October 2000 - XEmacs-Beta - XEmacs Mailing List Archives

Re: building xemacs-packages on unix 25 years

Gunnar Evermann

Adrian Aichner <adrian(a)xemacs.org> writes: > Gunnar, can you please give me execution-time numbers for a top-level > make clean > with and without following patch? Hm, I've just tried this on a real machine (Dual PII/550 + 1GB RAM) with an optimised 21.1.9 build and it was pretty fast either way... 63.23user 11.93system 1:20.22elapsed 93%CPU (0avgtext+0avgdata 0maxresident)k 2.06user 1.42system 0:04.15elapsed 83%CPU (0avgtext+0avgdata 0maxresident)k I guess it was only really slow on my laptop (PIII/450 64MB) because it ran out of memory for caching. I am gonna buy some more memory right now... > It still takes ages for me on NT, and it even rhymes! > make XEMACS_NATIVE_NT=t clean > Compilation started at Wed Oct 18 00:07:25 2000 +0200 (W. Europe Daylight Time) > > Compilation finished at Wed Oct 18 00:45:16 ouch, this sounds about like my laptop... Gunnar

1 participants
0 comments

Re: building xemacs-packages on unix 25 years

Gunnar Evermann

Adrian Aichner <adrian(a)xemacs.org> writes: > Setting it once in a hierarcical makefile structure isn't trivially > done cleanly. > > Let's make us Windows NT guys suffer, we're used to it :-) As Local.Rules needs to be modified before building anyway, I think having to set this manually is acceptable. > Gunnar, can you please give me execution-time numbers for a top-level > make clean > with and without following patch? If you really want to, I'll try to get them, but will have to use a different machine & xemacs this time. > How have you determined that my change is actually responsible for the > slowness? I first ran the unmodified version and saw that it was really slow and spawning lots of xemacs processes. I interrupted this after about 30 minutes and set the variable manually. Then it ran *much* faster. Of course this depends on what kind of xemacs you use (mine was mule pdump debug and no optimisation; which is kinda slow to start up). Gunnar

1 participants
0 comments

Re: RFC: xemacsweb - XEmacs vs. GNU Emacs 25 years

Michael Sperber

>>>>> "Martin" == Martin Buchholz <martin(a)xemacs.org> writes: Martin> I agree that members of the xemacs community should have a place to Martin> put their opinion. Our group is too diverse to have an "Our Side", Martin> however. So I suggest we have an "Opinion" web page, with opinions Martin> clearly identified by author. Consider this a place to put a Martin> "personal" web page for XEmacs developers, as long as the content is Martin> clearly xemacs-related. I agree. This would also give us a chance to link to http://www.jwz.org/doc/lemacs.html -- Cheers =8-} Mike Friede, Völkerverständigung und überhaupt blabla

1 participants
0 comments

Re: xemacsweb: Document Acrobat Reader PDF bug 25 years

Adrian Aichner

>>>>> "Stephen" == Stephen J Turnbull <turnbull(a)sk.tsukuba.ac.jp> writes: Hi Stephen, could you please try commiting this patch? Here is the command I would use (if it were my own patch) to submit it while sitting in .../xemacsweb: cvs -w -f -z3 -dadriana＠cvs.sourceforge.net:/cvsroot/xemacs commit -m \ "xemacsweb: Document Acrobat Reader PDF bug \ <14821.45049.395433.328650(a)turnbull.sk.tsukuba.ac.jp>" \ Documentation/ChangeLog Documentation/index.content Best regards, Adrian Stephen> Index: Documentation/ChangeLog Stephen> =================================================================== Stephen> RCS file: /cvsroot/xemacs/xemacsweb/Documentation/ChangeLog,v Stephen> retrieving revision 1.6 Stephen> diff -u -U0 -r1.6 ChangeLog Stephen> --- Documentation/ChangeLog 2000/08/20 21:50:41 1.6 Stephen> +++ Documentation/ChangeLog 2000/10/12 12:31:24 Stephen> ＠＠ -0,0 +1,4 ＠＠ Stephen> +2000-10-12 Stephen J. Turnbull <stephen(a)xemacs.org> Stephen> + Stephen> + * index.content: Document Acrobat Reader bug. Stephen> + Stephen> Index: Documentation/index.content Stephen> =================================================================== Stephen> RCS file: /cvsroot/xemacs/xemacsweb/Documentation/index.content,v Stephen> retrieving revision 1.4 Stephen> diff -u -r1.4 index.content Stephen> --- Documentation/index.content 2000/08/19 13:33:55 1.4 Stephen> +++ Documentation/index.content 2000/10/12 12:09:27 Stephen> ＠＠ -5,9 +5,10 ＠＠ Stephen> %main% Stephen> <h1>XEmacs Documentation</h1> Stephen> - <p> The following documents are available in Postscript, DVI Stephen> - and PDF (with permuted indices) formats for both US Letter and Stephen> - A4 sized paper:</p> Stephen> + <p> The following documents are available in Postscript, Stephen> + DVI and <a href="#AcrobatBug">PDF</a> (with permuted Stephen> + indices) formats for both US Letter and A4 sized Stephen> + paper:</p> Stephen> <table> Stephen> <tbody><tr> Stephen> ＠＠ -61,6 +62,26 ＠＠ Stephen> download them directly from Stephen> <a href="ftp://ftp.xemacs.org/pub/xemacs/docs/"><ftp://ftp.xemacs.org/pub...> Stephen> + <h3><a name="AcrobatBug" id="AcrobatBug">PDF Problems?</a></h3> Stephen> + Stephen> + <p> Many versions of Adobe Acrobat are known to be Stephen> + incapable of reading certain standard-conformant PDF Stephen> + files. Adobe has verified this, but hasn't done Stephen> + anything about it. The standard free PDF readers Stephen> + (<em>Ghostscript</em>,<em>gv</em>, and <em>xpdf</em>) do Stephen> + not have these limitations. <em>Ghostscript</em>, at Stephen> + least, is available for the Windows platform. Stephen> + Stephen> + <p>Unfortunately, both versions of <em>Ghostscript</em> Stephen> + prior to about 6.21 (ie, still in beta) and current Stephen> + versions of <em>pdflatex</em> emit such files (some of Stephen> + the problems were fixed in Aladdin Ghostscript's 6.0 Stephen> + release). There's not much we can do about it until Stephen> + updated versions of PDF writing software that cater to Stephen> + Acrobat's idiosyncracies become generally available. In Stephen> + particular, <em>pdflatex</em>, which we use for Stephen> + generating hardcopy of Texinfo documents.</p> Stephen> + Stephen> <h3>Additional documentation</h3> Stephen> <p>More recent material can be found in the <a href="../FAQ">FAQ</a> Stephen> -- Stephen> University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Stephen> Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091 Stephen> _________________ _________________ _________________ _________________ Stephen> What are those straight lines for? "XEmacs rules." -- Adrian Aichner <adrian(a)xemacs.org>, http://www.xemacs.org

2 participants
2 comments

Re: building xemacs-packages on unix 25 years

Adrian Aichner

>>>>> "Gunnar" == Gunnar Evermann <ge204(a)eng.cam.ac.uk> writes: Gunnar> I tried building the packages from source. Even doing a Gunnar> make clean in the root of the xemacs packages source tree Gunnar> takes ages. Gunnar> This is because the XEmacs.rules file contains: Gunnar> ----- Gunnar> XEMACS_NATIVE_NT = \ Gunnar> $(shell $(XEMACS) $(VANILLA) -batch \ Gunnar> -eval "(princ (eq system-type 'windows-nt))") Gunnar> ----- Gunnar> this is executed for every single directory in the whole Gunnar> source tree (i.e. for all 110 Makefiles). Gunnar> I am all in favour of compatability with different OSs, Gunnar> but could this please be done slightly more efficiently? Gunnar> e.g. only do this once instead of 110 time? I guess the best I can do is to make XEMACS_NATIVE_NT a variable to be specified by the user. Setting it once in a hierarcical makefile structure isn't trivially done cleanly. Let's make us Windows NT guys suffer, we're used to it :-) Gunnar, can you please give me execution-time numbers for a top-level make clean with and without following patch? How have you determined that my change is actually responsible for the slowness? It still takes ages for me on NT, and it even rhymes! cd d:\tmp\xemacs-packages\ make XEMACS_NATIVE_NT=t clean Compilation started at Wed Oct 18 00:07:25 2000 +0200 (W. Europe Daylight Time) for dir in libs comm games prog wp os oa; do \ make --unix -C ${dir} clean; \ done make[1]: Entering directory `/cygdrive/d/tmp/xemacs-packages/libs' <352 lines deleted by Adrian> rm -f strokes.elc ./auto-autoloads.elc \ ./custom-load.elc make[2]: Leaving directory `/cygdrive/d/tmp/xemacs-packages/oa/strokes' make[1]: Leaving directory `/cygdrive/d/tmp/xemacs-packages/oa' Compilation finished at Wed Oct 18 00:45:16 Best regards, Adrian Index: XEmacs.rules =================================================================== RCS file: /usr/CVSroot/XEmacs/xemacs-packages/XEmacs.rules,v retrieving revision 1.21 diff -u -u -r1.21 XEmacs.rules --- XEmacs.rules 2000/08/24 17:13:19 1.21 +++ XEmacs.rules 2000/10/17 22:58:31 ＠＠ -36,9 +36,23 ＠＠ # APA: XEmacs native NT currently (2000-06-23) has shortcomings we # need to temporarily work around. -XEMACS_NATIVE_NT = \ -$(shell $(XEMACS) $(VANILLA) -batch \ --eval "(princ (eq system-type 'windows-nt))") +# +# Please run make as follows for a native NT XEmacs: +# +# make XEMACS_NATIVE_NT=t ... +# +# since following test, which would be performed by every makefile +# including XEmacs.rules, slows down the make process terribly +# according to +# +# From: Gunnar Evermann <ge204(a)eng.cam.ac.uk> +# Subject: building xemacs-pacakges on unix +# To: XEmacs Developers <xemacs-beta(a)xemacs.org> +# Date: 14 Oct 2000 12:28:44 +0100 +# +# XEMACS_NATIVE_NT = \ +# $(shell $(XEMACS) $(VANILLA) -batch \ +# -eval "(princ (eq system-type 'windows-nt))") # Only in rare cases will you need to override this ifeq ($(AUTOLOAD_PATH),) Gunnar> Gunnar -- Adrian Aichner <adrian(a)xemacs.org>, http://www.xemacs.org

1 participants
0 comments

xemacs configuration problem with --prefix and --site-includes 25 years

John H Palmieri

I am not the sysadmin here, and so I'm basically installing my own private version of XEmacs. I've also installed copies of various libraries. I have - binary files in /scratch/palmieri/bin - library files in /scratch/palmieri/lib - include files in /scratch/palmieri/include But the command ./configure --prefix=/scratch/palmieri isn't good enough; I need to give the command ./configure --prefix=/scratch/palmieri --site-includes=/scratch/palmieri/include Why is this? As an example, there is a file called /scratch/palmieri/include/png.h, but without the --site-includes argument, config.log says configure:7891: checking for png.h configure:7899: gcc -E -I/usr/dt/include conftest.c >/dev/null 2>conftest.out configure:7895: png.h: No such file or directory -- John H. Palmieri Dept of Mathematics, Box 354350 palmieri(a)math.washington.edu University of Washington http://www.math.washington.edu/~palmieri Seattle, WA 98195-4350

2 participants
2 comments

Re: Package Freeze Melted. 25 years

Fabrice Popineau

Is it possible to build the packages from the cvs server under native win32 ? What is needed ? I have most of the gnu tools, but neither cygwin, neither bash. I have not much time to spend on this, so thanks for any answer (no chance, yes I have done it , whatever). -- Fabrice

2 participants
1 comments

Re: Package Freeze Melted. 25 years

Adrian Aichner

>>>>> "Fabrice" == Fabrice Popineau <Fabrice.Popineau(a)supelec.fr> writes: Fabrice> Is it possible to build the packages from the cvs server Fabrice> under native win32 ? What is needed ? I have most of the Fabrice> gnu tools, but neither cygwin, neither bash. Fabrice> I have not much time to spend on this, so thanks for any Fabrice> answer (no chance, yes I have done it , whatever). I have done so a little while back Here's the make version I used: GNU Make version 3.77, by Richard Stallman and Roland McGrath. Copyright (C) 1988, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. Report bugs to <bug-make(a)gnu.org>. Just type M-x compile RET make --unix bindist in the top-level (for all packages) or on a sub-level (for a particular one). I had to make some changes to makefiles along the way. I have commited all my changes to XEmacs CVS repository. Let me know if you have problems. Good Luck, Adrian Fabrice> -- Fabrice> Fabrice -- Adrian Aichner <adrian(a)xemacs.org>, http://www.xemacs.org

1 participants
0 comments

October sumo release 25 years

Steve Youngs

Due to unforeseen circumstances the sumo release that I'd hoped to do on the 15th, didn't happen. All going well, I'm hoping for the end of the week. -- |---<Regards, Steve Youngs>-----------[GnuPG KeyID: 59E4C4B2]---| | XEmacs - It's not just an editor... | | It's a way of life. | |----------------------------------<mailto:youngs＠xemacs.org>---|

1 participants
0 comments

Package Freeze Melted. 25 years

Steve Youngs

The other day, I called a package freeze, well disregard that. :-) It was pointed out to me that a freeze isn't really necessary, so from now on I won't be calling them. What I will do is announce upcoming sumo releases. I'll give about a week's notice to give people a chance to get any hacks in that they've got lying around. I'll send these announcements to xemacs-review, xemacs-beta and comp.emacs.xemacs. Is there anywhere else I should send them? Are there any package maintainers who'd like me to send a copy of these announcements directly to them? Proposed Release Schedule: Sumos - At least monthly, on or around the 15th of each month. Packages - As soon as updates have been applied and tested, unless the update happens close to a sumo release. In that instance I'll release the package along with the sumo. -- |---<Regards, Steve Youngs>-----------[GnuPG KeyID: 59E4C4B2]---| | XEmacs - It's not just an editor... | | It's a way of life. | |----------------------------------<mailto:youngs＠xemacs.org>---|

1 participants
0 comments

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

XEmacs-Beta October 2000