Gpc December 1997

gpc@gnu.de

24 participants
76 discussions

Re: Regular Expression
by Peter Gerwinski 22 Dec '97

22 Dec '97

Hallo, Hans! Hello, everybody! Hans, I like your ideas, but I have some problems to imagine how this should work in praxis. According to Hans Ecke: > * announce such an archive in the Homepage > * make a directory, where new items go inside. > They will stay there some time (2 weeks?) if after that time, nobody has > protested against them, they would be moved to the archive. Normally, nobody will take the time and download a file just for checking whether it is worth to be in such an archive. If it is the default to move it to the archive, I expect it to overflow relatively soon. I think that such an archive cannot work reasonably for a longer time without continuous human interaction. A "moderator" who feels responsible for it would be needed here. (Interested?;-) > This could be done rather easily. How? Are you thinking of some auxiliary program invoked by a `cron' script? > To stay in time, the files can be put in meantime to > the archive via symbolic links. You mean that there should be a readable `incoming' directory for the archive where files stay for some "evaluation time"; after that they can become members of the archive, and in the meantime they are symlinked into the archive? Schoene Gruesse, 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/

2 1

Re: unit init
by Peter Gerwinski 22 Dec '97

22 Dec '97

Hi! > it seems this little unit hangs GPC > > unit test; > interface > implementation > var pro:array[0..1] of ^procedure; > k:integer; > {*******************************} > procedure MC; > begin > writeln(' Merry Christmas ! '); > end; > {*******************************} > begin > for k:=0 to 1 do pro[k]:=@MC; > end. > > Maybe there are syntax errors , but I think it shouldn't > crash the compiler This bug is fixed meanwhile, but the result is not yet published. (There are no syntax errors in your unit, BTW.) > It works only if I use the option --fBorland , This is fixed, too. > I use the compiler which is in GPC-9710.zip, I see: It's high time for the release of gpc-2.1. > Bye and Merry Christmas everybody ! Thank you - as well! > PS : I hope you can understand my english... (Not a single word. ;-) 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: Last value in enum...
by Peter Gerwinski 21 Dec '97

21 Dec '97

According to Orlando Llanes: > How do you get the highest value from an enumerated type? Delphi defines > functions First() which gets the first value, and Last() which gets the > last value, which I believe is equivalent to High() and Low() in Borland > Pascal. Is there an equivalent function in GPC? There isn't, but since this does not break anything else (and is not too complicated), this is a good candidate to be implemented in the future. 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: gpc I/O & errors
by heckenb＠helena.hut.fi 20 Dec '97

20 Dec '97

Kevin A. Foss wrote: > >Well... but I think in most cases it would be more appropriate to return > >false for directories (I hadn't thought about this problem at all) because > >I'd usually call this functions before trying to open the file for reading > >or updating. > > Yeah, that's true -- I was thinking of using a check for file > existance to check and see if the filename was available for creating a > file with a unique filename. But your use is more common and good > code would check to see if the bind fails anyway. True, this seems like another use for the functions (and I just saw that I was using it in this way sometimes... ;-). OTOH, there can be other obstacles to creating a file, such as write-protected directories / file systems / volumes, so one would have to check the Bind, and perhaps the Rewrite as well. > >So, I think it would be better to return false if the binding fails. > > Yes I do too, now that I considered the actual usage of the function. > :) Just for the record, with the built-in Assign function, Exists could be shortened a little: function Exists(FileName:String):Boolean; var b:BindingType; f:Text; begin Assign(f,FileName); b:=Binding(f); Exists:=b.Bound and b.Existing; Unbind(f) end; Chief, perhaps it would be worth putting such a function into one of the BP compatibility units. Though BP does not have a built-in Exists, it could be hard for many ex-BP users (like me ;-) to figure out the bindings... -- Frank Heckenbach, Erlangen, Germany fjf(a)gmx.de http://home.pages.de/~fjf/links.htm

1 0

Last value in enum...
by Orlando Llanes 19 Dec '97

19 Dec '97

How do you get the highest value from an enumerated type? Delphi defines functions First() which gets the first value, and Last() which gets the last value, which I believe is equivalent to High() and Low() in Borland Pascal. Is there an equivalent function in GPC? See ya! Orlando Llanes "Hey, we all did the drug thing, we all did the money thing, and eventually you find out that none of that stuff fixes anything, and we have nowhere else to go except to evolve spiritually and intellectually" -- Meredith Brooks "Look out fo' flyeeng feet" O__/ a010111t(a)bc.seflin.org \/|____. O <__. \/\> / \ ____________|___\______ http://ourworld.compuserve.com/homepages/Monkey414

1 0

Re: gpc I/O & errors
by Kevin A. Foss 19 Dec '97

19 Dec '97

On Fri, 19 Dec 1997 22:00:21 +0100, Frank Heckenbach wrote: >Kevin A. Foss wrote: >> The problem (as noted in the >> comments) is that if the bind fails it thinks the file exists, which is >> a bit of a kludge. However I couldn't get around the problem that >> appeared when a user gives a name that is used by a directory. Frank's >> code using stat() correctly reports true for directories. > >Well... but I think in most cases it would be more appropriate to return >false for directories (I hadn't thought about this problem at all) because >I'd usually call this functions before trying to open the file for reading >or updating. Yeah, that's true -- I was thinking of using a check for file existance to check and see if the filename was available for creating a file with a unique filename. But your use is more common and good code would check to see if the bind fails anyway. >So, I think it would be better to return false if the binding fails. Yes I do too, now that I considered the actual usage of the function. :) -Kevin -- Kevin A. Foss --- kfoss(a)mint.net --

1 0

Re: gpc I/O & errors
by heckenb＠helena.hut.fi 19 Dec '97

19 Dec '97

Kevin A. Foss wrote: > The advantage is that you don't have to deal with any C-isms at all and > it should work on any GPC platform without worrying about the > underlying stat() implementation. Yes, that's better, thanks. I didn't know about these binding flags. > The problem (as noted in the > comments) is that if the bind fails it thinks the file exists, which is > a bit of a kludge. However I couldn't get around the problem that > appeared when a user gives a name that is used by a directory. Frank's > code using stat() correctly reports true for directories. Well... but I think in most cases it would be more appropriate to return false for directories (I hadn't thought about this problem at all) because I'd usually call this functions before trying to open the file for reading or updating. So, I think it would be better to return false if the binding fails. (As currently implemented, the binding will fail if the given file name is a directory, or the file doesn't exist and the directory isn't writable.) -- Frank Heckenbach, Erlangen, Germany fjf(a)gmx.de http://home.pages.de/~fjf/links.htm

1 0

Re: gpc I/O & errors
by Peter Gerwinski 19 Dec '97

19 Dec '97

According to Frank Heckenbach: > Laurentiu COCEA wrote: > > 2. Where do I get a list of compile-time and run-time errors? > > I fear such a list doesn't exist -- there's much work to do about the > documentation... :-( > > If you get the source distribution of GPC, you can find a *non-complete* list > of run-time errors in rts/rts-error.c. I don't know if the compile-time errors > are listed somewhere Such a list does not exist, but in my current GPC source (the `p' directory), grep -w "error (" *.c ../*.c | less grep -w "pedwarn (" *.c ../*.c | less grep -w "warning (" *.c ../*.c | less yields 887 error, 516 pedwarn, and 461 warning messages. ("Pedwarn" means that the warning turns into an error message in `--pedantic-errors' mode.) If you want to help us to produce a list with the most important compile-time and run-time messages, be welcome. 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

Re: gpc I/O & errors
by Kevin A. Foss 18 Dec '97

18 Dec '97

On Thu, 18 Dec 1997 20:19:16 +0100, Frank Heckenbach wrote: >Laurentiu COCEA wrote: > >> 1. How do I suspend I/O error messages in order to detect the >> presence/absence of a file? In Turbo Pascal, one does this as follows: >> >> {$I-} >> reset(file); >> {$I+} >> if IOResult <> 0 then... >Sorry, there's no {$I+-} directive in GPC yet, but some work has already >been done towards implementing it. > >As for checking if a file exists, there are other ways, e.g. searching the >directory, or (at least under Un*x systems, I don't know if also under Dos) >stat(2)ing the file. The following works on my Linux system, so it should >be alright for you: It is also possible to use gpc's built in extension to file binding that checks if a file exists, e.g. something like: program pas(input, output); var a : string(200); function exists(filename: string) : Boolean; { Checks if a file exists. It returns positive if b.existing is TRUE or if the bind failed. The bind will fail for example if the name given is actually a directory name. } var b : BindingType; f : bindable text; begin unbind(f); b := binding(f); b.name := filename; bind(f, b); b := binding(f); if b.bound then exists := b.existing else exists := TRUE; unbind(f); end; begin write('Enter a file name: '); readln(a); if exists(a) then writeln(a, ' exists.') else writeln(a, ' doesn''t exist.'); end. The advantage is that you don't have to deal with any C-isms at all and it should work on any GPC platform without worrying about the underlying stat() implementation. The problem (as noted in the comments) is that if the bind fails it thinks the file exists, which is a bit of a kludge. However I couldn't get around the problem that appeared when a user gives a name that is used by a directory. Frank's code using stat() correctly reports true for directories. -Kevin -- Kevin A. Foss --- kfoss(a)mint.net --

1 0

Re: gpc I/O & errors
by heckenb＠helena.hut.fi 18 Dec '97

18 Dec '97

Laurentiu COCEA wrote: > 1. How do I suspend I/O error messages in order to detect the > presence/absence of a file? In Turbo Pascal, one does this as follows: > > {$I-} > reset(file); > {$I+} > if IOResult <> 0 then... > > In SunWorkshop Pascal it's a little bit more complicated even if its > different. > Is there something similar to these in GNU Pascal? Sorry, there's no {$I+-} directive in GPC yet, but some work has already been done towards implementing it. As for checking if a file exists, there are other ways, e.g. searching the directory, or (at least under Un*x systems, I don't know if also under Dos) stat(2)ing the file. The following works on my Linux system, so it should be alright for you: program FileExists; type TStatBuf=Array[1..$4000] of Byte; {It's currently not easy to get the correct C structure into GPC, but since I don't use the contents of the structure here, it doesn't matter. The above should be big enough, anyway... ;-} function stat(Name:CString;var Buf:TStatBuf):Integer; C; function Exists(s:String):Boolean; var Buf:TStatBuf; begin Exists:=stat(s+#0,Buf)=0 end; var a:String(200); begin Write('Enter a file name: '); readln(a); if Exists(a) then Writeln(a,' exists.') else Writeln(a,' doesn''t exist.') end. > 2. Where do I get a list of compile-time and run-time errors? I fear such a list doesn't exist -- there's much work to do about the documentation... :-( If you get the source distribution of GPC, you can find a *non-complete* list of run-time errors in rts/rts-error.c. I don't know if the compile-time errors are listed somewhere -- Frank Heckenbach, Erlangen, Germany fjf(a)gmx.de http://home.pages.de/~fjf/links.htm

1 0

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 December 1997