16:23:23 1994 - --- src/document.h Wed Nov 9 13:19:14 1994 *************** *** 71,77 **** #if NeedFunctionPrototypes Protocol *ReadProtocolFiles(char *); ! Document *LoadDocument(URLParts *, Protocol *, MIMEType *); void DestroyDocument(Document *); Document *CreateDocument(); Document *BuildDocument(char *, int, char *, int, int); - --- 71,77 ---- #if NeedFunctionPrototypes Protocol *ReadProtocolFiles(char *); ! Document *LoadDocument(URLParts *, Protocol *, MIMEType *, int); void DestroyDocument(Document *); Document *CreateDocument(); Document *BuildDocument(char *, int, char *, int, int); diff -c ../orig-chimera-1.61/src/ftp.c src/ftp.c *** ../orig-chimera-1.61/src/ftp.c Thu Oct 27 08:37:36 1994 - --- src/ftp.c Wed Nov 9 13:13:49 1994 *************** *** 121,129 **** * necessary error checking being done. */ Document * ! ftp(up, mtlist) URLParts *up; MIMEType *mtlist; { Document *doc; char *query; - --- 121,130 ---- * necessary error checking being done. */ Document * ! ftp(up, mtlist, reload) URLParts *up; MIMEType *mtlist; + int reload; { Document *doc; char *query; diff -c ../orig-chimera-1.61/src/gopher.c src/gopher.c *** ../orig-chimera-1.61/src/gopher.c Thu Oct 27 08:30:52 1994 - --- src/gopher.c Wed Nov 9 13:13:18 1994 *************** *** 316,324 **** * This is the gopher+ frontend to gopher_main */ Document * ! gopherplus(up, mtlist) URLParts *up; MIMEType *mtlist; { return(gopher_main(up, mtlist, 1)); } - --- 316,325 ---- * This is the gopher+ frontend to gopher_main */ Document * ! gopherplus(up, mtlist, reload) URLParts *up; MIMEType *mtlist; + int reload; { return(gopher_main(up, mtlist, 1)); } *************** *** 329,337 **** * The plain gopher frontend to gopher_main */ Document * ! gopherplain(up, mtlist) URLParts *up; MIMEType *mtlist; { return(gopher_main(up, mtlist, 0)); } - --- 330,339 ---- * The plain gopher frontend to gopher_main */ Document * ! gopherplain(up, mtlist, reload) URLParts *up; MIMEType *mtlist; + int reload; { return(gopher_main(up, mtlist, 0)); } diff -c ../orig-chimera-1.61/src/http.c src/http.c *** ../orig-chimera-1.61/src/http.c Thu Oct 27 08:48:47 1994 - --- src/http.c Wed Nov 9 20:21:25 1994 *************** *** 574,606 **** * Sends an HTTP request */ int ! http_request(fd, up, type) int fd; URLParts *up; int type; { char *query; ! if (type == 0) query = build_request(up, NULL, NULL, 1); else if (type == 1) { char *filename; filename = MakeURL(up); ! query = build_request(up, filename, NULL, 1); free(filename); } else { char *extra_header; char *url; ! static char *format = "\ URI: %s\n\ X-protocol: %s\n\ X-hostname: %s\n\ X-port: %d\n\ X-filename: %s\n"; url = MakeURL(up); extra_header = alloc_mem(strlen(url) * 5 + strlen(format) + 1); sprintf (extra_header, - --- 574,624 ---- * Sends an HTTP request */ int ! http_request(fd, up, type, reload) int fd; URLParts *up; int type; + int reload; { char *query; ! if (type == 0) ! query = build_request(up, NULL, NULL, 1); else if (type == 1) { char *filename; filename = MakeURL(up); ! if(reload) ! query = build_request(up, filename, "Pragma: no-cache\n", 1); ! else ! query = build_request(up, filename, NULL, 1); free(filename); } else { char *extra_header; char *url; ! char *format; ! static char *format_normal = "\ ! URI: %s\n\ ! X-protocol: %s\n\ ! X-hostname: %s\n\ ! X-port: %d\n\ ! X-filename: %s\n"; ! static char *format_reload = "\ URI: %s\n\ + Pragma: no-cache\n\ X-protocol: %s\n\ X-hostname: %s\n\ X-port: %d\n\ X-filename: %s\n"; + if(reload) + format = format_reload; + else + format = format_normal; + url = MakeURL(up); extra_header = alloc_mem(strlen(url) * 5 + strlen(format) + 1); sprintf (extra_header, *************** *** 636,645 **** * Talk to an HTTP server. */ static Document * ! http_main(up, rup, mtlist) URLParts *up; URLParts *rup; MIMEType *mtlist; { Document *d; int s; - --- 654,664 ---- * Talk to an HTTP server. */ static Document * ! http_main(up, rup, mtlist, reload) URLParts *up; URLParts *rup; MIMEType *mtlist; + int reload; { Document *d; int s; *************** *** 650,656 **** s = net_open(up->hostname, up->port == 0 ? DEFAULT_HTTP_PORT:up->port); if (s < 0) return(NULL); ! if (http_request(s, rup != NULL ? rup:up, rup != NULL ? 1:0) == -1) { net_close(s); return(NULL); - --- 669,675 ---- s = net_open(up->hostname, up->port == 0 ? DEFAULT_HTTP_PORT:up->port); if (s < 0) return(NULL); ! if (http_request(s, rup != NULL ? rup:up, rup != NULL ? 1:0, reload) == -1) { net_close(s); return(NULL); *************** *** 699,709 **** * Frontend for regular HTTP */ Document * ! http(up, mtlist) URLParts *up; MIMEType *mtlist; { ! return(http_main(up, NULL, mtlist)); } /* - --- 718,729 ---- * Frontend for regular HTTP */ Document * ! http(up, mtlist, reload) URLParts *up; MIMEType *mtlist; + int reload; { ! return(http_main(up, NULL, mtlist, reload)); } /* *************** *** 712,721 **** * Frontend for proxy HTTP */ Document * ! http_proxy(up, rup, mtlist) URLParts *up; URLParts *rup; MIMEType *mtlist; { ! return(http_main(up, rup, mtlist)); } - --- 732,742 ---- * Frontend for proxy HTTP */ Document * ! http_proxy(up, rup, mtlist, reload) URLParts *up; URLParts *rup; MIMEType *mtlist; + int reload; { ! return(http_main(up, rup, mtlist, reload)); } diff -c ../orig-chimera-1.61/src/http.h src/http.h *** ../orig-chimera-1.61/src/http.h Mon Oct 24 14:11:47 1994 - --- src/http.h Wed Nov 9 15:04:29 1994 *************** *** 8,16 **** #if NeedFunctionPrototypes Document *ParseHTTP(int, int); ! Document *http(URLParts *, MIMEType *); ! Document *http_proxy(URLParts *, URLParts *, MIMEType *); ! int http_request(int, URLParts *, int); #else Document *ParseHTTP(); Document *http(); - --- 8,16 ---- #if NeedFunctionPrototypes Document *ParseHTTP(int, int); ! Document *http(URLParts *, MIMEType *, int); ! Document *http_proxy(URLParts *, URLParts *, MIMEType *, int); ! int http_request(int, URLParts *, int, int); #else Document *ParseHTTP(); Document *http(); diff -c ../orig-chimera-1.61/src/main.c src/main.c *** ../orig-chimera-1.61/src/main.c Tue Oct 25 10:04:56 1994 - --- src/main.c Wed Nov 9 20:21:50 1994 *************** *** 612,618 **** SetURL(nup); SetTitle("Downloading document..."); ! d = LoadDocument(nup, root.plist, root.mtlist); SetTitle(""); XDefineCursor(XtDisplay(root.toplevel), XtWindow(root.toplevel), - --- 612,618 ---- SetURL(nup); SetTitle("Downloading document..."); ! d = LoadDocument(nup, root.plist, root.mtlist, reload); SetTitle(""); XDefineCursor(XtDisplay(root.toplevel), XtWindow(root.toplevel), - ----Next_Part(Wed_Nov__9_20:50:45_1994)---- ------- Message 19 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa24183; 9 Nov 94 17:00 PST Received: from citi.umich.edu by citi.umich.edu for bug-chimera@cs.unlv.edu with SMTP; Wed, 09 Nov 94 19:59:02 -0500 From: Jim.Rees@umich.edu To: bug-chimera@cs.unlv.edu Date: Wed, 09 Nov 94 19:58:59 EST Subject: CUTILDIR Sender: rees@citi.umich.edu I think you should mention somewhere that if you change CUTILDIR in Common.tmpl, you have to make sure it's on your path, or add it to the PATH in options.h. ------- Message 20 Received: from canopus.cc.umanitoba.ca by JIMI.CS.UNLV.EDU id aa13432; 13 Nov 94 23:18 PST Received: from loopback (root@dyn2-056.cc.umanitoba.ca [130.179.244.56]) by canopus.cc.umanitoba.ca (8.6.9/8.6.9) with ESMTP id BAA11211 for ; Mon, 14 Nov 1994 01:18:50 -0600 Received: (from umlin000@localhost) by loopback (8.6.9/8.6.9) id UAA08955 for bug-chimera@cs.unlv.edu; Sun, 13 Nov 1994 20:02:48 -0600 From: umlin000@cc.UManitoba.CA Message-Id: <199411140202.UAA08955@loopback> Subject: Chimera 1.61, right mouse button To: bug-chimera@cs.unlv.edu Date: Sun, 13 Nov 1994 20:02:47 -0600 (CST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 611 I am using Linux 1.62 kernel, gcc 2.61 for ELF. Chimera is 1.61. The right mouse button doesn't show the URL. What could be wrong? - -- - ------------------------------------------------------------------------------ | Eric Lin < EricLin@ee.Umanitoba.CA > Home: (204) 783-2884 | | Computer Engineering FAX Modem: (204) 783-2884 | | University of Manitoba, Canada PGP public key via finger/keyserver | | WWW URL: | - ------------------------------------------------------------------------------ ------- Message 21 Received: from little-charlie.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa18542; 14 Nov 94 1:17 PST To: umlin000@cc.umanitoba.ca cc: bug-chimera@cs.unlv.edu Subject: Re: Chimera 1.61, right mouse button In-reply-to: Your message of "Sun, 13 Nov 1994 20:02:47 CST." <199411140202.UAA08955@loopback> Date: Mon, 14 Nov 1994 01:17:31 -0800 From: John Kilburg >I am using Linux 1.62 kernel, gcc 2.61 for ELF. > >Chimera is 1.61. >The right mouse button doesn't show the URL. What could be wrong? Oops. Yeah. Thats a new strategy. Take the feature out but leave it in the documentation. Sorry about that. I will add a new way to show the URL one of these days. -john ------- Message 22 Received: from forte.mathematik.uni-bremen.de by JIMI.CS.UNLV.EDU id aa06740; 14 Nov 94 13:05 PST Received: from forte (forte [134.102.232.1]) by forte (8.6.8/8.6.6) with SMTP id WAA16791 for ; Mon, 14 Nov 1994 22:05:14 +0100 Date: Mon, 14 Nov 1994 22:05:13 +0100 From: Thomas Eberhardt X-Sender: thomas@forte To: bug-chimera@cs.unlv.edu Subject: chimera-1.62 beta dated 7-nov on NetBSD Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi! I had to made a couple of changes to get the chimera-1.62 beta dated 7-nov working on my NetBSD/i386 1.0 system. Hopefully they don't break other systems. First one has to do with the fact, that on NetBSD off_t is 64bit: *** local.c- Sat Oct 29 02:36:37 1994 - --- local.c Sat Nov 12 04:33:05 1994 *************** *** 176,182 **** } if (S_ISDIR(s.st_mode)) d = LoadDir(filename); ! else d = LoadFile(filename, s.st_size, mtlist); return(d); } - --- 176,182 ---- } if (S_ISDIR(s.st_mode)) d = LoadDir(filename); ! else d = LoadFile(filename, (int) s.st_size, mtlist); return(d); } The second change accomodates that select(2) calls are interrupted by SIGCHLD's occuring from the system(3) call in PipeCommand: *** input.c- Fri Nov 4 10:56:48 1994 - --- input.c Sat Nov 12 16:52:41 1994 *************** *** 87,93 **** i++; } } ! else if (rval < 0) break; if (DisplayTransferStatus(0, 0) == 1) break; } - --- 87,93 ---- i++; } } ! else if (rval < 0 && errno != EINTR) break; if (DisplayTransferStatus(0, 0) == 1) break; } *************** *** 147,153 **** btlen = tlen; } } ! else if (rval < 0) { if (t != NULL) free(t); return(NULL); - --- 147,153 ---- btlen = tlen; } } ! else if (rval < 0 && errno != EINTR) { if (t != NULL) free(t); return(NULL); / Thomas Eberhardt \ Moo! (__) | Center for Complex Systems and Visualization (CeVis) |_________/--------(oo) | University of Bremen, FB 3, Bibliothekstr. 1 | * |______||\/ \_o_ D-28359 Bremen, Germany, FAX: +49 421 218-4236 _o_/ ^^ ^^ ------- Message 23 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa08075; 14 Nov 94 13:52 PST Received: from citi.umich.edu by citi.umich.edu for thomas@mathematik.uni-Bremen.de bug-chimera@cs.unlv.edu with SMTP; Mon, 14 Nov 94 16:51:12 -0500 From: Jim.Rees@umich.edu To: Thomas Eberhardt Cc: bug-chimera@cs.unlv.edu Date: Mon, 14 Nov 1994 16:51:12 -0500 Subject: Re: chimera-1.62 beta dated 7-nov on NetBSD Sender: rees@citi.umich.edu In-Reply-To: Thomas Eberhardt, Mon, 14 Nov 1994 22:05:13 +0100 The second change accomodates that select(2) calls are interrupted by SIGCHLD's occuring from the system(3) call in PipeCommand: Aha! I wonder if this causes the missing inline images we've been seeing? ------- Message 24 Received: from forte.mathematik.uni-bremen.de by JIMI.CS.UNLV.EDU id aa08766; 14 Nov 94 14:11 PST Received: from forte (forte [134.102.232.1]) by forte (8.6.8/8.6.6) with SMTP id XAA17314; Mon, 14 Nov 1994 23:11:22 +0100 Date: Mon, 14 Nov 1994 23:11:21 +0100 From: Thomas Eberhardt X-Sender: thomas@forte To: Jim.Rees@umich.edu cc: bug-chimera@cs.unlv.edu Subject: Re: chimera-1.62 beta dated 7-nov on NetBSD In-Reply-To: <199411142152.WAA12580@zeus> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Mon, 14 Nov 1994 Jim.Rees@umich.edu wrote: > The second change accomodates that select(2) calls are interrupted by > SIGCHLD's occuring from the system(3) call in PipeCommand: > > Aha! I wonder if this causes the missing inline images we've been seeing? > This is what this patch fixes (at least for me)! / Thomas Eberhardt \ Moo! (__) | Center for Complex Systems and Visualization (CeVis) |_________/--------(oo) | University of Bremen, FB 3, Bibliothekstr. 1 | * |______||\/ \_o_ D-28359 Bremen, Germany, FAX: +49 421 218-4236 _o_/ ^^ ^^ ------- Message 25 Received: from snooky.CS.UNLV.EDU by JIMI.CS.UNLV.EDU id aa29415; 15 Nov 94 21:40 PST To: bug-chimera@snooky.CS.UNLV.EDU Subject: HTML+ Support? Date: Tue, 15 Nov 1994 21:40:03 -0800 From: "Steve E. Lumos" What are the current plans re: HTML+ support in Chimera? If you aren't planning to add it anytime soon, I would like to request for a really REALLY important tag:
. I can't believe this wasn't included even in HTML 0.9. TTYL - Steve ------- Message 26 Received: from sam.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa00111; 15 Nov 94 22:29 PST To: "Steve E. Lumos" cc: bug-chimera@snooky.cs.unlv.edu Subject: Re: HTML+ Support? In-reply-to: Your message of "Tue, 15 Nov 1994 21:40:03 PST." Date: Tue, 15 Nov 1994 22:29:40 -0800 From: John Kilburg >What are the current plans re: HTML+ support in Chimera? If you aren't >planning to add it anytime soon, I would like to request for a really >REALLY important tag:
. I can't believe this wasn't >included even in HTML 0.9. Chimera uses the HTML Widget from Mosaic 2.4 which handles, I believe, HTML+. The
tag is a Netscape-ism as far as I can tell. I could be wrong. Its hard to predict when I'll get time to do anything about it but I plan on working on HTML3. I'm pretty sure the
tag is not a part of HTML3. HTML3 does something like: -john ------- Message 27 Received: from bibsyst.bibsyst.no by JIMI.CS.UNLV.EDU id aa01556; 16 Nov 94 0:02 PST Received: by bibsyst.bibsyst.no (Smail3.1.28.1 #15) id m0r7fK3-0001ygC; Wed, 16 Nov 94 09:02 WET Message-Id: Date: Wed, 16 Nov 94 09:02 WET From: Tore Morkemo To: bug-chimera@cs.unlv.edu Subject: compressed documents Hi! The convert file says something about handling gzip'ed files but it won't work for me. I'm using xc3-1.62. This URL fails. https://www.cs.cmu.edu:8001/afs/cs.cmu.edu/user/rgs/mosaic/drac-1.html.gz among others. - -- Tore Morkemo. *-----------------------------------------------------------------------------* Tore Morkemo, Bibliotek-Systemer A/S, Box 2093 Stubber|d, 3271 Larvik, Norway Tel: +47 33 11 68 00 Fax: +47 33 11 68 22 email: tore@bibsyst.no *-----------------------------------------------------------------------------* ------- Message 28 Received: from condor.CC.UMontreal.CA by JIMI.CS.UNLV.EDU id aa13521; 16 Nov 94 11:35 PST Received: from beryl.JSP.UMontreal.CA by condor.CC.UMontreal.CA with SMTP id AA28072 (5.65c/IDA-1.4.4 for bug-chimera@cs.unlv.edu); Wed, 16 Nov 1994 14:33:54 -0500 Received: by beryl.jsp.umontreal.ca (940406.SGI/5.17) id AA22127; Wed, 16 Nov 94 14:35:21 -0500 From: Lair Simon Message-Id: <9411161935.AA22127@beryl.jsp.umontreal.ca> Subject: [Netbsd-1.0/X11R6] Help ! To: bug-chimera@cs.unlv.edu Date: Wed, 16 Nov 1994 14:35:21 -0500 (EST) X-Mailer: ELM [version 2.4 PL21] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1537 Hello, Recently, I've upgraded my system from NetBSD-0.9 to NetBSD-1.0 to take advantage of the shared libraries. At the same time, I went from X11R5 to X11R6 for the XFree86 distribution. Way back then, Chimera was working just fine. After the upgrade, I decided it was recompiling time and Chimera was no exception. But silly me (i'd rather say stupid but anyway..), I've deleted my old version of Chimera, never expecting that I would get problems running it again. Damn was I wrong. So here is the scoop. I can compile Chimera-1.61 to be compatible with term-2.0.4 without any problems, but it won't run. The home documents cannot be opened. All permissions for the files are OK and they are in the good directory. Sometimes I get "Error Reached DisplayCurrent() without a document.", sometimes I get "Info/Error the file ......"and sometimes it just go segmentation fault and core dumped. I've tried to recompile netpbm without the shared libraries, I've switched to Pbmplus, tried with and without shared, tried compiling Chimera with or without shared, even singing "Love me tender" through my nose while standing on one foot didn't help. I really want Chimera to work cos It's simply great and Motif light. So if anyone can help on this one or point me to a document that would enlightened me for the rest of my life, go ahead don't just stand there reading and going "Blah,another moron that doesn't know f*ck all". Help me , maybe I'll send you candies. Simon Lair (lairs@jsp.umontreal.ca) ------- Message 29 Received: from kauri.vuw.ac.nz by JIMI.CS.UNLV.EDU id aa27123; 16 Nov 94 16:39 PST Received: by kauri.vuw.ac.nz id AA17974 (5.65c/IDA-1.4.4 for bug-chimera@cs.unlv.edu); Thu, 17 Nov 1994 13:39:02 +1300 Date: Thu, 17 Nov 1994 13:39:02 +1300 From: Nathan Torkington Message-Id: <199411170039.AA17974@kauri.vuw.ac.nz> To: bug-chimera@cs.unlv.edu Subject: For Your Information This isn't a bug, except possibly in the naming of the product ... | From: kanderso@mabillon.ICS.UCI.EDU (Ken Anderson) | Newsgroups: alt.hypertext,comp.lang.ada,comp.lang.c | Date: 15 Nov 94 23:31:38 GMT - ------------------------------------------------------------------------ Chimera Heterogeneous Hypermedia System Kenneth M. Anderson E. James Whitehead, Jr. Professor Richard N. Taylor Arcadia Project U.C. Irvine November 15, 1994 - ------------------------------------------------------------------------- ANNOUNCEMENT: The Arcadia Project at the University of California, Irvine, announces the release of version 1.0 of the heterogeneous hypermedia system, Chimera. GENERAL OVERVIEW: Chimera is an open, serverized, hypermedia system that supports n-ary links between heterogeneous tools and applications in a network. Objects manipulated by separate applications can be linked together through Chimera. From the user's standpoint, for example, while working with one object in one application, an anchor on the displayed object may be selected, causing another application to start up, displaying a related (linked) object. Chimera comes with bindings to C and Ada; bindings to several popular tools have been constructed. Chimera makes no assumptions or demands regarding user interface system employed or how or where objects are stored. Chimera embodies the following technical attributes: Heterogeneous object editor & viewer support: Many different applications can be integrated with Chimera allowing them to participate in a hyperweb. Anchors specialized to particular views: Chimera handles just the links between objects and not their display, thus the applications which display the objects can make the display of anchors customized to the views of the objects. Multiple-view, concurrent, and active displays: Chimera has a client-server architecture. The Chimera server handles connections from client applications and handles the routing of messages (e.g. link traversals) between them. Links across heterogeneous object managers: Since an application manages the access and display of an object, and since anchors are created with respect to an object's view and not directly on an object, links between objects stored in different object managers (or even a file system) are easily established. n-ary links: A link is not restricted to a source and destination. Instead, links are modeled as a set. One anchor can take a user to multiple destinations in Chimera and all of them can be viewed at once (limited only by screen real estate). Chimera comes with application programming interfaces (APIs) for clients written in the Ada and C programming languages. In future releases we plan to provide additional APIs for C++, SmallTalk, and tcl. HARDWARE REQUIREMENTS: We built and tested Chimera on Sun hardware running SunOS 4.1.3. SOFTWARE REQUIREMENTS: In order to compile the C API, you will need a C compiler such as gcc. The C API can be used along with the binary release of the server on platforms that do not have a SunAda compiler. In order to compile the source release and/or the Ada API, you will need the SunAda compiler version 1.1 or later. For the source release you will also need Q version 3.2, and Arpc version 402.3. Q and Arpc are systems built by the Arcadia project at the University of Colorado, Boulder. Instructions for obtaining these two systems are included in the INSTALL file of the release. You will need the program gzip to uncompress the tar files described below. gzip can be obtained from any GNU software ftp site. AVAILABILITY: Chimera is available via anonymous ftp: Host: liege.ics.uci.edu Directory: pub/arcadia/chimera Two files are available: chimera.1.0.tar.gz: A tar file compressed using gzip containing the full source release of Chimera. chimera.1.0.bin.tar.gz: A tar file compressed using gzip containing a binaries release of Chimera which includes the source code for the two APIs and the chimera server/support tools executables. Obtain the former if you want access to all of Chimera's source code. Obtain the later if you are more interested in integrating clients with Chimera and don't really care how the Chimera server is implemented. CONTACT INFORMATION: All feedback is welcome at chimera-local@ics.uci.edu. All bug reports are welcome at chimera-bugs@ics.uci.edu. We shall attempt to contact you as quickly as possible in response to your mail. WWW INFORMATION: The release and more information about Chimera and Arcadia in general is available on the World Wide Web. Information about: Arcadia in general: https://www.ics.uci.edu/Arcadia/ Arcadia at UCI : https://www.ics.uci.edu/Arcadia/atUCI.html Chimera : https://www.ics.uci.edu/Arcadia/Chimera/chimera.html Sincerely, Kenneth M. Anderson E. James Whitehead, Jr. Professor Richard N. Taylor ------- Message 30 Received: from mabon.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa28425; 16 Nov 94 17:07 PST To: bug-chimera@cs.unlv.edu Subject: Re: For Your Information In-reply-to: Your message of "Thu, 17 Nov 1994 13:39:02 +1300." <199411170039.AA17974@kauri.vuw.ac.nz> Date: Wed, 16 Nov 1994 17:07:02 -0800 From: John Kilburg I figured something like this would happen someday. So do I go and get a trademark for my Chimera before I talk to these guys? I've never had to deal with this before. -john >This isn't a bug, except possibly in the naming of the product ... > >| From: kanderso@mabillon.ICS.UCI.EDU (Ken Anderson) >| Newsgroups: alt.hypertext,comp.lang.ada,comp.lang.c >| Date: 15 Nov 94 23:31:38 GMT > >------------------------------------------------------------------------ > Chimera > Heterogeneous Hypermedia System > > Kenneth M. Anderson > E. James Whitehead, Jr. > Professor Richard N. Taylor > > Arcadia Project > U.C. Irvine > > November 15, 1994 >------------------------------------------------------------------------- >ANNOUNCEMENT: > >The Arcadia Project at the University of California, Irvine, announces >the release of version 1.0 of the heterogeneous hypermedia system, Chimera. > >GENERAL OVERVIEW: > >Chimera is an open, serverized, hypermedia system that supports n-ary >links between heterogeneous tools and applications in a network. >Objects manipulated by separate applications can be linked together >through Chimera. From the user's standpoint, for example, while >working with one object in one application, an anchor on the displayed >object may be selected, causing another application to start up, >displaying a related (linked) object. Chimera comes with bindings to C >and Ada; bindings to several popular tools have been constructed. >Chimera makes no assumptions or demands regarding user interface system >employed or how or where objects are stored. > >Chimera embodies the following technical attributes: > >Heterogeneous object editor & viewer support: > > Many different applications can be integrated with Chimera allowing > them to participate in a hyperweb. > >Anchors specialized to particular views: > > Chimera handles just the links between objects and not their display, thus > the applications which display the objects can make the display of anchors > customized to the views of the objects. > >Multiple-view, concurrent, and active displays: > > Chimera has a client-server architecture. The Chimera server handles > connections from client applications and handles the routing of messages > (e.g. link traversals) between them. > >Links across heterogeneous object managers: > > Since an application manages the access and display of an object, and > since anchors are created with respect to an object's view and not > directly on an object, links between objects stored in different object > managers (or even a file system) are easily established. > >n-ary links: > > A link is not restricted to a source and destination. Instead, links > are modeled as a set. One anchor can take a user to multiple > destinations in Chimera and all of them can be viewed at once (limited > only by screen real estate). > >Chimera comes with application programming interfaces (APIs) for clients >written in the Ada and C programming languages. In future releases >we plan to provide additional APIs for C++, SmallTalk, and tcl. > >HARDWARE REQUIREMENTS: > > We built and tested Chimera on Sun hardware running SunOS 4.1.3. > >SOFTWARE REQUIREMENTS: > > In order to compile the C API, you will need a C compiler such as gcc. > The C API can be used along with the binary release of the server on > platforms that do not have a SunAda compiler. > > In order to compile the source release and/or the Ada API, you will need th e > SunAda compiler version 1.1 or later. > > For the source release you will also need Q version 3.2, and Arpc version > 402.3. Q and Arpc are systems built by the Arcadia project at the > University of Colorado, Boulder. Instructions for obtaining these two > systems are included in the INSTALL file of the release. > > You will need the program gzip to uncompress the tar files described below. > gzip can be obtained from any GNU software ftp site. > >AVAILABILITY: > > Chimera is available via anonymous ftp: > > Host: liege.ics.uci.edu > Directory: pub/arcadia/chimera > > Two files are available: > > chimera.1.0.tar.gz: A tar file compressed using gzip containing the full > source release of Chimera. > > chimera.1.0.bin.tar.gz: A tar file compressed using gzip containing a > binaries release of Chimera which includes the > source code for the two APIs and the > chimera server/support tools executables. > > Obtain the former if you want access to all of Chimera's source code. > Obtain the later if you are more interested in integrating clients > with Chimera and don't really care how the Chimera server is implemented. > >CONTACT INFORMATION: > > All feedback is welcome at chimera-local@ics.uci.edu. > All bug reports are welcome at chimera-bugs@ics.uci.edu. > > We shall attempt to contact you as quickly as possible in response to your > mail. > >WWW INFORMATION: > > The release and more information about Chimera and Arcadia in general > is available on the World Wide Web. > > Information about: > > Arcadia in general: https://www.ics.uci.edu/Arcadia/ > Arcadia at UCI : https://www.ics.uci.edu/Arcadia/atUCI.html > Chimera : https://www.ics.uci.edu/Arcadia/Chimera/chimera.html > >Sincerely, > >Kenneth M. Anderson >E. James Whitehead, Jr. >Professor Richard N. Taylor ------- Message 31 Received: from cheviot.ncl.ac.uk by JIMI.CS.UNLV.EDU id aa13408; 17 Nov 94 1:51 PST Received: from aire.ncl.ac.uk by cheviot.ncl.ac.uk id (8.6.9/ for ncl.ac.uk) with SMTP; Thu, 17 Nov 1994 09:51:18 GMT From: "J.D.Coleman" Message-Id: Subject: Lost file descriptors & authentication To: Chimera Bugs Mailing List Date: Thu, 17 Nov 1994 08:52:59 +0000 (GMT) Organisation: University of Newcastle Computing Service X-Mailer: ELM [version 2.4 PL23] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1897 When you open a pipe to an external command (e.g. textencode) in 1.62 beta 3, the file descriptors aren't closed. Eventually you run out of descriptors. The patch is small - I've appended it to this message. The authenticate code changes have fixed a problem I was having where authenticated inlines weren't loaded because Chimera wasn't sending the authentication data - great!. Also, caching the authentication parameters is much nicer than having to type in the username/password for each page. However, a problem has appeared that if the authentication is wrong (e.g. mistyped password; I wish I could type), Chimera ends up looping and repeatedly tries to load the document with the bad username/password combination. I think this will also occur if you load a document that requires a different username or password, i.e. Chimera will repeatedly try to use the old username/password combination. I haven't tracked this done yet - I'll post a patch when/if I find the bug. - -- - -- Julian Coleman, Computing Service, | j.d.coleman@ncl.ac.uk - -- Claremont Tower, University of Newcastle, | as967@cleveland.freenet.edu - -- Newcastle upon Tyne, NE1 7RU, England. | - - - - - - - - - - - - - - - -- Tel +44-191-222-8068 Fax +44-191-222-8765 | Be wery, wery careful Wabbit ---8<---------------------------- Cut here ---------------------------->8--- - --- orig/src/convert.c Mon Nov 7 04:57:16 1994 +++ src/convert.c Tue Nov 15 14:30:33 1994 @@ -325,8 +325,8 @@ char *msg = GetFromStringDB("nopipe"); char *t; + close(fd[0]); close(fd[1]); - - close(fd[2]); t = alloc_mem(strlen(msg) + strlen(command) + 1); strcpy(t, msg); @@ -340,8 +340,8 @@ return(d); } + close(fd[0]); close(fd[1]); - - close(fd[2]); d = ConvertDocument(d, clist, altaccess, path); d->up = up; ------- Message 32 Received: from marvin.cdf.toronto.edu by JIMI.CS.UNLV.EDU id aa15311; 17 Nov 94 4:15 PST Received: from diamond by marvin.cdf.toronto.edu id <225>; Thu, 17 Nov 1994 07:15:24 -0500 From: Smarasderagd To: bug-chimera@cs.unlv.edu Subject: chimera on solaris on pokey hardware Content-Length: 681 Message-Id: <94Nov17.071524edt.225@marvin.cdf.toronto.edu> Date: Thu, 17 Nov 1994 07:15:23 -0500 I've been getting chimera off the ground on a network on Sparc IPCs running SunOS 5.3 (i.e. solaris 2.x) and ran into one problem regarding the behaviour of select(2), and System V's non-restarting system calls. As nearly as I can tell, a timer alarm or some caught signal (I think it may be SIGCLD/SIGCHLD) interrupts the select(2) in ReadInput() and causes it to fail, since select(2) returns -1 with errno == EINTR in such cases. I think all that needs to be done to cope with this is to add a check that errno != EINTR on a < 0 return from select(2). I've seen this for other code using select(2), and it causes no harm on systems with auto-restarting system calls. :;SD:; ------- Message 33 Received: from mabon.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa15713; 17 Nov 94 4:32 PST To: Smarasderagd cc: bug-chimera@cs.unlv.edu Subject: Re: chimera on solaris on pokey hardware In-reply-to: Your message of "Thu, 17 Nov 1994 07:15:23 EST." <94Nov17.071524edt.225@marvin.cdf.toronto.edu> Date: Thu, 17 Nov 1994 04:31:58 -0800 From: John Kilburg >I've been getting chimera off the ground on a network on Sparc IPCs >running SunOS 5.3 (i.e. solaris 2.x) and ran into one problem regarding >the behaviour of select(2), and System V's non-restarting system >calls. As nearly as I can tell, a timer alarm or some caught signal >(I think it may be SIGCLD/SIGCHLD) interrupts the select(2) in >ReadInput() and causes it to fail, since select(2) returns -1 with >errno == EINTR in such cases. > >I think all that needs to be done to cope with this is to add a check >that errno != EINTR on a < 0 return from select(2). I've seen this for >other code using select(2), and it causes no harm on systems with >auto-restarting system calls. Someone just posted a patch that works for 1.61 and 1.62. I can't find anywhere. To the list: Can someone dig it up and send it to the above address? -john ------- Message 34 Received: from mabon.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa15904; 17 Nov 94 4:44 PST To: bug-chimera@mabon.ISRI.UNLV.EDU Subject: popup source window Date: Thu, 17 Nov 1994 04:44:14 -0800 From: John Kilburg I was messing around with the idea of making a popup HTML source window that would allow the user to edit the HTML in the source window and apply the changes to the HTML display. This would allow one to edit HTML and see the changes right away. I can't remember where I got this idea...maybe Mosaic 2.5? Comments? -john p.s. I may have mentioned this once already...I forget now. ------- Message 35 Received: from mabon.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa13677; 17 Nov 94 22:11 PST To: bug-chimera@mabon.ISRI.UNLV.EDU Subject: 1.62 beta 11 Date: Thu, 17 Nov 1994 22:11:10 -0800 From: John Kilburg If you grabbed 1.62 beta 7 (?) then you will definitely want to grab this one. It adds the patches that were submitted here for 1.62 and some other code cleanups and fixes. NetBSD and AIX (and probably others) users will want to grab this one for sure. -john ------- Message 36 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa00883; 19 Nov 94 11:58 PST Received: from citi.umich.edu by citi.umich.edu for bug-chimera@mabon.ISRI.UNLV.EDU with SMTP; Sat, 19 Nov 94 14:57:42 -0500 From: Jim.Rees@umich.edu To: bug-chimera@mabon.ISRI.UNLV.EDU Date: Sat, 19 Nov 94 14:57:35 EST Subject: Re: 1.62 beta 11 Sender: rees@citi.umich.edu In-Reply-To: John Kilburg, Thu, 17 Nov 94 22:11:10 PST If you grabbed 1.62 beta 7 (?) then you will definitely want to grab this one. One small patch is needed to get it to compile: % diff -cb net.h- net.h *** net.h- Thu Nov 17 06:44:11 1994 - --- net.h Sat Nov 19 14:54:41 1994 *************** *** 7,16 **** */ #ifndef _NO_PROTO ! int net_open(char *, int); void net_close(int); char *net_gethostname(); ! int net_bind(int); int net_accept(int); #else int net_open(); - --- 7,16 ---- */ #ifndef _NO_PROTO ! int net_open(char *, short); void net_close(int); char *net_gethostname(); ! int net_bind(short); int net_accept(int); #else int net_open(); ------- Message 37 Received: from relay1.UU.NET by JIMI.CS.UNLV.EDU id aa08287; 19 Nov 94 20:49 PST Received: from krikkit.fsa.ca by relay1.UU.NET with SMTP id QQxqud15471; Sat, 19 Nov 1994 23:49:17 -0500 Received: by krikkit.fsa.ca (4.1/dino1.0) id AA05217; Sat, 19 Nov 94 21:49:32 MST Date: Sat, 19 Nov 94 21:49:32 MST From: Dean Huxley Message-Id: <9411200449.AA05217@krikkit.fsa.ca> To: bug-chimera@cs.unlv.edu Subject: xc11-1.62.tar.gz patch A couple of changes were needed for me to compile without proto-types: diff -r ./mxw/Imakefile ../chimera-mod/mxw/Imakefile 3c3 < DEFINES = $(XRELEASE) $(MYFLAGS) $(PROTO) - --- > DEFINES = $(XRELEASE) $(MYFLAGS) $(CPROTO) diff -r ./src/Imakefile ../chimera-mod/src/Imakefile 11c11 < DEFINES = $(XRELEASE) $(CDEFINES) $(PROTO) - --- > DEFINES = $(XRELEASE) $(CDEFINES) $(CPROTO) diff -r ./src/main.c ../chimera-mod/src/main.c 93c93 < int DisplayTransferStatus(int, int); - --- > int DisplayTransferStatus(); Dean Huxley (dean@fsa.ca) ------- Message 38 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa04991; 21 Nov 94 3:03 PST To: bug-chimera@cephas.ISRI.UNLV.EDU Subject: 1.62 beta 20 Date: Mon, 21 Nov 1994 03:03:46 -0800 From: John Kilburg I've corrected the bugs reported from the last beta and fixed some other bugs and rearranged some code just to make you crazy :) Also, I've incorporated the GIF code directly into chimera. Color and grayscale work fine but the monochrome code messes up most of the time. I grabbed the code from xloadimage and added support for transparent GIFs. I haven't had any problems with inline GIF and now chimera is much much much faster when inline GIFs are in a document. It is really cool. You can change the poorly named Chimera.maxColors X resource to change the maximum number of colors allowed in a single inline image. No, I didn't document it. Yes, I will forget to document it until 1.64. Note that the old convert code doesn't work right now so only GIFs are supported in beta 20. I tried to add XBM support but it doesn't work yet. I'm working on getting things straightened out. Fortunately, the vast majority of inline images are GIF. I also tested this version with Linux and term 2.2.8. It worked nicely (well, at least for me and I didn't test EVERYTHING and I actually tested beta 15). I also tested it on an Alpha running OSF/1 V3.0 but the cache didn't work...I'm working on that too. Make sure you check out the bottom of options.h when you get compile errors on "weird" machines. Feel free to curse my name. ftp://ftp.cs.unlv.edu/pub/chimera-misc/xc2-1.62.tar.gz -john p.s. I am trying to release a really good 1.62 before I work on the next big version... ------- Message 39 Received: from msa.tte.vtt.fi by JIMI.CS.UNLV.EDU id aa06399; 21 Nov 94 5:23 PST Received: (from msa@localhost) by msa.tte.vtt.fi (8.6.9/8.6.9) id PAA23214; Mon, 21 Nov 1994 15:21:21 +0200 Date: Mon, 21 Nov 1994 15:21:21 +0200 From: Markku Savela Message-Id: <199411211321.PAA23214@msa.tte.vtt.fi> To: john@cephas.ISRI.UNLV.EDU CC: bug-chimera@cephas.ISRI.UNLV.EDU In-reply-to: John Kilburg's message of Mon, 21 Nov 1994 03:03:46 -0800 <199411211138.NAA18177@hemuli.tte.vtt.fi> Subject: 1.62 beta 20 Reply-to: Markku Savela >... now chimera is much much much faster when inline GIFs are >in a document. It is really cool. Yes, now it works faster even for me. Though, it always could be yet faster, nothing is ever fast enough :-). It seems to do "jumpy" display on page which uses those coloured ball graphics, even while redisplaying (when it already should be using the cache). (for example https://www.vtt.fi/). There was one problem for me in the src/Imakefile: the NormalProgramTarget was split into two lines (with \), but this barfed for me at 'make depend' phase. Just made it one long line... - -- Markku Savela (msa@hemuli.tte.vtt.fi), Technical Research Centre of Finland Multimedia Systems, P.O.Box 1203, FIN-02044 VTT, https://www.funet.fi/~savela ------- Message 40 Received: from cheviot.ncl.ac.uk by JIMI.CS.UNLV.EDU id aa08368; 21 Nov 94 7:29 PST Received: from blagdon.ncl.ac.uk by cheviot.ncl.ac.uk id (8.6.9/ for ncl.ac.uk) with SMTP; Mon, 21 Nov 1994 15:28:39 GMT Received: (njkw@localhost) by blagdon.ncl.ac.uk (8.6.7/8.6.x-cf revision 8 for SunOS 4.1.x) id PAA10494; Mon, 21 Nov 1994 15:29:03 GMT Date: Mon, 21 Nov 1994 15:29:03 GMT From: Jim Wight Message-Id: <199411211529.PAA10494@blagdon.ncl.ac.uk> To: bug-chimera@cs.unlv.edu Subject: 1.62 beta 20 Reply-to: J.K.Wight@newcastle.ac.uk The new way of setting -D_NO_PROTO in Common.tmpl, viz. #if !defined(NeedFunctionPrototypes) CPROTO = -D_NO_PROTO #endif for use in libhtmlw doesn't work. The following extract from Imake.tmpl #ifndef NeedVarargsPrototypes #define NeedVarargsPrototypes NO #endif #ifndef NeedNestedPrototypes #define NeedNestedPrototypes NO #endif #ifndef NeedFunctionPrototypes #define NeedFunctionPrototypes (NeedVarargsPrototypes || NeedNestedPrototypes) #endif reveals that even if function prototypes are not needed NeedFunctionPrototypes IS defined (with a value of 0) and hence CPROTO doesn't get a value. Jim ------- Message 41 Received: from cheviot.ncl.ac.uk by JIMI.CS.UNLV.EDU id aa08538; 21 Nov 94 7:34 PST Received: from blagdon.ncl.ac.uk by cheviot.ncl.ac.uk id (8.6.9/ for ncl.ac.uk) with SMTP; Mon, 21 Nov 1994 15:33:59 GMT Received: (njkw@localhost) by blagdon.ncl.ac.uk (8.6.7/8.6.x-cf revision 8 for SunOS 4.1.x) id PAA10511; Mon, 21 Nov 1994 15:34:24 GMT Date: Mon, 21 Nov 1994 15:34:24 GMT From: Jim Wight Message-Id: <199411211534.PAA10511@blagdon.ncl.ac.uk> To: bug-chimera@cs.unlv.edu Subject: 1.62 beta 20 Reply-to: J.K.Wight@newcastle.ac.uk Why does test-chimera behave like this? ; ./test-chimera Set PWD to the full path of the current directory. Surely it could do PWD=`pwd` on my behalf. Jim ------- Message 42 Received: from concorde.inria.fr by JIMI.CS.UNLV.EDU id aa07582; 25 Nov 94 2:50 PST Received: from fantomas.inria.fr (fantomas.inria.fr [128.93.11.34]) by concorde.inria.fr (8.6.9/8.6.9) with ESMTP id LAA14164 for ; Fri, 25 Nov 1994 11:50:08 +0100 Received: (from lasgout@localhost) by fantomas.inria.fr (8.6.8/8.6.6) id LAA25952; Fri, 25 Nov 1994 11:50:06 +0100 Date: Fri, 25 Nov 1994 11:50:06 +0100 From: Jean-Marc Lasgouttes Message-Id: <199411251050.LAA25952@fantomas.inria.fr> To: bug-chimera@unlv.edu Subject: 2 notes about Chimera. Reply-to: Jean-Marc.Lasgouttes@inria.fr Two small possible improvements for Chimera: - - Concerning installation: would it be possible to add automatically CUTILDIR as defined in Common.tmpl to the Path variable of options.h? I guess it is a natural thing to do for everyone. - - When you save a document to a file (for example by ftp), it would be nice to have the name of the document as a default in the Popup. And one more to make three: - - In the default , I'd like the enter key to act as the "Submit query" button. Is it possible, desirable? And in general forms, is there a way to have "default" buttons? Jean-Marc. ------- Message 43 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa07977; 25 Nov 94 3:24 PST To: bug-chimera@cephas.ISRI.UNLV.EDU Subject: 1.62 beta 23 Date: Fri, 25 Nov 1994 03:24:04 -0800 From: John Kilburg This cleans up some of the silly installation problems that were reported recently and now chimera does not require the convert file for "basic" Web browsing. This means that chimera handles HTML, plain text, and GIF images without external programs. The convert file is still required for postscript inline images, decompression, and other fancy stuff but most people will never have to worry it (I hope). The convert file format has changed a bit (it should be simpler now). If someone wants to contribute static binaries of 1.62 beta 23, I will make them available on ftp.cs.unlv.edu or I will make links to them from the chimera home page (hint, hint). Note that inline images won't look right most of the time on monochrome machines and only GIF inline images are handled. I haven't added code for the other formats yet. ftp://ftp.cs.unlv.edu/pub/chimera-misc Thanks. -john ------- Message 44 Received: from concorde.inria.fr by JIMI.CS.UNLV.EDU id aa09505; 25 Nov 94 5:31 PST Received: from fantomas.inria.fr (fantomas.inria.fr [128.93.11.34]) by concorde.inria.fr (8.6.9/8.6.9) with ESMTP id OAA17194; Fri, 25 Nov 1994 14:30:59 +0100 Received: (from lasgout@localhost) by fantomas.inria.fr (8.6.8/8.6.6) id OAA26665; Fri, 25 Nov 1994 14:30:58 +0100 Date: Fri, 25 Nov 1994 14:30:58 +0100 From: Jean-Marc Lasgouttes Message-Id: <199411251330.OAA26665@fantomas.inria.fr> To: john@cephas.ISRI.UNLV.EDU CC: bug-chimera@cephas.ISRI.UNLV.EDU In-reply-to: <199411251147.MAA15214@concorde.inria.fr> (message from John Kilburg on Fri, 25 Nov 1994 03:24:04 -0800) Subject: Re: 1.62 beta 23 Reply-to: Jean-Marc.Lasgouttes@inria.fr It seems to work well. However, the change file has not been updated. Jean-Marc. ------- Message 45 Received: from citi.umich.edu by JIMI.CS.UNLV.EDU id aa10789; 25 Nov 94 7:10 PST Received: from citi.umich.edu by citi.umich.edu for bug-chimera@cephas.ISRI.UNLV.EDU with SMTP; Fri, 25 Nov 94 10:09:20 -0500 From: Jim.Rees@umich.edu To: bug-chimera@cephas.ISRI.UNLV.EDU Date: Fri, 25 Nov 94 10:09:12 EST Subject: Re: 2 notes about Chimera. Sender: rees@citi.umich.edu In-Reply-To: Jean-Marc Lasgouttes, Fri, 25 Nov 94 11:50:06 +0100 - When you save a document to a file (for example by ftp), it would be nice to have the name of the document as a default in the Popup. Yes, but how do you define "document name?" I've made this suggestion to the Mosaic people before, and they didn't like it. I think the application of some simple heuristics would be appropriate at least for the ftp case. - In the default , I'd like the enter key to act as the "Submit query" button. I've made this suggestion before too. I think it's a great idea. ------- Message 46 Received: from mail.cs.tu-berlin.de by JIMI.CS.UNLV.EDU id aa26809; 26 Nov 94 2:17 PST Received: from indium.cs.tu-berlin.de (czyborra@indium.cs.tu-berlin.de [130.149.24.24]) by mail.cs.tu-berlin.de (8.6.9/8.6.9) with ESMTP id LAA03577 for ; Sat, 26 Nov 1994 11:17:00 +0100 Received: (czyborra@localhost) by indium.cs.tu-berlin.de (8.6.7/8.6.6) id LAA15978; Sat, 26 Nov 1994 11:16:58 +0100 From: Roman Czyborra To: bug-chimera@cephas.ISRI.UNLV.EDU Subject: Re: 2 notes about Chimera. In-Reply-To: <199411251521.QAA12067@mail.cs.tu-berlin.de> by Jim.Rees@umich.edu dated 1994-11-25 10:09:12 MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Date: Sat, 26 Nov 1994 11:16:56 +0100 Message-ID: > - When you save a document to a file (for example by ftp), it would be > nice to have the name of the document as a default in the Popup. > Yes, but how do you define "document name?" (basename= strrchr(url, '/')) || (basename= url); ------- Message 47 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa00394; 27 Nov 94 23:47 PST To: bug-chimera@cs.unlv.edu Subject: Re: compressed documents In-reply-to: Your message of "Wed, 16 Nov 1994 09:02:00 +0700." Date: Sun, 27 Nov 1994 23:47:26 -0800 From: John Kilburg >The convert file says something about handling gzip'ed files but it won't work >for me. > >I'm using xc3-1.62. > >This URL fails. > >https://www.cs.cmu.edu:8001/afs/cs.cmu.edu/user/rgs/mosaic/drac-1.html.gz > >among others. This has been fixed and will appear in the next release (after the beta 24 that is out there now). -john ------- Message 48 Received: from STL-17SIMA.ARMY.MIL by JIMI.CS.UNLV.EDU id aa09001; 28 Nov 94 8:26 PST Received: from stl-17sima by 0.STL-17SIMA.ARMY.MIL id aa22372; 28 Nov 94 10:24 CST To: bug-chimera@cs.unlv.EDU Reply-To: gaspar@STL-17SIMA.ARMY.MIL Subject: Trouble loading urls Date: Mon, 28 Nov 1994 10:23:59 CST From: Al Gaspar I have downloaded xc23-1.62 and xc24-1.62. Both of them compiled with no problems, but I am having trouble getting them to work properly. Under xc23, I can reach an item that is a url on my home page, and I can reach urls that I have a bookmark for or that I reach with open; however, once I have reached a remote page, I can't seem to reach urls that are embedded on that remote page. With xc24, I get my own home page just fine; however, I can't reach the embedded urls on my home page, or sites using open, or sites from my bookmark file. Where should I start looking to see where I may have gone wrong? I am running on a Sun 690 under SunOs 4.1.3 with X11R5. Thanks in advance for the help. Cheers-- Al - -- Al Gaspar USAMC SIMA, ATTN: AMXSI-TTC, 1222 Spruce St., St. Louis, MO 63103-2834 COMMERCIAL: (314) 331-4354 AUTOVON: 555-4354 relay1.uu.net!stl-17sima.army.mil!gaspar ------- Message 49 Received: from rowe.williams.edu by JIMI.CS.UNLV.EDU id aa15840; 28 Nov 94 12:48 PST Received: from bigbird (bigbird.cc.williams.edu) by rowe.williams.edu with SMTP id AA18262 (5.65c/IDA-1.4.4 for ); Mon, 28 Nov 1994 15:47:58 -0500 Received: by bigbird (4.1/client-1.3) id AA15389; Mon, 28 Nov 94 15:47:57 EST Date: Mon, 28 Nov 1994 15:47:56 -0500 (EST) From: "Kyle F. Downey" Subject: compile blows on PS-Format To: bug-chimera@cs.unlv.edu Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I am using beta 24, compiling with GCC 2.6.0 on a SunOS 4.1.3_U1 system and linking with term 2.2.8 and X11R6 (haven't dared trying XAW3D yet). I can't get libhtmlw to compile. I get this error with the library: making all in ./libhtmlw... make[1]: Entering directory `/usr/share/src/chimera-1.62/libhtmlw' gcc -fpcc-struct-return -c -I/usr/X11R6/include - -DISINDEX_SUBMIT -DXRELEASE=6 -DFAKE_EXPOSE -D_NO_PROTO HTML-PSformat.c HTML-PSformat.c: In function `PSprintf': HTML-PSformat.c:211: void value not ignored as it ought to be make[1]: *** [HTML-PSformat.o] Error 1 I went into the file and found that in the PSprintf file it was assigning len to the return of a sprintf command. I went into the header files and found that, sure enough, said function returns a void. Is there a patch for this? How do you find len if the command isn't returning it? I didn't understand the intent of the code well enough to patch it myself. ------- Message 50 Received: from rowe.williams.edu by JIMI.CS.UNLV.EDU id aa16772; 28 Nov 94 13:09 PST Received: from bigbird (bigbird.cc.williams.edu) by rowe.williams.edu with SMTP id AA19217 (5.65c/IDA-1.4.4 for ); Mon, 28 Nov 1994 16:09:56 -0500 Received: by bigbird (4.1/client-1.3) id AA18003; Mon, 28 Nov 94 16:09:56 EST Date: Mon, 28 Nov 1994 16:09:55 -0500 (EST) From: "Kyle F. Downey" Sender: "Kyle F. Downey" Reply-To: "Kyle F. Downey" Subject: libhtmlw To: bug-chimera@cs.unlv.edu Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII I "patched" it by deleting the offending assignment to len and making the function return PS_len instead. I removed the conditional around the PS_len = strlen() bit to give it some meaning, but I'm not sure about this hack. However, test-chimera brings up a local file but I get "Error: Could no load document" with any bookmark and any link on that local home page. I am using term 2.2.8 and yes term was on at the time :). However, I got the same error _without_ the term daemon running. Anyway, when term fails it usually says "Network Unreachable." After I patched the sprintfp deal it compiled fine.... Any ideas? Also, there was something wrong with the Makefile under chimera: It said "Missing separator" I got it to work by running it all together and deleting RM $@ -- no trouble linking then -- but that was a hack. thanks for all your help kd ------- Message 51 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa20654; 28 Nov 94 14:49 PST To: "Kyle F. Downey" cc: bug-chimera@cs.unlv.edu Subject: Re: compile blows on PS-Format In-reply-to: Your message of "Mon, 28 Nov 1994 15:47:56 EST." Date: Mon, 28 Nov 1994 14:49:30 -0800 From: John Kilburg >I am using beta 24, compiling with GCC 2.6.0 on a SunOS 4.1.3_U1 system >and linking with term 2.2.8 and X11R6 (haven't dared trying XAW3D yet). >I can't get libhtmlw to compile. I get this error with the library: > >making all in ./libhtmlw... >make[1]: Entering directory `/usr/share/src/chimera-1.62/libhtmlw' >gcc -fpcc-struct-return -c -I/usr/X11R6/include >-DISINDEX_SUBMIT -DXRELEASE=6 -DFAKE_EXPOSE -D_NO_PROTO HTML-PSformat.c >HTML-PSformat.c: In function `PSprintf': >HTML-PSformat.c:211: void value not ignored as it ought to be >make[1]: *** [HTML-PSformat.o] Error 1 > >I went into the file and found that in the PSprintf file it was assigning >len to the return of a sprintf command. I went into the header files and >found that, sure enough, said function returns a void. Is there a patch >for this? How do you find len if the command isn't returning it? I didn't >understand the intent of the code well enough to patch it myself. Good question. I haven't messed around with that part of the HTML widget code. I will take a look tonight. I just compiled chimera on a machine with SunOS 4.1.3_U1, X11R5, and gcc 2.5.8 (its the development machine for chimera right now). -john ------- Message 52 Received: from cephas.ISRI.UNLV.EDU by JIMI.CS.UNLV.EDU id aa14512; 30 Nov 94 0:48 PST To: "Kyle F. Downey" cc: bug-chimera@cs.unlv.edu Subject: Re: compile blows on PS-Format In-reply-to: Your message of "Mon, 28 Nov 1994 15:47:56 EST." Date: Wed, 30 Nov 1994 00:48:35 -0800 From: John Kilburg >I am using beta 24, compiling with GCC 2.6.0 on a SunOS 4.1.3_U1 system >and linking with term 2.2.8 and X11R6 (haven't dared trying XAW3D yet). >I can't get libhtmlw to compile. I get this error with the library: > >making all in ./libhtmlw... >make[1]: Entering directory `/usr/share/src/chimera-1.62/libhtmlw' >gcc -fpcc-struct-return -c -I/usr/X11R6/include >-DISINDEX_SUBMIT -DXRELEASE=6 -DFAKE_EXPOSE -D_NO_PROTO HTML-PSformat.c >HTML-PSformat.c: In function `PSprintf': >HTML-PSformat.c:211: void value not ignored as it ought to be >make[1]: *** [HTML-PSformat.o] Error 1 > >I went into the file and found that in the PSprintf file it was assigning >len to the return of a sprintf command. I went into the header files and >found that, sure enough, said function returns a void. Is there a patch >for this? How do you find len if the command isn't returning it? I didn't >understand the intent of the code well enough to patch it myself. I'm a bit confused. The source that I have in front of me has a vsprintf() on line 211. Also, it is not defined in the header files (thanks Sun!) so it should (I think) default to int. -john ------- Message 53 Received: from sun2.nsfnet-relay.ac.uk by JIMI.CS.UNLV.EDU id aa22988; 30 Nov 94 9:17 PST Via: uk.ac.aston; Wed, 30 Nov 1994 16:25:08 +0000 Received: from caledfwlch.aston.ac.uk by email.aston.ac.uk with SMTP (PP) id <22757-0@email.aston.ac.uk>; Wed, 30 Nov 1994 16:21:58 +0000 Received: by caledfwlch.aston.ac.uk (Smail3.1.28.1 #2) id m0rCrpb-000FniC; Wed, 30 Nov 94 16:24 GMT Message-Id: To: bug-chimera@cs.unlv.edu Subject: Re: compile blows on PS-Format In-reply-to: Your message of Wed, 30 Nov 1994 00:48:35 PST X-Mailer: exmh version 1.5omega 10/6/94 Date: Wed, 30 Nov 1994 16:24:35 +0000 From: Gareth Owen > >I am using beta 24, compiling with GCC 2.6.0 on a SunOS 4.1.3_U1 system > >and linking with term 2.2.8 and X11R6 (haven't dared trying XAW3D yet). > >I can't get libhtmlw to compile. I get this error with the library: > > > >making all in ./libhtmlw... > >make[1]: Entering directory `/usr/share/src/chimera-1.62/libhtmlw' > >gcc -fpcc-struct-return -c -I/usr/X11R6/include > >-DISINDEX_SUBMIT -DXRELEASE=6 -DFAKE_EXPOSE -D_NO_PROTO HTML-PSformat.c > >HTML-PSformat.c: In function `PSprintf': > >HTML-PSformat.c:211: void value not ignored as it ought to be > >make[1]: *** [HTML-PSformat.o] Error 1 > > > >I went into the file and found that in the PSprintf file it was assigning > >len to the return of a sprintf command. I went into the header files and > >found that, sure enough, said function returns a void. Is there a patch > >for this? How do you find len if the command isn't returning it? I didn't > >understand the intent of the code well enough to patch it myself. > > I'm a bit confused. The source that I have in front of > me has a vsprintf() on line 211. Also, it is not defined > in the header files (thanks Sun!) so it should (I think) default to int. > > -john Gcc should use its fixed header files if present in preference to the Sun ones. In the fixed stdio.h vprintf is defined as extern int vsprintf (char *, const char *, __gnuc_va_list); They'll be in a directory /usr/local/lib/gcc-lib/sparc-sunos4.1.3/2.6.0/include or something close. If not you'll need to unpack the source and do a 'make install-fixincludes'. And beta 24 builds fine for me using GCC 2.6.0 , SunOS 4.1.3 Gareth - --------------- Gareth Owen,Information Systems ,Aston University , Birmingham , UK ------- End of Forwarded Messages