On Sat, 14 Mar 2009, Przemyslaw Czerpak wrote:
Hi,
> > set
> > PATH=C:\devl\cygwin\bin;C:\devl\cygwin\opt\mingw32ce\bin;C:\devl\hbvc-1.1\bin;%PATH%
^^
> > make -C doc install
> > make: make: Command not found
> > make: [doc.inst] E
On Sat, 14 Mar 2009, Phil Barnett wrote:
Hi Phil,
>>> The reason for the Harbour exception was so that we could include the
>>> macro compiler inside distributed executables without violating the GPL.
>> Can you explain it?
>> The macro compiler code is on exactly the same license as rest of Har
On Sat, 14 Mar 2009, Phil Barnett wrote:
Hi,
> LGPL only protects from GPL for dynamically linked code. If you staticly
> link LGPL, it becomes GPL and has the problem mentioned before.
> We spent months working out the license. I have the archives if we need to
> study.
Thank you for informat
On Sat, 14 Mar 2009, marek.horody...@interia.pl wrote:
> When I try compile this code :
[...]
>b := { |n, x|
>y := valtype( n)
>// test switch in extendend codeblok
>SWITCH y
>CASE 'N'
> x := Str( n)
>Break
^
>CASE 'D'
> x := DTOC( n)
> BR
On Sat, 14 Mar 2009, Phil Barnett wrote:
Hi,
> The exception was never about the compiler itself. It is all about users
> wishing to create proprietary code executables with GPL code built into it.
> The exception is for the users of the compiler, not the compiler itself.
I know though some pa
On Sun, 15 Mar 2009, Chen Kedem wrote:
Hi,
> > My question was why you were talking about macro compiler.
> > What is the difference between macro compiler and other
> > Harbour code like HVM or RTL
> The idea was that macro compiler contain parts of the compiler inside,
> and the user code is co
On Sat, 14 Mar 2009, Vailton Renato wrote:
Hi,
> I am migrating some routines for Harbour and noticed the absence of
> HB_FUNCPTR() that I used before. I came to develop the same function
> in Harbour and after researching the sources I found __DYNSN2SYM()
> which has the same purpose.
> I would
On Mon, 16 Mar 2009, Juan Gálvez wrote:
> Hi Przemek,
> You're right. It works.
> My fault is I thought hbcompat.ch would make the work and it seems it
> doesn't make it.
> Why ?
Because in your example PP has to know about extended codeblocks before it
begins any translations so before hbcompat.
Hi All,
I would like to ask VFP users to give me an information about some
timestamp/datetime transformations in this language.
I'm interesting in existing functions transform pictures, etc.
I'm ready to commit timestamp support to HVM but I do not want
to reinvent the wheel and create correspondi
Hi All,
I would like to ask someone who uses xBase++ and Harbour to make
some tests with tests/speedtst.prg.
Current tests/speedtst.prg code can be compiled and executed by Clipper
so I hope it should also work with xBase++ as least in single thread
mode which I set as default when __XPP__ macro
On Mon, 16 Mar 2009, Szak�ts Viktor wrote:
Hi,
> ---
> Xbase++ (R) Compiler 1.90.331 Apr 27 2006
> Copyright (c) Alaska Software. All rights reserved.
> File: speedtst.prg Line: 816
> speedtst.prg(182:0): error XBT0200: Syntax Error
> speedtst.prg(265:0): warning XBT0105: Function t030 does not
On Mon, 16 Mar 2009, Lorenzo Fiorini wrote:
> I'm trying to find out the best way to secure dbfs.
> I already know and use the SUID and SGID but they seem not suitable
> for all envs since the files created by the app are not owned by the
> user and this doesn't allow many typical import/export fea
s so I don't want
to change sth what can be left as is.
best regards,
Przemek
/*
* $Id: speedtst.prg 10578 2009-03-10 11:17:18Z druzus $
*/
/*
* Harbour Project source code:
* HVM speed test program
*
* Copyright 2008 Przemyslaw Czerpak
* www - http://www.harbour-project.org
*
*
On Mon, 16 Mar 2009, Andi Jahja wrote:
Hi,
> OWatcom: I suggest you have oWatcom Harbour flags standardized as what
> other compilers use as of now. Take a look at -s and -r switch (default
> is -r which is "use register calling convention"). Other Windoze
> compiler are using -s equivalent (-s =
-10 11:17:18Z druzus $
*/
/*
* Harbour Project source code:
*HVM speed test program
*
* Copyright 2008 Przemyslaw Czerpak
* www - http://www.harbour-project.org
*
*/
#define N_TESTS 55
#define N_LOOPS 100
#define ARR_LEN 16
#ifndef __HARBOUR__
#ifndef __XPP__
#if
* Harbour Project source code:
* HVM speed test program
*
* Copyright 2008 Przemyslaw Czerpak
* www - http://www.harbour-project.org
*
*/
#define N_TESTS 55
#define N_LOOPS 100
#define ARR_LEN 16
#ifndef __HARBOUR__
#ifndef __XPP__
#ifndef __CLIP__
#define __CL
On Mon, 16 Mar 2009, Szak�ts Viktor wrote:
Hi,
Thank you for your results.
> with 2 threads (The Harbour results look this weird on every run):
> 03/16/2009 20:08:01 Windows XP 05.01 Build 02600 Service Pack 3
> Xbase++ (R) Version 1.90 (MT)+
> THREADS: 2
> N_LOOPS: 100
Do you modified spee
On Mon, 16 Mar 2009, Mike Evans (Gmail) wrote:
Hi,
> As far as I know Xbase++ is always executed in MT mode cause at least the GC
> is executed in a different Thread.
Viktor also used HVM with MT support:
Harbour 1.1.0dev (Rev. 10601) (MT)+ Microsoft Visual C 14.0.24999 (32-bit)
On Mon, 16 Mar 2009, Szak�ts Viktor wrote:
Hi,
> > But so far speedtst code has not been updated to test MT performance in
> > XPP builds so always is executed only once by one thread.
> > Such modification will be next the step.
> > Now only Harbour and xHarbour builds can use more then one thre
On Mon, 16 Mar 2009, Szak�ts Viktor wrote:
Hi,
> Oh, just one :) an old one, it's still a P4 2.6 / 800 with Hyperthreading.
> [ Last week I was checking the market, but I just don't feel
> like assembling yet another junk PC, so for now I keep waiting,
> maybe a new Mac Mini will be the answer wi
On Mon, 16 Mar 2009, Szak�ts Viktor wrote:
Hi,
> I'm trying to map all xhb MT functions to Harbour equivalents
> to aid porting some MT dependent parts. Here's my rough
> results. Please extend, correct and if we reach a usable state,
> I'd like to add it to hbcompat.ch.
It may not be such simpl
On Tue, 17 Mar 2009, Szak�ts Viktor wrote:
Hi,
> Probably you missed my message yesterday, but
> this line will RTE in Xbase++:
> TEST t017 WITH o := errorArray() CODE x := o[2]
No I haven't and above is fixed version.
Now I would like to ask you and other xbase++ users to send here
the results
On Tue, 17 Mar 2009, Angel Pais wrote:
> Attached xbase++ test.
Thank you very much.
I would like to ask you also about Harbour results.
To compare them the test code should be executed on the same machine.
For Harbour compilation -gc3 should be used. AFAIK xbase++ does not
generate PCODE but cod
On Tue, 17 Mar 2009, Szak�ts Viktor wrote:
> Here are the results:
> >xpp speedtst.prg /n && alink speedtst
Thank you very much.
best regards,
Przemek
___
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo
On Tue, 17 Mar 2009, Przemyslaw Czerpak wrote:
Hi,
> > Here are the results:
> > >xpp speedtst.prg /n && alink speedtst
> Thank you very much.
I would like to ask you to repeat Harbour tests using current
speedtsts.prg code from S
On Tue, 17 Mar 2009, Szak�ts Viktor wrote:
> Here are the results with latest speedtst.prg:
Thank you very much.
best regards,
Przemek
___
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
On Tue, 17 Mar 2009, Szak�ts Viktor wrote:
Hi,
> Here are the results with latest speedtst.prg:
[...]
My last request. Can you also send Harbour results for ST mode?
I would like to compare some tests.
best regards and thank you for your tets.
Przemek
___
On Tue, 17 Mar 2009, Szak�ts Viktor wrote:
Hi,
> Here they are:
[...]
Again, thank you.
best regards,
Przemek
___
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
On Tue, 17 Mar 2009, Angel Pais wrote:
Hi,
> That behaviour is caused by this 3 commands:
> #command ? => outstd(EOL)
> #command ? => outstd(EOL);outstd()
> #command ?? => outstd()
> qout() sends output to console window
Yes it is.
> outstd() sends output to standar program output, and you on
On Tue, 17 Mar 2009, Szak�ts Viktor wrote:
> > For Harbour speedtst builds use -gc3 and -ko switches. -ko enables some
> > optimizations which breaks some strict Clipper compatibility, f.e.
> > different RT errors on wrong types in math operations but as I can see
> > XPP does not respect Clipper c
On Tue, 17 Mar 2009, Pritpal Bedi wrote:
Hi,
> Hello Przemek
> Here are the logs:
Thank you very much.
best regards,
Przemek
___
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
On Wed, 18 Mar 2009, vszak...@users.sourceforge.net wrote:
> 2009-03-18 01:04 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
> * include/hbatomic.h
> ! Typo (__inline__ -> __inline).
Thanks.
Can you check current speedtst -mt results in MSVC builds for:
speedtst --only=001,002,003,004,0
On Wed, 18 Mar 2009, dru...@users.sourceforge.net wrote:
Hi Viktor,
> 2009-03-18 02:16 UTC+0100 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
> * harbour/include/hbatomic.h
> ! fixed inline assembler code for atomic inc/dec operations.
> It's late and I haven&#x
On Wed, 18 Mar 2009, Chen Kedem wrote:
Hi,
> (rev10642)
> tests/speedtst.prg does not have svn:eol-style native.
> And complied with Clipper 5.2e I get:
> clipper speedtst /n
> SPEEDTST.PRG(186) Error C2002 Incomplete statement or unbalanced delimiters
> SPEEDTST.PRG(188) Error C2024 Unclosed c
On Wed, 18 Mar 2009, J. Lefebvre wrote:
Hi,
> J I like the idea of free discusion about general concept, but we first
> face a more basic problem.
It's working for me as it should.
> Either on full Borland or full MSVC (no mix of compiler) You can create
> a pure pcode dll, you can load it, but
On Wed, 18 Mar 2009, Szak�ts Viktor wrote:
Hi,
> >hbmk2 speedtst.prg -gc3 -mt
> >speedtst --only=001,002,003,004,005,006,007,008,009,010,011,012,043,044
> [ I've double checked that thread.obj contains asm vs. Interlocked*() calls.
> ]
Thanks and please ignore my repeated request.
Interesting. N
On Wed, 18 Mar 2009, Szak�ts Viktor wrote:
Hi,
> 1) One problem with this is that an .exp / .lib file will be generated
> for *every* Harbour executables created, at least with MSVC. This
> is because these two symbols will now be exported from all
> Harbour executables. [ I've been dealing with
On Wed, 18 Mar 2009, Chen Kedem wrote:
Hi,
> With rev10644 speedtst.prg can be complied with Clipper 5.2e
> There are still two problems with EOL:
> 1) speedtst.prg does not have svn:eol-style native (so the code is limited
> with LF instead od CRLF).
I've checked it and:
druzus:~/SVN/harbour/h
On Wed, 18 Mar 2009, Edmer wrote:
Hi,
> Has this been finalized already ?
Yes I implemented it. But before I'll commit it I would like to
cleanup function names and create some small set of functions
to operate on timestamp values.
I check in the internet functions used by ADS and VFP to operat
On Wed, 18 Mar 2009, Szak�ts Viktor wrote:
Hi,
> > Now I would like to ask your proposition for letter used to mark
> > milliseconds in _SET_TIME format. xHarbour use 'c', f.e.:
> > hh:mm:ss.ccc
> > I used 'f':
> > hh:mm:ss.fff
> I trust you that .fff is the better choice, 'c' seems to be use
On Wed, 18 Mar 2009, Edmer wrote:
Hi,
> Is this a confirmed bug ?
> >> dbskip(0) will move the record pointer to phantom record (honoring the
> >> scope). other RDDS don't do that
It's cause by ADS behavior.
> >> #include "ads.ch"
> >> proc main()
> >> rddregister( "ADS", 1 )
Not necessary fro
On Thu, 19 Mar 2009, vszak...@users.sourceforge.net wrote:
> 2009-03-19 12:58 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
> * config/rules.cf
> % Removed -ko option as it slightly decreases performance in -gc3 mode.
> (it's better in -gc[0-2] only)
Viktor,
Please compile speedtst wi
On Thu, 19 Mar 2009, Mario Simoes Filho wrote:
Hi,
> I would like to make a suggestion for the function dTOc (), a second
> parameter with the output format, something like this:
> ? dtoc (date (), "MM-DD-)
> ? dtoc (date (), "MM-)
> ? dtoc (date (), "-DD)
I already added such param
On Thu, 19 Mar 2009, vszak...@users.sourceforge.net wrote:
Hi,
> ---
> 2009-03-19 18:06 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
> * source/vm/maindll.c
> * source/vm/maindllp.c
> ! Fixed placement of HB_EXPORT in declaration.
Which compiler needs separate declaration for D
On Thu, 19 Mar 2009, Szak�ts Viktor wrote:
> I've added it for one of them... maybe it was ICC, or MSVC, otherwise it
> warned that the entry point won't be exported.
OK. Only one note:
BOOL HB_EXPORT WINAPI DllEntryPoint( HINSTANCE hInstance, DWORD fdwReason,
PVOID pvReserved );
always add
On Thu, 19 Mar 2009, Szak�ts Viktor wrote:
Hi Viktor,
> Attached the -gc2 results. Notice that the results aren't
> suited for -ko comparison as I've used different revisions.
> I hope this wasn't your goal. If it was I can redo -gc3
> tests with current build (-ko disabled).
Thank you very much
On Fri, 20 Mar 2009, Szak�ts Viktor wrote:
> > Thank you very much for test but I asked about reverted test. I want to
> > compare harbour-gc2 + speedtst-gc2 with harbour-gc3 + speedtst-gc2.
> > Can you send also harbour-gc3 + speedtst-gc2.
> Sorry, I didn't explain the filenames:
> Harbour -gc3 +
On Fri, 20 Mar 2009, Szak�ts Viktor wrote:
Hi,
> > I can believe in many different things but I do not belive that:
> > 03/19/09 23:05:24 Windows XP 5.1.2600 Service Pack 3
> > Harbour-gc2+speedtst-gc2 gives:
> > [ T000: empty loop overhead ]...0.06
> > Harbour-gc3
On Mon, 23 Mar 2009, Barry Jackson wrote:
Hi,
> Thanks Viktor - however I can't build latest SVN with mpkg_rpm.sh.
> I tried a clean build with no RPM folder and the script failed - I had to
> make the RPM tree manually before it would run.(I expected it to make a new
> one automatically)
It m
On Mon, 23 Mar 2009, Barry Jackson wrote:
Hi,
> [...@jackodesktop hbr]$ testsend
> testsend: error while loading shared libraries: libharbour.so: cannot open
> shared object file: No such file or directory
> What is now really needed in environment? I have nothing in there for
> harbour at pre
On Mon, 23 Mar 2009, Szak�ts Viktor wrote:
Hi,
> > 2009-03-22 18:16 UTC+0100 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
> > * harbour/include/hbstack.h
> >* added alternative asm inline function to __MINGW32__ builds which
> > operates on 64bit pointers. It
Hi All,
I've got a question about VFP and storing DATETIME values in memo FPT
files. Is is possible?
If yes then I would like to ask about DBF/FPT files created in VFP
by this code:
proc main()
dbCreate( "_tsttime", { { "F1", "M", 4, 0 } } )
use _tsttime
append blank
re
On Mon, 23 Mar 2009, Marcos Gambeta wrote:
Hi,
> -mwindows e mno-cygwin can be removed. Only -Wl,-subsystem,windows is
> necessary.
AFAIK -mwindows/-mconsole should be used instead of using directly LD
parameters. Of course both will probably make the same job in your case
but GCC knows about -
On Mon, 23 Mar 2009, Francesco Saverio Giudice wrote:
Hi,
> tested also after
> 2009-03-23 21:42 UTC+0100 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
> * harbour/contrib/hbodbc/odbc.c
> * harbour/contrib/hbtpathy/tpwin.c
> * harbour/contrib/hbwin/win_ole.c
> * har
On Wed, 25 Mar 2009, vszak...@users.sourceforge.net wrote:
Hi,
> * contrib/gtwvg/gtwvg.h
> * contrib/gtwvg/wvggui.h
> ! Fixed to _not_ override _WIN32_IE. Overriding these version
> numbers is generally a very bad idea. Instead the code should
> adapt to whatever values are de
On Fri, 06 Feb 2009, Przemyslaw Czerpak wrote:
Hi,
> > We should IMO never include windows.h from hbdefs.h. Besides
> > slowing down build time to a great deal, we don't do this for other
> > OS APIs either. This would also suggest that Windows API usage
> > is freel
On Thu, 26 Mar 2009, Szak�ts Viktor wrote:
Hi,
> > * harbour/contrib/gtwvg/gtwvg.h
>! moved _WIN32_IE declaration before #include ... to fix MinGW32
> compilation
> After this, mingw64 fails with these messages (previously it compiled
> cleanly):---
> make -C gtwvg install
> make[2]: En
On Thu, 26 Mar 2009, toni...@fwi wrote:
Hi,
> But with a more extensive test here I found that this code doesn´t
> work:
> procedure testolenew()
>local oWord, oText
>begin sequence
> oWord = GetActiveObject( "Word.Application" )
>recover
> begin sequence
> oWord
On Thu, 26 Mar 2009, Massimo Belgrano wrote:
Hi,
> gtxwc have one platform supported (linux)
> (so imo is related to this topics)
> If the rule for exist is related to number of platform supported
> gtwvg and gtxwc must be is same (irregolar) situation
GTXWC is supported by Linux, HPUX, SunOS, B
On Wed, 25 Mar 2009, Mindaugas Kavaliauskas wrote:
Hi,
> one of missing features is parameters passed by reference. Iterators is not
> implemented also, but I do not know if it is implemented in current win_ole
> code.
Current code is very short and simple. At least I understand it so far ;-)
On Fri, 27 Mar 2009, Saulius Zrelskis wrote:
Hi,
> > + added void hb_vmAtQuit( HB_INIT_FUNC pFunc, void * cargo );
> > AtQuit functions are executed after deallocating all HVM items and
> > disabling .prg destructors. They can make final cleanup at C level
> > but should not ree
Hi All,
I would like to ask xbase++ users to make MT performance tests with
this compiler and tests/speedtst.prg.
I'm interesting in results from:
speedtst.exe --thread=2 --scale
current version should create speedlog.txt file with results.
It would be perfect if someone can run this test on re
On Fri, 27 Mar 2009, Guy Roussin wrote:
> Thank you for the quick reply.
> But i think that -t"1970-01-01" is not "00:00 GMT"
> but "00:00 local time" isn't it ?
> So this is not exactly the result i expect ...
It doesn't matter. t"1970-01-01" is some timestamp constant value.
Important is base va
On Fri, 27 Mar 2009, Angel Pais wrote:
Hi,
> I'll do it in a while but AFAIK xbase++ only uses one processor per
> process, if so it does not scale.
> We'll confirm that soon
It's not such easy to use only one CPU. In most of operating system it's
not even possible. I would be very surprised if
On Fri, 27 Mar 2009, Angel Pais wrote:
Hi,
> It GPF'd with this log:
> FATAL ERROR LOG
> Not recoverable Error!
> SYS Thread-ID: 1224
> Module: MOM
> Error Codes: EH: 1006 Sub: 0(0) OS: 0 XPP: 15
> Call Stack of Thread 1 (1732):
> @notif...@i@SUBSCRIBE(0)
> HB_MUTEXSUBSCRIBE(0)
> TEST(0)
> MAIN(0
On Fri, 27 Mar 2009, Angel Pais wrote:
Hi,
> On second machine (dual core) it doesn't gpf'd but results look very
> strange to me.
I do not find anythinbg strange in the results.
> Pentium 4 3GZ 1GB RAM Dual Core
> 03/27/2009 16:55:40 Windows XP 05.02 Build 03790
> Xbase++ (R) Version 1.90 (MT
On Sat, 28 Mar 2009, Guy Roussin wrote:
> Hi Przemek,
> Why i get this :
> ? hb_ntot( HB_MILLISECONDS() )
> 00/00/00 00:00:00.000
Because HB_MILLISECONDS() returns julian time in milliseconds.
I guess you wanted to make sth like:
? hb_ntot( HB_MILLISECONDS() / ( 24 * 60 * 60 * 1000 ) )
And you
On Sat, 28 Mar 2009, toni...@fwi wrote:
Hi,
> >Il 26/03/2009 21.11, Przemyslaw Czerpak ha scritto:
> >> 5. The hbwin TOLEAUTO class supports iteration by FOR EACH loop.
> >Could you add FOR EACH and TIMESTAMP support ?
> >I've code like attached that doesn&#x
On Sat, 28 Mar 2009, Szak�ts Viktor wrote:
Hi,
> > more efficient to compile windows Harbour builds with -DHB_FM_WIN_ALLOC.
> Shouldn't we make this the default for hbvmmt lib?
Probably yes though it would be good to check the results with different
installations (single and multicpu machines, X
On Sat, 28 Mar 2009, Mike Evans (Gmail) wrote:
Hi,
> Even I don’t have the latest version of Xbase++ but an old one I'm sure
> that they lock the application threads to use only one CPU. They have a
> way to choose which CPU to use (as an example the less used CPU).
AFAIR some MS-Windows API fun
On Sat, 28 Mar 2009, Angel Pais wrote:
Hi,
> Here you have results on same computer 2 compilers.
> see this:
> xbase++ :
> [ TOTAL ]247.17 246.80 -> 1.00
>
> [ total application time: ]..
On Sat, 28 Mar 2009, Angel Pais wrote:
Hi,
> this is the .bat file I use to compile Harbour.
> All the rest is pure svn code.
> compile.bat
> SET PATH=C:\vsk\vc\bin;C:\vsk\bin;
> SET LIB=C:\vsk\vc\lib;C:\vsk\lib;
> SET INCLUDE=C:\vsk\vc\include;C:\vsk\include;
> set HB_COMPILER=msvc
> set HB_INST
On Sun, 29 Mar 2009, Przemyslaw Czerpak wrote:
Hi,
> > this is the .bat file I use to compile Harbour.
> > All the rest is pure svn code.
> > compile.bat
> > SET PATH=C:\vsk\vc\bin;C:\vsk\bin;
> > SET LIB=C:\vsk\vc\lib;C:\vsk\lib;
> > SET INCLUDE=C:\vs
On Sun, 29 Mar 2009, Angel Pais wrote:
Hi,
> Here new set of results with download link included...
> Please other windows users download and post results here
Thank you very much.
> Computer PIV Dual COre 3.0 win 2003 server, used tru VNC
> 03/29/2009 14:04:15 Windows XP 05.02 Build 03790
> Xb
On Sun, 29 Mar 2009, Angel Pais wrote:
Hi,
> I couldn't recompile harbour
> Here the error:
> cl.exe -nologo -I. -I../../../../../include -Gs -TP -W4 -wd4127 -Ot2b1
> -EHs-c- -MT -DHB_FM_STATISTICS_OFF -DHB_FM_DLMT_ALLOC -DHB_MT_VM -c
> ../../../fm.c -Fofm.obj
> fm.c
> ../../../fm.c(403) : e
On Sun, 29 Mar 2009, Angel Pais wrote:
Hi,
> I don't know if scalability is better but processor time went down from
> 100% to 62% in average on same machine.
> Also sacalability improved a little
> See attached...
> Compiler and Enviroment:
> set HB_COMPILER=msvc
> set HB_VISUALC_VER=90
> set H
On Sun, 29 Mar 2009, Angel Pais wrote:
Hi,
> Oops !
> Then I wil have to re compile harbour with the 3 mem allocator switches and
> to all tests again.
> No problem. It was my fault. Will try to finish them today.
My too. I've just added warning to speedtst result messages which informs
abo
On Sun, 29 Mar 2009, vouch...@users.sourceforge.net wrote:
Hi,
> ; REQUEST ( Przemek ) - Can you please provide a GTWVT.C like
> skeleton .cpp which I could use as a base for GTQTC ?
I'll try to make it in this week.
best regards,
Przemek
___
On Mon, 30 Mar 2009, Chen Kedem wrote:
> > 2009-03-29 11:02 UTC+0200 Viktor Szakats (harbour.01 syenar hu
> > * external/sqlite3/Makefile
> >- Disabled for bcc. Latest sqlite3 version breaks with this compiler:
> BCC does not use LL to mark LONG LONG constants.
> It needs suffix like i64 (or u
On Mon, 30 Mar 2009, vszak...@users.sourceforge.net wrote:
> Revision: 10738
>
> http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10738&view=rev
> Author: vszakats
> Date: 2009-03-30 11:15:54 + (Mon, 30 Mar 2009)
> Log Message:
> ---
> 2009-03-30 13:10 UT
On Mon, 30 Mar 2009, Szak�ts Viktor wrote:
Hi,
> I didn't want to kill it, just wanted to tweak it for MT.. we're using such
> MT switch for most compilers to force MT friendly runtine/code generation/
> whatever. Isn't this needed for owatcom?
We were discussing about it when OW OS2 port was ad
On Mon, 30 Mar 2009, Francesco Saverio Giudice wrote:
Hi,
> here are speedtst logs for Vista Ultimate with Intel(R) Core(TM)2 Duo
> CPU T8300 @ 2.40GHz
> and with 2 x Intel(R) Xeon(TM) CPU 3.00GHz Dual Core (4 Core in
> total) Microsoft Windows Server 2003, Standard Edition Service Pack 2
> (Buil
On Sun, 29 Mar 2009, Szak�ts Viktor wrote:
Hi,
> > BTW: in hbsetup.h you added:
> > ( defined(__POCC_TARGET__) && __POCC_TARGET__ == 2 )
> > to detect HB_OS_WIN_CE builds. Maybe it's documented method in POCC
> > and in such case should be left but if not then I suggest to rather
On Tue, 31 Mar 2009, Szak�ts Viktor wrote:
Hi,
> After recent owatcom tweaks, its performance went back to a very good level
> in ST mode.
> msvc ST: 29s MT: 49s
> mingw ST: 34s
> owatcom ST: 38s MT: 75s
> pocc ST: 44s
> bcc ST: 46s
> owatcom (yesterday) ST: 54s
> And this is without register cal
On Tue, 31 Mar 2009, Szak�ts Viktor wrote:
> Attached results.
> Both with -gc3, one with r10748 (stack calls), and one
> with r10756 + HB_USER_CFLAGS=-6r. ST and MT,
> two runs with each combination.
> [ The previously posted quick MT results seems wrong, now
> with repeated and saved tests it's a
On Tue, 31 Mar 2009, Szak�ts Viktor wrote:
Hi,
> Can you upload them to some public host?
> Sure:
> http://www.syenar.hu/harbour/owst.zip
I've just tested them with WINE on AMD Phenom(tm) 8450 Triple-Core 2100MHZ.
Results below.
Tomorrow I'll make tests with real Windows also with mixed version
On Wed, 01 Apr 2009, Massimo Belgrano wrote:
Hi,
> Will be intresting this discussion on fivetech forum?
> http://forums.fivetechsupport.com/viewtopic.php?f=3&t=15076
No,
>
> Speedtest CLIPPER vs. xHarbour - COMMIT
> Is CLIPPER still faster in database management?
No,
> I am testing the COMM
On Wed, 01 Apr 2009, Szak�ts Viktor wrote:
Hi,
> Second issue of the unified and self-contained Windows release
> package, which is able to create x86, x64, WinCE/ARM executables
> (both shared and static) out of the box without the need of any
> external tools or settings. This time also in the
On Wed, 01 Apr 2009, Szak�ts Viktor wrote:
> I've added clipper/rtlink recognition, but I miss the .lnk parser
> code, and to be frank I don't really know its exact details (used
> to use "FI hello1, hello2"), so if you happen to have some/any
> piece of such parser code I'll can tackle it to hbmk2
On Wed, 01 Apr 2009, Przemyslaw Czerpak wrote:
> I've just written such simple parser.
[...]
Below is corrected version.
I removed one redundant line I left by mistake in copy and past
and added some basic translation for well know Clipper, Blinker,
ExoSpace, SIX, Class(y) files.
For sure
On Thu, 02 Apr 2009, Pritpal Bedi wrote:
Hi,
> I am not asking for how I am returned the pointer.
> My question is : once I am returned a pointer how can I
> pass it back to some another function which is expecting
> .
> pFont := QT_QFont() // say I am returned default font
> at this point is
On Thu, 02 Apr 2009, Pritpal Bedi wrote:
Hi,
> It is as simple as:
> hbqt_par_QWebView( 1 )->setUrl( *hbqt_par_QUrl( 2 ) );
Yes, but I suggest to make it a little bit more complicated and
add at least some basic parameter validation and add RT errors
on wrong parameters before use.
Otherwise suc
On Fri, 03 Apr 2009, Adam Lubszczyk wrote:
Hi,
> > Problem:
> > In Clipper or FoxPro samebody create index like:
> > INDEX ON Name_and_Last TO xxx
> > but field name in DBF is 'NAME_AND_L'.
> > In Clipper and FoxPro it's work OK
> > but Harbour when open index show error:
> > variable 'Name_and_L
On Sun, 05 Apr 2009, vszak...@users.sourceforge.net wrote:
> 2009-04-05 09:53 UTC+0200 Viktor Szakats (harbour.01 syenar hu)
> * utils/hbmk2/hbmk2.prg
> + Enabled rtlink/blinker parsing code, so now hbmk2 will
> work as a plugin replacement for clipper, rtlink/blinker/exospace
> w
On Mon, 06 Apr 2009, Szak�ts Viktor wrote:
Hi,
> I'll add this. Can we rename these switches to '-hbdir[inc|bin|lib]' ?
> they'd align a bit better and would allow make future extensions
> a little bit easier.
> I have a new dir for dynamic libs in hbmk2, so maybe we could
> also have an -hbdirdy
On Mon, 06 Apr 2009, Szak�ts Viktor wrote:
> In fact I didn't change anything in above change for owatcom. The result
> is still:
> Making harbour-11.dll...
> Making harbourmt-11.dll...
> hbmk: Processing configuration:
> C:\harbour\_hb_install_temp\harbour-win-owatcom\bin\hbmk.cfg
> Open Watcom Li
On Mon, 06 Apr 2009, snaipe...@users.sourceforge.net wrote:
Hi,
> 2009-04-06 19:35 UTC+0200 Mindaugas Kavaliauskas (dbtopas/at/dbtopas.lt)
> * harbour/source/rtl/hbstrfmt.c
> + added precision support for %s and %d.
> COMMENT: I've used GNU printf() function for test. Some of
>
On Mon, 06 Apr 2009, vszak...@users.sourceforge.net wrote:
Hi,
> 2009-04-06 18:12 UTC+0200 Viktor Szakats (harbour.01 syenar hu)
> * bin/hb-mkdyn.bat
> * owatcom tweaks. Build is still going dunno if this will be okay.
What's the problem?
I created manually harbour.dll and it works.
best
On Mon, 06 Apr 2009, Szak�ts Viktor wrote:
Hi,
> Now only this warning remains:
> Warning! W1171: Locally defined symbol hb_stricmp is imported.
> (why only this one, I don't know. It's caused by '-lhbcplr -lhbpp
> -lhbcommon' parameters when building hbrun-dll, hbmk2-dll)
Not only this one. In
On Mon, 06 Apr 2009, Szak�ts Viktor wrote:
Hi,
> > BTW in Linux shared builds are broken in hbmk2 and still -llibharbour
> > is used instead of -lharbour.
> -llibharbour is used since the beginning.
So from the beginning it is broken. -llibharbour cannot work.
> The problem was arch/comp
> lib
401 - 500 of 2668 matches
Mail list logo
