| Commit message (Collapse) | Author | Files | Lines |
|---|
|
from locking up when running BulletSim on a separate thread.
|
|
If the C# column can't be found in the positionMap (but the line can),
use the map immediately after it while correcting for the offset,
unless that results in an LSL position before the previous LSL position
in the positionMap.
The idea behind this heuristic is that in most, if not all cases C#
consumes more characters than LSL (for example LSL_Types.LSLInteger
instead of just 'integer').
Thus if the distance between the columns of two markers differ in
the C# and LSL file, the distance in the C# file will be larger.
Moreover, we can assume that every time this happens we will have
a marker at the beginning of the longer 'keyword', because those
keywords were generated by us in the first place.
For example:
C#: LSL_Types.LSLInteger f2(LSL_Types.LSLString s)
^ ^
1 2
will always have markers at the beginning of the long keywords
'LSL_Types.LSLInteger' and 'LSL_Types.LSLString'.
If an error is generated in between (for example at the beginning
of the function name 'f2') then the correct position is found
by using an offset relative to 2 rather than 1.
Note that a case where this isn't working correctly is
when the user adds extra spaces. For example:
LSL: integer f2( string s)
would still use the start of 'string' as reference and
then go backwards 3 characters only because the corresponding
C# still looks like
C#: LSL_Types.LSLInteger f2(LSL_Types.LSLString s)
^ ^
only 3 chars difference
and the reported error at 'f2' would be here:
LSL: integer f2( string s)
^
This can only be fixed by generating a mapping for 'f2' itself, or
generating a mapping whenever the amount of spaces is changed.
|
|
When a compile error reports a colum/error that is not an exact
match in the positionMap dictionary, the last position in the
map with a line number and position before the reported error
should be returned.
The old code had the following problems:
1) It returns l,c - which are line and column of the C# file, not LSL.
2) It doesn't set l to 'line' when the map has an entry with 'line'.
3) It sorts the map without taking columns into account, which may
result in a random order of the columns. With my mono implementation
the columns were reversed in order.
For example, if the map contains the following lines:
99,5,49,10
100,30,50,10
100,40,1,0
101,5,51,10
and a translation of 100,35 was requested,
then the old code would compare '100' with the keys in
the first column - setting l to that key while it is
smaller. Hence, l is set to 99.
Then it finds the key 100 and doesn't update l.
Because of the reversed sort order, it first compares
the column 35 with 40, finding that it is smaller
and therefore it stops; returning 99,1 instead of finding
the correct 100,30 entry and returning 50,10.
This patch causes 50,10 to be returned.
The remaining problems after this patch are:
1) The sorting might not be necessary at all.
2) The is code duplication (I fixed both instances,
but really there should be no code duplication
imho).
|
|
child doesn't have a physical shape. Another attempt at fixing Mantis 7191.
|
|
AvatarFlyingGroundMargin and AvatarFlyingGroundUpForce set to 5.0 and
2.0 respectively which seems to give about the same action as in SL.
Also moved force addition to before the velocity to force computation
so the upward velocity is properly applied to the avatar mass.
|
|
|
|
avoid any possibility of inter-test inter-ference
|
|
position map and leading to invalid error line numbers/columns
This is because jump statement generation was mistakenly inserting its own line without updating the csharp positions in CSCodeGenerator.
This is Aleric Inglewood's patch in http://opensimulator.org/mantis/view.php?id=7195 but applied to opensim itself rather than the defunct code generation in opensim-libs. Thanks!
This patch also adds a regression test for this case from myself.
|
|
(swimming)
Signed-off-by: Michael Cerquoni <nebadon2025@gmail.com>
|
|
flop to the floor when flying with bullet physics and acts more like ODE and SL flight.
Signed-off-by: Michael Cerquoni <nebadon2025@gmail.com>
|
|
|
|
|
|
this with a warning.
|
|
exception and move to the next request rather than terminate the simulator.
|
|
don't try to extract materials data rather than throw an exception
|
|
Setting it for foreign users does not make sense, since cntrl+shift+H always teleports them back to their original grid.
|
|
the fix was good) Thanks FreakyTech.
|
|
that simulator statistics is output to the console. Setting to zero turns
stats logging off.
|
|
|
|
HG gods are not safe at this point. It's better to disallow this until
they can be made safe.
This reverts commit e86c765be3b0d94c94ff1c5f15a3949ecc857627.
|
|
|
|
|
|
This should get around the exception reported in Mantis 7191 and 7204
by checking for the unbuilt child and rebuilding the linkset the next tick.
A warning message is output when this rebuild happens and this message is
clamped to 10 times in case there is a problem with a loop.
|
|
0.7.6 to a varregion running in 0.8 and above will be denied teleport, rather than be allowed and crash the viewer.
|
|
|
|
|
|
Signed-off-by: Michael Cerquoni <nebadon2025@gmail.com>
|
|
given parameters.
|
|
draw distance optimization is enabled. Makes terrain editting a lot
snappier.
|
|
being given permission to rez on group land.
|
|
|
|
old viewers around anymore.
|
|
Specifically, StoreEstateSettings was not being used anywhere; instead EstatSetting.Save was being called, but that method is a trigger to the DB-layer code directly, which, besides being wrong, was making it impossible to replace the service with a remote connector.
Also added more packing/unpacking code.
|
|
region creation thread. For varregions or simulators with many regions,
this will speed up simulator startup and elimiate some thread timeout
warnings.
|
|
|
|
Still has problems with child avatars.
|
|
[Terrain]SendTerrainUpdatesByViewDistance=true.
This tracks which patches have been sent to each client and outputs the
patches as the avatar moves.
|
|
erroneous information.
Also, added conversion of EstateSettings from/to key-value pairs in preparation for robust net work connectors.
|
|
|
|
ScenePresence.SendTerseUpdateToAllClients() which is extremely helpful when investigating presence update triggers.
|
|
RawVelocity update threshold for now in BSCharacter.UpdateProperties().
For some reason as yet unidentified (feedback?) a threshold above 0.4 here causes the RawVelocity to move between a lower and upper bound rather than remaining constant.
The RawVelocity increased until it triggered the threshold update, at which point it started to decrease until it again triggered the threshhold update.
This delta-v was enough to exceed the checks in ScenePresence.SendTerseUpdateToAllClients() and produce jittery avatar flight because of the fluctuating velocity.
With a threshold of 0.4 (or 0, as with ODE), the RawVelocity remains constant in BulletSim and so avatar flight becomes mostly smooth - remaining occasional glitches appear to be a result of errors in distance extraploation.
There are no obvious problems with commenting out the threshold.
Misterblue, if this is wrong or I've missed some subtlety here, please feel free to revert and/or correct.
The same considerations may or may not apply to object velocity updates.
|
|
required. This preserves API compatibility for external modules using this function.
|
|
|
|
which logs every request it receives.
|
|
bitmap generation in WorldMapModule throw an exception on next startup.
This commit replaces the hardcoded region sizes in WorldMapModule.GenerateOverlay() with numbers pulled from m_scene.RegionInfo
|
|
other assembly versions
|
|
|
|
messages instead of a warning message.
Same justification as earlier commit 996a6c2. These are not warnings but should still be visible to the user at any log level.
|
|
|
|
the issue with different AppDomain BaseDirectory in local and Jenkins test runs
|
