And this is my reply to the previous post:
Jimmy
>> :-) Which means that you cannot work on the same area from different
>> threads.
>
> you can use Xbase++ ZeroSpace() to switch Workspace if you want ...
Absolutely right. Today Przemek has committed exactly same behaviour as of
Xbase++
*
Yet another post from Jimmy, an Xbase++ developer:
hi,
> :-) Which means that you cannot work on the same area from different
> threads.
you can use Xbase++ ZeroSpace() to switch Workspace if you want ...
> Concretely, unless the thing has been changed beyond my knowledge, it is
> just a matt
2008-09-20 05:37 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/common.mak
* harbour/source/rdd/Makefile
! restored dbdetach.* in makefiles
+ harbour/source/rdd/dbdetach.c
+ added file I forgot to add in previous commit
* harbour/source/vm/thread.c
+ added supp
Hello Guys
Here is the message transcript on xHarbour NG:
***
Pritpal Bedi ha scritto:
> Hello Gian
>
>>> Development of the trunk continues at version 1.1.0 with a major update
>>> to the project to allow multi-threading.
>>>
>> Just by the w
On Friday 19 September 2008 08:51:29 pm Bill Smith wrote:
> BCC32 (WinXP) reports it doesn't know how to make obj/b32/dbdetach.obj-
>
> On Sat, 2008-09-20 at 03:38 +0200, Przemyslaw Czerpak wrote:
> > 2008-09-20 03:38 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
> > * harbour/source/vm/hv
BCC32 (WinXP) reports it doesn't know how to make obj/b32/dbdetach.obj-
On Sat, 2008-09-20 at 03:38 +0200, Przemyslaw Czerpak wrote:
> 2008-09-20 03:38 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
> * harbour/source/vm/hvm.c
> * minor modification: #ifdef ... => #if defined( ... )
>
2008-09-20 03:38 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/source/vm/hvm.c
* minor modification: #ifdef ... => #if defined( ... )
* harbour/source/rtl/gtcrs/gtcrs.c
* harbour/source/rtl/gttrm/gttrm.c
! fixed reentrance counter resetting
best regards
Przemek
Message: 4
Date: Fri, 19 Sep 2008 19:06:21 +0200
From: "Massimo Belgrano" <[EMAIL PROTECTED]>
Subject: RE: [Harbour] Unresolved External with BCC582
To: "Harbour Project Main Developer List."
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset="US-ASCII"
You have to l
On Wed, 17 Sep 2008, Przemyslaw Czerpak wrote:
Hi Pritpal,
> > Here are two functions in Xbase++ which amply
> > demonstrate the functionality of ZERO space where
> > two threads can share a common WA.
> Many thanks for the documentation.
I just commited them but I had to make one thing which
i
2008-09-20 00:42 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
+ harbour/source/rdd/wadetach.c
+ added .prg functions:
hb_waDetach( [|], [] ) ->
hb_waRequest( [], [], [<@xCargo>], [] )
->
The first one detach workarea from current thread and move i
On Fri, 19 Sep 2008, Pritpal Bedi wrote:
Hi Pritpal,
> Function HB_EXEC() is included in HBTIP.LIB.
> Is not this function a candidate for RTL?
Rather not. This function uses hack which is
strictly oriented for current HVM low level
code. I'd rather prefer sth cleaner for such
jobs or implement
Przemek,
Przemyslaw Czerpak-2 wrote:
>
> OK, I'll look at RDD code and check what can be optimal solution.
>
>> BTW clause gives RTE, can you please look into?
>
> When I was working on new PP and std.ch I removed them and
> haven't restored so far. Maybe even good because now we can
> colle
On Fri, 19 Sep 2008, Pritpal Bedi wrote:
Hi Pritpal,
> I re-request :
> 1) Extend clause to be of type PHB_ITEM
> OR
> 2) Provide new clause like of type PHB_ITEM in
> core of RDD.
> I understand that this feature may be used by new RDDs in
> different way but probably new RDDs may have addit
Hello Przemek
I re-request :
1) Extend clause to be of type PHB_ITEM
OR
2) Provide new clause like of type PHB_ITEM in
core of RDD.
I understand that this feature may be used by new RDDs in
different way but probably new RDDs may have additional features
which cannot be covered uniformally
2008-09-19 22:27 UTC+0200 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/source/vm/set.c
! fixed casting
* harbour/contrib/xhb/xhbcopyf.c
* replaced TABs with SPACEs
* harbour/include/hbapirdd.h
* harbour/source/rdd/workarea.c
* harbour/source/rdd/wacore.c
* harbour/sou
Hello Guys
Function HB_EXEC() is included in HBTIP.LIB.
Is not this function a candidate for RTL?
HBWhat.lib calls this function. In xHarbour it is
included in eval.c.
Regards
Pritpal Bedi
--
View this message in context:
http://www.nabble.com/HB_EXEC%28%29---Candidate-for-RTL---tp19578811p1
On Fri, 19 Sep 2008, maurilio longo wrote:
Hi Maurilio,
> > This function cannot be used to join given thread because it waits for
> > _any_ thread termination.
> No, you can wait for a particular thread.
> See:
> ptid (PTID) - in/out Address of the thread identification.
> Input The address of t
Przemyslaw,
> This function cannot be used to join given thread because it waits for
> _any_ thread termination.
No, you can wait for a particular thread.
See:
ptid (PTID) - in/out Address of the thread identification.
Input The address of the ThreadID of the thread of interest. If ptid is 0,
On Fri, 19 Sep 2008, Maurilio Longo wrote:
Hi Maurilio,
> this is what doswaitthread() does:
> Places the current thread into a wait state until another thread in the
> current process has ended. It then returns the thread identifier of the
> ending thread.
[...]
> I hope this helps and thanks fo
On Fri, 19 Sep 2008, Maurilio Longo wrote:
Hi Maurilio,
> no, there is not mmap on OS/2. Several years ago I wrote a .dll to emulate it
> for the msql (miniSQL) OS/2 port and it worked ok, but was not the real thing
> :)
> You can still find it on http://hobbes.nsmu.edu, but I think it would be
I vote also for not implement "feature" which breaks C52 compatibility
HB_strictlyCOMPAT_C53_and_imcompatible_52
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Szakáts Viktor
Sent: Friday, September 19, 2008 7:21 PM
To: Harbour Project Main Developer List
Hi Miguel,
(remember that OrdKeyVal and others doesn't exists at CL52 as you
know perfectly)
You've added two new RTEs in CDX code in two commits.
Both of these went into the index creation code
(hb_cdxOrderCreate()), so I'm not sure what do these have
to do with 5.3-only OrdKeyVal(). What
You have to link your application. with cw32mt.lib instead of cw32.lib
and with hbvmmt.lib instead of hbvm.li
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Mario H.
Sabado
Sent: Friday, September 19, 2008 1:47 AM
To: harbour@harbour-project.org
Subject:
Przemyslaw Czerpak wrote:
> In theory such situation is possible if OS2 does not keep
> thread return results for parent process when thread terminates.
> In such case DosWaitThread() will always fail and hb_threadJoin()
> will not store thread return value in second parameter. I'll
> add workaroun
Przemyslaw,
no, there is not mmap on OS/2. Several years ago I wrote a .dll to emulate it
for the msql (miniSQL) OS/2 port and it worked ok, but was not the real thing :)
You can still find it on http://hobbes.nsmu.edu, but I think it would be
better not to use it at all.
Best regards.
Maurilio
Szakáts Viktor escribió:
Hi Miguel,
we are talking about CDX not six.
the question is create incompatible CDX or not create ?
Incompatible with what? I don't really understand what
existing bugs are you trying to fix here to be honest.
(remember that OrdKeyVal and others doesn't exists a
Hi Alex,
That's perfect, thanks a lot.
Brgds,
Viktor
On 2008.09.19., at 17:08, Alex Strickland wrote:
Szakáts Viktor wrote:
Not at all, very good idea.
I basically used your release wording, and added a little:
To all
Harbour: 1.0.1 released (2008-09-18 18:41)
The Harbour Project is a
Szakáts Viktor wrote:
Not at all, very good idea.
I basically used your release wording, and added a little:
To all
Harbour: 1.0.1 released (2008-09-18 18:41)
The Harbour Project is a Free Open Source Software effort to build a
multiplatform Clipper language compiler. Harbour consists of t
Hi Miguel,
we are talking about CDX not six.
the question is create incompatible CDX or not create ?
Incompatible with what? I don't really understand what
existing bugs are you trying to fix here to be honest.
Instead of revert, i think is better to work by setting. Then people
as you tha
I trust Alex's judgment on the best wording on this
announcement.
Regarding yours, I'd definitely give a lesser marketing
tone to it, and I'd definitely refrain from advertising
not-yet existing features in Harbour (like multi-window),
however cool they might be, since there is no need to spread
rg
http://lists.harbour-project.org/mailman/listinfo/harbour
__ Información de ESET Smart Security, versión de la base de
firmas de virus 3455 (20080919) __
ESET Smart Security ha comprobado este mensaje.
http://www.eset.com
___
H
Absolutly but is only a suggestion
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Lorenzo
Fiorini
Sent: Friday, September 19, 2008 2:48 PM
To: Harbour Project Main Developer List.
Subject: Re: [Harbour] comp.lang.clipper
On Fri, Sep 19, 2008 at 2:26 PM,
>As Przemek says, SIX for example doesn't RTE. So this seem
>rather an RDD driver (and/or version) specific property,
>and as such I'd suggest to replicate it that way in Harbour.
FYI ADS doesn't RTE too.
Regards,
Toninho.
__
Faça ligações para ou
On Fri, Sep 19, 2008 at 2:26 PM, Massimo Belgrano <[EMAIL PROTECTED]> wrote:
> What do you thiunk of :
>
> Harbour community is pleased to announce that a major release of the
> harbour: vesion 1.0.0.
> ...
> http://lists.harbour-project.org/mailman/listinfo/harbour-users
Massimo, don't you thin
Good idea
What do you thiunk of :
Harbour community is pleased to announce that a major release of the
harbour: vesion 1.0.0.
Harbour is robust, highly tested, and valued by developers all over the
globe for its capabilities. Available on a wide range of Unix and
Windows platforms.
now we are w
Hi Alex,
Not at all, very good idea.
Brgds,
Viktor
On 2008.09.19., at 14:15, Alex Strickland wrote:
Hi
Does anyone have any objections to me putting a short note on
comp.lang.clipper outlining the significant milestones recently
passed and the new mt stuff?
Unless I missed something al
Hi Miguel,
Only CL53/COMIX CDX driver generate RT error for keys longer then
maximal
supported length. All other like NTX, CL52CDX, SIX3 (NTX,CDX,NSX)
and
[x]Harbour current code simply cut the index keys to the maximal
length
without any RT errors so I do not know it's good idea. People
Hi
Does anyone have any objections to me putting a short note on
comp.lang.clipper outlining the significant milestones recently passed
and the new mt stuff?
Unless I missed something already.
Regards
Alex
___
Harbour mailing list
Harbour@harbour-p
Only CL53/COMIX CDX driver generate RT error for keys longer then maximal
supported length. All other like NTX, CL52CDX, SIX3 (NTX,CDX,NSX) and
[x]Harbour current code simply cut the index keys to the maximal length
without any RT errors so I do not know it's good idea. People may have
code which
Only CL53/COMIX CDX driver generate RT error for keys longer then maximal
supported length. All other like NTX, CL52CDX, SIX3 (NTX,CDX,NSX) and
[x]Harbour current code simply cut the index keys to the maximal length
without any RT errors so I do not know it's good idea. People may have
code which
Hi Miguel, Przemek,
? "Length of OrdKeyVal() :" + Str( Len(OrdKeyVal()) )
WHILE ! EOF()
? Right( FIELD->Test, 1 )
DbSkip()
ENDDO
USE
FERASE( "__MYTEST.DBF" )
FERASE( "__MYTEST.CDX" )
? "end"
RETURN NIL
Only CL53/COMIX CDX driver generate RT error for keys longe
On Thu, 18 Sep 2008, David Arturo Macias Corona wrote:
Hi David,
> > FAILED mttest05, mttest06
> > something appear/disappear flashing in screen
> mttest05 are failing in random way
> Results of previous message were after building Harbour
> After reboot
> Rebuilding mttest05
> Failed firs
On Fri, 19 Sep 2008, Miguel Angel Marchuet wrote:
> OrdKeyVal can't be more large than 240 at cdx almoust in clipper.
Just like in all other languages.
240 bytes is forced by internal CDX structure for 512 bytes
index page length.
You have to be able to store in the worst case at least two keys
in
All fine
My suggest is have one page for download for easy approach, and one
Third-party Tools and Libraries
5 colunn Nape of platform,compiler, Harbour Binary 1.1 + C Compiler +
Note
Binary C compiler Notes
- Windows
BCC 5.51
BCC 5.8
2008-09-19 12:21 UTC+0100 Miguel Angel Marchuet <[EMAIL PROTECTED]>
* contrib/hbbmcdx/bmdbfcdx.c
* source/rdd/dbfcdx/dbfcdx1.c
! fixed return value after rt error.
* Added rt error as in clipper when OrdkeyVal is greater than 240, with
default
capabilities as in clipper.
cod
Budyanto Dj. ha scritto:
On 18 Sep 2008 at 20:06, Francesco Saverio Giudice wrote:
FYI, today sf.net reports Harbour in Overall Rank at position 172 and at
10th under Database category.
In Compilers category it's #4
http://sourceforge.net/softwaremap/trove_list.php?form_cat=48
Great!
Best
If i build without HB_BUILD_ST=Y I can create A MT application linking
HBVMmt.lib instead hbbm.lib
If I Build with HB_BUILD_ST=N only hbvm.lib is created
I am using bcc
Wich difference is between HB_BUILD_ST=N+hbvmmt.lib and HB_BUILD_ST=Y ?
Massimo Belgrano
2008-09-19 11:00 UTC+0200 Viktor Szakats (harbour.01 syenar hu)
* contrib/hbct/fcopy.prg
* contrib/hbnf/dfile.prg
* contrib/hbpgsql/pgrdd.prg
* contrib/hbtpathy/telepath.prg
* contrib/rddado/adordd.prg
+ Some STATICs converted to THREAD STATIC.
* contrib/hbtip/ftpcln.prg
* STAT
Hi Przemek,
Many thanks, this is perfectly inline with what I've expected.
Brgds,
Viktor
On 2008.09.18., at 22:25, Przemyslaw Czerpak wrote:
On Thu, 18 Sep 2008, Szak�ts Viktor wrote:
Hi Viktor,
I also have one:
If a given thread gets a non-handled RTE, or, an internal
error, will the whol
or better
/* Make sure KEY has proper type and iLen lower than 240 */
if ( uiLen > CDX_MAXKEY )
{
if( hb_cdxErrorRT( pArea, bType == 'U' ? EG_DATATYPE : EG_DATAWIDTH,
EDBF_INVALIDKEY, NULL, 0, uiLen > CDX_MAXKEY ?
EF_CANDEFAULT : 0 ) == FAILURE )
{
OrdKeyVal can't be more large than 240 at cdx almoust in clipper.
If you are agree I upload this changes:
I want to do to this changes :
bType = hb_cdxItemType( pResult );
switch ( bType )
{
case 'N':
case 'D':
uiLen = 8;
break;
case 'L':
uiL
memtst.prg 9410 2008-09-15 21:09:27Z druzus $
09/19/08 09:56:39 Harbour 1.1.0dev (Rev. 9444) (MT), Windows XP
5.1.2600 Service Pack 2
Sep 19 2008 09:49:15 Borland C++ 5.8.2 (32 bit)
09/19/08 10:00:03 Harbour 1.1.0dev (Rev. 9444) (MT), Windows XP
5.1.2600 Service Pack 2
Sep 19 2008 08:53:57 Open W
Viktor,
I agree, publics should be public :) between threads with synchronization left
to the user.
This has the advantage to be xbase++ compatible.
I think that on mt issues we should be compatible with xbase++ which is now
the "de facto" standard.
Best regards.
Maurilio.
Szakáts Viktor wro
53 matches
Mail list logo
