Gpc April 2000

gpc@gnu.de

22 participants
44 discussions

Re: case sensitive defines
by Frank Heckenbach 10 Apr '00

10 Apr '00

Russ Whitaker wrote: > Suggestion: > > define (all lower case letters) case insensitive > > Define (One or more upper case) case sensitive I don't like this very much, I'm afraid. The other GPC directives are case-insensitive, so I think `$define' should also be. (Besides, it wouldn't be BP compatible...) If there are no objections, I'm implementing: Case insensitive: {$define} {$cidefine} --cidefine Case sensitive: {$csdefine} -D --define --csdefine Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

2 1

Re: faq-version 3
by Frank Heckenbach 10 Apr '00

10 Apr '00

Russ Whitaker wrote: > At long last here's vers 3 of the proposed FAQ A few more typos: compatable -> compatible proir -> prior kernal -> kernel nee -> need > and you can find the units directory with > find /usr/local/lib -name units Better: and you can find the units directory with gpc --print-file-name=units after installation. If this doesn't give the correct directory (or just says `units'), your installation is not quite right yet. > flags : short; This should be ShortInt. > Pascal syntax related questions may be appropriate in: > > comp.lang.pascal.ansi-iso Pascal according to ANSI and ISO > standards. > comp.lang.pascal.borland Borland Pascal questions. I'm wondering if we should suggest c.l.p.b at all. Is anyone monitoring it for GPC related questions regularly? (I'm not reading it anymore because of the high noise/signal ratio...) Also, the new gpc-doc mailing list could be mentioned here. It's still in the starting phase, but it will be the place for documentation related discussions (such as this FAQ :-). The last two changes apply to the upcoming release, due to some recent changes: > - A few unportable routines are not supported in the Dos unit (see > the comments there). I think there's an older, unportable, Dos > unit which supports them (or else it's easy to write one) if you > really need those routines... - A few unportable routines in the Dos unit are only supported under DJGPP if `__BP_UNPORTABLE_ROUTINES__' is defined (i.e., when compiling with `-D__BP_UNPORTABLE_ROUTINES__'). > Then, add this to your `djgpp.env' file: > > --------------------------------------------------------- > [gpc-cpp] We renamed the preprocessor to gpcpp, so this line should read then: [gpcpp] When the FAQ is complete so far, let us know so we can include it into the upcoming release distribution. Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

1 0

Re: faq-version 3
by Frank Heckenbach 10 Apr '00

10 Apr '00

Russ Whitaker wrote: > At long last here's vers 3 of the proposed FAQ A few more typos: compatable -> compatible proir -> prior kernal -> kernel nee -> need > and you can find the units directory with > find /usr/local/lib -name units Better: and you can find the units directory with gpc --print-file-name=units after installation. If this doesn't give the correct directory (or just says `units'), your installation is not quite right yet. > flags : short; This should be ShortInt. > Pascal syntax related questions may be appropriate in: > > comp.lang.pascal.ansi-iso Pascal according to ANSI and ISO > standards. > comp.lang.pascal.borland Borland Pascal questions. I'm wondering if we should suggest c.l.p.b at all. Is anyone monitoring it for GPC related questions regularly? (I'm not reading it anymore because of the high noise/signal ratio...) Also, the new gpc-doc mailing list could be mentioned here. It's still in the starting phase, but it will be the place for documentation related discussions (such as this FAQ :-). The last two changes apply to the upcoming release, due to some recent changes: > - A few unportable routines are not supported in the Dos unit (see > the comments there). I think there's an older, unportable, Dos > unit which supports them (or else it's easy to write one) if you > really need those routines... - A few unportable routines in the Dos unit are only supported under DJGPP if `__BP_UNPORTABLE_ROUTINES__' is defined (i.e., when compiling with `-D__BP_UNPORTABLE_ROUTINES__'). > Then, add this to your `djgpp.env' file: > > --------------------------------------------------------- > [gpc-cpp] We renamed the preprocessor to gpcpp, so this line should read then: [gpcpp] Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

1 0

Re: faq-version 3
by Russ Whitaker 10 Apr '00

10 Apr '00

That's 3 typos and 3 spelling errors Either way, wouldn't have found them without your help Thanks russ ---------- > From: Tom Schneider <toms(a)ncifcrf.gov> > To: russwhit(a)mind.net > Subject: Re: faq-version 3 > Date: Sunday, April 09, 2000 5:14 PM > > Russ: > > YEA!! A Faq!! > > A quick glance showed up some spelling errors (or more likely typos) ; so > here are my $0.02: > > definately -> definitely > disriminants -> discriminants > distrubution -> distribution > recomended -> recommended > replacemnet -> replacement > requiremnent -> requirement > saparately -> separately > > That's enough for a quick pass ... > > Tom > > Dr. Thomas D. Schneider > National Cancer Institute > Laboratory of Experimental and Computational Biology > Frederick, Maryland 21702-1201 > toms(a)ncifcrf.gov > permanent email: toms(a)alum.mit.edu > http://www.lecb.ncifcrf.gov/~toms/ >

1 0

Re: 6 MOD -90 Problem on Windows
by Mariusz Zynel 07 Apr '00

07 Apr '00

Peter Gerwinski <peter(a)gerwinski.de> wrote: > > What I'd like to have checked is whether GNU C's "mod" operation > produces different results on different platforms. If so, we > have detected a bug in the backend, otherwise in the frontend. My 0.02$ in this thread... Under Solaris 7 on both Intel and Sparc machines, GCC gives that 6 mod -90 is 6, while GPC gives -84. Regards -- . Mariusz Zynel Institute of Mathematics | phone: +48 (85) 745 75 52 University of Bialystok | Akademicka 2 St | mailto:mariusz@math.uwb.edu.pl 15-267 Bialystok | Poland | http://math.uwb.edu.pl/~mariusz/

2 1

Re: GPC: how to patch gcc 2.95.2 ?
by Martin Maechler 07 Apr '00

07 Apr '00

>>>>> "Peter" == Peter Gerwinski <peter(a)gerwinski.de> writes: Peter> You can patch it with the same diff as 2.95.1. MM> MM> but there's no such diff in the gpc sources! Peter> --> get current (alpha) distribution, gpc-19991030. Peter> (the next one is in preparation) Okay, I got that one, I found the proper patch file in diffs/p applied and was able to "configure & make" successfully. Next step "install" does not work according to the documentation in gpc/p/INSTALL >> 8. When everything compiles, you can verify what the `make -n >> pascal.install' command does, and if you are satisfied run it >> without the `-n' option to install the compiler `gpc1', front end >> `gpc', run time system `libgpc.a' to the same place where `gcc' >> was installed. This installation process does *not* overwrite >> existing copies of `libgcc.a' or `specs', should they exist. >> >> You can also install GPC as part of the complete GNU compiler >> suite using "make install" as described in the GCC installation >> docs. ....% make pascal.install ------------------- make: *** No rule to make target `pascal.install'. Stop. Now there's really no "pascal.install" target in the Makefile. The corresponding section in the Makefile is >> INSTALL_MODULES = \ >> install-apache \ >> install-ash \ >> install-autoconf \ >> install-automake \ >> ......... (lots omitted) ........... >> install-tgas \ >> install-time \ >> install-uudecode \ >> install-wdiff \ >> install-zip \ >> $(EXTRA_TARGET_HOST_INSTALL_MODULES) and $(EXTRA_TARGET_HOST_INSTALL_MODULES) is empty >> # These can be overridden by config/mt-*. >> # The _TARGET_ is because they're specified in mt-foo. >> # The _HOST_ is because they're programs that run on the host. >> EXTRA_TARGET_HOST_ALL_MODULES = >> EXTRA_TARGET_HOST_INSTALL_MODULES = >> EXTRA_TARGET_HOST_CHECK_MODULES = I've used a build directory different from source, and written ../gpc-gcc-src/gcc-2.95.2/configure which gave Configuring for a i686-pc-linux-gnu host. Created "Makefile" in /usr/local/app/gcc-build using "mt-frag" Configuring libiberty... creating cache ../config.cache ................ What would have been the correct configure statement ? How do I (non-manually) install gpc now ? BTW: Tried "make check" also which ran quite a few things and ended in errors (but not within gpc testing I think). Should I report this somewhere ? ------------------------ Peter> A new version of the FAQ is in preparation. Please search the Peter> list archives for it. MM> MM> no url? Peter> http://www.gnu.de/archive/wilma.cgi/gpc. (also on homepage) not quite. The above is a "Search" interface to the gpc mailing lists, and the homepages has a link to a FAQ; but that FAQ is more than three years old. Thanks again

3 3

Case-insensitive defines
by Frank Heckenbach 07 Apr '00

07 Apr '00

Hi, as you might know, GPC supports "defines" for two purposes, as conditionals for `ifdef' etc. (like some other Pascal compilers do), and as macros (like the C preprocessor). However, currently they're case-sensitive (like in C) which is different from other Pascal compilers (e.g., Borland Pascal) and generally not very "Pascalish", so I want to introduce case-insensitive defines. But there are also a few valid uses for case-sensitive defines, so I'd like to keep both. Actually, I've already done most of the work. The question that remains is about the syntax. Currently, GPC understands: #define foo -- inherited from the C preprocessor, but not recommended {$d foo} -- we plan to change this. More precisely, we want to drop all the one-letter directives except those compatible to BP as well as $M (not the same as in BP, but no problem as it will only write a harmless message for the BP memory directives which are unnecessary in GPC) and $W (warnings -- quite widely used). {$define foo} -- this is what BP uses. So for BP compatibility, we should probably make {$define} case-insensitive, and rename {$d} to something longer and leave it case-sensitive. Any suggestions? I thought of maybe {$def}, but this wouldn't make the difference to {$define} very clear. OTOH, something like {$define-case-sensitive} would make it clear, but it's very long. Peter has suggested {$cdefine} (c for `C' or for `case'). We're not very fond of either of them, perhaps someone has got a better idea... Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

7 8

Vs: building GPC compiler + FAQ
by Pernille Bech & Klaus F. Østergaard 07 Apr '00

07 Apr '00

The new FAQ is hard to find, only through mailing archive. Should it not be at the ftp site? and on the GPC homepage? regarding building a GPC compiler, on a redhat 6.0, I followed the instructions from the new FAQ: I have installed a gcc-2.95.2.i686.rpm which is working using gcc-2.95.2.tar.gz and gpc-19991030.tar.gz unpacked them in /usr/local/src moved /p to /gcc applied the patch and when I come to configure it does not work and I get this: I use this command : ../gcc-2.95.2/configure --enable-languages=pascal "i686" it returns this: Created "Makefile" in /usr/local/src/gcc-build /usr/bin/ld: cannot open crt1.o: No such file or directory collect2: ld returned 1 exit status *** The command 'gcc -o conftest -g -O2 conftest.c' failded. *** You must set the enviroment variable CC to a working compiler. What is wrong? kind regards, klaus

3 2

Re: GPC: how to patch gcc 2.95.2 ?
by Frank Heckenbach 07 Apr '00

07 Apr '00

Peter Gerwinski wrote: > > BTW: Tried "make check" also which ran quite a few things and ended in > > errors (but not within gpc testing I think). > > Should I report this somewhere ? > > There "should" be four (IIRC) known bugs in the GPC test suite > (tests not ending up in "TEST foo.pas OK"). If there are > significantly more failures, then something went wrong, and you > should report it here. But `make check' doesn't run the "real" test suite, but the old dejagnu stuff. Try `make check2-gpc' instead. Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

1 0

6 MOD -90 Problem on Windows
by Tim Kaulmann 07 Apr '00

07 Apr '00

Hello ! I found out that the result of 6 MOD -90 in my little Pascal Program on Windows is -84 and on Solaris is 6 (in my opinion the right one). Whats up there ? BTW : My Calculator HP48 says that -84 was right ??? A matter of definition ? Hope you can help me. Tim

5 11

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

1997

1996

1995

Gpc April 2000