13c9fe7923
Merge branch 'master' of ssh://opensimulator.org/var/git/opensim
2016-06-23 06:39:15 -07:00
648cf51411
Fix failing test
2016-06-23 06:38:34 -07:00
a2e4beea86
Update HttpServer_OpenSim.dll to fix an inventory loading bug.
2016-06-23 15:20:07 +02:00
56e2358ec2
Because there's no such thing as too much culture. (looking forward to .NET 4.5 to set this globally)
2016-06-22 20:31:01 -07:00
cc9ff12988
Set culture also in object crossings.
2016-06-22 20:18:16 -07:00
c94c929ec1
Amend to previous commit -- set the culture also on the receiving end.
2016-06-22 18:35:14 -07:00
33efc91342
Mantis #7802 : set culture to US English on the thread that does the agent crossing, not that it's async and needs to parse decimal numbers.
2016-06-22 18:10:12 -07:00
1e213f5150
Mantis #7802 : added debug message to diagnose the problem.
2016-06-22 15:55:39 -07:00
94616510f2
Missing AsBoolean
2016-06-22 15:39:09 -07:00
193a3afa64
Make a debug message less misleading.
2016-06-22 15:28:18 -07:00
f8bd3d2f2e
Mantis #7935 : regression on transferring attachments over the HG
2016-06-22 11:52:53 -07:00
0b32cf0e8f
Can't do it do it this low in the stack. Attachments on teleports apparently rely on
...
State being != 0.
Revert "Mantis #7932 : Some objects were being serialized with PCode=9 and State != 0 (meaning attachments). That should never happen."
This reverts commit f19199cf4a
2016-06-22 06:21:23 -07:00
74c79d1f88
Merge branch 'master' of ssh://opensimulator.org/var/git/opensim
2016-06-21 20:28:44 -07:00
f19199cf4a
Mantis #7932 : Some objects were being serialized with PCode=9 and State != 0 (meaning attachments). That should never happen.
2016-06-21 20:28:25 -07:00
5c81e03738
Fix a dumb mistake
2016-06-21 01:49:47 +02:00
8e40bb6fb7
Make some more modules inheritable
2016-06-21 01:46:38 +02:00
9de8cefee5
Split IM handling so an inheriting module can cange names if required.
2016-06-21 01:20:21 +02:00
34b308d5b6
Guard against multiple calls to SetOwnerId on one logical ownership change.
2016-06-20 16:02:37 -07:00
bc0fc16f44
Still mantis #7929 : apply next owner mask to the new owner, the group.
2016-06-20 15:32:22 -07:00
359908fa88
Related to mantis #7929 : "For sale" was not being accepted, because of permissions fail. This adds a new permissions check CanSellGroupObject. THIRD-PARTY PERMISSIONS MODULES TAKE NOTE OF THIS NEW EVENT.
2016-06-20 10:08:15 -07:00
623a2f859f
Mantis #7929 : objects deeded to the group were loosing some important properties. This makes the "For sale" be visible, but is not the end of the story.
2016-06-20 10:05:17 -07:00
ff65248b9c
Removed a couple of verbose messages
2016-06-19 20:13:32 -07:00
8d99fe095a
Mantis #7920 : group info was not being updated in many situations. (regression) Putting back the heavy messaging.
2016-06-19 17:35:11 -07:00
4f3da0a643
AgentGroupData was being sent by both UDP and EQ -- an overkill. Turn off UDP messaging.
2016-06-19 16:28:27 -07:00
76ab1ed10c
Restored sending group membership data via EventQueue -- this had been a regression. Took the opportunity to refactor that code, so that both Groups V2 and XmlRpcGroups can use the same function.
2016-06-19 16:27:12 -07:00
d51fe01ed8
Mantis #7915 : bring XmlRpcGroups/GroupsModule.cs up to speed with Groups V2
2016-06-19 15:15:57 -07:00
a9fb3d0e18
Minor -- deregister an event handler in Groups V2
2016-06-19 15:14:50 -07:00
ea3fe90ddb
Mantis #7928 : show grid size and show regions now working. Integers were being overflown.
2016-06-19 13:57:01 -07:00
12b73603cc
Still related to mantis #7917 , parcel prims info. Now sending the correct numbers for "Simulator primitive usage" related to the parcel owner, not the entire region.
2016-06-19 09:17:04 -07:00
ce7fa728ef
Mantis #7917 : regression on prim limits module.
2016-06-18 22:07:01 -07:00
206faf0218
Guard against occurring null ref.
2016-06-18 17:00:15 -07:00
fe1f01defb
Merge branch 'master' of ssh://opensimulator.org/var/git/opensim
2016-06-17 12:40:41 -07:00
8a070f16dd
Mantis #7848 : something wrong with a data field in Groups V2
2016-06-17 12:40:04 -07:00
ed168473af
Merge branch 'master' of opensimulator.org:/var/git/opensim
2016-06-16 01:48:27 +02:00
d2f61f5320
Also make the HGUserManagementModule more amenable to subclassing.
2016-06-16 01:47:52 +02:00
f800f5ba57
Make the UserManagementModule inheritable
2016-06-15 16:51:02 +02:00
e0a82a82de
"reverting" the keep alive changes brought in from avination, to check if that's what's making osgrid fail. This is not really reverting, it's more manually resetting some places (not all) where keep alive is set.
2016-06-14 08:58:37 -07:00
4be4e98525
Revert "Add a few more debug messages to diagnose problems in OSGrid's proxy use"
...
This reverts commit 2d2414adcf
2016-06-13 20:40:47 -07:00
1e9a944633
Revert "More debug to diagnose osgrid"
...
This reverts commit b5f29e93f6
2016-06-13 20:39:52 -07:00
b5f29e93f6
More debug to diagnose osgrid
2016-06-13 20:35:40 -07:00
467de7a2f0
Merge branch 'master' of ssh://opensimulator.org/var/git/opensim
2016-06-13 18:04:50 -07:00
2d2414adcf
Add a few more debug messages to diagnose problems in OSGrid's proxy use
2016-06-13 18:04:28 -07:00
cdfbf78d07
Fields should be protected unless private is required
2016-06-14 01:27:48 +02:00
68c73951ef
Make the LLUDPServerShim inheritable. There is absolutely no reason for
...
this to be sealed!
2016-06-14 01:21:47 +02:00
141205532b
Make chat and instant message modules fully subclassable
2016-06-14 00:03:40 +02:00
3ef5e1a24b
Mantis #7923 . EDIT BEAMS ARE BACK!
2016-06-12 22:01:36 -07:00
c2e63e807e
Mantis #7861 . llEdgeOfWorld not working. Two reasons: (1) edge was being filled but not used; (2) region size was being ignored when computing the neighbors' coordinates.
...
Not sure if this is the intended semantics of this function, but it made sense in my varregion test with neighbors all around except in the South.
2016-06-12 14:48:30 -07:00
d0c65d15dc
Still mantis #7858 : send kill object / update to child agents too
2016-06-12 14:29:30 -07:00
634d85a30d
Detection of Cylinder and Prism for flexi prims
...
Signed-off-by: Mandarinka Tasty <mandarinka.tasty@gmail.com>
Signed-off-by: Diva Canto <diva@metaverseink.com>
2016-06-12 13:58:20 -07:00
42a9afdc43
Mantis #7858 : DeleteSceneObject done slightly differently. ProcessEntities now checks whether the objects have been deleted and, if so, sends an extra kill object packet, in order to compensate for potential race conditions encountered by the first one.
...
Note: I still cannot reproduce this problem, but I was able to emulate it by adding an artificial delay on ProcessEntities, which did, indeed, result in objects not being deleted. This fix fixed my emulated scenario.
2016-06-12 12:23:52 -07:00
Diva Canto