Re: [fpc-devel]Stable tree broken for FreeBSD? [maintenance ramblings]

2003-03-20 Thread Peter Vreman
Hi, using the fixes branch from this morning I try: OS_TARGET=freebsd gmake all which ends in: ppc386 -dUNIX -dNOMOUSE -dNOGPM -Xs -OG2p3 -n -Fi../inc -Fi../i386 -Fi../unix -FE. -di386 -dRELEASE -I../objpas ../unix/sysutils.pp fina.inc(101,11) Error: Identifier not found DOS

Re: [fpc-devel]Re: some m68k releated stuff

2003-03-31 Thread Peter Vreman
10.Work:Develop/FPC/fixes/fpc/compiler make compiler Makefile:17: *** You need the GNU utils package to use this Makefile. Stop. Ideas? :) I'll take a look again on my GNU utils stuff, but i think everything is available, and on the path... starting make fpc_info might help you (at least

Re: [fpc-devel]RPMs and DEBs

2003-09-12 Thread Peter Vreman
Hi, How is maintaining the RPMs and DEBs? For example the spec and rules files? There is no official maintainer. I guess that i made the last changes to the files. Peter ___ fpc-devel maillist - [EMAIL PROTECTED]

Re: [fpc-devel]FCL/Packages/Lazarus compilation

2003-09-28 Thread Peter Vreman
Hi, Circular reference: fpgtk requires the FCL while the FCL requires some packages to compile (different ones though, if you already installed the packages up to this point the FCL does compile). Anyway, this is imo not nice. ^^ There are 2 sub directories under packages: base and extra The

Re: [fpc-devel]proposition changes in unit TYPES

2003-10-02 Thread Peter Vreman
I try use FPC 1.1 for Windows and I have problem with some of types declarations for example THandle is declared many times in WINDOWS.PP (base.inc) is: Thandle= HANDLE = type Cardinal; in CLASSES (classesh.inc) THANDLE = longint; in SYSUTILS with other types like POINT RECT I

RE: [fpc-devel]fpcmake

2003-10-13 Thread Peter Vreman
At 13:51 13-10-2003, you wrote: I agree with the complainant to some extent... the fpcmake files are very complicated, and make itself isn't the easiest utility to use for non (l)unix'ites -that's putting it mildly!-. However, I suppose whatever we do is going to be complicated, so we have to

Re: [fpc-devel]fpcmake

2003-10-13 Thread Peter Vreman
At 23:43 11-10-2003, you wrote: If we change FPCMAKE to generate code like this: - #add dir of make.exe to PATH override PATH +=$(dir $(MAKE))- override PATH:=$(subst \,/,$(PATH)) #$(error $(PATH)) ifeq ($(findstring ;,$(PATH)),) inUnix=1

Re: [fpc-devel]gdb info problems

2003-10-24 Thread Peter Vreman
Hi, There are some problems with gdb info. Because line 1 contains code and by contrast line 3 does not (where are only 2 lines)(i tested with more examples, there was same problems). I am interesting of their origin. begin end. gdbinfo line 1 Line 1 of test.pp is at address $40cab8

Re: [fpc-devel]Operator overloading question.

2003-11-07 Thread Peter Vreman
Hello. While trying to use the operator overload feature of FPC, I find out a restriction not mentioned on the documentation. One can overload several operators but not the comparison one (=); this, at least, while operating on classes. The code below shows this situation (commenting out

Re: [fpc-devel]Bug or missing documentation?

2003-11-07 Thread Peter Vreman
Hello. The code below gives: foo.pas(13,33) Error: Duplicate identifier NAME when compiled with -S2. It compiles ok with -Sd though. The thing is, I coudn't find any reference to this difference of behavior under the Compiler Modes section of FPC docs, neither in other sections. Is this a

Re: [fpc-devel]FreePascal and AnsiStrings

2003-11-07 Thread Peter Vreman
1. Is the order purely incidental or is there a reason why FreePascal chose to use a different order than Delphi? incidental 2. I know it is officially discouraged to use 'internal formats', but for optimization's sake, would it be safe to assume that the internal format for FreePascal's

Re: [fpc-devel]Bug in AssignPipe() on Linux

2003-11-07 Thread Peter Vreman
Using the latest version of the compiler, and I am having a problem with the popen() function on Linux. It looks like popen() fails because AssignStream() fails, because AssignPipe() fails, and that's about as far as I got. ( I'm lost when it comes to the do_SysCall() stuff. ) FAILS:

Re: [fpc-devel]Bug in AssignPipe() on Linux

2003-11-08 Thread Peter Vreman
At 10:38 8-11-2003, you wrote: Hello, Peter - thanks for your response. Did you only test the 1.9.0 compiler or also distributed the 1.9.0 RTL ? I tested with each compiler version with its own RTL. Please check for a leftover syscallo.inc in fpc/rtl/linux/. It should only be available in in

Re: [fpc-devel]daily binary snapshots

2003-11-10 Thread Peter Vreman
Hi folks, daily binary snapshots of fpc and lazarus without source, doc, etc would be useful. binary snapshots are already available: http://www.freepascal.org/develop.html#snapshots ___ fpc-devel maillist - [EMAIL PROTECTED]

Re: [fpc-devel]TThread hara-kiri

2003-11-13 Thread Peter Vreman
Does the example program threads.pp (doc/examples/fcl/threads.pp) work for anybody under Linux? It aborts for me as soon as the first thread is created, printing Killed. When I compile it with 1.9.1 I get the same effect - program is immediately killed (the message Killed is printed by

Re: [fpc-devel]TThread hara-kiri

2003-11-13 Thread Peter Vreman
Please no more clone() :-) BeginThread uses pthread_create :-) Btw -- shouldn't tthread.inc then be moved to inc/ instead of {sys}/? It would probably be platform-independent itself if using BeginThread. Ofcourse. That is the whole point of making tthread.inc dependent on SysThrds. The

Re: [fpc-devel]Some idea of joint units

2003-11-13 Thread Peter Vreman
But first there is smartlinking, second there could be smartuniting Impossible, you can't have the compiler remove a unit just because nothing from it is used, the unit could contain an object that only registers itself with a factory on initialization, thus nothing from the unit is ever

Re: [fpc-devel]threads in linux

2003-11-14 Thread Peter Vreman
But I do think that using BeginThread and such functions would be far better. I'd be willing to implement this, but I only have access to a Linux 2.6 (with NPTL) system (my own) and some Linux 2.4 (with LinuxThreads) systems (at university). I thought that you were already rewriting it to

[fpc-devel]default calling convention change for i386

2003-12-23 Thread Peter Vreman
Hi all, From today the default calling convention for i386 is changed from stdcall (the default since 1.9.0) to register calling. This means that you have to look at how assembler code loads the arguments and maybe store them yourself in local variables. The register calling is compatible with

Re: [fpc-devel]default calling convention change for i386

2003-12-26 Thread Peter Vreman
On Wed, 24 Dec 2003, Peter Vreman wrote: }Most important was compatibility, second is speed. In the drystone test }there was a performance gain of at least 10%. Okay, I think this answers the question :) } Register convention saves opcode space in the called function, because } within

Re: [fpc-devel]Bug on writing value into file of byte

2003-12-27 Thread Peter Vreman
Hi, compiling this program with 1.0.10 give me error Type mismatch. var outf: file of byte; begin assign(outf, 'test.bug'); rewrite(outf); write(outf, 10); {only explicit typecasting helps: byte(10)} close(outf); end. Is this fixed in 1.9? Fixed

Re: [fpc-devel]wrong MAKE all

2004-01-11 Thread Peter Vreman
At 23:10 9-1-2004, you wrote: Hi I've try to do make all but I recieve ***No rule to make target clean my home dir is c:\fp, and I use WIN98 I nottice that UTILS clean good, from log : MAKE.EXE[5]: Leaving directory `C:/fp/source/compiler/utils' but for rtl MAKE looking for c:\pp

Re: [fpc-devel]More units for MacOS X port?

2004-01-26 Thread Peter Vreman
On 26 jan 2004, at 11:40, Karl-Michael Schindler wrote: With the MacOS X port available, i played (maybe fooled) around in order to port some of the basic units, such as termio crt video keyboard serial. I managed to get it compiling by transferring missing definitions of constants and

Re: [fpc-devel]Determining overloads between Single and Double

2004-02-15 Thread Peter Vreman
Jonas Maebe wrote: On 13 feb 2004, at 22:17, Michalis Kamburelis wrote: but sometimes it's a pain trying to write code in ObjFpc mode that should compile under Delphi too In that case, why do you want to use that single(1.0) construct? That won't compile in Delphi either... I'm not sure

Re: [fpc-devel]Determining overloads between Single and Double

2004-02-15 Thread Peter Vreman
Delphi mode I can't use p(1.0) and I can't use p(Single(1.0)) too. Add a new overload for extended and it compiles also in delphi mode. Bug report 2129 is the source that single(1.0) typecasts does not work as expected. A solution can be to do conversion for float constants. Conversion is

Re: [fpc-devel]fpImage

2004-02-15 Thread Peter Vreman
Hi, Is there a maintainer for fpImage ? And what is the development status of fp(read|write)bmp ? Is anyone working on adding other encodings than 32 bit ? If not, than I'll have a try to add the missing encodings. Michael van Canneyt ([EMAIL PROTECTED]) is the maintainer. patches can be

Re: [fpc-devel]Question

2004-02-24 Thread Peter Vreman
Hi ! Thank you for Free Pascal ! I have small question. Possible insert binary dump (text) as code in procedure ? (Similarity of operator 'INLINE' in old Turbo Pascal.) And if yes then as ? You can use the data2inc tool to convert binary data to a const type. inline is a

Re: [fpc-devel]fpc l10n issue

2004-03-03 Thread Peter Vreman
Hello, 1) did you thought ever about creation l10n framework for fpc? I think contributors could submit translated fpc messages files (.msg) and other locale dependent stuff there. It needs tool for merging translated messages from one version to higher also. resourcestrings and the

Re: [fpc-devel]Linux readdir syscall is superseded by getdents

2004-04-04 Thread Peter Vreman
Hello all, I'm new to this list and I coudn't find out how to send in patches so i do it here... On my system FPC doesn't work because the fpreaddir-function failes on larger directories. I saw that fpreaddir still uses the linux readdir-call, wich is superseded by getdents. (also the

Re: [fpc-devel]faster parser

2004-04-15 Thread Peter Vreman
Hi, I'll try to write a faster parser/compiler, which also supports UTF8. Are you sure that you want to write a parser? Because that is the part that does the semantics and handles all the statements and does not known about UTF8. What is wrong with the current scanner? Peter

Re: [fpc-devel]a bug is found

2004-04-22 Thread Peter Vreman
I submitted a bug in the FPC bug list (as #3061). I would discuss it a little because now I write a stuff to radically solve problems with indexes export. Maybe, these two problems need to be solved as one complex? But I'm understandig that big changes in the compiler code may cause new

Re: Re[2]: [fpc-devel]about the bug 2364

2004-05-14 Thread Peter Vreman
Hello Peter, Wednesday, April 28, 2004, 4:02:59 PM, you wrote: Hello all, I have some ideas about the bug #2363 (Unfixed Error export problem). Second way: storing exported items info in PPU. PV That is possible. It needs a new ppuentry that will contain the following PV items per

Re: [fpc-devel]Probably three new bugs

2004-05-09 Thread Peter Vreman
At 14:15 9-5-2004, you wrote: I found three bugs: 1)When I am compiling program with {$LINKLIB c} directive, I have got error, that cannot find gcc library (d:/maciek/fpc/bin/go32v2/ld.exe: cannot find -lgcc). 2)When I am using printf function, I have got error message like this:

Re: [fpc-devel]FileIsReadOnly problem

2004-04-26 Thread Peter Vreman
At 20:11 24-4-2004, you wrote: The sense of the result of FileIsReadOnly seems to be inverted - see attched patch Colin Thanks, applied Peter ___ fpc-devel maillist - [EMAIL PROTECTED] http://lists.freepascal.org/mailman/listinfo/fpc-devel

Re: [fpc-devel]about the bug 2364

2004-04-28 Thread Peter Vreman
Hello all, I have some ideas about the bug #2363 (Unfixed Error export problem). I see two ways to solve this problem but I'm not sure that I'm able to realize them. First way (may be wrong): generating edata sections (or their analogs for non-Windows platforms) vor every unit that

Re: [fpc-devel]faster parser

2004-04-29 Thread Peter Vreman
At 19:07 29-4-2004, you wrote: Hi, I wrote a Red-Black tree, which seems to be faster than TDictionary, even without spreading items into multible slots. The Algorithm for rearanging the nodes is faster as for an AVL tree. The disadvantage to a normal search tree is that every node requires an

Re: [fpc-devel]Recompile the RTL to trace bugs

2004-05-03 Thread Peter Vreman
I guess the problem with dynarrays and openarrays (also as with overloaded functions wich are also very buggy for a long time) is the fact that FPC core team are old stylish programmers, and do not use these modern features themselves. Thus these features are for users and are not so solid.

Re: [fpc-devel]Recompile the RTL to trace bugs

2004-05-03 Thread Peter Vreman
It is also not possible to use new features in the compiler. The compiler needs to be able to be compiled with 1.0.x for bootstrapping. This may be a dumb question but, can you explain what you mean by this, and why this is not possible? When building a new compiler you always start with

Re: [fpc-devel]Many new warnings

2004-05-29 Thread Peter Vreman
At 20:17 29-5-2004, you wrote: On Fri, 28 May 2004 23:24:51 +0200 (CEST) Peter Vreman [EMAIL PROTECTED] wrote: The current cvs fpc 1.9.3 gives many new warnings. The following code demonstrates 6 types, where I'm not sure why they now need a warning. Maybe the compiler cracks can tell me

Re: [fpc-devel]fpc/snapshot/v19/win32-i386 build

2004-05-30 Thread Peter Vreman
Hello all, can somebody run the script for a new fpc/snapshot/v19/win32-i386 build. The current snap is dated 18.05.2004. Only John Lee has access to the machine. ___ fpc-devel maillist - [EMAIL PROTECTED]

Re: [fpc-devel]Fix for Make / Error with space in PATH env var

2004-06-28 Thread Peter Vreman
I originally posted this to fpc-pascal but didn't get much of a response. The problem where having spaces in the environmental variable will error out the Make process has been described. Here is a work around I came up with: It seems to work on my system. Could someone check this out and

Re: [fpc-devel]Fix for Make / Error with space in PATH env var

2004-06-28 Thread Peter Vreman
It looks like a hack. Maybe it can fix this issue for pwd.exe, but it occurres more often in Makefiles. Those other parts will then still fail. Peter, I agree with you. However, Make is unwilling to fix the problem. They feel that having spaces in the path or in filenames is bad

Re: [fpc-devel]smartlink for Win32 dlls fails again

2004-06-29 Thread Peter Vreman
Hello all, Today I built last compiler (1.9.5) from sources. Then I tried to compile a small test win32 dll with -CX -XX and got a binary without export functions entries. That is interesting that using -a option (using external assembler) solves this problem. These are all know issues.

Re: [fpc-devel]FPC 2.0

2004-07-11 Thread Peter Vreman
At 14:20 11-7-2004, you wrote: Hello all, is there a release date for FPC 2.0 ? When it is ready. We can't give a date as all developpers have limited time. In the meanwhile you can use the 1.9.x beta's. They are already more stable than 1.0.x. BTW: Can somebody produce a new fpc win32 binary

Re: [fpc-devel]Make clean fails for cvs.

2004-07-12 Thread Peter Vreman
Hi, If I do make distclean in the fpc root dir, I get the following error: make[2]: *** No rule to make target `dbase_clean', needed by `clean'. Stop. make[1]: *** [db_clean] Error 2 make: *** [fcl_distclean] Error 2 Make fcl gives: make[2]: *** No rule to make target `dbase_all',

Re: [fpc-devel]Version numbers in the CVS

2004-07-12 Thread Peter Vreman
Hello, is there a reason that the version numbers in the CVS are different? Some parts has version number 1.9.5, some has 1.9.4. Just a simple search-replace doesn't work since you then also change the change-logs. So maybe this just has still to be done? It gave problems when i've tried

Re: [fpc-devel]Building 1.9.4 on a Windows box

2004-07-16 Thread Peter Vreman
Hi All, Maybe its my lack of sleep or too much coffee but I just cant seem to get 1.9.4 to build. I have the 1.9.2 binary installed and it is working fine. I keep getting the error: system.pp(32,2) Fatal: Can't open include file systemh.inc make[2]: *** [msg2inc.exe] Error 1 make[2]:

Re: [fpc-devel]Amd64?

2004-08-26 Thread Peter Vreman
Hi, is it possible to compile pascal programs for amd64? We have a Delphi program that needs more of 3Gb ram user space... it'd be a perfect excuse to begin to use fpc at work. Current CVS can be compiled for amd64. See for example the testsuite run at

Re: [fpc-devel]bug in Changes.XXX file generation?

2004-08-26 Thread Peter Vreman
On Thu, 26 Aug 2004 10:31:29 +0200, Vincent Snijders wrote I noticed that sometimes cvs-changes are not included in the Changes.lazarus file in fpc/logs. For example there was no entry for the adding of: http://www.freepascal.org/cgi-bin/viewcvs.cgi/projects/lazarus/components/im

Re: [fpc-devel]patch for fpcmake (web bug 3276)

2004-08-27 Thread Peter Vreman
At 22:07 27-8-2004, you wrote: Hi, Attached patch implements the support for .lpr sourcefiles in fpcmake, just like it supports .dpr source files. See also web bug 3276. Applied. Thanks Peter ___ fpc-devel maillist - [EMAIL PROTECTED]

Re: [fpc-devel]1.9.4 String comparison problem

2004-08-27 Thread Peter Vreman
At 16:33 26-8-2004, you wrote: Hi! In FPC 1.9.4 there is a bug concerning comparison of a string if it is empty. The program to test is Program Bug; Var SS : ShortString; AS : AnsiString; Begin SS := 'asdf'; AS := 'asdf'; WriteLn('SS : ', SS ''); WriteLn('Length(SS) 0:

Re: [fpc-devel]Issues with GDB on Mac OS X with mixed case file names

2004-08-28 Thread Peter Vreman
Many thanks for the fix. Unfortunately, I discovered too late, that this enables GDB support on Mac OS X only for the main program file. In unit files it still includes the lowercase filenames into the .stabs. However, I found the problem and a possible solution. In globals.pas there are

Re: [fpc-devel]Issues with GDB on Mac OS X with mixed case file

2004-08-29 Thread Peter Vreman
At 14:00 29-8-2004, you wrote: On Sat, 28 Aug 2004 22:26:08 +0200 (CEST) Peter Vreman [EMAIL PROTECTED] wrote: However, I found the problem and a possible solution. In globals.pas there are three FindFile routines, namely: TSearchPathList.FindFile, FindFile, and FindFilePchar. All

Re: [fpc-devel]Issues with GDB on Mac OS X with mixed case file

2004-08-29 Thread Peter Vreman
One other topic: Why there are *.pp files? The handling is always very nonpractical and boring. Should we rename them to *.pas? I don't see problems. You rename all your sources to *.pas. And you can set highlighting for *.pp also to pascal. Under linux a lot of editors already support *.pp for

Re: [fpc-devel]Generic Types in FPC

2004-09-12 Thread Peter Vreman
According to Robert Love's blog Delphi will provide generic types support. Will FPC support the feature in the same way? Delphi Syntax for Generic Types will be: type TFooT = class private data1: T; public function SomMethod(param1: INteger; Param2 :T) : Integer; end;

Re: [fpc-devel]Proposal for audo-destructors

2004-09-12 Thread Peter Vreman
The matter has been discussed several times by the FPC core group, at great length. No satisfying solution on which everyone agreed was reached. I would really appreciate some comments on my question about patches that would implement optional, unofficial or pluggable features. I would like

Re: [fpc-devel] bug report

2004-10-11 Thread Peter Vreman
http://www.freepascal.org states: Version 1.0.10 is the latest stable version the Free Pascal. http://www.us.freepascal.org/develop.html states: Download Daily Update of Release Tree (version 1.0.x) These snapshots contain the latest bug fixes, without any major new feature. This should be

[fpc-devel] path macro's changed

2004-11-01 Thread Peter Vreman
Since yesterday then possible macro's that can be used in the paths are changed. The following macro's are now supported, with examples of the value: $fpcversion = 1.9.4 $fpcdate = compiler date $fpccpu = i386 $fpcos= linux $fpctarget= i386-linux Also the * can now be

Re: [fpc-devel] gui console in windows

2004-11-04 Thread Peter Vreman
At 01:07 5-11-2004, you wrote: Hello, i am interesting, what do you think about redirecting text streams not to windows console, but creating free pascal's own gui console (for example implementing it under -WX switch)? I mean to do something like graph unit is done now or unit wincrt in Virtual

[fpc-devel] new go32v2 and win32 as/ld and make utils

2004-11-05 Thread Peter Vreman
There are new as and ld versions available (asld*.zip) for go32v2 and win32. Also the tools needed for the new makefiles are provided (make*.zip). These packages will be also part of the next release and are therefor recommended when you want to use the current 1.9.5 development compiler.

Re: [fpc-devel] new go32v2 and win32 as/ld and make utils

2004-11-05 Thread Peter Vreman
There are new as and ld versions available (asld*.zip) for go32v2 and win32. Also the tools needed for the new makefiles are provided (make*.zip). These packages will be also part of the next release and are therefor recommended when you want to use the current 1.9.5 development compiler.

Re: [fpc-devel] (no subject)

2004-11-06 Thread Peter Vreman
the new directory structure has broken something!? D:\Lazarusmake process_begin: CreateProcess((null), ppc386.exe -iVSPTPSOTO, ...) failed. process_begin: CreateProcess((null), ppc386.exe -iSP, ...) failed. process_begin: CreateProcess((null), ppc386.exe -iTP, ...) failed. process_begin:

[fpc-devel] Memavail, Maxavail, Heapsize removed use GetHeapStatus instead

2004-11-22 Thread Peter Vreman
The following functions are removed: - MemAvail - MaxAvail - HeapSize With the new heap manager the values that the functions returned are invalid and can't be used to make any decisions whether there is enough memory. There is a new call added to replace the functionality: procedure

Re: [fpc-devel] external problem

2004-11-25 Thread Peter Vreman
- ... implementation {$LINK dfc.ow} //GCC object file style procedure dfCrap(setoperation: Tso); cdecl; external; ... - Hi! The lines above do compile and work fine

[fpc-devel] New rules for External

2004-11-25 Thread Peter Vreman
- Interface declaration has External Every implementation declaration will give an error: (03210_E_Procedure is already declared EXTERNAL) - Interface declaration has no External Implementation has External - For Win32 (maybe OS/2 also?) this is allowed when the import is done

Re: [fpc-devel] external problem

2004-11-25 Thread Peter Vreman
interface procedure dfCrap(setoperation: Tso); cdecl; implementation {$LINK dfc.ow} //GCC object file style procedure dfCrap(setoperation: Tso); cdecl; external; ... The declaration does exist in the interface! BTW: I am working under Win32. You aren't importing from a DLL, so

Re: [fpc-devel] New rules for External

2004-11-26 Thread Peter Vreman
This is not Delphi compatible ... Ofcourse we check this before changing things like this. Delphi only accepts the next directive for importing from DLLs: external 'dllname' name 'functionname' FPC supports besides importing from DLLs also referencing other public accessible assembler

Re: [fpc-devel] problem making docs

2004-12-20 Thread Peter Vreman
Hi, Two days ago, I could make the docs by doing: make htmldist in the fpc\doc directory. Now I get these errors: make htmlinstall DOCINSTALLDIR=doc make[1]: Binnengaan van map `/home/vsds/lazarus/source/fpc-1.9/fpc/docs' make[1]: *** Geen regel voor aanmaken doel `includes', nodig voor

Re: [fpc-devel] THandle and 64bit platforms

2004-12-20 Thread Peter Vreman
Hi, From what I understand from the definition of THandle in sysunixh.inc it is defined as a LongInt, even on 64 bit platforms. Currently I'm porting Lazarus to x64 and there I need a 64bit THandle, how to solve it ? File descriptors are still 32bit on x86_64, therefor Thandle=32bit.

RE: [fpc-devel] THandle and 64bit platforms

2004-12-20 Thread Peter Vreman
Hi, From what I understand from the definition of THandle in sysunixh.inc it is defined as a LongInt, even on 64 bit platforms. Currently I'm porting Lazarus to x64 and there I need a 64bit THandle, how to solve it ? File descriptors are still 32bit on x86_64, therefor Thandle=32bit. A

RE: [fpc-devel] THandle and 64bit platforms

2004-12-20 Thread Peter Vreman
Hi, From what I understand from the definition of THandle in sysunixh.inc it is defined as a LongInt, even on 64 bit platforms. Currently I'm porting Lazarus to x64 and there I need a 64bit THandle, how to solve it ? File descriptors are still 32bit on x86_64, therefor Thandle=32bit. A

Re: [fpc-devel] THandle and 64bit platforms

2004-12-20 Thread Peter Vreman
That said, the RTL should also avoid confusion with the Windows/Delphi THandle type, and introduce a cross-platform and opaque TFileHandle type. It's text/file/file of in pascal ;) Which will happen to be equal to THandle on 32-bit windows. On 32-bit Linux, the definition of THandle will

Re: [fpc-devel] TList slowness in classes

2004-12-24 Thread Peter Vreman
I'm attaching a simple demo program that shows this. When compiled like fpc -OG -O2 -Op2 demo_resourcestring_slow.pas (to get maximum optimizations) sample output of it is Time of Foo_Normal: 16 Time of Foo_ResourceString: 106 So time difference is really noticeable.

Re: [fpc-devel] TList slowness in classes

2004-12-24 Thread Peter Vreman
This is because there is an extra (implicit) Try/Finally block. Thank you and Peter for answers. This way I was able to see how try...finally section looks in assembler :) Anyway, I understand that the answer is can't be speed up. OK, I can live with that. That is not correct. For your own

Re: [fpc-devel] Re: compiler bug?

2004-12-31 Thread Peter Vreman
How about this one: program problem; begin Write('Test One: '); if ( boolean(255) ) then WriteLn(True) else WriteLn(False); Write('Test Two: '); WriteLn( boolean(255) ); end. FPC 1.9.4 output: Test One: FALSE Test Two: TRUE Kylix 1.0 output: Test One: TRUE Test

Re: [fpc-devel] Two classes.pp - why?

2005-01-09 Thread Peter Vreman
There are two classes.pp in the fpc sources for every OS. First question: Why? Second question: Are there any other double units? Third: Do I need to add a workaround in Lazarus to find the right one, or will one of them be removed soon anyway? It has only one classes.pp per os in the

Re: [fpc-devel] The new 1.9.6 release

2005-01-12 Thread Peter Vreman
There are many problems in this version - at least under the GO32V2 target. Mainly the things asociated with the FreeVision. Even the installer doesn't work properly. The mouse is not visible and the 'OK' button is in strange way hidden. Nor in the IDE is the mouse not shown. Interresting is,

Re: [fpc-devel] WINDRES WS_CAPTION

2005-01-12 Thread Peter Vreman
Hello! I used FPC 1.0.4 for a long time. Now I switched to 1.9.6. I have some programs which has a window without standard caption. The resource file defines the windows style az WS_BORDER, but the window (program is compiled with 1.9.6) has a caption! I opened the exe with Resource Hacker

RE: [fpc-devel] WINDRES WS_CAPTION

2005-01-12 Thread Peter Vreman
I used FPC 1.0.4 for a long time. Now I switched to 1.9.6. I have some programs which has a window without standard caption. The resource file defines the windows style az WS_BORDER, but the window (program is compiled with 1.9.6) has a caption! I opened the exe with Resource Hacker

Re: [fpc-devel] Lazarus problems with current FPC 1.9.7

2005-01-28 Thread Peter Vreman
Since a week I get strange, sporadic errors with current fpc. For example: TMainIDE.DoSaveAll Program received signal SIGSEGV, Segmentation fault. $0a9a0c08 in ?? () (gdb) bt #0 $0a9a0c08 in ?? () First symbol in segment of executable not a source symbol That's why I think, it's a

Re: [fpc-devel] msg to po conversion

2005-02-01 Thread Peter Vreman
Hello, could you tell me, are there tools for Free Pascal's .msg file conversion to .po format and backwards? I have traslated 1.0.6 messages and want to move them to 2.0. I guess you mean the error*.msg that are used by the compiler. There are no conversion tools for it. There is a msgdif

Re: [fpc-devel] bug 3626

2005-02-02 Thread Peter Vreman
im not going to argue whether this is right or wrong but in delphi at least $R only affects run time range checking. In delphi that code will give an error (constant expression violates subrange bounds) even with range checking off. I think this is a special case with for-loops. Delphi also

[fpc-devel] LongString type disabled

2005-02-04 Thread Peter Vreman
I've disabled the Longstring type in the System unit. There are no helper/conversion functions implemented to support this string type. ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel

Re: [fpc-devel] porting

2005-02-06 Thread Peter Vreman
peter green wrote: i want to atempt to port FPC to a new CPU target (PIC18F452) there don't seem to be any guides to starting a port. any idea where one sohuld start when porting fpc. note: the cpu architecture in question has nowhere near enough power to support the compiler so all

Re: [fpc-devel] Internal error in Delphi mode

2005-02-11 Thread Peter Vreman
Hi All, I've just recently started using Free Pascal through the Lazarus project, and I've found a bit of my code which is causing problems with the compiler. I've attached the complete sample program below. In objfpc mode everything works fine; in delphi mode the Length(c.IntArray[0])

Re: [fpc-devel] Bug in PWidechar and refcounting

2005-02-13 Thread Peter Vreman
Hi all, the bug-report on the website doesn't work, so I write a mail: If you assign a widechar to a pwidechar, the refcount isn't increased. This construct is often used in the variants-code, which is allowed, according to the comments at the top of wstrings.inc. An example: Program

Re: [fpc-devel] Bug in PWidechar and refcounting

2005-02-13 Thread Peter Vreman
Hi all, the bug-report on the website doesn't work, so I write a mail: If you assign a widechar to a pwidechar, the refcount isn't increased. This construct is often used in the variants-code, which is allowed, according to the comments at the top of wstrings.inc. An example: Program

Re: [fpc-devel] Re: Free Pascal Bug 3640 modification.

2005-02-14 Thread Peter Vreman
Hi, I write here because bug database is unavailable. Peter wrote that fpc and IDE fp contain the same compiler, that means that fpc also generates sub directory info. I don't know how it is possible but there are differences between symbols for gdb generated by fpc and fp. Compile with the

Re: [fpc-devel] Video unit -- bug? Tracedump -- request?

2005-02-16 Thread Peter Vreman
Video doesn't restore terminal settings on Linux after exiting -- the command line doesn't show any typed in text, and doesn't process CR characters... Is this a bug? And if not, what's the cure? (except of course manual reset of the console after execution) Do you call donevideo?

Re: [fpc-devel] Bug in PWidechar and refcounting

2005-02-16 Thread Peter Vreman
Alexey Barkovoy wrote: PAnsiChar, PChar are just pointers and not garbage collected by compiler. But AnsiString and WideString are compiler managed types. So, as Peter mentioned, behaviour you are seeing is by design. In Delphi WideString is not reference counted at all. The layout of the

Re: [fpc-devel] Abbrevia and Delphi compatibility

2005-02-16 Thread Peter Vreman
After eliminating dozens of bugs now I have a working version of Abbrevia for Delphi. Unfortunately this version is not usable with FPC, primarily because FPC doesn't support properties for the Object type. In my code I like to use Object for records with methods and properties, which need

Re: [fpc-devel] Abbrevia and Delphi compatibility

2005-02-16 Thread Peter Vreman
After eliminating dozens of bugs now I have a working version of Abbrevia for Delphi. Unfortunately this version is not usable with FPC, primarily because FPC doesn't support properties for the Object type. In my code I like to use Object for records with methods and properties, which need

Re: [fpc-devel] TShiftState as enum

2005-02-17 Thread Peter Vreman
Older Delphi (and TP?) versions implemented sets of subranges (e.g. set of 510..515) by stripping unused bytes in front of the set. The lowest bit in a set variable always had an ordinal value of 2^n, and above set would occupy 2 bytes, equaling an set of (504..519). In newer Delphi versions

Re: [fpc-devel] Can't declare procedure as EXTERNAL

2005-02-20 Thread Peter Vreman
Hi, I am porting CLX to FPC. Some time ago, when I created a Qt/Embedded binding for FPC, Qt.pas did not compile anymore due to some changes in FPC. The error was: Can't declare procedure as EXTERNAL See also: http://www.nl.freepascal.org/lists/fpc-pascal/2004-December/007738.html I

Re: [fpc-devel] compiling fpc from cvs

2005-02-21 Thread Peter Vreman
So let's assume I'm a completely idiot and I have downloaded fpc from cvs to compile it on a windows system. Do I have any hope to compile it? Install fpc 1.0.10 Download fpc from cvs Start cmd.exe Go to the new fpc directory Type 'make all' Not much different from compiling any other C

Re: [fpc-devel] Improving Ref Counting

2005-02-27 Thread Peter Vreman
Why are you looking at GC/Refcounting when the problem is the try..finally? It is better to rewrite the try..finally code using the C++ ABI for exception handling. Where do you see improvements in the C++ ABI? Or even differences? Windows implements this ABI, and every language should use

Re: [fpc-devel] Patch for bug 3595 and 3758

2005-03-07 Thread Peter Vreman
Here is the anounced patch. Thanks, Applied ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-devel

Re: [fpc-devel] Local procedures as procedural parameter

2005-03-13 Thread Peter Vreman
With the above method, it should be relatively simple to implement local procedures (functions) as procedural parameters, without affecting the code generation of the current compiler. The default calling convention is with registers. How do you want to handle that? At beforehand from the

Re: [fpc-devel] Local procedures as procedural parameter

2005-03-16 Thread Peter Vreman
That won't work on x86 when the static link would be on the stack, because there the callee removes the parameters from the stack (so if it's a global function, it will remove sizeof(pointer) bytes too little from the stack. On x86, you can generate code that checks the static link pointer

Re: [fpc-devel] Local procedures as procedural parameter

2005-03-16 Thread Peter Vreman
On 16 Mar 2005, at 21:46, Peter Vreman wrote: Then you can do it just as well on ppc and other processors, but the point of the trick was to avoid having to do this. Implementing this sort of hacks will complicate the code generator (I'm not even immediately sure how it would be implemented

  1   2   3   4   >