From kevinfishburne at ...590...  Fri Apr  1 05:12:18 2011
From: kevinfishburne at ...590... (Kevin Fishburne)
Date: Thu, 31 Mar 2011 23:12:18 -0400
Subject: [Gambas-devel] gb3 build 3712: crash while typing in IDE,
	project attached
Message-ID: <4D954292.6090007@...590...>

This application has raised an unexpected
error and must abort.

[6] Type mismatch: wanted FForm, got FTextEditor instead.
CComponent.GetClassSymbols.1165


Try renaming "Public Sub Form_Open()" to "Public Sub Main()". I'd just 
switched it to use the SDL component to test some structure capabilities 
when it went nuclear on me.

-- 
Kevin Fishburne
Eight Virtues
www: http://sales.eightvirtues.com
e-mail: sales at ...590...
phone: (770) 853-6271

-------------- next part --------------
A non-text attachment was scrubbed...
Name: gb3_test.tar.bz2
Type: application/x-bzip
Size: 5570 bytes
Desc: not available
URL: <http://lists.gambas-basic.org/pipermail/devel/attachments/20110331/59b17501/attachment.bin>

From david.villalobos.c at ...176...  Tue Apr  5 15:35:58 2011
From: david.villalobos.c at ...176... (David Villalobos Cambronero)
Date: Tue, 5 Apr 2011 07:35:58 -0600
Subject: [Gambas-devel] Release of Gambas 3
Message-ID: <BANLkTimv9oPo+7=75yGDBhub78n=oBuheg@...178...>

Hi Benoit, and all.

Some of us are trying to help the Gambas project with some code lines and
translations, I've read that Gambas 3 is getting closer, so I want to know
if you have some comments for the projects that we have or you have any
ideas about what our projects should be able to do before Gambas 3 is
released. I want my project to be part of Gambas 3, but I don't know if it
fits the Gambas 3 requirements that's why I'm asking.

It would be nice for translators to know when Gambas 3 will be released so
we can have all translations up to date.

Best regards
---
David
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gambas-basic.org/pipermail/devel/attachments/20110405/2b706baa/attachment.html>

From gambas at ...1...  Tue Apr  5 17:45:00 2011
From: gambas at ...1... (=?utf-8?q?Beno=C3=AEt_Minisini?=)
Date: Tue, 5 Apr 2011 17:45:00 +0200
Subject: [Gambas-devel] Release of Gambas 3
In-Reply-To: <BANLkTimv9oPo+7=75yGDBhub78n=oBuheg@...178...>
References: <BANLkTimv9oPo+7=75yGDBhub78n=oBuheg@...178...>
Message-ID: <201104051745.00257.gambas@...1...>

> Hi Benoit, and all.
> 
> Some of us are trying to help the Gambas project with some code lines and
> translations, I've read that Gambas 3 is getting closer, so I want to know
> if you have some comments for the projects that we have or you have any
> ideas about what our projects should be able to do before Gambas 3 is
> released. I want my project to be part of Gambas 3, but I don't know if it
> fits the Gambas 3 requirements that's why I'm asking.
> 
> It would be nice for translators to know when Gambas 3 will be released so
> we can have all translations up to date.
> 
> Best regards
> ---
> David

What project?

-- 
Beno?t Minisini



From david.villalobos.c at ...176...  Tue Apr  5 19:09:30 2011
From: david.villalobos.c at ...176... (David Villalobos Cambronero)
Date: Tue, 5 Apr 2011 11:09:30 -0600
Subject: [Gambas-devel] Release of Gambas 3
In-Reply-To: <201104051745.00257.gambas@...1...>
References: <BANLkTimv9oPo+7=75yGDBhub78n=oBuheg@...178...>
	<201104051745.00257.gambas@...1...>
Message-ID: <BANLkTi=5xP=M1yCxSS_ujG2nZYAVOu9ojQ@...178...>

GB.MYSQL

Regards
---
David


2011/4/5 Beno?t Minisini <gambas at ...1...>

> > Hi Benoit, and all.
> >
> > Some of us are trying to help the Gambas project with some code lines and
> > translations, I've read that Gambas 3 is getting closer, so I want to
> know
> > if you have some comments for the projects that we have or you have any
> > ideas about what our projects should be able to do before Gambas 3 is
> > released. I want my project to be part of Gambas 3, but I don't know if
> it
> > fits the Gambas 3 requirements that's why I'm asking.
> >
> > It would be nice for translators to know when Gambas 3 will be released
> so
> > we can have all translations up to date.
> >
> > Best regards
> > ---
> > David
>
> What project?
>
> --
> Beno?t Minisini
>
>
> ------------------------------------------------------------------------------
> Xperia(TM) PLAY
> It's a major breakthrough. An authentic gaming
> smartphone on the nation's most reliable network.
> And it wants your games.
> http://p.sf.net/sfu/verizon-sfdev
> _______________________________________________
> Gambas-devel mailing list
> Gambas-devel at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/gambas-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gambas-basic.org/pipermail/devel/attachments/20110405/11b543f7/attachment.html>

From gambas at ...1...  Tue Apr  5 19:15:26 2011
From: gambas at ...1... (=?utf-8?q?Beno=C3=AEt_Minisini?=)
Date: Tue, 5 Apr 2011 19:15:26 +0200
Subject: [Gambas-devel] Release of Gambas 3
In-Reply-To: <BANLkTi=5xP=M1yCxSS_ujG2nZYAVOu9ojQ@...178...>
References: <BANLkTimv9oPo+7=75yGDBhub78n=oBuheg@...178...>
	<201104051745.00257.gambas@...1...>
	<BANLkTi=5xP=M1yCxSS_ujG2nZYAVOu9ojQ@...178...>
Message-ID: <201104051915.26300.gambas@...1...>

> GB.MYSQL
> 
> Regards
> ---
> David
> 

The sources of gb.mysql are located inside the gambas 3 subversion repository, 
so gb.mysql is automatically part of Gambas 3.

I thought it was evident. :-)

-- 
Beno?t Minisini



From david.villalobos.c at ...176...  Tue Apr  5 19:28:35 2011
From: david.villalobos.c at ...176... (David Villalobos Cambronero)
Date: Tue, 5 Apr 2011 11:28:35 -0600
Subject: [Gambas-devel] Release of Gambas 3
In-Reply-To: <201104051915.26300.gambas@...1...>
References: <BANLkTimv9oPo+7=75yGDBhub78n=oBuheg@...178...>
	<201104051745.00257.gambas@...1...>
	<BANLkTi=5xP=M1yCxSS_ujG2nZYAVOu9ojQ@...178...>
	<201104051915.26300.gambas@...1...>
Message-ID: <BANLkTin=L2BwLhEMUivg=Cnxe_nEptGKxA@...178...>

Ok, nice.

I'll check the code for errors and work in the documentation.

Regards
---
David


2011/4/5 Beno?t Minisini <gambas at ...1...>

> > GB.MYSQL
> >
> > Regards
> > ---
> > David
> >
>
> The sources of gb.mysql are located inside the gambas 3 subversion
> repository,
> so gb.mysql is automatically part of Gambas 3.
>
> I thought it was evident. :-)
>
> --
> Beno?t Minisini
>
>
> ------------------------------------------------------------------------------
> Xperia(TM) PLAY
> It's a major breakthrough. An authentic gaming
> smartphone on the nation's most reliable network.
> And it wants your games.
> http://p.sf.net/sfu/verizon-sfdev
> _______________________________________________
> Gambas-devel mailing list
> Gambas-devel at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/gambas-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gambas-basic.org/pipermail/devel/attachments/20110405/1db53de1/attachment.html>

From gambas at ...1...  Wed Apr 13 00:34:11 2011
From: gambas at ...1... (=?utf-8?q?Beno=C3=AEt_Minisini?=)
Date: Wed, 13 Apr 2011 00:34:11 +0200
Subject: [Gambas-devel] gb3 build 3712: crash while typing in IDE,
	project attached
In-Reply-To: <4D954292.6090007@...590...>
References: <4D954292.6090007@...590...>
Message-ID: <201104130034.11523.gambas@...1...>

> This application has raised an unexpected
> error and must abort.
> 
> [6] Type mismatch: wanted FForm, got FTextEditor instead.
> CComponent.GetClassSymbols.1165
> 
> 
> Try renaming "Public Sub Form_Open()" to "Public Sub Main()". I'd just
> switched it to use the SDL component to test some structure capabilities
> when it went nuclear on me.

Fixed in revision #3758.

Regards,

-- 
Beno?t Minisini



From david.villalobos.c at ...176...  Mon Apr 18 04:15:40 2011
From: david.villalobos.c at ...176... (David Villalobos Cambronero)
Date: Sun, 17 Apr 2011 20:15:40 -0600
Subject: [Gambas-devel] Error using Regexp
Message-ID: <BANLkTimUJC_wUXYW_wtOzyPzaC+aqe02Nw@...178...>

Hi,

In my project I use Regexp, but now I get a error. I try to find it but I
can't.

I got the "Out of bounds" error. See the attached example.

Gracias
---
David
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.gambas-basic.org/pipermail/devel/attachments/20110417/4a3cd2f8/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Error.Regexp-0.0.1.tar.gz
Type: application/x-gzip
Size: 4768 bytes
Desc: not available
URL: <http://lists.gambas-basic.org/pipermail/devel/attachments/20110417/4a3cd2f8/attachment.bin>

From kevinfishburne at ...590...  Mon Apr 18 09:13:16 2011
From: kevinfishburne at ...590... (Kevin Fishburne)
Date: Mon, 18 Apr 2011 03:13:16 -0400
Subject: [Gambas-devel] Wait statement changes in recent revision
Message-ID: <4DABE48C.20605@...590...>

Alright Beno?t, what changed in the recent revision? ;)

I noticed before that supposedly recursive Wait statements were 
not-so-recursive, and that I had to call them multiple times to allow 
various events to trigger. If I had two events that had queued, for 
example, I'd have to call Wait twice per frame for each to execute. Now 
calling Wait several times seems to trigger the first event over and 
over without triggering the second. I even had a stack overflow early 
on, which I hadn't seen since my earliest days of programming. I can 
adapt to the new way of things, but I'm not sure exactly what is 
happening now. All I could think of was a GOSUB without a RETURN, haha.

Specifically the Draw event of the SDL screen executes each time Wait is 
called, without regard to queued device events from the gamepad. If the 
SDL screen Draw event takes 0.125 seconds to execute, but the gamepad 
Read event occurs 30 times per second, is it possible to call Wait 
between frames to drop the excessive gamepad input? This worked before 
but now it just queues the gamepad input for eternity, executing only 
the SDL screen refresh.

I'm thinking I need to add code to the gamepad processing procedure to 
ignore input if the rendering frame hasn't been updated since the 
previous gamepad input. Good idea, or am I missing something?

-- 
Kevin Fishburne
Eight Virtues
www: http://sales.eightvirtues.com
e-mail: sales at ...590...
phone: (770) 853-6271




From lordheavym at ...176...  Mon Apr 18 09:54:11 2011
From: lordheavym at ...176... (Laurent Carlier)
Date: Mon, 18 Apr 2011 09:54:11 +0200
Subject: [Gambas-devel] Wait statement changes in recent revision
In-Reply-To: <4DABE48C.20605@...590...>
References: <4DABE48C.20605@...590...>
Message-ID: <201104180954.11763.lordheavym@...176...>

Le lundi 18 avril 2011 09:13:16, Kevin Fishburne a ?crit :
> Alright Beno?t, what changed in the recent revision? ;)
> 
> I noticed before that supposedly recursive Wait statements were
> not-so-recursive, and that I had to call them multiple times to allow
> various events to trigger. If I had two events that had queued, for
> example, I'd have to call Wait twice per frame for each to execute. Now
> calling Wait several times seems to trigger the first event over and
> over without triggering the second. I even had a stack overflow early
> on, which I hadn't seen since my earliest days of programming. I can
> adapt to the new way of things, but I'm not sure exactly what is
> happening now. All I could think of was a GOSUB without a RETURN, haha.
> 
> Specifically the Draw event of the SDL screen executes each time Wait is
> called, without regard to queued device events from the gamepad. If the
> SDL screen Draw event takes 0.125 seconds to execute, but the gamepad
> Read event occurs 30 times per second, is it possible to call Wait
> between frames to drop the excessive gamepad input? This worked before
> but now it just queues the gamepad input for eternity, executing only
> the SDL screen refresh.
> 
> I'm thinking I need to add code to the gamepad processing procedure to
> ignore input if the rendering frame hasn't been updated since the
> previous gamepad input. Good idea, or am I missing something?

The behaviour of Wait has changed, in fact, it's fixed now to work properly, as 
specified in the wiki doc see http://gambasdoc.org/help/lang/wait?v3

Regards,



From gambas at ...1...  Mon Apr 18 10:31:43 2011
From: gambas at ...1... (=?utf-8?q?Beno=C3=AEt_Minisini?=)
Date: Mon, 18 Apr 2011 10:31:43 +0200
Subject: [Gambas-devel] Wait statement changes in recent revision
In-Reply-To: <4DABE48C.20605@...590...>
References: <4DABE48C.20605@...590...>
Message-ID: <201104181031.43406.gambas@...1...>

> Alright Beno?t, what changed in the recent revision? ;)
> 
> I noticed before that supposedly recursive Wait statements were
> not-so-recursive, and that I had to call them multiple times to allow
> various events to trigger. If I had two events that had queued, for
> example, I'd have to call Wait twice per frame for each to execute. Now
> calling Wait several times seems to trigger the first event over and
> over without triggering the second. I even had a stack overflow early
> on, which I hadn't seen since my earliest days of programming. I can
> adapt to the new way of things, but I'm not sure exactly what is
> happening now. All I could think of was a GOSUB without a RETURN, haha.
> 
> Specifically the Draw event of the SDL screen executes each time Wait is
> called, without regard to queued device events from the gamepad. If the
> SDL screen Draw event takes 0.125 seconds to execute, but the gamepad
> Read event occurs 30 times per second, is it possible to call Wait
> between frames to drop the excessive gamepad input? This worked before
> but now it just queues the gamepad input for eternity, executing only
> the SDL screen refresh.
> 
> I'm thinking I need to add code to the gamepad processing procedure to
> ignore input if the rendering frame hasn't been updated since the
> previous gamepad input. Good idea, or am I missing something?

Now WAIT should behave as specified in the documentation :

1) With no argument, it will process only (one?) refresh event. Input events 
(mouse, keyboard, joystick) are queued.

2) With a time argument, it will process all events during the specified time.

Normally, joystick events are queud by SDL. Why do you need to ignore them 
from time to time?

-- 
Beno?t Minisini



From kevinfishburne at ...590...  Wed Apr 20 21:51:42 2011
From: kevinfishburne at ...590... (Kevin Fishburne)
Date: Wed, 20 Apr 2011 15:51:42 -0400
Subject: [Gambas-devel] Wait statement changes in recent revision
In-Reply-To: <201104181031.43406.gambas@...1...>
References: <4DABE48C.20605@...590...>
	<201104181031.43406.gambas@...1...>
Message-ID: <4DAF394E.3070001@...590...>

On 04/18/2011 04:31 AM, Beno?t Minisini wrote:
>> Alright Beno?t, what changed in the recent revision? ;)
>>
>> I noticed before that supposedly recursive Wait statements were
>> not-so-recursive, and that I had to call them multiple times to allow
>> various events to trigger. If I had two events that had queued, for
>> example, I'd have to call Wait twice per frame for each to execute. Now
>> calling Wait several times seems to trigger the first event over and
>> over without triggering the second. I even had a stack overflow early
>> on, which I hadn't seen since my earliest days of programming. I can
>> adapt to the new way of things, but I'm not sure exactly what is
>> happening now. All I could think of was a GOSUB without a RETURN, haha.
>>
>> ...
>
> Now WAIT should behave as specified in the documentation :
>
> 1) With no argument, it will process only (one?) refresh event. Input events
> (mouse, keyboard, joystick) are queued.
>
> 2) With a time argument, it will process all events during the specified time.
>
> Normally, joystick events are queud by SDL. Why do you need to ignore them
> from time to time?

Thanks for the clarification Laurent and Beno?t. What's funny is I had 
thought there was a problem with Wait from the beginning, but it had 
been "wrong" for so long I thought it was "right". I've adjusted the 
code so that it works properly now.

I still need to try out the new SDL joystick support, as I'm still using 
the old /dev/input/js0 device (which also queues input). The reason I 
need to flush or ignore some gamepad input is because the controls are 
dual-analog (Playstation-style). The analog sticks create a rapid flood 
of events when they're moved, which by far exceeds the frame rate of the 
game. Since these stick movement events are queued, any movement of an 
analog stick results in those events being spread out over dozens of 
frames, taking several seconds to finish. If you move, the player just 
keeps on moving. If you rotate, the player just keeps on rotating.

I'll figure it out...no worries. :)

-- 
Kevin Fishburne
Eight Virtues
www: http://sales.eightvirtues.com
e-mail: sales at ...590...
phone: (770) 853-6271




From kevinfishburne at ...590...  Thu Apr 21 06:04:50 2011
From: kevinfishburne at ...590... (Kevin Fishburne)
Date: Thu, 21 Apr 2011 00:04:50 -0400
Subject: [Gambas-devel] Wait statement changes in recent revision
In-Reply-To: <4DAF394E.3070001@...590...>
References: <4DABE48C.20605@...590...>	<201104181031.43406.gambas@...1...>
	<4DAF394E.3070001@...590...>
Message-ID: <4DAFACE2.9040904@...590...>

On 04/20/2011 03:51 PM, Kevin Fishburne wrote:
> On 04/18/2011 04:31 AM, Beno?t Minisini wrote:
>>> Alright Beno?t, what changed in the recent revision? ;)
>>>
>>> I noticed before that supposedly recursive Wait statements were
>>> not-so-recursive, and that I had to call them multiple times to allow
>>> various events to trigger. If I had two events that had queued, for
>>> example, I'd have to call Wait twice per frame for each to execute. Now
>>> calling Wait several times seems to trigger the first event over and
>>> over without triggering the second. I even had a stack overflow early
>>> on, which I hadn't seen since my earliest days of programming. I can
>>> adapt to the new way of things, but I'm not sure exactly what is
>>> happening now. All I could think of was a GOSUB without a RETURN, haha.
>>>
>>> ...
>>
>> Now WAIT should behave as specified in the documentation :
>>
>> 1) With no argument, it will process only (one?) refresh event. Input events
>> (mouse, keyboard, joystick) are queued.
>>
>> 2) With a time argument, it will process all events during the specified time.
>>
>> Normally, joystick events are queud by SDL. Why do you need to ignore them
>> from time to time?
>
> Thanks for the clarification Laurent and Beno?t. What's funny is I had
> thought there was a problem with Wait from the beginning, but it had
> been "wrong" for so long I thought it was "right". I've adjusted the
> code so that it works properly now.
>
> I still need to try out the new SDL joystick support, as I'm still using
> the old /dev/input/js0 device (which also queues input). The reason I
> need to flush or ignore some gamepad input is because the controls are
> dual-analog (Playstation-style). The analog sticks create a rapid flood
> of events when they're moved, which by far exceeds the frame rate of the
> game. Since these stick movement events are queued, any movement of an
> analog stick results in those events being spread out over dozens of
> frames, taking several seconds to finish. If you move, the player just
> keeps on moving. If you rotate, the player just keeps on rotating.
>
> I'll figure it out...no worries. :)

*** NOTE ***
I've solved the problem. I'm sending this anyway in the event it may be 
helpful to others. I was experimenting with code the whole time I was 
writing it. :) The answer is that I encapsulated the entirety of the 
gamepad read event code with "Do Until Eof(pad_device)"..."Loop". Before 
I was just enclosing the "Read new gamepad data" section, but I needed 
to enclose the entire procedure that processed the gamepad data. Here's 
the original request for help:

---

I guess I thought I was smarter than I really was. Damn...

The SDL Draw event (Screen_Draw) triggers at 60 fps, but the rate at 
which the program can actually draw a frame is more around 15 fps. The 
gamepad read event (Gamepad_READ) executes once per Screen_Draw event, 
and queues gamepad events that don't get processed. The gamepad code 
looks like this (abbreviated):

' Gamepad device data.
Public pad_device As Process
Public pad_time As Integer
Public pad_data As Short
Public pad_type As Byte
Public pad_number As Byte

' Set up the gamepad.
pad_device = Exec ["cat", "/dev/input/js0"] For Read As "Gamepad"

' General declarations.
Dim s As String     ' Stores gamepad device input.

' Read new gamepad data.
Read #pad_device, s, 4
pad_time = Integer@(s)
Read #pad_device, s, 2
pad_data = Short@(s)
Read #pad_device, s, 1
pad_type = Byte@(s)
Read #pad_device, s, 1
pad_number = Byte@(s)

I've tried multiple ways of getting the Gamepad_READevent to be 
processed repeatedly without the Screen_Drawevent executing each time, 
but have failed. I can't call Wait, because it will keep calling the 
Screen_Drawevent.

The high frame rate of the gamepad events mixed in with the low frame 
rate of the rendering event cause analog stick movements to be spread 
across several seconds of rendering frames, which of course is bad. Any 
ideas of how to get the gamepad event to keep processing until empty, 
while only allowing the rendering event to execute once?

---

If anyone notices anything odd with the use of the Wait statement, let 
the mailing list know since this is a pretty significant change to 
fundamental operations and should be examined closely. Of course, read 
the documentation first and make sure it's in fact misbehaving:

http://gambasdoc.org/help/lang/wait?v3

-- 
Kevin Fishburne
Eight Virtues
www: http://sales.eightvirtues.com
e-mail: sales at ...590...
phone: (770) 853-6271