Revision: 10356
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10356&view=rev
Author: druzus
Date: 2009-02-20 03:11:51 + (Fri, 20 Feb 2009)
Log Message:
---
2009-02-20 04:15 UTC+0100 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/include/hbgti
Revision: 10355
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10355&view=rev
Author: vouchcac
Date: 2009-02-20 00:52:31 + (Fri, 20 Feb 2009)
Log Message:
---
2009-02-19 16:50 UTC-0800 Pritpal Bedi (prit...@vouchcac.com)
* harbour/source/rtl/gtwvt/gt
Revision: 10354
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10354&view=rev
Author: vszakats
Date: 2009-02-19 23:45:22 + (Thu, 19 Feb 2009)
Log Message:
---
2009-02-20 00:44 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
* utils/hbmk2/hbmk2.prg
I not have gd lib package to test uhttpd
and test seem already done
2009/2/19 Francesco Saverio Giudice :
> Hi Maurilio,
>
> Il 19/02/2009 17.12, Maurilio Longo ha scritto:
>>
>> Hi,
>>
>> I've built uhttpd on OS/2 using inet, it works ok, but every new
>> connection
>> uses 4 Kb of memory, more o
moved from c:\vc9.bat c:\harbour\vc9.bat and removed line with cd \harbour
Give follow error but continue without problem and all works fine (Thanks )
make[1]: [install] Error 1 (ignored)
I not have changed starting directory using HB_INSTALL_PREFIX
It is very strange for a windows user be con
Hi Maurilio,
Il 19/02/2009 17.12, Maurilio Longo ha scritto:
Hi,
I've built uhttpd on OS/2 using inet, it works ok, but every new connection
uses 4 Kb of memory, more or less, which is not freed until after the process
ends.
Question: did anybody make some test with uhttpd build with inet on w
On Thu, 19 Feb 2009, Szak�ts Viktor wrote:
> Let me include here our conversation from autumn I was referring to,
> back then the situations looked cleaner than what it seems now:
[...]
> -- vszakats says:
> No, if possible let's please remove -n1 and keep only -n/-n-.
>
OK. When hbmk2 will b
Hello.
I am trying to compile harbour in Fedora 5 (I need to rotate in him).
But he asks to change Client Subversion (very old).
Does some exist it sorts things out of me to outline that?
Best Regards.
{}s
Joel Bernardes
Veja quais são os assuntos do momento no Yahoo! +Buscados
On Wed, 18 Feb 2009, Lorenzo Fiorini wrote:
Hi,
> > Are there restrictions I can't see?
> Maybe I found the problem.
> If I replace XWC_DEFAULT_FONT_ENCODING "iso10646-1" with "*-1" I get
> many fonts working.
Yes though this are not unicode fonts so not in all countries it's an
option.
Looks
-Messaggio Originale-
Da: "Viktor Szakáts"
A: "Harbour Project Main Developer List."
Data invio: giovedì 19 febbraio 2009 22.20
Oggetto: Re: [Harbour] TGet: setting cursor position to ::nMaxLen + 1
Hi Enrico,
You can subclass the GET() class and create your custom setPosMy() metho
Hi Enrico,
You can subclass the GET() class and create your custom setPosMy() method,
or maybe you can even override the
original and continue to use the name.
Brgds,
Viktor
On Thu, Feb 19, 2009 at 9:56 PM, Enrico Maria Giordano <
e.m.giord...@emagsoftware.it> wrote:
>
> -Messaggio Original
-Messaggio Originale-
Da: "Viktor Szakáts"
A: "Harbour Project Main Developer List."
Data invio: giovedì 19 febbraio 2009 20.35
Oggetto: Re: [Harbour] TGet: setting cursor position to ::nMaxLen + 1
Is this a Clipper compatibility fix?
No, as far as I know. It is just a personal n
Revision: 10353
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10353&view=rev
Author: vouchcac
Date: 2009-02-19 20:48:17 + (Thu, 19 Feb 2009)
Log Message:
---
2009-02-19 12:42 UTC-0800 Pritpal Bedi (prit...@vouchcac.com)
* harbour/contrib/gtwvg/wvgwi
It's created, but it's not installed, since your script is still wrong.
You've added 'cd \harbour', which broke the HB_INSTALL_PREFIX
setting.
Either remove 'cd \harbour' and run the script from the
harbour root dir, as I've recommended here a few times,
or tweak
'set HB_INSTALL_PREFIX=%~dp0'
to
'
Hi Przemek,
> Use .a files just like .o ones and pass them to compiler as is.
> It will now what to do with them.
This is in SVN already. For -l I'm stripping all extensions,
not just dll/so/a. Pls tell if this is a problem, and I'll solve it.
> Yes it also supports it. DJGPP LD has also uses
Hi Przemek,
> > I know, but this isn't a Linux tool :) For Linux it isn't hardcoded,
> > just the default is set to gcc. You can set it just like for hbmk
> > script HB_COMPILER=gpp|owatcom. The new command
>
> For hbmk script I do not have to set anything because it's build
> dynamically in inst
Revision: 10352
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10352&view=rev
Author: vszakats
Date: 2009-02-19 19:46:12 + (Thu, 19 Feb 2009)
Log Message:
---
2009-02-19 20:44 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
* utils/Makefile
! Fixe
Is this a Clipper compatibility fix?
Brds,
Viktor
On Thu, Feb 19, 2009 at 8:09 PM, Enrico Maria Giordano <
e.m.giord...@emagsoftware.it> wrote:
> Dear friends, I need to place the GET's cursor position to ::nMaxLen + 1. I
> changed the current code from setPos() method
>
> ::nPos := iif( ::nMaxLe
I have resolved , was a problem due to old files in my repository that
svn update not update
Thanks
Il 19 febbraio 2009 20.01, Pritpal Bedi ha scritto:
>
> Massimo
>
> Can you send the errors log such as below with latest SVN download.
>
> Pritpal Bedi
>
>
> Massimo Belgrano-3 wrote:
>>
>> i Not
Dear friends, I need to place the GET's cursor position to ::nMaxLen + 1. I
changed the current code from setPos() method
::nPos := iif( ::nMaxLen == 0, 1, ::nMaxLen )
to
::nPos := iif( ::nMaxLen == 0, 1, nPos )
and it seems to work fine.
What is the better way to do this change, other than
Revision: 10351
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10351&view=rev
Author: druzus
Date: 2009-02-19 19:08:14 + (Thu, 19 Feb 2009)
Log Message:
---
2009-02-19 20:13 UTC+0100 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/source/vm/thr
Revision: 10350
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10350&view=rev
Author: tfonrouge
Date: 2009-02-19 19:06:11 + (Thu, 19 Feb 2009)
Log Message:
---
2009-02-19 12:59 UTC-0600 Teo Fonrouge (teo/at/windtelsoft/dot/com)
* common.mak
* cha
Massimo
Can you send the errors log such as below with latest SVN download.
Pritpal Bedi
Massimo Belgrano-3 wrote:
>
> i Not able to create gtwvg.lib in c:\harbour\lib
> C:\Program Files\\Microsoft SDKs\Windows\v6.0A\include\olectl.h(205) :
> warning C4201: utilizzata estensione non standard:
Warning W8080 source\vm\thread.c 1078: 'fResult' is declared but never used
in function hb_threadWait
EMG
--
EMAG Software Homepage: http://www.emagsoftware.it
The EMG's ZX-Spectrum Page: http://www.emagsoftware.it/spectrum
The Best of Spectrum Games: http://www.emagsoftware.it/tbosg
The EM
On Wednesday 18 February 2009 21:09:59 Viktor Szakáts wrote:
> Hi all,
> After rechecking the situation with gtcrs, gtsln,
> I'd like to ask group whether we could move
> these two GTs to the contrib area.
>
> The reasons are:
> 1) Dependency on optional external libs,
> with the same issues ou
On Thu, 19 Feb 2009, Szak�ts Viktor wrote:
Hi,
> Right now every filename passed as is (without -l) with the
> extension .a is considered a library. Would it help if hbmk would
> simply handle .a files just like object files? This would mean
> that libs with .a ending must have to be passed with
Massimo,
contrib/examples/uhttpd is the source code, you build it using inet (instead
of socket, which is default), then you start it.
>From a different pc or even the same, open IE, and ask for
localhost/index.html if on the same pc or ip-of-the-other-pc/index.html.
Now keep refreshing it (F5 o
Sorry i can't.
i not have uhttpd server and don't know How i write a simple query
Have only windows and harbour
2009/2/19 Maurilio Longo :
> A sample?
>
> I simply query http://ip-of-pc-with-uhttpd/index.html from a different pc,
> on the pc with the uhttpd server memory keeps going down.
>
>
On Thu, 19 Feb 2009, Szak�ts Viktor wrote:
Hi,
> I know, but this isn't a Linux tool :) For Linux it isn't hardcoded,
> just the default is set to gcc. You can set it just like for hbmk
> script HB_COMPILER=gpp|owatcom. The new command
For hbmk script I do not have to set anything because it's b
A sample?
I simply query http://ip-of-pc-with-uhttpd/index.html from a different pc,
on the pc with the uhttpd server memory keeps going down.
I've rebuilt without HB_FM_STATISTICS_OFF but when uhttpd stops it does not
show leaked memory blocks.
Maurilio.
Massimo Belgrano wrote:
> if you po
Revision: 10349
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10349&view=rev
Author: vszakats
Date: 2009-02-19 15:44:59 + (Thu, 19 Feb 2009)
Log Message:
---
2009-02-19 16:44 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
* utils/hbmk/hbmk2.prg
if you post a sample i will try on windows (uhttpd with inet)
2009/2/19 Maurilio Longo :
> Hi,
>
> I've built uhttpd on OS/2 using inet, it works ok, but every new connection
> uses 4 Kb of memory, more or less, which is not freed until after the process
> ends.
>
> Question: did anybody make some
Horodyski Marek (PZUZ) pisze:
> On Windows and Linx works DosNawigator. For my better than MC (edito like in
> ME).
Cool, do you know, why keyboard does not work under linux ?
Adam
--
ABIX - Linuksowe Systemy Wspomagania Biznesu www.abix.info.pl
Gadu-Gadu: 302315
Skype: abix_adamj
Wsparcie: htt
Hi,
I've built uhttpd on OS/2 using inet, it works ok, but every new connection
uses 4 Kb of memory, more or less, which is not freed until after the process
ends.
Question: did anybody make some test with uhttpd build with inet on windows?
did he experienced such 'memory leaks'?
I'm not sure it
I've done a little test
speedtst --thread=50
takes 93,98 secs with current code and 93,16 with previous, so current code is
as efficient as previous :)
Maurilio.
Maurilio Longo wrote:
> Ok, now it works, all mttest?? do work ok and even speedtst --thread=40
> reaches the end.
>
> Thank you s
On Thu, 19 Feb 2009, Maurilio Longo wrote:
Hi Maurilio,
> Ok, now it works, all mttest?? do work ok and even speedtst --thread=40
> reaches the end.
> Thank you so much!
Thank you for confirmation. Now it should work keeping the exact pthreads
semantic but there is one important note for develop
This is from yesterday. You'll have wait for the next nightly.
Brgds,
Viktor
2009/2/19 Horodyski Marek (PZUZ)
>
>
>From: Viktor Szakáts [mailto:harbour...@syenar.hu]
> Sent: Thursday, February 19, 2009 3:25 PM
> To: Harbour Project Main De
>
> Sorry I still don't see any real problem.
> The number of the files says nothing about the quality of a solution.
> 1 big file, many small files? It depends.
Scattering logic for one feature to several independent
places can work, it's just error prone and much more
difficult to prove (test)
From: Viktor Szakáts [mailto:harbour...@syenar.hu]
Sent: Thursday, February 19, 2009 3:25 PM
To: Harbour Project Main Developer List.
Subject: Re: [Harbour] Log from build.
Hi Marek,
Is possible made an harbour-updater written in harbour with tips for make
more simple operation?
2009/2/19 Lorenzo Fiorini :
> On Thu, Feb 19, 2009 at 2:57 PM, Viktor Szakáts wrote:
>
>> I hope you're not serious telling this to me ;)
>
> What I can say is that today I can do:
>
> svn co http://
On Thu, Feb 19, 2009 at 2:57 PM, Viktor Szakáts wrote:
> I hope you're not serious telling this to me ;)
As I told you, english can be "dangerous", but unfortunately my german
and my spanish are even worse.
> Certain logic is spanning across several different files,
> which makes maintenance di
Ok, now it works, all mttest?? do work ok and even speedtst --thread=40
reaches the end.
Thank you so much!
Maurilio.
dru...@users.sourceforge.net wrote:
> Revision: 10348
>
> http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10348&view=rev
> Author: druzus
> Date:
Revision: 10348
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10348&view=rev
Author: druzus
Date: 2009-02-19 14:31:59 + (Thu, 19 Feb 2009)
Log Message:
---
2009-02-19 15:37 UTC+0100 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/source/vm/thr
Revision: 10347
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10347&view=rev
Author: vszakats
Date: 2009-02-19 14:28:38 + (Thu, 19 Feb 2009)
Log Message:
---
got the date wrong again. sorry.
Modified Paths:
--
trunk/harbour/ChangeLog
Hi Marek,
As for hbmk2, pls try again after:
2009-02-18 13:14 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
Brgds,
Viktor
On Thu, Feb 19, 2009 at 3:11 PM, Horodyski Marek (PZUZ) <
m.horody...@pzuzycie.com.pl> wrote:
> >-Original Message-
> >From: Massimo Belgrano [mailto:mbelgr...@delta
Thanks for the tip Petr, I'll lay the infrastructure for this.
But before we add anything like this, I'd like to hear about
misdetections or actual problems with current method.
Brgds,
Viktor
On Thu, Feb 19, 2009 at 2:57 PM, Petr Chornyj wrote:
>
>
>
> vszakats wrote:
> >
> > Revision: 10344
>
Revision: 10346
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10346&view=rev
Author: vszakats
Date: 2009-02-19 14:22:58 + (Thu, 19 Feb 2009)
Log Message:
---
2009-02-18 15:18 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
* utils/hbmk/hbmk2.prg
Przemyslaw,
I'm sorry, it still does not work.
mttest04 never ends with a single thread running
1 200 blocked
2 200 blocked
3 200 blocked
4 200 blocked
5 200 running
6 200 blocked
I did not run the other tests.
Maurilio.
Przemyslaw Czerpak wrote:
>
> I've got it. Yet another stup
>-Original Message-
>From: Massimo Belgrano [mailto:mbelgr...@deltain.it]
>Sent: Thursday, February 19, 2009 11:37 AM
>To: Harbour Project Main Developer List.
>Subject: Re: [Harbour] Log from build.
>
>Hi Marek
>please use the -trace option and post your outputs.
Here is todays log :
>
> > I beg you pardon?
>
> *nix is used to describe Unix derivatives like Solaris, AIX, HPUX,
> BSD, Linux, ...
I hope you're not serious telling this to me ;)
> > To me it seems it's all about your distrust towards my work
>
> Please don't read too much inside my poor english :)
>
> > so I in
vszakats wrote:
>
> Revision: 10344
>
> http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10344&view=rev
> Author: vszakats
> Date: 2009-02-19 12:46:19 + (Thu, 19 Feb 2009)
>
> Log Message:
> ---
> 2009-02-18 13:14 UTC+0100 Viktor Szakats (harbour.01 s
On Thu, Feb 19, 2009 at 2:18 PM, Viktor Szakáts wrote:
> I beg you pardon?
*nix is used to describe Unix derivatives like Solaris, AIX, HPUX,
BSD, Linux, ...
> To me it seems it's all about your distrust towards my work
Please don't read too much inside my poor english :)
> so I invite you to
>
> > [ On Linux I can test, so it's not blind modification nor
> > something I've haven't done before, in case this wasn't
> > obvious from my activity here and the binary builds I've
> > so far produced for *nix systems (for Lorenzo). ]
>
> *nix is not only Linux.
I beg you pardon?
> For exam
Revision: 10345
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10345&view=rev
Author: druzus
Date: 2009-02-19 13:13:36 + (Thu, 19 Feb 2009)
Log Message:
---
2009-02-19 14:18 UTC+0100 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/source/vm/thr
On Thu, 19 Feb 2009, Maurilio Longo wrote:
Hi Maurilio,
> With your latest commit (thread.c at revision 10343):
> mttest04 never ends, two threads keep running, the others are all blocked.
> mttest05 blocks after a little while, 4 threads blocked
> mttest06 the same, 5 threads blocked
> mttest07
Revision: 10344
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10344&view=rev
Author: vszakats
Date: 2009-02-19 12:46:19 + (Thu, 19 Feb 2009)
Log Message:
---
2009-02-18 13:14 UTC+0100 Viktor Szakats (harbour.01 syenar hu)
* utils/hbmk/hbmk2.prg
Przemyslaw Czerpak wrote:
>
> So for two threads it works OK.
>
it seems so, speedtst even works when only two threads are used.
>> mttest04 blocks with 5 blocked threads.
>> mttest05 blocks with 4 blocked threads
>> same for mttest06
>> mttest07 blocks with five blocked threads
> [...]
>> Waiti
On Thu, 19 Feb 2009, Maurilio Longo wrote:
> Przemyslaw,
> mttest01, 02, 03 work ok.
> mttest08 and 09 run to completion
> mttest11 runs ok with a running clock
> mttest12 runs ok
So for two threads it works OK.
> mttest04 blocks with 5 blocked threads.
> mttest05 blocks with 4 blocked threads
>
I am providing.
2009/2/18 Viktor Szakáts :
> Hi Renato,
> Looks good!
> One more thing, not design related, but it'd be nice to have a news item:
> "Harbour v1.0.1 is released!" on 2008-09-18.
>
> Brgds,
> Viktor
> On Wed, Feb 18, 2009 at 2:05 PM, Vailton Renato wrote:
>>
>> I have posted the cha
On Thu, Feb 19, 2009 at 10:55 AM, Viktor Szakáts wrote:
> [ On Linux I can test, so it's not blind modification nor
> something I've haven't done before, in case this wasn't
> obvious from my activity here and the binary builds I've
> so far produced for *nix systems (for Lorenzo). ]
*nix is not
Done.
___
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
Przemyslaw,
mttest01, 02, 03 work ok.
mttest04 blocks with 5 blocked threads.
(E:\REPOSITORY\HARBOUR\tests\mt)mttest04
Harbour 1.1.0dev (Rev. 10342)
Main start
Starting threads: <1><2><3><4><5>
Wait 5 seconds or hit any key...
Waiting for threads...
mttest05 blocks with 4 blocked threads
(E:
Revision: 10343
http://harbour-project.svn.sourceforge.net/harbour-project/?rev=10343&view=rev
Author: druzus
Date: 2009-02-19 11:44:36 + (Thu, 19 Feb 2009)
Log Message:
---
2009-02-19 12:49 UTC+0100 Przemyslaw Czerpak (druzus/at/priv.onet.pl)
* harbour/source/vm/thr
On Thu, 19 Feb 2009, Maurilio Longo wrote:
Hi Maurilio,
> it starts misbehaving with three threads, look at this:
> 1 200 blocked fd504d64
> 2 200 blocked fd5080c4
> 3 200 running fd507008
> 4 200 ready fd507008
> threads 3 and 4 are running and run to completion, thread 2 is blocked si
On Thu, 19 Feb 2009, Maurilio Longo wrote:
Hi Maurilio,
> it does not work, I did a full rebuild then I did build tests/speedtst with
> hbmk2 -mt speedtst then I did a few tests.
> 1) speedtst --thread=2 runs ok
> 2) speedtst --thread=20 deadlock after some time
> 3) speedtst --thread=40 deadlock
Przemyslaw,
it starts misbehaving with three threads, look at this:
1 200 blocked fd504d64
2 200 blocked fd5080c4
3 200 running fd507008
4 200 ready fd507008
threads 3 and 4 are running and run to completion, thread 2 is blocked since
it started, thread 1 is blocked waiting for the ot
Hi Przemek,
(answers part 2)
> > > 8. When -l gcc option is used then should
> > > not contain prefix (lib) and extension (.so,.dll,.a)
> >
> > hbmk does pass the stripped lib name to gcc -l option
> > already. The extension is already stripped (all of them,
> > maybe that's too harsh?). So i
Przemyslaw,
it does not work, I did a full rebuild then I did build tests/speedtst with
hbmk2 -mt speedtst then I did a few tests.
1) speedtst --thread=2 runs ok
2) speedtst --thread=20 deadlock after some time
3) speedtst --thread=40 deadlocks as it starts, it creates 40 threads but all
threads
Hi Marek
please use the -trace option and post your outputs.
2009/2/19 Horodyski Marek (PZUZ) :
>
>
>From: Viktor Szakáts [mailto:harbour...@syenar.hu]
>Sent: Wednesday, February 18, 2009 4:37 PM
>To: Harbour Project Main Developer List.
>
From: Viktor Szakáts [mailto:harbour...@syenar.hu]
Sent: Wednesday, February 18, 2009 4:37 PM
To: Harbour Project Main Developer List.
Subject: Re: [Harbour] Log from build.
Thanks Marek.
These are
Hi Przemek,
> Now in Linux is 100% predictable - GCC is hardcoded as the only
> one choice :-( and it's necessary to use HB_COMPILER envvar.
> BTW -cc= option will be also usable feature.
I know, but this isn't a Linux tool :) For Linux it isn't hardcoded,
just the default is set to gcc. You c
Hi Przemek,
Okay, let's leave them there, that's why I was asking.
However I'd like to make some changes here to move
their solution closer to the standard we use in Harbour.
Particularly the inclusion logic and some misc GT
settings, and specifically to move GTSLN/GTCRS
specific logic to their re
Hi Pritpal,
> >1) You're introducing xhb dependency for gtwvg, which isn't good.
>
> As stated in previous post, I have deffered this decision.
> So until we reach an agreement WAPI_*() functions category
> will contain only those functions not needing structures.
That's fine good, but now you
Hello
<<<
Thanks, I think you should look at it, the other thing is GC
collected pointers which is essential for a lot of tasks here.
I'd suggest to try it, it's not that bad believe me.
>>>
I tried it after posting above message. But I get only
pointers at PRG level which I cannot exploit the
Hi Pritpal,
> I tried but am unable to use it in a generic way. May be
> I am not familiar with this concept since begining.
> Francesco was trying to make it generic, so I am waiting
> till he comes up with this solution. I, at my level best, can do
> with which I am familiar.
Thanks, I think
>-Original Message-
>From: ABIX Adam Jurkiewicz [mailto:a.jurkiew...@abix.info.pl]
>Sent: Wednesday, February 18, 2009 6:38 PM
>To: Harbour Project Main Developer List.
>Subject: Re: [Harbour] Linux development environment
>
>Viktor Szakáts pisze:
>> Hi all,
>>
>> Has anyone some recommen
I invite everybody give a reply to this poll
What is your favourite 32-bit Clipper-compatible Compiler ?
http://cch4clipper.blogspot.com
>I am conducting a poll as above at http://cch4clipper.blogspot.com.
>This is to enable me to determine the future contents of Clipper...
>Clipper... Clipper.
77 matches
Mail list logo
