January 2002 - Gpc

StringUtils, FileUtils
by Frank Heckenbach 31 May '02

31 May '02

Hi, in the current GPC version, I've included a StringUtils unit for general string handling routines (on a higher level than the built-in and RTS routines). Up to now, it only contains routines for a string hash table, but I'm planning to add some more routines. Perhaps some of you also have some routines to contribute (or features to request). Conditions for inclusion into this unit should be, IMHO: - Generally useful (not meant specifically for a certain program or something), - self-contained (not relying on an external library etc.), - new functionality compared to the existing routines (whereas the same or similar functionality under a different interface for compatibility purposes should go into other units, like GPCUtil for compatibility to `Turbo Power' units), - and, of course, compilable with the current GPC version, portable, bug-free, etc. ;-) The same goes for the FileUtils unit which has been there for some months already. 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 10

Re: Complex_Arctan
by da Silva, Joe 10 Feb '02

10 Feb '02

Please see below ... Joe. > -----Original Message----- > From: Emil Jerabek [SMTP:ejer5183@artax.karlin.mff.cuni.cz] > Sent: Friday, January 25, 2002 5:23 AM > To: gpc(a)gnu.de > Subject: Complex_Arctan > > Hi all, > > I've had a look at the complex arctangent function in the RTS. It turned > out that > > - it raises a runtime error or returns an infinite imaginary part on > arguments too close > to the singularities +/- i, for example > ArcTan (Cmplx (1e6 * MinReal, +/- 1)) > > - it often returns a real number when it should return something with a > small imaginary part > instead (example: ArcTan (Cmplx (1e17, 1e17)) on IA32, YMMV), and in > general, > the imaginary part of the result is very unreliable when |Im (z)| << > |Re (z)|^2 > (because it is essentially computed as `Ln (something close to 1)'; > example: > Im (Arctan (Cmplx (1000, 0.0001))) gives 9.999989726e-11 instead of > 9.999990000e-11, > again this is machine-dependent) > > The following patch to p/rts/math.pas solves these problems (I hope it > doesn't introduce > others). > > ------------------------------ > > function Log1P (x: Real): Real; asmname 'log1p'; > > function Complex_Arctan (z: Complex): Complex; > const Ln2 = 0.6931471805599453094172321214581765680755; > var b, c, d: Real; > begin > c := Sqr (Re (z)); > b := 1 - Abs (Im (z)); > d := ((1 + Abs (Im (z))) * b - c) / 2; > [Joe da Silva] I wonder, would it be better to use the following instead, to reduce problems with overflow for large values (magnitude) of Re(z) and/or Im(z)? : c := Re(z) * 0.5 * Re(z); b := 1 - Abs(Im(z)); d := (1 + Abs(Im(z))) * 0.5 * b - c; Also, as a general question, is division more expensive than multiplication, or does the complier's optimization take care of such matters anyway? > if Abs (b) > Abs (Re (z)) then > b := Log1P (4 * (Abs (Im (z)) / b) / (b + Re (z) * (Re (z) / b))) / 4 > else if c <= MinReal then > b := (Ln2 - Ln (Abs (Re (z)))) / 2 > else > b := Log1P (4 * (Abs (Im (z)) / Re (z)) / (b * (b / Re (z)) + Re (z))) > / 4; > if Im (z) < 0 then b := -b; > Complex_Arctan := Cmplx (ArcTan2 (Re (z), d) / 2, b) > end; > ------------------------ > > > Emil Jerabek >

3 8

gcc 2.96.2 and 3.0.3
by Guillaume Rousse 04 Feb '02

04 Feb '02

Hi all. I'm trying to compile latest gpc 20011222 snapshot for Mandrake, which ships either gcc-2.96 and gcc-3.03. Latest message on this topic (early december) says gpc is not supported on these versions. Is this still true ? -- Guillaume Rousse <rousse(a)ccr.jussieu.fr> GPG key http://lis.snv.jussieu.fr/~rousse/gpgkey.html

6 23

Files
by berend＠beard.nest.nl 01 Feb '02

01 Feb '02

* Moved (from: gpc) by Berend de Boer using timEd 1.10.b3+. Peter Gerwinski wrote in a message to Berend de Boer: PG> I tried to use Extended Pascal "binding" to assign a file PG> variable to an external -- existing -- file. I just copied PG> the "Assign" pro- cedure by Berend as published in PG> borland2ep.doc. It does not always work: I often get a PG> runtime error saying that the file did not exist. If you can repeat it, compile with another Extended Pascal compiler, for example Prospero's one. If that fails too, there is really something wrong with the code. But I suspect gpc. You're using which OS? Groetjes, Berend (-: fido: 2:281/527.23 email: berend(a)beard.nest.nl

4 3

operators pow and **
by Maurice Lombardi 01 Feb '02

01 Feb '02

Hi all, It seems that operators pow and ** cannot be redefined in gpc. Is that expected ? I work with djgpp / w98 dos box (should be irrelevant I think) Thanks Maurice -- Maurice Lombardi Laboratoire de Spectrometrie Physique, Universite Joseph Fourier de Grenoble, BP87 38402 Saint Martin d'Heres Cedex FRANCE Tel: 33 (0)4 76 51 47 51 Fax: 33 (0)4 76 51 45 44 mailto:Maurice.Lombardi@ujf-grenoble.fr

4 4

Re: Mingw [was GPC newbie]
by gerdunne＠eircom.net 31 Jan '02

31 Jan '02

Hi chief. I carried out all your instructions. Downloaded MinGW-1.1.tar. I got a bit further but now I have a new problem. When I 'GPC *.pas -o *.exe ' I get 'ld: cannot find -lgpc' and when I do the GPC.EXE --verbose I get the following:- C:\>gpc.exe --verbose Using builtin specs. gpc version 19990118, based on egcs-2.91.60 19981201 (egcs-1.1.1 release) ld \lib\crt1.o -L\mingw32\lib -lgpc -lcrtdll -lmingw32 -lgcc -lmoldname -lcrtdl l -lkernel32 -ladvapi32 -lshell32 -lmingw32 -lgcc -lmoldname -lcrtdll ld: cannot find -lgpc C:\> My autoexec.bat Path is pointing at all the directories Any help would be appreciated. Regards Ger. african_chief(a)bigfoot.com wrote: < < BTW: you can get the full Mingw release from this URL: < http://sourceforge.net/project/showfiles.php?group_id=2435 < < < Best regards, The Chief < --------- < Prof. Abimbola Olowofoyeku (The African Chief) < Email: African_Chief(a)bigfoot.com < http://www.bigfoot.com/~african_chief/ <

2 1

Re: How to Handle Variable Names in Both C and Pascal
by Gloria, Jing 30 Jan '02

30 Jan '02

Chuck F, thanks for the prompt response and for the utility that you sent. I will give it a try. I will have to FTP my files from the Solaris system to my desktop PC. Everyone, I apologize for any inconvenience that HTML in my mail may cause. I am using Microsoft Outlook and I have selected Plain Text format when composing my e-mail. I do not know why it is not turning off HTML. As I am typing out this e-mail, I see that it says "Plain Text" on the header. I hope HTML is turned off now. Regards, Jing

2 1

How to Handle Variable Names in Both C and Pascal
by Gloria, Jing 30 Jan '02

30 Jan '02

I would appreciate getting suggestions on how to approach the following situation. As mentioned in previous messages to this group, I am porting code from SPARC/Sun_OS/Sun_compilers (development and runtime) to Intel/Solaris_7/GNU_compilers (also development and runtime). Source code are in C, C++, and Pascal. One requirement is to keep the source code so that it can be built in either platforms. I have used #ifdef/#else/#endif with asmname and wrapper functions/ subroutines to handle this. My problem is with several global variables. For example, I have one called "emulator_error" and it is used in 250 files with around 100 in C and 150 in Pascal source files. Is there a way to avoid editing all 150 Pascal files to put the #ifdef/#else/#endif with asmname directive? I believe writing a script to do this will be complicated since the variable names can be declared in a line where several variables are declared. For example: Var test_val, emulator_error, sum : integer; and only emulator_error is used in both C and Pascal; test_val and sum are referenced in Pascal source code only. Regards, Jing Gloria Texas Instruments Sherman, Texas

2 1

Complex_Arctan
by Emil Jerabek 29 Jan '02

29 Jan '02

Hi all, I've had a look at the complex arctangent function in the RTS. It turned out that - it raises a runtime error or returns an infinite imaginary part on arguments too close to the singularities +/- i, for example ArcTan (Cmplx (1e6 * MinReal, +/- 1)) - it often returns a real number when it should return something with a small imaginary part instead (example: ArcTan (Cmplx (1e17, 1e17)) on IA32, YMMV), and in general, the imaginary part of the result is very unreliable when |Im (z)| << |Re (z)|^2 (because it is essentially computed as `Ln (something close to 1)'; example: Im (Arctan (Cmplx (1000, 0.0001))) gives 9.999989726e-11 instead of 9.999990000e-11, again this is machine-dependent) The following patch to p/rts/math.pas solves these problems (I hope it doesn't introduce others). ------------------------------ function Log1P (x: Real): Real; asmname 'log1p'; function Complex_Arctan (z: Complex): Complex; const Ln2 = 0.6931471805599453094172321214581765680755; var b, c, d: Real; begin c := Sqr (Re (z)); b := 1 - Abs (Im (z)); d := ((1 + Abs (Im (z))) * b - c) / 2; if Abs (b) > Abs (Re (z)) then b := Log1P (4 * (Abs (Im (z)) / b) / (b + Re (z) * (Re (z) / b))) / 4 else if c <= MinReal then b := (Ln2 - Ln (Abs (Re (z)))) / 2 else b := Log1P (4 * (Abs (Im (z)) / Re (z)) / (b * (b / Re (z)) + Re (z))) / 4; if Im (z) < 0 then b := -b; Complex_Arctan := Cmplx (ArcTan2 (Re (z), d) / 2, b) end; ------------------------ Emil Jerabek

2 4

Re: Accessing Mailing List Mail - Newbie Questions
by da Silva, Joe 29 Jan '02

29 Jan '02

Please see below ... Joe. > -----Original Message----- > From: Frank Heckenbach [SMTP:frank@g-n-u.de] > Sent: Monday, January 21, 2002 2:39 AM > To: gpc(a)gnu.de > Subject: Re: Accessing Mailing List Mail - Newbie Questions > > Gloria, Jing wrote: > > > I am able to get to the mailing list archives thru my > > Web browser. I have also registered to subscribe to > > this mailing list. Pls point me to any documentation > > on how to use this system. > > There's none available. I thought it was more or less > slef-explaining. > > Or rather, as they say, the best documentation is the source which > can be found at http://fjf.gnu.de/crystal/. :-) > > > On the Web page, I can read the mail that have been > > submitted using the link from the GPC documentation. > > > > 1. How do I reply to a particular mail? There is no > > option to reply to the entry that I am reading. > > No, I haven't enabled this feature yet. It's mostly a policy > decision. We don't want anonymous replies on the list (and it might > also create another entry for spammers, though I'm not sure how many > spammers try to exploit web interfaces yet). And since the archive > (technically) independent of the list distribution, we'd have to do > a parallel user management (including passwords) for the archive. > > > 2. I sometimes see a section in the page that says > > "Reply to" with a list of previous messages. However, > > if I click on a mail entry in the list, that particular > > mail entry is displayed. > > Yes, this means as much as "The present mail is a reply to the > following mails". If you have a better idea how to name it, let me > know. > [Joe da Silva] How about "In reply to" ? > > 3. There is a login button at the end of the page for > > registered Crystal users. Do I need to register before > > I can reply to the mail in the mailing list? > > Registration is currently only used for administrators, but I don't > think it would be a problem to make it available for other users as > well. I didn't expect anyone to want to reply via this way, so I > didn't provide this options yet, but if there's some interest ... > > I'd like to know if anyone else is interested, so I can decide if I > should do the registrations manually or set up an automated system > (if there are many) ... > [Joe da Silva] I for one, only use it for reading messages. > Frank > > -- > Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/, 7977168E > GPC To-Do list, latest features, fixed bugs: > http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

2 1