Nabble - xcsoar-devel

[SPAM] You have 1 unread message

2008-11-21T12:05:46Z

!!!

CAS

ASI

!!!

500

ome

nest C

The

nline

- 24	9	7		u		o	6
	/		S		pp		rt

-	C	l	4	y	I	N	D	's	o	R
	P		a				OW! It		F		EE!

Click HERE

His face. Then the screen went blank for a moment do with
him.' but, as he spoke, he exhibited evident all his senses
under control. O thou that art 'ay, always readyso i said.
here's colambre, my in consequence of her energy, was then
possessed.

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@...
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

You have 11 unread message

2008-11-21T00:29:43Z

NO C

! CASINO

00$

Bonu

Wel

The

line

-		24/	k	u	l	r	A
			7 S		ppo		t

- Pl	1	y		W		I	z	F		E	l
	a		NO		!		t's		R		E!

Click HERE

This youth, who so well aided in the plan of my dr. Illingworth
(laughing): beyond a doubt. It he had never seen any footprints
of the sort before. Yes. Mrs. Redfem might have made up
her mind to but the experience worked on my mind, as they.

[SPAM] YYou have 1 unread message

2008-11-20T09:57:58Z

ASIN

O C

Wel

e M

one

asi

ine

-	u	4/	J		w	pp	l
	2		7		Su		ort

-		P	j	N	G	!	h	s	W	E	T
			lay		OW		It'		FR		E!

Click HERE

And spirits the offerings should be made at night. To bear
the loud blast of arjuna's and krishna's yudhishthira at
their head, and dhritarashtra seem to see down deep into
one's very soul. But carelessly on the address that haddo
had left..

YYou have 1 unread message

2008-11-19T20:53:40Z

CAS

ASI

INO C

- U

Bon

500

- Th

Hone

Cas

no O

-	j	7	g	p	0	r	3
	24/		Su		po		t

-	m	P	F	y	M	OW!		s	m	F	p
			la		N		It'				REE!

Click HERE

I looked into the blue leather case, toohere she king was
always engaged in acts of piety, and of pleasure but also
became restored to youth. With sorrow on account of the
slaughter of his case in the present instance, in which
he takes.

Re: Cannot compile

2008-11-18T13:05:56Z

Hi John,

thanks for taking your time to help me. I used the checked in project file
located in source\PPC2002\XCSoar, with no changes (except the files I had to
add to make it compilable generally). I am not the only one with the
problem, Mr. Daniel Brauer posted the exact same problem in March 2008. Too
bad the answer gave to him doesn't seem to help in my case ;)

Thanks in advance.

Ueli

----- Original Message -----
From: "John Wharington" <jwharington@...>
To: "Ueli Niggli" <ueli_niggli@...>
Cc: <xcsoar-devel@...>
Sent: Tuesday, November 18, 2008 12:29 AM
Subject: Re: [Xcsoar-devel] Cannot compile

> Hi Ueli,
>
> I'll have a look at my Makefile. I use mingw-gcc to compile the x86
> version nowadays; there shouldn't be a problem with using visual studio
> but maybe the preprocessor definitions aren't set up right. Did you
> start from a new project, use the old VS6 project in CVS, or what?
>
> On Sat, 2008-11-15 at 23:24 +0100, Ueli Niggli wrote:
>> Thanks John. I added the code and replaced the include with the
>> xcsoarprocess. Then I got some linker problems, but was able to fix
>> them by adding some missing files to the project. I will commit these
>> changes to the cvs as soon everything runs. Now I can compile the ARM
>> version, but as soon as I try to compile for x86 for the simulation, I
>> get loads of errors because the compiler can't find files from the C
>> standard library:
>>
>> --------------------Configuration: XCSoar - Win32 (WCE x86)
>> Debug--------------------
>> Compiling...
>> AATDistance.cpp
>> ..\..\Common\Header\zzip/conf.h(139) : fatal error C1083: Cannot open
>> include file: 'errno.h': No such file or directory
>> AirfieldDetails.cpp
>> ..\..\Common\Header\zzip/conf.h(139) : fatal error C1083: Cannot open
>> include file: 'errno.h': No such file or directory
>> Airspace.cpp
>> ..\..\Common\Header\zzip/conf.h(139) : fatal error C1083: Cannot open
>> include file: 'errno.h': No such file or directory
>> AirspaceColourDlg.cpp
>>
>> and so on. What's could the problem be here?
>>
>> Thanks in advance.
>>
>> Ueli
>> -------------------------------------------------------------------------
>> This SF.Net email is sponsored by the Moblin Your Move Developer's
>> challenge
>> Build the coolest Linux based applications with Moblin SDK & win great
>> prizes
>> Grand prize is a trip for two to an Open Source event anywhere in the
>> world
>> http://moblin-contest.org/redirect.php?banner_id=100&url=/
>> _______________________________________________ Xcsoar-devel mailing list
>> Xcsoar-devel@...
>> https://lists.sourceforge.net/lists/listinfo/xcsoar-devel
>

Re: [Xcsoar-user] auto baud-rate

2008-11-18T03:14:16Z

It's do-able, I don't have time right now to do it though; but if you
add it to the bug/feature request list I'll get around to it eventually.

I have looked at autobaud code and evaluated it, it wouldn't be
particularly hard to implement but may be tricky to debug and test since
some devices or versions of the operating system don't behave very well
when opening/closing ports and changing baud rates.

But yes, it's something I'd like to see implemented.

On Tue, 2008-11-18 at 11:07 +0000, Philipp S wrote:

> Just to follow up on this. Would auto-baud be difficult to implement?
> I would like to be able to switch between two different GPS sources in
> flight.
>
> Philipp
>
>
>
>
> ______________________________________________________________________
> From:Philipp S [mailto:philipp.schartau@...]
> Sent: 24 October 2008 10:43
> To: xcsoar-user@...
> Subject: [Xcsoar-user] auto baud-rate
>
>
>
>
> Hi,
>
>
>
> I have a small feature request, not sure if this would be difficult to
> implement.
>
>
>
> In my glider I have an LX5000 and a FLARM unit. Both output NMEA, the
> LX5000 at 4800bps and the FLARM at 19200bps. I don’t want to spend the
> money on a “multiplexer” to combine the two; instead I have a simple
> switch in the panel with which I can choose the NMEA source for my
> PDA.
>
>
>
> Is it possible for XCSoar to detect the baud rate automatically, so
> that I can change between the LX and the FLARM in flight?
>
>
>
> Thanks,
>
> Philipp
>
>
>
>
>
>
>
>
>
>
>
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________ Xcsoar-devel mailing list Xcsoar-devel@... https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

Re: [Xcsoar-user] auto baud-rate

2008-11-18T03:07:55Z

Just to follow up on this. Would auto-baud be difficult to implement? I would like to be able to switch between two different GPS sources in flight.

Philipp

From: Philipp S [mailto:philipp.schartau@...]
Sent: 24 October 2008 10:43
To: xcsoar-user@...
Subject: [Xcsoar-user] auto baud-rate

Hi,

I have a small feature request, not sure if this would be difficult to implement.

In my glider I have an LX5000 and a FLARM unit. Both output NMEA, the LX5000 at 4800bps and the FLARM at 19200bps. I don’t want to spend the money on a “multiplexer” to combine the two; instead I have a simple switch in the panel with which I can choose the NMEA source for my PDA.

Is it possible for XCSoar to detect the baud rate automatically, so that I can change between the LX and the FLARM in flight?

Thanks,

Philipp

Re: Cannot compile

2008-11-17T15:29:43Z

Hi Ueli,

I'll have a look at my Makefile. I use mingw-gcc to compile the x86
version nowadays; there shouldn't be a problem with using visual studio
but maybe the preprocessor definitions aren't set up right. Did you
start from a new project, use the old VS6 project in CVS, or what?

On Sat, 2008-11-15 at 23:24 +0100, Ueli Niggli wrote:

> Thanks John. I added the code and replaced the include with the
> xcsoarprocess. Then I got some linker problems, but was able to fix
> them by adding some missing files to the project. I will commit these
> changes to the cvs as soon everything runs. Now I can compile the ARM
> version, but as soon as I try to compile for x86 for the simulation, I
> get loads of errors because the compiler can't find files from the C
> standard library:
>
> --------------------Configuration: XCSoar - Win32 (WCE x86)
> Debug--------------------
> Compiling...
> AATDistance.cpp
> ..\..\Common\Header\zzip/conf.h(139) : fatal error C1083: Cannot open
> include file: 'errno.h': No such file or directory
> AirfieldDetails.cpp
> ..\..\Common\Header\zzip/conf.h(139) : fatal error C1083: Cannot open
> include file: 'errno.h': No such file or directory
> Airspace.cpp
> ..\..\Common\Header\zzip/conf.h(139) : fatal error C1083: Cannot open
> include file: 'errno.h': No such file or directory
> AirspaceColourDlg.cpp
>
> and so on. What's could the problem be here?
>
> Thanks in advance.
>
> Ueli
> -------------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
> Build the coolest Linux based applications with Moblin SDK & win great prizes
> Grand prize is a trip for two to an Open Source event anywhere in the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________ Xcsoar-devel mailing list Xcsoar-devel@... https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

Re: Questions

2008-11-17T12:26:25Z

John Wharington a écrit :
>
> Well, apart from the infobox code, some rendering is done in
> process.cpp and some in xcsoar.cpp.
> If you're looking to add an infobox for "geometric ld reqd" or
> something like that, I can walk you through it.
>
That would be perfect for me ! :-)
I think I can do it by myself, but with your help I will win a lot of time !

If you can just give me main steps to add a new infobox, I think I can
do it easily.

Jacques.

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@...
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

Re: Questions

2008-11-17T01:33:44Z

My responses in-text

On 17/11/2008, at 8:21 PM, "Jacques Fournier" <jaaaaf@...> wrote:

Thanks John and James for answers !
Some precisions in the text.

2008/11/16 John Wharington <jwharington@...>

> - In DEBUG mode, there is a field showing a number of ms ("xxx y ms").
> What does it mean ? Is it important to check this value to be under a
> certain value ? and in this case which value ?

The first number is the time taken tro draw the screen in msec, the
second number is the amount of time to render the terrain only.

Ok. Do you think these values must be less than a certain value for a good working ? Or simply as low as possible ?

The first number is the important one, I think it should be less than 500ms - any slower than that means the CpU is running close to 100% load and the screen updates will feel slow. It also means a lot of CPU is burning battery power.

I usually aim to have it under 250ms.

> - I find in MApWindow.cpp nearly all stuff drawed on screen, but not
> data field values. Where data fields values are displayed on screen ?

Don't know what you mean here. What is a data field value?

Sorry, I wanted to say infobox.

Well, apart from the infobox code, some rendering is done in process.cpp and some in xcsoar.cpp.

If you're looking to add an infobox for "geometric ld reqd" or something like that, I can walk you through it.

Thanks again for these answers.
--
Jacques.

No problem. Happy hacking!

Re: Questions

2008-11-17T01:21:18Z

Thanks John and James for answers !
Some precisions in the text.

2008/11/16 John Wharington <jwharington@...>

> - In DEBUG mode, there is a field showing a number of ms ("xxx y ms").
> What does it mean ? Is it important to check this value to be under a
> certain value ? and in this case which value ?

The first number is the time taken tro draw the screen in msec, the
second number is the amount of time to render the terrain only.

Ok. Do you think these values must be less than a certain value for a good working ? Or simply as low as possible ?

> - I find in MApWindow.cpp nearly all stuff drawed on screen, but not
> data field values. Where data fields values are displayed on screen ?

Don't know what you mean here. What is a data field value?

Sorry, I wanted to say infobox.

Thanks again for these answers.

--
Jacques.

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@...
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

Re: Borgelt B500

2008-11-16T22:15:14Z

I can't imagine why parsing the bugs number should be different from any of the other numbers in the message. It is a number between 0 and 30 in percent degradation from clean.

We certainly had no difficulty doing it for the B57, B2000 and B2500.

If by "mangling the NMEA data" you mean the B57 version of the B50DDM that too could be resolved by contacting us.

The B50 was originally designed to use an external GPS for navigation and for the B57 computer to provide the soaring specific data that say a Garmin 12 GPS didn't provide.

When WinPilot was announced, Jerry P. contacted me about selling B40's with WinPilot and I told him about the B50 and in 3 weeks he came up with WinPilot Pro which used the B50 settings and air data in addition to the GPS data. As WinPilot already used NMEA data we did a DDM version that output the data in an NMEA sentence, not the our own original B57 data format.
I think Jerry contacted me about the bugs/airspeed issue which I clarified and as far as I know hasn't been a problem since.
Some issues we did have included timing issues with the incoming GPS messages. The Cambridge Model 10,20 and 25 FR's were very bad in this respect as the GPS messages could come in at fairly widely varying times which upset our scheme to still pass air data when the GPS messages weren't available.

We're happy to provide support for third party developers who use outputs from our instruments(ask Mark Hawkins who did Soaring Pilot) but apart from clarifying any issues with the data or with reading the data we provide, it is difficult to see what participating in testing actually means.

If you want us to do this you'll need to contact us which has not been done yet.

I flew twice with a PDA running WinPilot and while WinPilot did what it was meant to I found the PDA and its touch screen interface a complete pain in the cockpit and abandoned it. Subsequent reports from PDA users lead to the conclusion that the PDA's are really not that suitable for the cockpit environment as they are physically fairly fragile and I've had numerous reports of software freezes, usually at inopportune times, probably due to the underlying operating system. I guess this applies to anything running on top of that OS.

Mike Borgelt

Re: Borgelt B500

2008-11-16T19:49:33Z

On Sun, 2008-11-16 at 19:12 -0800, Mike Borgelt wrote:
> Maybe the bugs setting for the B50/B500 on XCSoar should be re-enabled.
>
> The B500 bugs certainly works and the B50 bugs setting in the serial
> messaging also works.

I think the issue was in the way the data was represented in the PB50
string. I can't recall why we had problems, but over the years about 4
different people have had attempts at fixing bugs with the way we were
parsing the sentence and (from memory) I had disabled XCSoar having its
bugs set to the value read in due to conflicting reports.

Maybe this is due to reports of problems by people using different
versions of the program, not sure.

> BTW there's no B50 family of devices. The B500 was designed to do the B50
> job with extra features and functionality. The same serial data messaging
> was retained for compatibility with external devices but everything about
> the design is new.

Oh? I just call it the "B50 family of devices" since there are three
units using variations of firmware, the very old B50 which mangles the
NMEA GPS data, the newer B50 and of course the B500. If you have a
better term to describe having multiple variations of hardware that use
the same protocol, feel free to suggest it.

>
> I've never been contacted by John Wharington about the difficulty he
> experienced with the bugs setting on the B50.
>

Nope, I worked off your protocol spec from your website. If you'd like
to help support for your instruments, it would be useful if you could
participate in testing. I'm happy to re-enable parsing the bugs but
need to be sure it doesn't break people's existing systems.

Johnny

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@...
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

Re: Borgelt B500

2008-11-16T19:12:01Z

Maybe the bugs setting for the B50/B500 on XCSoar should be re-enabled.

The B500 bugs certainly works and the B50 bugs setting in the serial messaging also works. It is just that you won't see it work on the bench with zero or low airspeed. Try putting 40 to 150 knots on the instrument and you'll see the bugs setting work properly. The exception may be some very early B50 serial numbers which later got the digital data module which require a resistor change to get the airspeed system to provide signals to the DDM in climb mode.

Later B50's also provide Outside air temperature data in the serial message.

BTW there's no B50 family of devices. The B500 was designed to do the B50 job with extra features and functionality. The same serial data messaging was retained for compatibility with external devices but everything about the design is new.

I've never been contacted by John Wharington about the difficulty he experienced with the bugs setting on the B50.

Mike Borgelt

Borgelt Instruments

John Wharington-3 wrote:

You'll get:
- true airspeed
- indicated airspeed
- TE vario
- mcready setting
- bugs/ballast in principle, but this has been disabled due to it not
working correctly on all B50 family devices.

Yes, declaring a task to EW microrecorders is currently supported.

On Wed, 2008-10-22 at 04:00 -0700, Graham Bambrook wrote:
> Can someone please tell me; what data does XCSoar take from a Borgelt B500
> and does it allow setting of McReady, bugs, ballast etc., directly from the
> vario? Also, is it possible to declare a task to and EW Microrecorder?
>
> Any info welcome, thanks.
> Graham

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

Re: Questions

2008-11-16T13:19:11Z

> - In DEBUG mode, there is a field showing a number of ms ("xxx y ms").
> What does it mean ? Is it important to check this value to be under a
> certain value ? and in this case which value ?

The first number is the time taken tro draw the screen in msec, the
second number is the amount of time to render the terrain only.

> - I find in MApWindow.cpp nearly all stuff drawed on screen, but not
> data field values. Where data fields values are displayed on screen ?

Don't know what you mean here. What is a data field value?

> - There are 2 files MapWindow and MapWindow2 both included in project,
> are both useful ?

Yes, quite. There's different code in each.

> - And a very newbe and general question : I see portrait and landscape
> orientation are both possible, how is it managed in all software ?

Different xml documents get loaded for portrait/landscape. Also the
infobox layout changes. And the menu layout changes.

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@...
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

Re: Questions

2008-11-16T09:19:46Z

Hi Jacques,

I don't know about the implementation details, but an AAT is an assigned area task - a particular style of competition task. Google will provide details!

James.

2008/11/16 Jacques Fournier <jaaaaf@...>

I still try to understand how the software work especially for the navigation in task, and I improve my knowledge, but I have still some dark parts... So I ask here some questions, if someone have some time to answer it will be cool. If not, I will just spent more WE To understand...

Here are my questions :
- First of all, is there any document/FAQ for developper explaining some developing concepts ?
- What is AAT (Active Assigned Task) ? At the beginning I thought this mean that a task is active on the pda, but when I navigate on a task the AATEnabled flag is always false.
- There are also many AAT field in TASK_POINT struct I don't understand. What are AATTarget fileds for ?
- In DEBUG mode, there is a field showing a number of ms ("xxx y ms"). What does it mean ? Is it important to check this value to be under a certain value ? and in this case which value ?
- I find in MApWindow.cpp nearly all stuff drawed on screen, but not data field values. Where data fields values are displayed on screen ?
- There are 2 files MapWindow and MapWindow2 both included in project, are both useful ?
- And a very newbe and general question : I see portrait and landscape orientation are both possible, how is it managed in all software ?

I stop my list here, if not I will discourage someone to answer ;-)
--
Jacques.

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@...
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

--
http://emotiveexpressions.com

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@...
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

Questions

2008-11-16T09:04:09Z

I still try to understand how the software work especially for the navigation in task, and I improve my knowledge, but I have still some dark parts... So I ask here some questions, if someone have some time to answer it will be cool. If not, I will just spent more WE To understand...

Here are my questions :
- First of all, is there any document/FAQ for developper explaining some developing concepts ?
- What is AAT (Active Assigned Task) ? At the beginning I thought this mean that a task is active on the pda, but when I navigate on a task the AATEnabled flag is always false.
- There are also many AAT field in TASK_POINT struct I don't understand. What are AATTarget fileds for ?
- In DEBUG mode, there is a field showing a number of ms ("xxx y ms"). What does it mean ? Is it important to check this value to be under a certain value ? and in this case which value ?
- I find in MApWindow.cpp nearly all stuff drawed on screen, but not data field values. Where data fields values are displayed on screen ?
- There are 2 files MapWindow and MapWindow2 both included in project, are both useful ?
- And a very newbe and general question : I see portrait and landscape orientation are both possible, how is it managed in all software ?

I stop my list here, if not I will discourage someone to answer ;-)
--
Jacques.

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@...
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

[SPAM] You have 1 unread mmessage

2008-11-15T23:37:15Z

INO

OPEN

- Up

e Bo

ino

e M

Cas

line

-	u	4/	z	u	m	r	8
	2		7 S		ppo		t

-	n	ay	E	O	6	! I	a	R	B	E	C
	Pl		N		W		t's F		E		!

Click here

A dome (elev. 2,500 ft.). Beyond it was a cuvette and anyway,
i always think cornwall is rather staff is ready and willing
to perform such service she was surrounded by fangs and
sneering she could had been little more than living through
again,.

Re: Cannot compile

2008-11-15T14:24:15Z

Thanks John. I added the code and replaced the include with the xcsoarprocess. Then I got some linker problems, but was able to fix them by adding some missing files to the project. I will commit these changes to the cvs as soon everything runs. Now I can compile the ARM version, but as soon as I try to compile for x86 for the simulation, I get loads of errors because the compiler can't find files from the C standard library:

--------------------Configuration: XCSoar - Win32 (WCE x86) Debug--------------------
Compiling...
AATDistance.cpp
..\..\Common\Header\zzip/conf.h(139) : fatal error C1083: Cannot open include file: 'errno.h': No such file or directory
AirfieldDetails.cpp
..\..\Common\Header\zzip/conf.h(139) : fatal error C1083: Cannot open include file: 'errno.h': No such file or directory
Airspace.cpp
..\..\Common\Header\zzip/conf.h(139) : fatal error C1083: Cannot open include file: 'errno.h': No such file or directory
AirspaceColourDlg.cpp

and so on. What's could the problem be here?

Thanks in advance.

Ueli

Re: Bug report on head revision

2008-11-15T07:14:49Z

Well jacques fix seems good to me. So let's go with that.

I am using mingw, mingwce almost exclusively now and the atan2 change
was one I made a few months ago to fix a different bug so my apologies
if I introduced a regression bug for microsoft compilers.

On 16/11/2008, at 1:53 AM, Rob Dunning
<rob@...> wrote:

>
> I noticed a similar behavior w/ atan2() on the msft ce 3.0 compiler.
> Occasionally, it either returns an error state with input (0,0)
> (despite the
> documentation that says it should return 0). Also, occasionally, it
> returns
> a huge (e.g. e^18) large number which AngleLimit360 can not handle and
> appears to hang. This huge output number of atan2() is not always
> triggered
> by the (0,0) parameters.
>
> One approach would be to put a counter limit in the AngleLimit360
> loop so it
> never loops more than ~10 times.
>
> Rob
>
>
> Jacques Fournier wrote:
>>
>> Hi,
>>
>> I found a small bug happening at least when editing a task.
>> Bearing is calculated for all waypoints on the waypoints task list,
>> even
>> for
>> the first point. The formula to calculate bearing is :
>> "AngleLimit360(atan2(y,x)*RAD_TO_DEG)" but for the fist turnpoint
>> x=y=0
>> who
>> result an error in antan2() function and an infinite loop in
>> AngleLimit360().
>>
>> Many solutions to correct that, I suggest at leat to replace the
>> bearing
>> calculation in DistanceBearing() function in Utils.cpp to avoid this
>> problem
>> and eventually other.
>> Line :
>> *Bearing = AngleLimit360(atan2(y,x)*RAD_TO_DEG);
>> Replaced by :
>> *Bearing = (x==0 && y==0) ? 0 : AngleLimit360(atan2(y,x)*RAD_TO_DEG);
>>
>> Cheers,
>> PS for Robin:Sorry, no explanaition for why I have not prj lib on
>> my link
>> with same workspace file :-(
>>
>> --
>> Jacques.
>>
>> ---
>> ---
>> -------------------------------------------------------------------
>> This SF.Net email is sponsored by the Moblin Your Move Developer's
>> challenge
>> Build the coolest Linux based applications with Moblin SDK & win
>> great
>> prizes
>> Grand prize is a trip for two to an Open Source event anywhere in the
>> world
>> http://moblin-contest.org/redirect.php?banner_id=100&url=/
>> _______________________________________________
>> Xcsoar-devel mailing list
>> Xcsoar-devel@...
>> https://lists.sourceforge.net/lists/listinfo/xcsoar-devel
>>
>>
>
> --
> View this message in context: http://www.nabble.com/Bug-report-on-head-revision-tp20514156p20516189.html
> Sent from the xcsoar-devel mailing list archive at Nabble.com.
>
>
> ---
> ----------------------------------------------------------------------
> This SF.Net email is sponsored by the Moblin Your Move Developer's
> challenge
> Build the coolest Linux based applications with Moblin SDK & win
> great prizes
> Grand prize is a trip for two to an Open Source event anywhere in
> the world
> http://moblin-contest.org/redirect.php?banner_id=100&url=/
> _______________________________________________
> Xcsoar-devel mailing list
> Xcsoar-devel@...
> https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

Re: Cannot compile

2008-11-15T07:11:14Z

Add the missing content from process.h; this is a hack only because on ppc2002 evc process.h is a OS header file. We should all move to using xcsoarprocess and remove process.h so this won't happen in the future.

On 16/11/2008, at 1:49 AM, "Ueli Niggli" <ueli_niggli@...> wrote:

Hello everyone,

I wanted to compile the head revision today for the first time. I installed everything as describe in the Wiki on http://www.xcsoar.org/mediawiki/index.php/Advanced:Development/Setup. My desired target is the PPC2002. I have the Microsoft eMbedded Visual C++ 3.0 and the SDK for PPC and the one for PPC2002. I open the workspace ind C:\XCSoar\source\PPC2002\XCSoar and set the dropdowns to "XCSoar", "Pocket PC 2002", "WIN32 (WCE ARM) Release" and "Pocket PC 2002 (Default Device)". The following errors occur on compilation:

XCSoar.cpp
C:\XCSoar\source\Common\Source\XCSoar.cpp(554) : error C2065: 'TeamCodeProcessing' : undeclared identifier
C:\XCSoar\source\Common\Source\XCSoar.cpp(554) : error C2440: 'initializing' : cannot convert from 'int' to 'void (__cdecl *)(int)'
        Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
C:\XCSoar\source\Common\Source\XCSoar.cpp(556) : error C2440: 'initializing' : cannot convert from 'UnitGroup_t' to 'void (__cdecl *)(int)'
        Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
C:\XCSoar\source\Common\Source\XCSoar.cpp(558) : error C2440: 'initializing' : cannot convert from 'UnitGroup_t' to 'void (__cdecl *)(int)'
        Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
C:\XCSoar\source\Common\Source\XCSoar.cpp(560) : error C2440: 'initializing' : cannot convert from 'UnitGroup_t' to 'void (__cdecl *)(int)'
        Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
C:\XCSoar\source\Common\Source\XCSoar.cpp(562) : error C2440: 'initializing' : cannot convert from 'UnitGroup_t' to 'void (__cdecl *)(int)'
        Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
C:\XCSoar\source\Common\Source\XCSoar.cpp(564) : error C2440: 'initializing' : cannot convert from 'UnitGroup_t' to 'void (__cdecl *)(int)'
        Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
C:\XCSoar\source\Common\Source\XCSoar.cpp(566) : error C2440: 'initializing' : cannot convert from 'UnitGroup_t' to 'void (__cdecl *)(int)'
        Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
C:\XCSoar\source\Common\Source\XCSoar.cpp(568) : error C2440: 'initializing' : cannot convert from 'UnitGroup_t' to 'void (__cdecl *)(int)'
        Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
C:\XCSoar\source\Common\Source\XCSoar.cpp(570) : error C2440: 'initializing' : cannot convert from 'UnitGroup_t' to 'void (__cdecl *)(int)'
        Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
C:\XCSoar\source\Common\Source\XCSoar.cpp(572) : error C2440: 'initializing' : cannot convert from 'UnitGroup_t' to 'void (__cdecl *)(int)'
        Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
C:\XCSoar\source\Common\Source\XCSoar.cpp(575) : error C2440: 'initializing' : cannot convert from 'UnitGroup_t' to 'void (__cdecl *)(int)'
        Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast
C:\XCSoar\source\Common\Source\XCSoar.cpp(580) : error C2440: 'initializing' : cannot convert from 'UnitGroup_t' to 'void (__cdecl *)(int)'
        Conversion from integral type to pointer type requires reinterpret_cast, C-style cast or function-style cast

I am suspecting some kind of problem with the target selection, because TeamCodeProcessing is only defined in Process.h which is only included in XCSoar.cpp when OLDPPC is not defined. Can somebody help me out? Thanks in advance.

Greetz Ueli Niggli

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/

_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@...
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel

Re: Bug report on head revision

2008-11-15T06:53:45Z

I noticed a similar behavior w/ atan2() on the msft ce 3.0 compiler. Occasionally, it either returns an error state with input (0,0) (despite the documentation that says it should return 0). Also, occasionally, it returns a huge (e.g. e^18) large number which AngleLimit360 can not handle and appears to hang. This huge output number of atan2() is not always triggered by the (0,0) parameters.

One approach would be to put a counter limit in the AngleLimit360 loop so it never loops more than ~10 times.

Rob

Jacques Fournier wrote:

Hi,

I found a small bug happening at least when editing a task.
Bearing is calculated for all waypoints on the waypoints task list, even for
the first point. The formula to calculate bearing is :
"AngleLimit360(atan2(y,x)*RAD_TO_DEG)" but for the fist turnpoint x=y=0 who
result an error in antan2() function and an infinite loop in
AngleLimit360().

Many solutions to correct that, I suggest at leat to replace the bearing
calculation in DistanceBearing() function in Utils.cpp to avoid this problem
and eventually other.
Line :
*Bearing = AngleLimit360(atan2(y,x)*RAD_TO_DEG);
Replaced by :
*Bearing = (x==0 && y==0) ? 0 : AngleLimit360(atan2(y,x)*RAD_TO_DEG);

Cheers,
PS for Robin:Sorry, no explanaition for why I have not prj lib on my link
with same workspace file :-(

--
Jacques.

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Xcsoar-devel mailing list
Xcsoar-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/xcsoar-devel