Re: [Flightgear-devel] Debugging
Hi Chris, You may have an incorrect line break in a XML file, in a property name. You should write property/my/own/property/property instead of property /my/own/property /property Regards, -Fred - Chris Wilkinson blobster...@yahoo.com.au a écrit : Hi there, I've downloaded and compiled fgfs this morning from git, and am trying to make a modified a/c I've worked on in the past play nice with this git code, but I get this message trying to run the sim with my a/c. Fatal error: ' ' found in propertyname after '' name must begin with alpha or '_' I've tried changing debug level to bulk, debug or warn to see if I can get more info about which of the config files for the a/c causes the segfault, but the debug output does not tell me which file it comes from. Does anyone know of a easy way I can determine this? I'm sure it must be a simple typo somewhere, but there are oodles of text to pore through - where would I start?!? :-) Kind regards, Chris Wilkinson, Brisbane, Australia. -- Virtualization is moving to the mainstream and overtaking non-virtualized environment for deploying applications. Does it make network security easier or more difficult to achieve? Read this whitepaper to separate the two and get a better understanding. http://p.sf.net/sfu/hp-phase2-d2d ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel -- Frédéric Bouvier http://my.fotolia.com/frfoto/ Photo gallery - album photo http://www.youtube.com/user/fgfred64 Videos -- Virtualization is moving to the mainstream and overtaking non-virtualized environment for deploying applications. Does it make network security easier or more difficult to achieve? Read this whitepaper to separate the two and get a better understanding. http://p.sf.net/sfu/hp-phase2-d2d___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] Debugging
Thanks Fred, I will take a look at the files soon, but if you know a way of easily finding incorrect syntax, without needing to open the xml files and reading them thru one by one - you'll be my new best friend if you can tell me... :-) Regards, Chris W., Brisbane, Australia. From: Frederic Bouvier fredfgf...@free.fr To: FlightGear developers discussions flightgear-devel@lists.sourceforge.net Sent: Mon, 4 October, 2010 4:28:21 PM Subject: Re: [Flightgear-devel] Debugging Hi Chris, You may have an incorrect line break in a XML file, in a property name. You should write property/my/own/property/property instead of property /my/own/property /property Regards, -Fred - Chris Wilkinson blobster...@yahoo.com.au a écrit : Hi there, I've downloaded and compiled fgfs this morning from git, and am trying to make a modified a/c I've worked on in the past play nice with this git code, but I get this message trying to run the sim with my a/c. Fatal error: ' ' found in propertyname after '' name must begin with alpha or '_' I've tried changing debug level to bulk, debug or warn to see if I can get more info about which of the config files for the a/c causes the segfault, but the debug output does not tell me which file it comes from. Does anyone know of a easy way I can determine this? I'm sure it must be a simple typo somewhere, but there are oodles of text to pore through - where would I start?!? :-) Kind regards, Chris Wilkinson, Brisbane, Australia. -- Virtualization is moving to the mainstream and overtaking non-virtualized environment for deploying applications. Does it make network security easier or more difficult to achieve? Read this whitepaper to separate the two and get a better understanding. http://p.sf.net/sfu/hp-phase2-d2d ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel -- Frédéric Bouvier http://my.fotolia.com/frfoto/ Photo gallery - album photo http://www.youtube.com/user/fgfred64 Videos -- Virtualization is moving to the mainstream and overtaking non-virtualized environment for deploying applications. Does it make network security easier or more difficult to achieve? Read this whitepaper to separate the two and get a better understanding. http://p.sf.net/sfu/hp-phase2-d2d___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] Debugging
Hi Gene, - Gene Buckle a écrit : On Mon, 4 Oct 2010, Frederic Bouvier wrote: Hi Chris, You may have an incorrect line break in a XML file, in a property name. You should write property/my/own/property/property instead of property /my/own/property /property Fred, aren't both forms valid XML? Is there a way the filename of the offender could be included in the error message? It would be a huge help to folks chasing down these things. Yes, both form as valid as propertygarbage /my/own/property garbage/property is valid XML too. The point is that line breaks are not allowed in property names, so you should avoid to put line breaks or others white space combination between the element markers (between and ). Sure the code could be amended to print the file name. In the mean time, I would start under gdb and put a breakpoint on the instruction that print this message. -Fred -- Frédéric Bouvier http://my.fotolia.com/frfoto/ Photo gallery - album photo http://www.youtube.com/user/fgfred64 Videos -- Virtualization is moving to the mainstream and overtaking non-virtualized environment for deploying applications. Does it make network security easier or more difficult to achieve? Read this whitepaper to separate the two and get a better understanding. http://p.sf.net/sfu/hp-phase2-d2d ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
[Flightgear-devel] FGCom server replacement
A donation of hosting and suitable bandwidth to handle the FGCom server has been agreed and we will be working on this tonight. This is a medium term deal until a permanent solution and the financial resources to cover it are available. That's a debate for later but it needs folk to start thinking about it now. We will be sucking it to see how it works out for bandwidth. If it becomes too heavy and impacts on normal business operations then we will need to look into putting a cap on users and/or tower freqs. I am assuming (always dangerous) that we have never had 50 FGCom users at any one time on 10 freqs. The proposed solution should be able to handle several times those numbers but until we get it up and running and do some stress testing we will not know for certain. We will get the DNS records transferred so everything should Just Work as before. Stand by for announcements of testing and hopefully service restoration over the next couple of days. I would be really grateful if we could get some folks who are willing to help us test this as we fumble our way towards an optimal set-up. Basically we'd like you to log on and keep talking when we announce testing and not be too disappointed when the service suddenly dies for any reason. No music please, if there is any music or similar abuse then the deal will be terminated without notice. We need stress testing but please leave music out. So please keep monitoring this list, the FGCom topic on the multiplayer section of the forum and #flightgear on irc.flightgear.org for announcements of testers required. Thanks :-) -- Best Regards Willie -- Virtualization is moving to the mainstream and overtaking non-virtualized environment for deploying applications. Does it make network security easier or more difficult to achieve? Read this whitepaper to separate the two and get a better understanding. http://p.sf.net/sfu/hp-phase2-d2d ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] recent environment changes break airport elevation with metar altimeter settings
On Tue, Oct 5, 2010 at 1:28 AM, dave perry skida...@mindspring.com wrote: Before I set the altimeter to 30.35 (i.e. as launched at 29.92), the altimeter reads aprox. 643 and it noticeably drifts down from that acceptable value. If I set the altimeter to 30.35, the altimeter reads approximately 982 not near 653 or 670. It should have drifted down to the proper value if you had waited a little longer. It certainly has stabilized at 643 feet for me. See this picture: http://imagebin.ca/view/LyxG93.html (Note: the needle animation seems a little off) I think this is due to the weather interpolation code, which tries to smooth out the sudden changes that used to occur when you crossed METAR boundaries. -- Csaba/Jester -- Virtualization is moving to the mainstream and overtaking non-virtualized environment for deploying applications. Does it make network security easier or more difficult to achieve? Read this whitepaper to separate the two and get a better understanding. http://p.sf.net/sfu/hp-phase2-d2d ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] recent environment changes break airport elevation with metar altimeter settings
On 10/04/2010 05:54 PM, Csaba Halász wrote: On Tue, Oct 5, 2010 at 1:28 AM, dave perryskida...@mindspring.com wrote: Before I set the altimeter to 30.35 (i.e. as launched at 29.92), the altimeter reads aprox. 643 and it noticeably drifts down from that acceptable value. If I set the altimeter to 30.35, the altimeter reads approximately 982 not near 653 or 670. It should have drifted down to the proper value if you had waited a little longer. It certainly has stabilized at 643 feet for me. See this picture: http://imagebin.ca/view/LyxG93.html (Note: the needle animation seems a little off) I think this is due to the weather interpolation code, which tries to smooth out the sudden changes that used to occur when you crossed METAR boundaries. After 18 minutes sitting on the runway with the altimeter setting matching the current METAR, the altimeter is still showing 100 ft too high. Questions: Q1: Since there has not been a METAR change, why is interpolation smoothing necessary at launch. Q2: The rate of interpolation seems excruciatingly slow. Can this be increased? -- Beautiful is writing same markup. Internet Explorer 9 supports standards for HTML5, CSS3, SVG 1.1, ECMAScript5, and DOM L2 L3. Spend less time writing and rewriting code and more time creating great experiences on the web. Be a part of the beta today. http://p.sf.net/sfu/beautyoftheweb ___ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
[Flightgear-devel] Terragear terrafit memory leak
Hello.
Latest GIT from Terragear has 2 memory leaks in terrafit.cc and
Terra/greadyinsert.
Heap size increases from 0 to hero in under 2 minutes given enough files
are thrown at it.
Patches attached. Notify upstream, close main shuttle bay doors.
--
Adrian
diff --git a/src/Prep/Terra/GreedyInsert.cc b/src/Prep/Terra/GreedyInsert.cc
index 2659bb1..3c60c9f 100644
--- a/src/Prep/Terra/GreedyInsert.cc
+++ b/src/Prep/Terra/GreedyInsert.cc
@@ -20,6 +20,10 @@ void TrackedTriangle::update(Subdivision s)
+GreedySubdivision::~GreedySubdivision()
+{
+ delete heap;
+}
GreedySubdivision::GreedySubdivision(Map *map)
{
diff --git a/src/Prep/Terra/GreedyInsert.h b/src/Prep/Terra/GreedyInsert.h
index 3b4b816..23588be 100644
--- a/src/Prep/Terra/GreedyInsert.h
+++ b/src/Prep/Terra/GreedyInsert.h
@@ -68,7 +68,7 @@ protected:
public:
GreedySubdivision(Map *map);
-
+~GreedySubdivision();
array2char is_used;
Edge *select(int sx, int sy, Triangle *t=NULL);
diff --git a/src/Prep/TerraFit/terrafit.cc b/src/Prep/TerraFit/terrafit.cc
index 6a09bc7..b02f304 100644
--- a/src/Prep/TerraFit/terrafit.cc
+++ b/src/Prep/TerraFit/terrafit.cc
@@ -203,7 +203,8 @@ void fit_file(const std::string path) {
gzprintf(fp,%+03.8f %+02.8f %0.2f\n,vx,vy,vz);
}
}
-
+delete mesh;
+delete DEM;
gzclose(fp);
}
--
Beautiful is writing same markup. Internet Explorer 9 supports
standards for HTML5, CSS3, SVG 1.1, ECMAScript5, and DOM L2 L3.
Spend less time writing and rewriting code and more time creating great
experiences on the web. Be a part of the beta today.
http://p.sf.net/sfu/beautyoftheweb___
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel
Re: [Flightgear-devel] Terragear terrafit memory leak
On Tue, 2010-10-05 at 07:29 +0300, Adrian Musceac wrote:
Hello.
Latest GIT from Terragear has 2 memory leaks in terrafit.cc and
Terra/greadyinsert.
Heap size increases from 0 to hero in under 2 minutes given enough files
are thrown at it.
Patches attached. Notify upstream, close main shuttle bay doors.
--
Adrian
Actually, the hamsters were still loose. But I think I managed to
recapture most of them. Sg_gzifstream was not closed for each array file
and memory allocation size grew beyond the solar system. Turns out it
does have a close() method, just that it's implemented in the header
file.
Patch attached.
--
Adrian
diff --git a/src/Lib/Array/array.cxx b/src/Lib/Array/array.cxx
index 419b28d..f8220b2 100644
--- a/src/Lib/Array/array.cxx
+++ b/src/Lib/Array/array.cxx
@@ -100,10 +100,11 @@ bool TGArray::open( const string file_base ) {
bool
TGArray::close() {
// the sg_gzifstream doesn't seem to have a close()
-
+// yes it does, it's just defined in the header file
+array_in-close();
+fitted_in-close();
delete array_in;
delete fitted_in;
-
return true;
}
--
Beautiful is writing same markup. Internet Explorer 9 supports
standards for HTML5, CSS3, SVG 1.1, ECMAScript5, and DOM L2 L3.
Spend less time writing and rewriting code and more time creating great
experiences on the web. Be a part of the beta today.
http://p.sf.net/sfu/beautyoftheweb___
Flightgear-devel mailing list
Flightgear-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/flightgear-devel
