Gpc

gpc@gnu.de

5400 discussions

Re: symbol out of scope
by Frank Heckenbach 15 Jul '99

15 Jul '99

Peter Gerwinski wrote: > Leonardo A. Saravia wrote: > > I have just installed gpc-19990118.i386-pc-djgppv201.zip and try to compile > > a program designed for VMS (VAX) or CMS (IBM) operative systems. I will > > appreciate some help, it seems to be a configuration problem, and I am new > > with Pascal and GNU Pascal. > > > > autocor.pas: In function `Add': > > autocor.pas:311: warning: constructing limited integer set `0..255' > > These warnings are coming from sets like "[ i ]" where `i' is a > variable of plain integer type. GPC cannot currently handle > "set of Integer" and warns about the restriction of the valid > scope for `i'. > > Maybe we should take out this warning - it is somewhat too > verbose ... I don't think so, since it can be a real problem, if I understand it correctly. If i is < 0 or > 255, the result will be wrong, right? However, if i is declared as 0 .. 255 rather than Integer, this is safe (modulo GPC's missing range checking :-/), and there is no warning. However^2, GPC will soon know `Include' and `Exclude' (compatible to BP 7) which will be able to handle out of range values correclty (i.e., cause a runtime error or ignore them, but not cause a wrong result). And perhaps, some day, GPC will be able to handle sets of dynamic size, and `[i]' with i : Integer will work without problems... 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

symbol out of scope
by Leonardo A. Saravia 14 Jul '99

14 Jul '99

Hi, I have just installed gpc-19990118.i386-pc-djgppv201.zip and try to compile a program designed for VMS (VAX) or CMS (IBM) operative systems. I will appreciate some help, it seems to be a configuration problem, and I am new with Pascal and GNU Pascal. autocor.pas: In function `Add': autocor.pas:311: warning: constructing limited integer set `0..255' autocor.pas:313: warning: constructing limited integer set `0..255' autocor.pas: In function `Sub': autocor.pas:320: warning: constructing limited integer set `0..255' autocor.pas:322: warning: constructing limited integer set `0..255' autocor.pas: In function `Ajoute': autocor.pas:329: warning: constructing limited integer set `0..255' autocor.pas: In function `Enleve': autocor.pas:336: warning: constructing limited integer set `0..255' autocor.pas: In function `Initbit': autocor.pas:1602: warning: constructing limited integer set `0..255' c:/djgpp/tmp\ccdaaaaa: Assembler messages: c:/djgpp/tmp\ccdaaaaa:17201: Fatal error: C_EFCN symbol out of scope Thanks Leonardo A. Saravia - Programa de Invest. en Ecologia Matematica Universidad Nacional De Lujan - C.C. 221 - (6700) - Lujan - Argentina Tel: 54 2323 423171/421030 Fax: 54 2323 425975 e-mail: lsaravia(a)mail.retina.ar

2 1

Re: Otherwise
by Peter Gerwinski 13 Jul '99

13 Jul '99

Hi! Jonathan Ashley wrote: > > Well, I get: > > other.pas: In function `program_Testcase': > other.pas:15: parse error before `Otherwise' > [...] Ah - this bug was fixed on 9 Apr 1999, probably after your first posting about this. Please get, compile and install a recent snapshot from ftp://agnes.dida.physik.uni-essen.de/gnu-pascal/alpha/ to solve the problem. Hope this helps, Peter -- Peter Gerwinski, Essen, Germany, http://home.pages.de/~Peter.Gerwinski/ Maintainer GNU Pascal - http://home.pages.de/~GNU-Pascal/ - gpc-19990118 PGP key on request - 6C 94 45 BE 28 A4 96 - 0E CC E9 12 47 25 82 75 *** Vote against SPAM! ********* http://www.politik-digital.de/spam/ ***

1 0

RE: Otherwise
by Jonathan Ashley 13 Jul '99

13 Jul '99

>What's the problem? For me, this program outputs "value wasn't 1" >as it should be. > >If I change the first assignment to "test := 1" I get "subvalue >was 5", so everything okay here, too. Damn! :-) Well I get: other.pas: In function `program_Testcase': other.pas:15: parse error before `Otherwise' other.pas:15: undeclared identifier `Otherwise' (first use this function) other.pas:15: (Each undeclared identifier is reported only once other.pas:15: for each function it appears in.) other.pas:16: missing `..' other.pas:16: parse error before `End' I'm using gpc 2.8.1 on Solaris. I downloaded a binary distrib. a few weeks ago. Very strange... regards, -- Jon Ashley ********************************************************************** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. This email has been swept for the presence of known computer viruses. **********************************************************************

1 0

Otherwise
by Jonathan Ashley 13 Jul '99

13 Jul '99

Hi all, I posted to this list before, but we were in the middle of switching mail systems and I'm not sure my message got out. The gist of my message was: there seems to be a bug in the way GPC handles the 'otherwise' clause of a case statement. If one of the previous cases contains a case statement itself, the compiler falls over. For example: ----------- program TestCase(input, output); var test: integer; subtest: integer; begin test := 3; subtest := 5; case test of 1: case subtest of 5: writeln("subvalue was 5") end otherwise writeln("value wasn't 1") end end. ----------- Has anyone else noticed this or is it my mistake? regards, -- Jon Ashley ********************************************************************** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. This email has been swept for the presence of known computer viruses. **********************************************************************

2 1

Re: "<>" operator (was: First question)
by Ken Linder (KC7RAD) 09 Jul '99

09 Jul '99

Hi Hilohi, Yes, GPC strings require single quotes `'` not double quotes. Your assumption is correct... Use the single quotes. And by the way... Don't be ashamed of asking anything on this mailing list. You should see some of the stupid questions I have asked! :-) All the best! Ken Linder ken(a)kc7rad.lvcm.com or pas(a)kc7rad.lvcm.com http://kc7rad.lvcm.com/ At 08:26 PM 7/9/99 +0300, Hilohi Gabriel wrote: >Well ... it's a much simple program .. I'm quite ashame of showing it to >such experienced programmers as i think you are ... >But it was a simple instruction : > >while s <> " " do > >and s was s:string[1000] > >Do i have to say > while s <> ' ' ??? > >

1 0

TEST
by Hilohi Gabriel 09 Jul '99

09 Jul '99

Is this the mailing list ? I subscribed to some gnu-pascal list but it had a different email address ... Please reply as soon as possible ... Thank you

2 5

Re: Support for gcc-2.8.1?
by Frank Heckenbach 08 Jul '99

08 Jul '99

"Russ Whitaker" wrote: > Receintly I purchased the latest distribution of SuSE: 6.1 > It contains: > > GCC 2.7.2.3-36 > GPC 990118-0 > egcs 990315-9 Then obviously Peter's protest has had an effect. The 6.1 shipped in Germany still contained a prehistoric version, AFAIK, but apparently the international version, which might have come out a little later, now contains a current GPC. :-) > Their comment on egcs: > > "This is the present state of the egcs-compiler-develement > Unfortunately it is not advisable to compile a 2.0.x kernel > with this compiler yet." I don't think this is going to change, since the bugs are in the kernel, not the compiler (it relies on undocumented behaviour of GCC-2.7.x) -- except that the 2.0 kernels will fade out, as 2.2 spreads, which is safe to compile with newer GCC versions and EGCS, AFAIK. > Would expect the other distributions to be somewhat simular. > Therefore, why not bundle a cut-down version of gcc-xxx with > gpc? Then YOU control which version of gcc is being used and > all the receiver would have to do ( hopefully ) is unpack and compile I don't think we should include it in the GPC source archive, because that would blow up the download size substantially for those who have it already, but we could (and IMHO should) always put the source of the "correct" GCC version besides the GPC source in the same directory on Agnes (or alternatively, one archive with GPC only and one with GPC and GCC). 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: Support for gcc-2.8.1?
by Russ Whitaker 08 Jul '99

08 Jul '99

Hi Peter On Wednesday, June 30, 1999 5:34 PM you wrote > > Hello, folks! > > The EGCS is going to release egcs-1.2 as gcc-2.95. > > How much sense does it make to continue support for > (a) gcc-2.8.1 and (b) egcs-1.1.x? > > Since it causes work to continue with (a) and (b), > I'd prefer to switch completely to gcc-2.95. > > Comments? Objections? Receintly I purchased the latest distribution of SuSE: 6.1 It contains: GCC 2.7.2.3-36 GPC 990118-0 egcs 990315-9 Their comment on egcs: "This is the present state of the egcs-compiler-develement Unfortunately it is not advisable to compile a 2.0.x kernel with this compiler yet." According to the file list egcs 990315 is egcs-2.91.66 Would expect the other distributions to be somewhat simular. Therefore, why not bundle a cut-down version of gcc-xxx with gpc? Then YOU control which version of gcc is being used and all the receiver would have to do ( hopefully ) is unpack and compile Thanks, Russ <russwhit(a)mind.net>

2 1

GPC mailing list archive?
by Broadey, Kevin M 07 Jul '99

07 Jul '99

Is this mailing list archived anywhere? The FAQ doesn't say so, and the links from the GPC Home Page only point off to UNIX mailbox files for traffic up to the end of 1998. It would be handy to have the list searchable online. Kevin

3 2

Jump to page:

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

1997

1996

1995

Gpc