Hi David,
Please try to remove OS/2 exception handling stuff
in source/vm/extrap.c and see if it helps.
Brgds,
Viktor
On 2008.06.02., at 10:05, David Arturo Macias Corona wrote:
Viktor
2008-06-01 23:56 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
* include/hbvm.h
* source/vm/hvm.c
2008-06-02 11:34 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
* source/vm/extrap.c
! Blind fix for OS/2.
--
Brgds,
Viktor
___
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
Viktor
Please try to remove OS/2 exception handling stuff
in source/vm/extrap.c and see if it helps.
These are SVN browse of last change (rev 8553, 8562)
What code should I to change ?
-
{
167 hb_errInternal(
2008-06-02 12:42 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/contrib/hbsqlit3/tests/blob.prg
* harbour/contrib/hbsqlit3/tests/sqlite3_test.prg
* harbour/contrib/hbsqlit3/sqlite3/sqlite3.c
* harbour/contrib/hbsqlit3/sqlite3/sqlite3.h
* converted TABs to SPACEs
best
Viktor
2008-06-02 11:34 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
* source/vm/extrap.c
! Blind fix for OS/2.
David, please test with current repo.
No change, exactly same error:
Exception c005 at address 0x1c021b0d
( everything was rebuilt with new extrap.c )
2008-06-02 13:32 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/include/hbexprop.h
* harbour/include/hbexprb.c
* harbour/source/common/expropt2.c
+ added hb_compExprReducePower() and compile time optimization for:
NumConst1 ^ NumConst2
+ added compile time
Hi David,
I have run out of ideas then, sorry.
You could still try to remove all OS/2 stuff from
extrap.c and see if it helps. In all OS/2 stuff
I mean everything which is guarded with HB_OS_OS2.
Brgds,
Viktor
On 2008.06.02., at 12:50, David Arturo Macias Corona wrote:
Viktor
2008-06-02
Great, thanks a lot Przemek.
Some other candidates for optimizations that sprang
to mind (after I wrote that NOTE unfortunately), are
the hb_bit*() functions. Not a priority, but they would
be useful, when using these functions in #defines
to not cause any additional runtime cost. If you agree
2008-06-02 15:40 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/contrib/hbsqlit3/hbsqlit3.c
! use char * instead of LPSTR to compile in non Windows environment
! fixed one printf() parames
! fixed casting and parameter type in sqlite3_blob_open()
*
Hi all,
I am trying to build RDDADS using the current CVS - I set HB_DIR_ADS
as indicated but I am getting the following build error using MSVC v6:
..\mtpl_vc.mak(112) : fatal error U1023: syntax error in expression
Stop.
Randy.
___
Harbour
Hi Randy,
Are you sure this file is unmodified?
(Line 112 is a comment, and the only change
in the recent month was a few flags removed.)
Brgds,
Viktor
On 2008.06.02., at 16:08, Randy Portnoff wrote:
Hi all,
I am trying to build RDDADS using the current CVS - I set HB_DIR_ADS
as
Perfect!
Brgds,
Viktor
On 2008.06.02., at 15:41, Przemyslaw Czerpak wrote:
2008-06-02 15:40 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/contrib/hbsqlit3/hbsqlit3.c
! use char * instead of LPSTR to compile in non Windows environment
! fixed one printf() parames
!
Hi Przemek,
One observation:
I'd expect HB_CURLPAUSE_ALL to be optimized
to one integer value.
HB_CURLPAUSE_ALL is
hb_bitOr( A, B )
where
A = ( 2 ^ 0 )
B = ( 2 ^ 2 )
In the resulting pcode, hb_bitOr() is not optimized,
and it's being pushed two doubles with the (width set
to 255, 255).
Hi Randy,
I think you miss the HB_VISUALC_VER envvar, and
the line no is misreported by nmake (yes, and they
didn't bother to fix even in 9.0 - compatibility
must be the reason ;)
So, please use this:
set HB_VISUALC_VER=60
I'll add something to default to a safely looking value
instead of
Hi all,
IMO, any of the environment variables that are set to a value like
yes (eg. HB_BUILD_ST) should be case insensitive. I had mine set as
HB_BUILD_ST=YES and it had no effect - I took me a while to realize
what the problem was.
Randy.
___
Hi Viktor,
I run MAKE_VC.BAT and this is the output...
Using this .dll: C:\AXS32\32bit\ace32.dll
Microsoft (R) Program Maintenance Utility Version 6.00.9782.0
Copyright (C) Microsoft Corp 1988-1998. All rights reserved.
..\mtpl_vc.mak(112) : fatal error U1023: syntax error in expression
Thanks - That worked!
At 11:03 AM 6/2/2008, you wrote:
Hi Randy,
I think you miss the HB_VISUALC_VER envvar, and
the line no is misreported by nmake (yes, and they
didn't bother to fix even in 9.0 - compatibility
must be the reason ;)
So, please use this:
set HB_VISUALC_VER=60
I'll add
On Mon, 02 Jun 2008, Szakáts Viktor wrote:
Hi Viktor,
We currently also use the '()' non-optimization for
HB_SYMBOL_UNUSED().
It is and was optimized. For all expressions like:
(exp) op (exp)
our expression optimizer strips parenthesis.
It's also not Clipper compatible and maybe we should
Hi Przemek,
In general there is a problem with compile time optimizations for
CHR( integer_value * 256 )
and $ operator in expressions like:
$ exp
caused by bugs in Clipper compiler. It gives different results then
runtime functions/operations. Keeping strict Clipper compatibility
is hard
2008-06-02 18:59 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
* contrib/xhb/hbcompat.ch
! Minor fix in hb to xhb branch.
* contrib/hbcurl/hbcurl.ch
* '( 2 ** n )' expressions changed to actual values.
; We now have these optimized, but I'm committing anyway,
as some
What about new harbour's web site and fidonet group?
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Phil Barnett
Sent: Monday, June 02, 2008 5:01 AM
To: Harbour@harbour-project.org
Subject: [Harbour] Release Time
I know I promised to make the release
Hi Przemek,
We currently also use the '()' non-optimization for
HB_SYMBOL_UNUSED().
It is and was optimized. For all expressions like:
(exp) op (exp)
our expression optimizer strips parenthesis.
It's also not Clipper compatible and maybe we should disable
this optimization when -kc is
3152 (20080602) __
The message was checked by ESET Smart Security.
http://www.eset.com
___
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
On Mon, 02 Jun 2008, Szakáts Viktor wrote:
Hi Viktor,
One thing I forgot: To do this we'd need a switch
which would toggle the parenthesised expression
optimization only.
OK, I can add it.
BTW, for me it's unclear by looking at the 'harbour /k?'
help, which options are enabled by default,
On Mon, 02 Jun 2008, Szakáts Viktor wrote:
Hi Viktor,
It's not possible to make a perfect decision here,
unless investing extreme energy (like those flags)
to keep bug compatibility, not to mention what an
ugly inconsistency this is. My personal opinion in this
case is to fix this Clipper
It is possible that I lost but .-
2006-03-09 16:30 UTC+0100 Ryszard Glab rglab//imid.med.pl
...
* source/common/hbstr.c
* source/rtl/strings.c
* added hb_strRemEscSeq() function that removes C-like Esc
sequences (\n \t \r \b currently)
On the source/rtl/strings.c is not and on the
Hi Przemek,
It's not possible to make a perfect decision here,
unless investing extreme energy (like those flags)
to keep bug compatibility, not to mention what an
ugly inconsistency this is. My personal opinion in this
case is to fix this Clipper bug in Harbour, and make it
available using a
On Mon, 02 Jun 2008, Guillermo Varona Silupú wrote:
Any suggestions/tip for this?
I have a small console application in which I want to show the symbol
of the square root, but the screen shows me at number 1 as superscript.
What should I set to display the symbol of the square root?
Env.:
-DHB_FM_STATISTICS_OFF
Valid.
-DHB_GUI ?
Still exists. I think we should somehow
remove this (meaning: replace with some superior
solution) in the future. Since this is
yet another thing to remember at build-time and
yet another thing to badly interact with
a multi GT app.
Actually
Hi Przemek,
2008-06-02 19:55 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/include/hbexprop.h
* harbour/include/hbexprb.c
* harbour/source/common/expropt2.c
+ added compile time INT( num ) optimization
* harbour/source/compiler/hbusage.c
* updated -k? description
Which of the following symbols are still useful/mandatory?
-DHB_FM_STATISTICS_OFF
-DHB_GUI
-DHB_NO_PROFILER
-DHB_NO_TRACE
-DHB_WIN32_IO
Thank you in advance.
EMG
--
EMAG Software Homepage: http://www.emagsoftware.it
The EMG's ZX-Spectrum Page: http://www.emagsoftware.it/spectrum
The Best
On Mon, 02 Jun 2008, Enrico Maria Giordano wrote:
Which of the following symbols are still useful/mandatory?
-DHB_FM_STATISTICS_OFF
It's still functional. Default is HB_FM_STATISTICS
-DHB_GUI
It disables inkey polling in main HVM loop.
In normal programs it means that K_ALT_C and K_ALT_D
Plus, rather than saying 'lowercase/uppercase significant'
it would be more intuitive to just use lowercase
for 'J' and 'M', and make casing not to matter, like
for the rest of Harbour switches. I can make this, but
I wonder if there was any special reason doing it like
this?
Probably Ryszard
On Mon, 02 Jun 2008, Enrico Maria Giordano wrote:
Hi Enrico,
Please look at the following sample:
FUNCTION MAIN()
LOCAL bTest := { | Param | MyFunc() }
RETURN NIL
STATIC FUNCTION MYFUNC()
RETURN NIL
The compiler shows this warning if /w2 or /w3 are used:
Warning W0004 Codeblock
Please look at the following sample:
FUNCTION MAIN()
LOCAL bTest := { | Param | MyFunc() }
RETURN NIL
STATIC FUNCTION MYFUNC()
RETURN NIL
The compiler shows this warning if /w2 or /w3 are used:
Warning W0004 Codeblock parameter: 'PARAM' declared but not used in
function: 'MAIN'
Is it
On Mon, 02 Jun 2008, Szakáts Viktor wrote:
Since we already have a TRACE functionality on the
C level which has nothing to do with this, I'd like
to rename this to HB_NO_TRACEPRG. Or rather, shouldn't
we tie this functionality to PROFILER? Opinions?
HB_NO_TRACEPRG is OK for me though I'd
Przemyslaw Czerpak escribió:
On Mon, 02 Jun 2008, Guillermo Varona Silupú wrote:
Any suggestions/tip for this?
I have a small console application in which I want to show the symbol
of the square root, but the screen shows me at number 1 as superscript.
What should I set to display the
-Messaggio Originale-
Da: Szakáts Viktor [EMAIL PROTECTED]
A: Harbour Project Main Developer List. harbour@harbour-project.org
Data invio: lunedì 2 giugno 2008 22.04
Oggetto: Re: [Harbour] Warning on unused codeblock parameters
We may, but as the one having went through almost all
What do you think of making this a runtime option?
(a Set())
The idea of using this macro was eliminating all possible
overhead from main HVM loop. Using and RT switch like new
set() does not make it. Anyhow current code can/should be
updated. See the note about hb_set.HB_SET_KEYPOLL I left
in
On Mon, 02 Jun 2008, Guillermo Varona Silupú wrote:
Hi Guillermo,
It depends on used GT and font.
What GT driver do you use?
Using the GT and font default. I do not set any GT and font.
How do I know that GT and font that I am using?
to check GT driver you can run you program with //info
Przemyslaw Czerpak escribió:
On Mon, 02 Jun 2008, Guillermo Varona Silupú wrote:
Hi Guillermo,
It depends on used GT and font.
What GT driver do you use?
Using the GT and font default. I do not set any GT and font.
How do I know that GT and font that I am using?
to check GT
2008-06-03 01:31 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
* source/vm/extend.c
! hb_storclen_buffer() fixed to free the passed buffer
in case it couldn't be stored. This situation was
easy to miss and difficult to detect for the caller,
and it might cause
On Monday 02 June 2008 01:05:15 pm Massimo Belgrano wrote:
What about new harbour's web site and fidonet group?
What about it?
I thought fidonet died about 15 years ago...
--
Waiting for sunspots.
___
Harbour mailing list
Harbour@harbour-project.org
Thanks Viktor,
Why not complete?
In order to send out to full printers and terminals may be useful.
HB_EXPORT char * hb_strRemEscSeq( char *str, ULONG *pLen )
{
ULONG ul = *pLen, ulStripped = 0;
char *ptr, *dst, ch;
ptr = dst = str;
while( ul )
{
if( *ptr == '\\' )
44 matches
Mail list logo
