Thanks Massimo!
Another question: RDDSQL for MySQL also need MyODBC?
Sérgio
--
From: "Massimo Belgrano"
Sent: Thursday, May 14, 2009 3:59 AM
To: "Harbour Project Main Developer List."
Subject: Re: [Harbour] Mini-tutorial MySQL connection
ADO like
I have much code writing using something like this:
HB_FUNC (TEST)
(
some *pPtr;
if (! ISPOINTER (1))
return;
pPtr = (add *) hb_parptr (1);
... more code here
)
As I understand it then from now on the right would add a further validation as:
HB_FUNC (TEST)
(
some *pPtr;
I have much code writing using something like this:
HB_FUNC (TEST)
(
some *pPtr;
if (! ISPOINTER (1))
return;
pPtr = (add *) hb_parptr (1);
... more code here
)
As I understand it then from now on the right would add a further validation as:
HB_FUNC (TEST)
(
some *pPtr;
On Thu, 14 May 2009, Przemyslaw Czerpak wrote:
Hi,
> The bug is not recently introduced.
> It has been existing for years because was hidden by someone who disabled
> checking for AE_SUCCESS value returned by AdsGet*() functions and hardcoded
> the size of returned value to maximum size of declar
newest ads * $Id: ads1.c 11050 2009-05-14 17:12:31Z druzus $
len( items->desc) generating error
ERROR BASE/ Argument error: LEN
--
View this message in context:
http://n2.nabble.com/newest-ads-len%28-area-%3Ecfield-%29--generate-RTE-tp2902639p2902639.html
Sent from the harbour-devel maili
On Thu, 14 May 2009, Luis Krause Mantilla wrote:
Hi,
> The Harbour list claims there's a bug recently introduced
> to rddads. Char fields larger than 23 display one extra
> char at the end. I can't try the latest xharbour from cvs
> but with ads1.c v. 1.140 there aren't any problems.
The bug
In my code in cases like this I return NIL instead of NULL
pointer. This
is better for future compatibility, if we are going to change to
collectible pointers in the future. Since the value of the
collectible
pointer will not be NULL.
My vote goes to returning same type for both e
Try change it:
IF Empty (hLib) ---> IF Valtype(hLib) <> "P"
if hLib <= 32 ---> IF Valtype(hLib) <> "P"
I'm not sure this would work. 'P' type is returned
even in case of error, but its content is NULL.
To solve this specific and similar cases, I've
added WIN_P2N() function, so such code can
Viktor Szakáts wrote:
Hi Mindaugas,
if hLib <= 32 && <<<-- Here error
Probably 'IF Empty( hLib )' should be used here, and
LoadLibrary() modified to return NULL when pointer is <= 32.
But now I know why we had this code full of numeric pointers.
Mindaugas:
I'm using 1.140. I'm in the middle of heave code changes on our
app and I won't be able to try 1.141 anytime soon but will see if
someone else can test it and confirm/deny.
Regards,
Mindaugas Kavaliauskas wrote:
Luis Krause Mantilla wrote:
Przemek:
FWIW, I tested with xHarbour wi
Hi,
great! The fix below should be enough (not tested), but maybe we should
add a better error check for ads function return values...
Index: C:/harboursvn/contrib/rddads/ads1.c
===
--- C:/harboursvn/contrib/rddads/ads1.c (revis
Try change it:
IF Empty (hLib) ---> IF Valtype(hLib) <> "P"
if hLib <= 32 ---> IF Valtype(hLib) <> "P"
___
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
Hi
This program works very fine:
#include "adordd.ch"
REQUEST ADORDD
function Main()
clear screen
setmode(25,80)
USE giovanni VIA "ADORDD" TABLE "clienti" MYSQL FROM "db4free.net" USER "user" PASSWORD "password"
Luis Krause Mantilla wrote:
Przemek:
FWIW, I tested with xHarbour with ads 8.1, 9.0 and 9.1 dlls
and it does not add an extra dummy char. I changed the field
length to 40 and it still worked as expected.
Hi,
Have you tried with ads1.c file revision 1.141 (Mon May 4 08:25:44 2009
UTC) or ne
Revision: 11049
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=11049&view=rev
Author: vszakats
Date: 2009-05-14 17:04:54 + (Thu, 14 May 2009)
Log Message:
---
2009-05-14 19:01 UTC+0200 Viktor Szakats (harbour.01 syenar hu)
* utils/hbmk2/hbmk2.prg
Hi ppl.
I´d posted this message in Advantage.General:
---cut---
I´m a Harbour user and our group found a problem when use
AdsGetField(..., ADS_NONE): character fields longer then 23 bytes read
from tables increase pulLen by one adding one garbage character to the
end of string. Is this an expecte
Hi ppl.
I´d posted this message in Advantage.General:
---cut---
I´m a Harbour user and our group found a problem when use
AdsGetField(..., ADS_NONE): character fields longer then 23 bytes read
from tables increase pulLen by one adding one garbage character to the
end of string. Is this an expecte
> Would you please add this little icon to our webpage:
>
> href="http://sourceforge.net/community/cca09/nominate/?project_name=Harbour&project_url=http://www.harbour-project.org/";> src="http://sourceforge.net/images/cca/cca_nominate.png"; border="0"/>
>
> [ If someone has any objection, please s
> Would you please add this little icon to our webpage:
>
> href="http://sourceforge.net/community/cca09/nominate/?project_name=Harbour&project_url=http://www.harbour-project.org/";> src="http://sourceforge.net/images/cca/cca_nominate.png"; border="0"/>
>
> [ If someone has any objection, please s
Hi Rossine,
> Hello Viktor,
>
> The same would be true to use this: 'IF Empty (hLib)', but this in itself
> will fivewin about some changes for this work and in my application also
> been several changes :(.
Yes I know, and I'm happy that we see know what sort of problems
we need to deal with
>
> > In this case you can copy hbmk.cfg to one of these location neutral dirs:
> > /etc/harbour or ~/.harbour.
>
> The problem is that such hbmk.cfg will be common to all Harbour
> installations I will have in the system.
True.
> > If there is a way to detect that softlink condition we can add
Przemek:
FWIW, I tested with xHarbour with ads 8.1, 9.0 and 9.1 dlls
and it does not add an extra dummy char. I changed the field
length to 40 and it still worked as expected.
Regards,
It seems to be ACE bug. Not Harbour.
Here is reduced example:
REQUEST ADS
proc main()
ad
Can someone who has older ACE libraries check if the problem is local to
version 9.00 and does not exist in 8.10 and/or in older versions?
Not an issue in ACE 6.2
___
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.or
Hello Viktor,
The same would be true to use this: 'IF Empty (hLib)', but this in itself
will fivewin about some changes for this work and in my application also
been several changes :(.
I suggest you make a test for when using fivewin together with the harbour,
that the "LoadLibrary" return nu
On Thu, 14 May 2009, Petr Chornyj wrote:
Hi,
> > If you use zlib compiled with this macro then it does not contain
> > any functions used for compression like gzwrite() or gzflush() so
> > you cannot even link final binaries if code use such functions.
> > Do you want to remove HB_GZWRITE()/HB_GZ
On Thu, 14 May 2009, Szak�ts Viktor wrote:
> In this case you can copy hbmk.cfg to one of these location neutral dirs:
> /etc/harbour or ~/.harbour.
The problem is that such hbmk.cfg will be common to all Harbour
installations I will have in the system.
> If there is a way to detect that softlink
Rossine
You have to apply patch in olecore.c itself.
You cannot club this function in #prgma begin/end
construct. Rebuild olecore.c.
Also try Przemek's refinements. May be any could help.
Regards
Pritpal Bedi
--
View this message in context:
http://www.nabble.com/Errors-with-11032-tp23521549
Przemyslaw Czerpak wrote:
Here is reduced example:
REQUEST ADS
proc main()
adsSetServerType( 1 )
rddSetDefault( "ADS" )
dbCreate( "test.dbf", { { "F", "C", 24, 0 } } )
USE test
while lastrec() < 10
dbappend()
field->F := repl( "12345678
Hello Pritpal,
Including this in my example: (WEBEXP.PRG)
[CODE]
#include "FiveWin.ch"
function Main()
local oWnd, oActiveX
local cEvents := ""
DEFINE WINDOW oWnd TITLE "FiveWin ActiveX Support"
oActiveX = TActiveX():New( oWnd, "Shell.Explorer" )
oWnd:oClient = oActiveX // T
Antonio Linares can perhaps give us a opnião about this ...
___
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
Hello Przemek and All,
Przemyslaw Czerpak-2 wrote:
>
> In summary I guess that you wanted to make sth like:
>
>if( lOleError == S_OK )
> lOleError = CoCreateInstance( HB_ID_REF( ClassID ), NULL,
> CLSCTX_SERVER, fIID ? HB_ID_REF( iid ) : HB_ID_REF( IID_IDispatch ), (
> void**
Thank you. Please also Chens request about the sf.net logo update, plus IMO
we should also have a link from the sf.net logo (in the
middle left of homepage) to sf.net as per sf.net instructions.
I think right now we're missing page loads from sf.net stats because
of the missing/old link.
Brgds,
Vi
Thanks Alex, I'll commit it in next session.
Brgds,
Viktor
On Thu, May 14, 2009 at 2:02 PM, Alex Strickland wrote:
> Hi Viktor
>
> Some time ago you tidied up my code and added it to the hbwin library. I
> found a typo I made, it is at line 237 :
>
> NewTimeouts.ReadTotalTimeoutConstant = ( s_
On Thu, May 14, 2009 at 2:01 PM, Rossine wrote:
>
> Hello,
>
> In class TIMAGE de FWH, when using:
>
>
> #ifdef __CLIPPER__
> hLib = LoadLib32( "freeimage.dll" )
> #else
> hLib = LoadLibrary( "freeimage.dll" )
> #endif
>
> msgstop( valtype( hlib ) + "-"+ valtoprg( hlib ) )
>
> i
Revision: 11048
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=11048&view=rev
Author: vszakats
Date: 2009-05-14 12:17:47 + (Thu, 14 May 2009)
Log Message:
---
2009-05-14 14:17 UTC+0200 Viktor Szakats (harbour.01 syenar hu)
* utils/hbmk2/hbmk2.prg
During the day now available...
___
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
On Thu, 14 May 2009, Szak�ts Viktor wrote:
I made some test with recent hbmk2 and it stop to automatically
add [n]curses, slang and x11 libraries in link phase.
I'll try to make some tests, but it'd help if you'd post your
hbmk.cfg
cmdline and -trace output.
I've found the reason.
Because I
Hi Viktor
Some time ago you tidied up my code and added it to the hbwin library. I found a
typo I made, it is at line 237 :
NewTimeouts.ReadTotalTimeoutConstant = ( s_ReadIntervalTimeout == -1 ? 0 :
s_ReadTotalTimeoutConstant );
It should be:
NewTimeouts.ReadTotalTimeoutConstant = (
>
> Yes, in few cases I want to remove HB_GZWRITE()/HB_GZFLUSH(), as, as I do
> guess majority from the users of Windows never use by the functions of
> HB_GZ* family and it can reduce the size of final binary file (little
> piece). I understand usually, that this is not important vitally or eve
Hello,
In class TIMAGE de FWH, when using:
[CODE]
hLib = LoadLibrary( "freeimage.dll" )
msgstop( hLib )
[ENDCODE]
...with harbour 11021, i see:
[RESULT]
Type = N Value = 61997056
[ENDRESULT]
...and harbour 11047:
[RESULT]
Type = P value = 0x03D2
[ENDRESULT]
...and i see this er
Hi,
Apply this patch in olecore.c and let me know it solves this issue:
but there is a question if we want to still tolerate code with
C pointer <-> HVM number item conversions and keep workarounds
for it. I'm leaving the decision to Mindaugas.
Usage of numbers instead of pointers is one
On Thu, 14 May 2009, Szak�ts Viktor wrote:
> > I made some test with recent hbmk2 and it stop to automatically
> > add [n]curses, slang and x11 libraries in link phase.
> I'll try to make some tests, but it'd help if you'd post your hbmk.cfg
> cmdline and -trace output.
I've found the reason.
Beca
Przemek, thanks!
>Please inform Extended System that AdsGetField(..., ADS_NONE)
>for character fields longer then 23 bytes read from DBF tables increase
>pulLen by one adding one garbage character to the end of string.
Done.
I´ll post any news here.
Best regards,
Toninho.
_
Przemyslaw Czerpak-2 wrote:
>
>
> If you use zlib compiled with this macro then it does not contain
> any functions used for compression like gzwrite() or gzflush() so
> you cannot even link final binaries if code use such functions.
> Do you want to remove HB_GZWRITE()/HB_GZFLUSH()/... from rt
Hello, Pritpul,
Apply this patch in olecore.c and let me know it solves this issue:
Mindaugus, please check if this can be implemented like this or
some else construct is needed.
HB_FUNC( OLECREATEOBJECT ) /* ( cOleName | cCLSID [, cIID ] ) */
{
...
if( HB_IS_NUMBER( 1 ) )
{
We will
Przemek, thanks!
>Please inform Extended System that AdsGetField(..., ADS_NONE)
>for character fields longer then 23 bytes read from DBF tables increase
>pulLen by one adding one garbage character to the end of string.
Done.
I´ll post any news here.
Best regards,
Toninho.
___
>
> if( lOleError == S_OK )
> lOleError = CoCreateInstance( HB_ID_REF( ClassID ), NULL,
> CLSCTX_SERVER, fIID ? HB_ID_REF( iid ) : HB_ID_REF( IID_IDispatch ), (
> void** ) ( void * ) &pDisp );
>}
> + else if( ISNUM( 1 ) )
> + {
> + pDisp = ( IDispatch * ) ( HB_PTRDIFF )
Revision: 11047
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=11047&view=rev
Author: vszakats
Date: 2009-05-14 10:18:55 + (Thu, 14 May 2009)
Log Message:
---
2009-05-14 12:18 UTC+0200 Viktor Szakats (harbour.01 syenar hu)
* contrib/hbwin/tests/testd
Hi Przemek,
I can revert it although IMO we should rather think about a proper
alternative, these options were missing from default builds since
last year and no one reported them missing, which suggest it
isn't that important. I've since removed all HB_EXTENSION from
code, so readding this one
On 14 May 2009 11:09, I wrote:
> Hmm, I don't know if this GZ had something to do with the LZ algorithm...
Total nonsense, I should had looked at the code.
Its a flag to avoid linking parts which are not used, to create a smaller code.
Chen.
___
Harbo
On Thu, 14 May 2009, Szak�ts Viktor wrote:
> Thanks, please test it now.
Thanks, looks OK.
I made some test with recent hbmk2 and it stop to automatically
add [n]curses, slang and x11 libraries in link phase.
Can I set some default libraries which should be used with all
projects in hbmk.cfg?
b
Revision: 11046
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=11046&view=rev
Author: vszakats
Date: 2009-05-14 09:19:28 + (Thu, 14 May 2009)
Log Message:
---
2009-05-14 11:18 UTC+0200 Viktor Szakats (harbour.01 syenar hu)
* contrib/rddads/rddads.h
Revision: 11045
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=11045&view=rev
Author: vszakats
Date: 2009-05-14 09:04:50 + (Thu, 14 May 2009)
Log Message:
---
2009-05-14 10:55 UTC+0200 Viktor Szakats (harbour.01 syenar hu)
* utils/hbmk2/hbmk2.prg
On Thu, 14 May 2009, Szak�ts Viktor wrote:
Hi,
> > BTW I also found that aceapi-9.00.0.7.tar.gz is wrongly autodetected as
> > 9.10 not 9.00. Viktor Can you look at it?
> Of course, could you post me ace.h included in aceapi-9.00.0.7.tar.gz?
In attachment you a have diff for version 9.00.0.7 and
On Thu, 14 May 2009, Szak�ts Viktor wrote:
Hi,
> > In this case you removed important functionality without leaving any
> > alternative for users. _SET_EOL is not a solution or replacement at all.
> > Now it's not possible to use multiple EOLs in this set of functions what
> > is very important i
Przemek,
> Can you explain what you want to do and what does it mean for you
> "respect a NO_GZCOMPRESS definition"?
> If you use zlib compiled with this macro then it does not contain
> any functions used for compression like gzwrite() or gzflush() so
> you cannot even link final binaries if code
On Wed, 13 May 2009, Pritpal Bedi wrote:
Hi,
> > METHOD New( oWnd, cProgID, nRow, nCol, nWidth, nHeight ) CLASS TActiveX
> > ::oOleAuto = TOleauto():New( ActXPdisp( ::hActiveX ) )
> Apply this patch in olecore.c and let me know it solves this issue:
> Mindaugus, please check if this can be
Hi,
> I've just made test in Linux using aceapi-9.10.0.0.tar.gz and it gives
> wrong results so also needs above workaround.
> So I'll re-enable the hack I added. I do not know if newer ACE client
> exists. Please inform Extended System that AdsGetField(..., ADS_NONE)
> for character fields longe
Hello Przemek,
>> I had that problem. I solved using the latest ADS 9.10 SDK.
>
> Thanks for the information. So it's already fixed.
> Can someone who has older ACE libraries check if the problem is local to
> version 9.00 and does not exist in 8.10 and/or in older versions?
I was using before AC
>
> Apply this patch in olecore.c and let me know it solves this issue:
> Mindaugus, please check if this can be implemented like this or
> some else construct is needed.
>
> HB_FUNC( OLECREATEOBJECT ) /* ( cOleName | cCLSID [, cIID ] ) */
> {
> wchar_t*cCLSID;
> GUIDClassID, iid;
>
> > * source/rtl/mlcfunc.c
> > - Deleted HB_EXTENSION branch. Please modify your program to
> > use Set( _SET_EOL ) instead of passing EOLs explicitly to
> > memo functions. This works also in xhb.
> > INCOMPATIBLE when for HB_EXTENSION builds.
>
> In this case you removed
Thanks very much for the quick fix, the problem is now gone.
Brgds,
Viktor
On Wed, May 13, 2009 at 11:49 PM, Przemyslaw Czerpak wrote:
> On Wed, 13 May 2009, Szak�ts Viktor wrote:
>
> Hi,
>
> > An even simpler one:---
> > PROCEDURE Main()
> >LOCAL oBrowse, tmp
> >oBrowse := TBrowseNew( 0,
- Forwarded message from Przemyslaw Czerpak -
From: Przemyslaw Czerpak
Subject: Re: [Harbour] hbzlib and NO_GZCOMPRESS
To: "Harbour Project Main Developer List."
Date: Tue, 12 May 2009 15:43:05 +0200
Lines: 26
On Tue, 12 May 2009, Petr Chornyj wrote:
Hi,
> It is possible to change ou
ADO like ODBC need a driver in each client
2009/5/14 Sérgio Kondo :
> Hi,
> ADORDD need MyODBC in all Client?
> Thanks
> Sérgio
>
>
> From: calimer...@freemail.it
> Sent: Wednesday, May 13, 2009 12:15 PM
> To: harbour@harbour-project.org ; harbour@harbour-project.org
> Subject: [Harbour] Mini-tut
Revision: 11044
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=11044&view=rev
Author: druzus
Date: 2009-05-14 06:48:19 + (Thu, 14 May 2009)
Log Message:
---
2009-05-14 08:56 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/contrib/rddad
On Thu, 14 May 2009, Przemyslaw Czerpak wrote:
Hi
> > > Fernando reported that 9.10 is fixed.
> > I'm not confirm this information. 9.10 still bugged without your fix.
> So it means that it has to be reported to Extended System and the workaround
> activated also for other version.
I've jus
Hello
Rossine wrote:
>
> METHOD New( oWnd, cProgID, nRow, nCol, nWidth, nHeight ) CLASS TActiveX
> ::oOleAuto = TOleauto():New( ActXPdisp( ::hActiveX ) )
>
Apply this patch in olecore.c and let me know it solves this issue:
Mindaugus, please check if this can be implemented like this or
- Forwarded message from Przemyslaw Czerpak -
From: Przemyslaw Czerpak
Subject: Re: [Harbour] hbmk2: any features still missing?
To: "Harbour Project Main Developer List."
Date: Tue, 12 May 2009 14:47:55 +0200
Lines: 12
On Tue, 12 May 2009, Szak�ts Viktor wrote:
Hi,
> Hi All,
> Is th
- Forwarded message from Przemyslaw Czerpak -
From: Przemyslaw Czerpak
Subject: Re: [Harbour] SF.net SVN: harbour-project:[11023] trunk/harbour
To: "Harbour Project Main Developer List."
Date: Tue, 12 May 2009 17:54:36 +0200
Lines: 20
On Tue, 12 May 2009, vszak...@users.sourceforge.net
On Wed, 13 May 2009, toni...@fwi wrote:
Hi
> Thanks for your fix, but I'm an ADS 9.10 user and I can see that if I
> use ads1.c 10193 2009-02-06 18:23:08Z vszakats, I see no bug.
> The bug appear in ads1.c 10946 2009-04-28 19:35:27Z druzus.
> Can you check it please? Maybe I'm wrong.
It's possi
(sending this message again as I can't see it in the archive, with a direct CC
just to make sure)
On Thu, 14 May 2009 00:58:16 +0200, Szak?ts Viktor wrote:
> Hi Vailton,
> Would you please add this little icon to our webpage:
While you do that, if you can update the SF logo as I wrote in this m
71 matches
Mail list logo
