]> git.mxchange.org Git - simgear.git/log
simgear.git
20 years agoFrederic Bouvier:
ehofman [Sun, 21 Mar 2004 21:54:51 +0000 (21:54 +0000)]
Frederic Bouvier:

If the input string of sgSplitPath is empty, it returned a list with one empty string, not an empty list.

20 years agoFrederic Bouvier:
ehofman [Sat, 20 Mar 2004 22:41:40 +0000 (22:41 +0000)]
Frederic Bouvier:

The message 'Alert: catching up on tile delete queue'
comes from the fact that 48 tiles are scheduled and
added to the cache at startup before the plane location
is initialized. My proposed patch is to initialize
SGLocation with an invalid position and detect this
fact before scheduling tiles. I prefer to do that
rather than testing for lon and lat being 0,0 because
it is a valid position and someone could want to fly
near Accra.

20 years agoFrederic Bouvier:
ehofman [Sat, 20 Mar 2004 22:38:13 +0000 (22:38 +0000)]
Frederic Bouvier:

This patch is for windows only. It hides the console window
until there is a message to print. It only support SG_LOG,
that I think is the right way to display something in FG.

20 years agoAnother Cygwin fix. This seems to take care of things proeprly.
ehofman [Thu, 18 Mar 2004 09:25:28 +0000 (09:25 +0000)]
Another Cygwin fix. This seems to take care of things proeprly.

20 years agoDisable deselecting the branch if it becomes translucent because it crashes the UFO.
ehofman [Wed, 17 Mar 2004 15:20:20 +0000 (15:20 +0000)]
Disable deselecting the branch if it becomes translucent because it crashes the UFO.

20 years agoRemove an unused library since FlightGear dropped support for WeatherCM. Users are...
ehofman [Wed, 17 Mar 2004 10:45:32 +0000 (10:45 +0000)]
Remove an unused library since FlightGear dropped support for WeatherCM. Users are highly encouraged to use environment/metar instead.

20 years agoAttempt to fix the Cygwin build problem for once and for all
ehofman [Wed, 17 Mar 2004 10:31:34 +0000 (10:31 +0000)]
Attempt to fix the Cygwin build problem for once and for all

20 years agoVarious changes in preparation for the 0.3.5 release.
curt [Wed, 17 Mar 2004 04:22:09 +0000 (04:22 +0000)]
Various changes in preparation for the 0.3.5 release.

20 years agoHopefully the last Cygwin fix
ehofman [Tue, 16 Mar 2004 14:25:15 +0000 (14:25 +0000)]
Hopefully the last Cygwin fix

20 years agoFurther refinement of the Cygwin problem as suggested by Frederic.
ehofman [Tue, 16 Mar 2004 09:39:22 +0000 (09:39 +0000)]
Further refinement of the Cygwin problem as suggested by Frederic.

20 years agoAttempt to fix a nasty Cygwin problem.
ehofman [Tue, 16 Mar 2004 09:24:14 +0000 (09:24 +0000)]
Attempt to fix a nasty Cygwin problem.

20 years agoLimmit the change in sun color due to visibility to a saner range.
ehofman [Fri, 12 Mar 2004 18:55:22 +0000 (18:55 +0000)]
Limmit the change in sun color due to visibility to a saner range.

20 years agoUse a more clever way to deselect a fully translucent leaf.
ehofman [Fri, 12 Mar 2004 17:38:57 +0000 (17:38 +0000)]
Use a more clever way to deselect a fully translucent leaf.

20 years agoUnselect the branch if the object has become fully translucent.
ehofman [Fri, 12 Mar 2004 10:09:09 +0000 (10:09 +0000)]
Unselect the branch if the object has become fully translucent.

20 years agoUpdate for non-conformal (older) compilers
ehofman [Fri, 12 Mar 2004 08:59:12 +0000 (08:59 +0000)]
Update for non-conformal (older) compilers

20 years agoMacOS X refinement
ehofman [Mon, 8 Mar 2004 08:59:58 +0000 (08:59 +0000)]
MacOS X refinement

20 years agoSilently ignore texture files that are not present.
ehofman [Sun, 7 Mar 2004 18:47:37 +0000 (18:47 +0000)]
Silently ignore texture files that are not present.

20 years agoMSVC .NET 2003 fix
ehofman [Sun, 7 Mar 2004 09:36:31 +0000 (09:36 +0000)]
MSVC .NET 2003 fix

20 years agoMacOS X 10.3 fix
ehofman [Sun, 7 Mar 2004 09:28:48 +0000 (09:28 +0000)]
MacOS X 10.3 fix

20 years agoWork around a limitation of gc_calc_course_dist(). When the start and end
curt [Wed, 3 Mar 2004 21:37:41 +0000 (21:37 +0000)]
Work around a limitation of gc_calc_course_dist().  When the start and end
points are too close together, this routine can return a course of "nan"
but the distance is valid.  Someday someone who understands the math should
rewrite the gc_calc_course_dist() routine so it is more robust, but for now
it's easiest to simply check for a nan result and code around the limitation.

20 years ago#define isnan _isnan for MSVC and Mingwin.
curt [Wed, 3 Mar 2004 21:35:38 +0000 (21:35 +0000)]
#define isnan _isnan for MSVC and Mingwin.

20 years agoRemove a typo
ehofman [Wed, 3 Mar 2004 20:06:28 +0000 (20:06 +0000)]
Remove a typo

20 years agoPut Curt's cloud movement fix back in
ehofman [Wed, 3 Mar 2004 20:05:18 +0000 (20:05 +0000)]
Put Curt's cloud movement fix back in

20 years agoMSVC .NET 2003 fix.
ehofman [Wed, 3 Mar 2004 19:59:44 +0000 (19:59 +0000)]
MSVC .NET 2003 fix.

20 years agoAdd const back in the function
ehofman [Wed, 3 Mar 2004 19:54:24 +0000 (19:54 +0000)]
Add const back in the function

20 years agoSome small fixes
ehofman [Wed, 3 Mar 2004 19:48:28 +0000 (19:48 +0000)]
Some small fixes

20 years agoA real MSVC fix this time
ehofman [Tue, 2 Mar 2004 15:18:18 +0000 (15:18 +0000)]
A real MSVC fix this time

20 years agoRevert the last change, MSVC still doesn't like it.
ehofman [Tue, 2 Mar 2004 14:58:58 +0000 (14:58 +0000)]
Revert the last change, MSVC still doesn't like it.

20 years agoMSVC fixes
ehofman [Tue, 2 Mar 2004 14:49:01 +0000 (14:49 +0000)]
MSVC fixes

20 years agoUse a round-robin scheme rather than returning a random textured state.
ehofman [Tue, 2 Mar 2004 13:28:12 +0000 (13:28 +0000)]
Use a round-robin scheme rather than returning a random textured state.

20 years agoMake it possible to define multiple texture for a material (multiple <texture> entrie...
ehofman [Tue, 2 Mar 2004 10:51:13 +0000 (10:51 +0000)]
Make it possible to define multiple texture for a material (multiple <texture> entries in the materials.xml file). The code can return a random texture, or a pre defined texture upon request (default is random)

20 years agoAdd front() methods SGQueue, SGLockedQueue, and SGBlockingQueue so that the
curt [Sat, 28 Feb 2004 18:51:20 +0000 (18:51 +0000)]
Add front() methods SGQueue, SGLockedQueue, and SGBlockingQueue so that the
can be used more interchangably with a regular STL queue.

20 years agoLoad submodels before animations, so that we can animate submodels.
david [Fri, 27 Feb 2004 03:30:01 +0000 (03:30 +0000)]
Load submodels before animations, so that we can animate submodels.

Allow submodels to be named when they are loaded.

20 years agoMelchior FRANZ:
ehofman [Thu, 26 Feb 2004 09:46:36 +0000 (09:46 +0000)]
Melchior FRANZ:

Add proxy support to the metar class. Authorization is untested, but everything else works. Martin will have to tell us ...

20 years agoMelchior FRANZ:
ehofman [Mon, 23 Feb 2004 20:07:20 +0000 (20:07 +0000)]
Melchior FRANZ:
Here are some updates for the metar class:

- support for empty visibility group
- support for 4-alnum IACO ids (rather than only 4-alpha)
- better diagnosis in exception messages
- check for 404 http response
- replace last constant by constants.h definition

20 years agoComment out an improperly written constructor.
curt [Mon, 23 Feb 2004 01:34:31 +0000 (01:34 +0000)]
Comment out an improperly written constructor.

20 years agoAn ugly hack to get MipsPro 7.4.1 working on IRIX 6
ehofman [Fri, 20 Feb 2004 16:07:51 +0000 (16:07 +0000)]
An ugly hack to get MipsPro 7.4.1 working on IRIX 6
.5.20 (Yuck)

20 years agoFix from Richard Harke for 64 bit systems. The reftag was left
andy [Fri, 20 Feb 2004 15:10:49 +0000 (15:10 +0000)]
Fix from Richard Harke for 64 bit systems.  The reftag was left
uninitialized by naNum().  If it happened to be constructing it on the
stack in a location previously occupied by a real reference, it would
generate a corrupt naRef.

20 years agoChanges to tke sky dome coloring
ehofman [Wed, 18 Feb 2004 14:33:16 +0000 (14:33 +0000)]
Changes to tke sky dome coloring

20 years agoRemove an extra sgSetVec call
ehofman [Tue, 17 Feb 2004 14:40:05 +0000 (14:40 +0000)]
Remove an extra sgSetVec call

20 years agoBe a little quieter at the default debug level.
david [Sat, 7 Feb 2004 21:36:20 +0000 (21:36 +0000)]
Be a little quieter at the default debug level.

20 years agoMove the new metar class from FlightGear to SimGear
ehofman [Mon, 2 Feb 2004 10:12:31 +0000 (10:12 +0000)]
Move the new metar class from FlightGear to SimGear

20 years agoYank the MSVC special handling. It turns out it was becuase "strlen"
andy [Sun, 1 Feb 2004 17:47:08 +0000 (17:47 +0000)]
Yank the MSVC special handling.  It turns out it was becuase "strlen"
has special voodoo in the parser.  That's much more cleanly handled by
renaming the function than by #ifdef'ing.

20 years agoFix a slight ambiguity in variable names.
curt [Sat, 31 Jan 2004 19:50:35 +0000 (19:50 +0000)]
Fix a slight ambiguity in variable names.

20 years agoActivate the driver fog workaround again. It doesn't seem to be solved yet.
ehofman [Thu, 29 Jan 2004 18:25:55 +0000 (18:25 +0000)]
Activate the driver fog workaround again. It doesn't seem to be solved yet.

20 years agoFrederic BOUVIER:
curt [Tue, 27 Jan 2004 15:55:47 +0000 (15:55 +0000)]
Frederic BOUVIER:

Win32 serial port communication fixes.

20 years agoMake sure all libraries are used to test for certain functions
ehofman [Tue, 27 Jan 2004 09:41:33 +0000 (09:41 +0000)]
Make sure all libraries are used to test for certain functions

20 years agoFix a mistake
ehofman [Mon, 26 Jan 2004 19:59:36 +0000 (19:59 +0000)]
Fix a mistake

20 years agoClean up the Pre-, and PostDraw functions a bit. Especially the glPushAttrib has...
ehofman [Sat, 24 Jan 2004 12:08:12 +0000 (12:08 +0000)]
Clean up the Pre-, and PostDraw functions a bit. Especially the glPushAttrib has had some attention. The NVidia hack is commented out for now.

20 years agoFix an initialization problem
ehofman [Mon, 19 Jan 2004 13:37:26 +0000 (13:37 +0000)]
Fix an initialization problem

20 years agoOops, I originally had ramped the vasi/papi color transition the wrong way.
curt [Fri, 16 Jan 2004 17:37:05 +0000 (17:37 +0000)]
Oops, I originally had ramped the vasi/papi color transition the wrong way.
So as you passed through the target glide slope from low to high it would
be colored: red -> white -> small range of transition to red -> white.
Now it goes the right way so you get: red -> smooth transition to -> white.
You can tell you are getting high if you see the bottom vasi start to turn
pink ... etc. etc. hopefully just like in real life.

20 years agoTry to prevent a devide by zero situation
ehofman [Thu, 15 Jan 2004 14:23:06 +0000 (14:23 +0000)]
Try to prevent a devide by zero situation

20 years agoDon't do any calculations when thay are not needed
ehofman [Thu, 15 Jan 2004 14:22:03 +0000 (14:22 +0000)]
Don't do any calculations when thay are not needed

20 years agoMake sure MispPro compilers < 7.4 still work
ehofman [Thu, 15 Jan 2004 14:21:31 +0000 (14:21 +0000)]
Make sure MispPro compilers < 7.4 still work

20 years agoThe emissive values just don't seem like the right thing to do. You suddenly
curt [Wed, 14 Jan 2004 18:02:58 +0000 (18:02 +0000)]
The emissive values just don't seem like the right thing to do.  You suddenly
see the dark side of the moon quite clearly, which usually isn't the case.

The rest of the moon still seems a bit oversaturated right now ...

20 years agoMake the vasi lights slightly larger/brighter.
curt [Fri, 9 Jan 2004 16:49:13 +0000 (16:49 +0000)]
Make the vasi lights slightly larger/brighter.

20 years agoUse the same coloring scheme for the moon and the sun, add a bit of light reflection...
ehofman [Fri, 9 Jan 2004 10:19:00 +0000 (10:19 +0000)]
Use the same coloring scheme for the moon and the sun, add a bit of light reflection to the moon

20 years agoUpdates to the alpha-test animation class
ehofman [Thu, 8 Jan 2004 10:38:29 +0000 (10:38 +0000)]
Updates to the alpha-test animation class

20 years agoUpdates to the alpha-test animation class
ehofman [Thu, 8 Jan 2004 10:25:23 +0000 (10:25 +0000)]
Updates to the alpha-test animation class

20 years agoFrederic Bouvier:
ehofman [Wed, 7 Jan 2004 09:07:57 +0000 (09:07 +0000)]
Frederic Bouvier:

The attached patch adds a new animation class, called SGAlphaTestAnimation, to enable alpha test in 3D models.
This will remove many artefact caused by painting order of translucent models, and I need it for an upcoming model. In addition, I added a min-factor and a max-factor to the range animation to modulate min-m, min-property, max-m or max-property.

20 years agoSmarter vasi coloring.
curt [Tue, 30 Dec 2003 07:04:05 +0000 (07:04 +0000)]
Smarter vasi coloring.

20 years agoTrack some additional data required for computing vasi/papi colors.
curt [Tue, 30 Dec 2003 05:53:50 +0000 (05:53 +0000)]
Track some additional data required for computing vasi/papi colors.

20 years agoGo with an all white texture for vasi/papi lights. We will vary the
curt [Tue, 30 Dec 2003 05:53:03 +0000 (05:53 +0000)]
Go with an all white texture for vasi/papi lights.  We will vary the
background color externally to change from white to red.

20 years agoAdd a number of Metar related constants
ehofman [Sat, 27 Dec 2003 13:41:02 +0000 (13:41 +0000)]
Add a number of Metar related constants

20 years agoResolve some dependency problems
ehofman [Sat, 27 Dec 2003 09:31:02 +0000 (09:31 +0000)]
Resolve some dependency problems

20 years agoUpdates to support sgi's MIPSpro compiler version 7.4 and newer. This also fixes...
ehofman [Fri, 26 Dec 2003 13:55:09 +0000 (13:55 +0000)]
Updates to support sgi's MIPSpro compiler version 7.4 and newer. This also fixes a number of C++ issues where FlightGear didn't follow the standard.

20 years agoParse/GC interaction fixed.
andy [Mon, 22 Dec 2003 19:27:51 +0000 (19:27 +0000)]
Parse/GC interaction fixed.
Remove the OP_NEWARGS "optimization" (it wasn't).

20 years agoMelchior FRANZ:
ehofman [Fri, 19 Dec 2003 17:44:30 +0000 (17:44 +0000)]
Melchior FRANZ:
There are constructors for sg_io_exception that don't initialize the
location. In this case getFormattedMessage does still output an extra
line with " at" followed by ... nothing. Looks silly.

20 years agoRewrite sg_geodesy. The new version is more accurate to the WGS84
andy [Fri, 19 Dec 2003 02:40:33 +0000 (02:40 +0000)]
Rewrite sg_geodesy.  The new version is more accurate to the WGS84
standard and includes a sgCartToGeod() function which is 100%
symmetric (down to the precision of a double) with sgGeodToCart().

20 years agoMelchior FRANZ:
ehofman [Sun, 14 Dec 2003 14:45:29 +0000 (14:45 +0000)]
Melchior FRANZ:

There's another problem with sg_socket.cxx: the timeout value is
hardcoded as 0. This is appropriate for servers as they continuously
poll the socket and never want to wait for clients. But clients
*do* want to wait a few seconds for a server response.

20 years agoMake sure both client and server situations are hendled properly
ehofman [Sun, 14 Dec 2003 10:07:51 +0000 (10:07 +0000)]
Make sure both client and server situations are hendled properly

20 years agoRevert to the previous version
ehofman [Thu, 11 Dec 2003 15:26:23 +0000 (15:26 +0000)]
Revert to the previous version

20 years agofgrun fixes
ehofman [Thu, 11 Dec 2003 13:17:45 +0000 (13:17 +0000)]
fgrun fixes

20 years agoExtra i++ led to a "double skip" when parsing a \r\n pair. So a DOS
andy [Mon, 8 Dec 2003 16:55:50 +0000 (16:55 +0000)]
Extra i++ led to a "double skip" when parsing a \r\n pair.  So a DOS
file would look like it was all on one line, and a comment would skip
to the end of the file.

20 years agoBug fixes. The priority queue wasn't handling boundary conditions at
andy [Sun, 7 Dec 2003 19:53:34 +0000 (19:53 +0000)]
Bug fixes.  The priority queue wasn't handling boundary conditions at
the edge of the table properly.  The new code is half the size -- it
*has* to be correct, right?

20 years agoProperty interpolator subsystem. A utility, primarily for Nasal scripts
andy [Fri, 5 Dec 2003 01:49:44 +0000 (01:49 +0000)]
Property interpolator subsystem.  A utility, primarily for Nasal scripts

20 years agoAdded setsize() and subvec() library functions
andy [Fri, 5 Dec 2003 01:49:02 +0000 (01:49 +0000)]
Added setsize() and subvec() library functions

21 years agoA first attempt on detecting endianness and number of bits
ehofman [Mon, 1 Dec 2003 18:57:32 +0000 (18:57 +0000)]
A first attempt on detecting endianness and number of bits

21 years agoRename IRIX to irix_string in simgear/compatibility and assign "simgear/compatibility...
ehofman [Mon, 1 Dec 2003 18:56:53 +0000 (18:56 +0000)]
Rename IRIX to irix_string in simgear/compatibility and assign "simgear/compatibility/irix_string" to STL_STRING for IRIX

21 years agoAdd Nasal Vs. 1.5
ehofman [Mon, 1 Dec 2003 14:33:45 +0000 (14:33 +0000)]
Add Nasal Vs. 1.5

21 years agoAttempt to fix a bug in SGSocket::read() affecting TCP style sockets.
curt [Thu, 27 Nov 2003 13:24:26 +0000 (13:24 +0000)]
Attempt to fix a bug in SGSocket::read() affecting TCP style sockets.

21 years agoRevert an IRIX/O2 only setting again
ehofman [Thu, 27 Nov 2003 10:23:13 +0000 (10:23 +0000)]
Revert an IRIX/O2 only setting again

21 years agoPortability fix
ehofman [Thu, 27 Nov 2003 10:22:29 +0000 (10:22 +0000)]
Portability fix

21 years agoMake sure autogenerated files don't show up
ehofman [Wed, 26 Nov 2003 14:42:05 +0000 (14:42 +0000)]
Make sure autogenerated files don't show up

21 years agoPortability fixes
ehofman [Wed, 26 Nov 2003 14:39:46 +0000 (14:39 +0000)]
Portability fixes

21 years agorename libnasal.a to libsgnasal.a to match the existing SimGear library
curt [Tue, 25 Nov 2003 22:49:02 +0000 (22:49 +0000)]
rename libnasal.a to libsgnasal.a to match the existing SimGear library
naming convention.

21 years agoMSVC > 6.0 fix
ehofman [Tue, 25 Nov 2003 21:45:56 +0000 (21:45 +0000)]
MSVC > 6.0 fix

21 years agoNasal and event manager updates from Andy Ross.
curt [Tue, 25 Nov 2003 21:26:01 +0000 (21:26 +0000)]
Nasal and event manager updates from Andy Ross.

21 years agoVersion 1.3 of Andy Ross's "NASAL" embedded scripting language.
curt [Tue, 25 Nov 2003 20:16:28 +0000 (20:16 +0000)]
Version 1.3 of Andy Ross's "NASAL" embedded scripting language.

21 years agoMinor changes to logging messages (to STDOUT).
david [Mon, 24 Nov 2003 17:41:43 +0000 (17:41 +0000)]
Minor changes to logging messages (to STDOUT).

21 years agoMacOS X fixes
ehofman [Sun, 23 Nov 2003 10:04:45 +0000 (10:04 +0000)]
MacOS X fixes

21 years agoFIx a typo
ehofman [Fri, 21 Nov 2003 21:56:57 +0000 (21:56 +0000)]
FIx a typo

21 years agoFrederic Bouvier wrote:
ehofman [Wed, 19 Nov 2003 15:16:21 +0000 (15:16 +0000)]
Frederic Bouvier wrote:
> When a c172 is on one machine, I only got segfault on an animation
> not found. This animation is named ControlsGroup and I guess
> that one object name referenced in this null animation no
> longer exist in the .ac model. I say that because the
> preceding one is analogous and works.
>
> So I would say the multiplayer works, except when there is a c172.

It seems that this animation refers to 'PanelInstruments' that is
a panel outside the model, so, when loaded with SGModelLib::load_model
the object is not found in the model and there is a non tested
read access through a null pointer in sgMakeAnimation.

21 years agoChange istringstream back to a typedef so there is no need to redefine every class...
ehofman [Sun, 9 Nov 2003 08:56:09 +0000 (08:56 +0000)]
Change istringstream back to a typedef so there is no need to redefine every class memeber.

21 years agoSave on a number of CPU costly strcmp calls when using the blend function
ehofman [Tue, 4 Nov 2003 13:25:23 +0000 (13:25 +0000)]
Save on a number of CPU costly strcmp calls when using the blend function

21 years agoUpdates for the official 0.3.4 version.
curt [Wed, 22 Oct 2003 19:21:24 +0000 (19:21 +0000)]
Updates for the official 0.3.4 version.

21 years agoShoot, I was trying to hunt down a bug that wasn't even caused by the sstream impleme...
ehofman [Mon, 20 Oct 2003 19:53:35 +0000 (19:53 +0000)]
Shoot, I was trying to hunt down a bug that wasn't even caused by the sstream implementation! Back out some of the previous patches

21 years agoAnd don't forget to free up the used memory.
ehofman [Mon, 20 Oct 2003 19:38:27 +0000 (19:38 +0000)]
And don't forget to free up the used memory.

21 years agoMake it easy on myself (and make it work as a bonus)
ehofman [Mon, 20 Oct 2003 19:32:50 +0000 (19:32 +0000)]
Make it easy on myself (and make it work as a bonus)

21 years agoLast fixes
ehofman [Mon, 20 Oct 2003 12:14:10 +0000 (12:14 +0000)]
Last fixes