Gpc

gpc@gnu.de

5400 discussions

Re: BP -> GPC: High()/Low()...
by Kevin A. Foss 17 Feb '98

17 Feb '98

On Mon, 16 Feb 1998 14:40:21 -0500 (EST), Orlando Llanes wrote: > Another thing to note is that GPC will return the actual first and >last element number, but BP starts it at 0 and ends on the last element. >For example, if the index goes from -1 to 4, BP will treat it as 0..6 >whereas GPC (and any Standard Pascal compliant Pascal Compiler?) will >treat it as -1..4 (don'tcha just LOVE GPC?!?!?!?! :) Um, so Low()/High() return the indices of the lower/upper bound of the array, right? [Just making sure I understood the examples, as I have no experience with these BP functions] It seems to me that the GPC behavior makes more sense (and would return more useful values), or was that your point? What good does Low() do in BP if it always returns a 0? -Kevin -- Kevin A. Foss --- kfoss(a)mint.net

2 1

BP -> GPC: High()/Low()...
by Orlando Llanes 16 Feb '98

16 Feb '98

I didn't see this at first, but for those of you who use High() and Low() in BP can use it in GPC with 2 "work-arounds": First a BP function (to demonstrate it's use so you understand how the work around functions): PROCEDURE BPFoo( B : ARRAY OF Integer ); BEGIN WriteLn( 'ARRAY[ ', Low( B ), '..', High( B ), ' ] OF Integer;' ); END; The 2 GPC work arounds are as follows: PROCEDURE GPCFooParam( B : ARRAY[ Low..High : Integer ] OF Integer ); BEGIN WriteLn( 'ARRAY[ ', Low, '..', High, ' ] OF Integer;' ); END; -=[ OR ]=- FUNCTION GPCHighInt( B : ARRAY[ Low..High : Integer ] OF Integer ): Integer; BEGIN GPCHighInt := High; END; FUNCTION GPCLowInt( B : ARRAY[ Low..High : Integer ] OF Integer ): Integer; BEGIN GPCLowInt := Low; END; In the first example of the GPC work around, there's no need for the High()/Low() functions as Low and High already mark the first and last element, respectively. In the second example, it's a work-around that prolly isn't worth it, but another way to do it nonetheless (especially if you need to know the first and last element index from the main program). I don't know if there's a way to do the latter GPC work-around so that it can take an array of any type (I don't know how to do that yet if it's possible), but as far as I know, you need a function for each type you will be using. Another thing to note is that GPC will return the actual first and last element number, but BP starts it at 0 and ends on the last element. For example, if the index goes from -1 to 4, BP will treat it as 0..6 whereas GPC (and any Standard Pascal compliant Pascal Compiler?) will treat it as -1..4 (don'tcha just LOVE GPC?!?!?!?! :) See ya! Orlando Llanes "Meine Damen und Herren, Elvis hat soeben das Gebaeude verlassen!" "Look out fo' flyeeng feet" O__/ a010111t(a)bc.seflin.org \/|____. O <__. \/\> / \ ____________|___\______ http://ourworld.compuserve.com/homepages/Monkey414

1 0

x86 i/o port access
by Frank Heckenbach 12 Feb '98

12 Feb '98

Following a request on a German Pascal newsgroup, I put together a little unit for accessing the i/o ports on the x86 from code posted here last year, together with specific Linux functions to enable port access, and a demo program. I put them at http://home.pages.de/~fjf/programs.htm in case someone here is interested or has any comments. Frank -- Frank Heckenbach, Erlangen, Germany fjf(a)gmx.de http://home.pages.de/~fjf/links.htm

1 0

Manuals...
by croy＠osiris.com 12 Feb '98

12 Feb '98

Hello, I need a better source of documentation for GPC... I did read the documentation on agnes but I still have a feeling of missing something when I use the compiler.... Most of the time I go by intuition, having used for many years various Pascal implementation I end up with a trial and error process to functions, libraries and RTL function calls... Is there a place where we have a list of the function and there call parameter structure. I am using right now a generic Pascal reference and an old Borland Pascal language definition with a grain of salt and lots of little test programs but I find that so time consuming. Hopefully some people on the list are more experience with the GPC more specific issues and can help guiding me to a better reference. Thanks Claude.

2 1

Re: GPC problem with DJGPP
by Peter Gerwinski 11 Feb '98

11 Feb '98

According to Johan Larsson: > > Furthermore, the following program does not compile (parse > error at the class model `Class' etc. are not yet implemented into GPC; only the "old" OOP model from BP7 (`object' etc.) is supported at the moment. > though -fobject-pascal -fextended-pascal is enabled): These options do not enable anything; in contrast they disable everything outside the specified Pascal dialects. Greetings, Peter -- Peter Gerwinski, Essen, Germany, free physicist and programmer Maintainer GNU Pascal - http://home.pages.de/~GNU-Pascal/ - 1 Oct 1997 PGP key fingerprint: AC 6C 94 45 BE 28 A4 96 0E CC E9 12 47 25 82 75 Fight the SPAM! - http://maps.vix.com/

1 0

GPC problem with DJGPP
by Johan Larsson 11 Feb '98

11 Feb '98

Hi, I've tried to install GPC together with DJGPP and RHIDE. I get some very odd results. When attempting to compile .pas files the compiler report: "cannot open crt0.o: No such file or directory (ENOENT)" It appears as the lib path cannot be reached. Furthermore, the following program does not compile (parse error at the class model though -fobject-pascal -fextended-pascal is enabled): program classes; type efClass = class constructor Create; end; efMetaClass = class end; constructor efClass.Create; begin end; begin end. Can someone help me here? Is there detailed configuration information available on the Internet? Best Regards, Johan Larsson URL: http://www.acc.umu.se/~jola/ MAIL: jola(a)acc.umu.se Also, REMEMBER to check out EFLIB, the free object-oriented framework for EVERY Pascal programmer, at http://ftp.sunet.se/EFLIB/. Simply SOPHISTICATED!

2 1

Re: BP and BO question...
by Peter Gerwinski 09 Feb '98

09 Feb '98

According to Jim Roland: > > Thanks for the help. I got a successful compile now (no errors), however, > when I run the program i compiled (with gcc cgi-postsplit.pas -o > cgi-postsplit), You mean `gpc cgi-postsplit.pas -o cgi-postsplit'? > I get a bash error telling me the file is not there. Please try the following: * Make sure that this is no trivial error, for instance the executable is in the current directory, but the current directory is not in your path. (This is - or should be - the case when you are working as `root' (which you shouldn't do anyway.;-)) To check it, try `./postsplit' instead of just `postsplit'. * Use GPC only to compile, but link with GCC: `gpc -c postsplit.pas'; `gcc postsplit.o -o postsplit'. * If this does not help, add the `-v' option to your call to `gpc' and post the result here: `gpc cgi-postsplit.pas -v -o cgi-postsplit'. BTW, what system do you use? Linux probably (from your signature), but we must be sure about this in order to offer any help. > I tried running configure and make (even a make distclean before i > started) and I get an error 126 on ../gcc-2.7.2.2/specs! Thoughts? No idea, sorry. :-( > Right > now (I don't know what happened to cause this not to work nor compile > properly), but now I can't compile a program. A simple program that > assigns a string to S and writeln's it out doesn't error on compile, but > runs w/o output (test.pas). [...] As you can imagine, my GPC has no problem with this. Good luck, Peter -- Peter Gerwinski, Essen, Germany, free physicist and programmer Maintainer GNU Pascal - http://home.pages.de/~GNU-Pascal/ - 1 Oct 1997 PGP key fingerprint: AC 6C 94 45 BE 28 A4 96 0E CC E9 12 47 25 82 75 Fight the SPAM! - http://maps.vix.com/

1 0

Re: BP and BO question... (fwd)
by Jim Roland 09 Feb '98

09 Feb '98

Thanks for the help. I got a successful compile now (no errors), however, when I run the program i compiled (with gcc cgi-postsplit.pas -o cgi-postsplit), I get a bash error telling me the file is not there. When I compile a test program test.pas, I get no output at all Attached are those files. I tried running configure and make (even a make distclean before i started) and I get an error 126 on ../gcc-2.7.2.2/specs! Thoughts? Right now (I don't know what happened to cause this not to work nor compile properly), but now I can't compile a program. A simple program that assigns a string to S and writeln's it out doesn't error on compile, but runs w/o output (test.pas). a program that I have attached (cgi-postsplit.pas) give me a bash error telling me there is no such file or directory...HELP! Thanks! program stringsplit; var x,y : integer; s : string(255); begin s:='This is a string'; writeln('The String is = ',s) end. program stringsplit; #include "strpas.pas" var x, y : integer; s, newenv, curenv, newenvval, curenvval : string(255); teststr: CString; type PCString = ^CString; Function getenv(curenv:CString): PCString; C; Function setenv(newenv,newenvval:Tstring; x:integer): integer; C; begin s:='Test string'; writeln('The String is = ',s) end.

1 0

Re: Pointer to Method...
by Peter Gerwinski 07 Feb '98

07 Feb '98

According to Orlando Llanes: > Ok, this is a tricky one. I need to know how to get a pointer to a > method from an object. I've tried a million and one different things, but > was not able to do so. Is it even possible? Not without tricks. If you carefully cast pointers and take into account the implicit `Self' parameter of each method and pass it explicitly when you invoke the method through the pointer, you might succeed. > The reason I need this is > because I'm toying with the idea of making an object method an interrupt > routine. I suspect this is hopeless because you cannot expect the interrupted process to put the `Self' parameter into a certain register or a certain stack slot. You "must" use an "ordinary" procedure as the interrupt handler which may then call the method. (* And, of course, you must be very careful with interrupts in 32-bit protected mode anyway. ;*) Good luck, Peter -- Peter Gerwinski, Essen, Germany, free physicist and programmer Maintainer GNU Pascal - http://home.pages.de/~GNU-Pascal/ - 1 Oct 1997 PGP key fingerprint: AC 6C 94 45 BE 28 A4 96 0E CC E9 12 47 25 82 75 Fight the SPAM! - http://maps.vix.com/

1 0

Re: BP and BO question...
by Peter Gerwinski 07 Feb '98

07 Feb '98

According to Jim Roland: > So the convention to access a system call or part of the C Library is to > declare a function with same parameters and return data-type and put a C; > at the end of it? Yes. > Seems too easy! :) Okay, if you prefer something more difficult: The `C' directive is only to declare functions that are lowercase only. If the C declaration is in mixed case, for instance void MyCFunction ( int foo, long bar, long long baz ); write Procedure MyCFunction ( foo: Integer; bar: MedInt; baz: LongInt ); AsmName 'MyCFunction'; (* For more about `MedInt' and friends, see `info -f gpc -n "Integer types"'. This requires something more recent than gpc-2.0; see ftp://agnes.dida.physik.uni-essen.de/gnu-pascal/beta/ ftp://agnes.dida.physik.uni-essen.de/gnu-pascal/alpha/ *) Have fun, Peter -- Peter Gerwinski, Essen, Germany, free physicist and programmer Maintainer GNU Pascal - http://home.pages.de/~GNU-Pascal/ - 1 Oct 1997 PGP key fingerprint: AC 6C 94 45 BE 28 A4 96 0E CC E9 12 47 25 82 75 Fight the SPAM! - http://maps.vix.com/

1 0

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