aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/OpenSim/Region/CoreModules/World/Archiver (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Added more debug messages under the -verbose option.Diva Canto2011-01-081-2/+6
|
* Drop current OAR version back to 0.5 since these are currently now ↵Justin Clark-Casey (justincc)2010-12-113-35/+34
| | | | | | compatible with older OpenSim versions again. Removed ability to choose 0.4 to avoid having to write code that polices this properly. Please shout on the ml if you really need this.
* Preservation of creator information now also working in IARs. Cleaned up ↵Diva Canto2010-11-292-7/+30
| | | | usage help. Moved Osp around, deleted unnecessary OspInventoryWrapperPlugin, added manipulation of SOP's xml representation in a generic ExternalRepresentationUtils function.
* Fix: Allow use of parameters without specifying a file path with command ↵Marck2010-11-231-1/+1
| | | | "save oar"
* Added creator info across the board -- TaskInventoryItems and InventoryItems ↵Diva Canto2010-11-211-3/+7
| | | | | | themselves. Tested. Seems to be working, main tests pass. Nothing done for IARs or HG transfers yet -- this only works for OARs for the time being. New migration in inventory table in order to make CreatorID varchar(255).
* Global creator information working on MySQL DB and on load/save OARs. ↵Diva Canto2010-11-212-2/+20
| | | | | | | Creator name properly shown on the viewer as first.last @authority. New option added to save oar -profile=url. Migration on RegionStore making CreatorID be 255 chars. Moved Handling of user UUID -> name requests to a new module UserManagement/UserManagementModule.
* Knock V0_2 decals off archive tests since they're misleadingJustin Clark-Casey (justincc)2010-11-201-8/+8
|
* Fix unit tests and save problemsJustin Clark-Casey (justincc)2010-11-203-3/+4
| | | | | | (note; contrary to previous commits, saving 0.4 OARs is actually broken since serialized objects in inventories are not yet converted properly). Firstly, if you can save a version 1.0 OAR then you can load it too. Secondly, closing the save stream even on success before assets/objects are saved is a bad idea
* correct build break. for some reason, xbuild didn't rebuild test dllsJustin Clark-Casey (justincc)2010-11-201-2/+2
|
* Bump oar version to 1.0 from 0.5Justin Clark-Casey (justincc)2010-11-201-83/+119
| | | | | | | | If oar contents are being changed such that older versions of opensim can't load them, then the major version must be increased This also locks version parameters to either 1.0 or 0.4, so that arbitrary 'versions' cannot be saved Also closes save stream properly in the event of an error Version 1.0 OARs are currently incompatible with OpenSim 0.7.0.2 and earlier. However, you can still save compatible version 0.4 OARs by specifing --version=0 on the save oar command line e.g. save oar --version=0 oars/test.oar
* Old deserialization can't deal with commas in flag fields. Making use of ↵Diva Canto2010-11-022-16/+10
| | | | -version option on save oar command. Bumped archives version to 0.5; version < 0.5 generates flag fields without commas. Everything else is identical.
* Formatting cleanup.Jeff Ames2010-10-271-3/+3
|
* Implement guard against trying to load incompatible version IARsJustin Clark-Casey (justincc)2010-10-221-1/+1
|
* start parsing iar control fileJustin Clark-Casey (justincc)2010-10-221-1/+1
| | | | This change requires a prebuild[.sh|.bat] since a System.Xml.Linq reference is added to prebuild.xml
* slightly simplify oar control file loading codeJustin Clark-Casey (justincc)2010-10-221-10/+3
|
* write IAR control file first in the archive rather than lastJustin Clark-Casey (justincc)2010-10-221-1/+2
|
* minor: add some text to the oar abort messages on assets timeout to make it ↵Justin Clark-Casey (justincc)2010-10-221-1/+1
| | | | clear that the oar will not be usable
* Drop max oar loading version back to 0 from 1Justin Clark-Casey (justincc)2010-10-161-1/+1
|
* Rip out version option since it turns out that the changed object ↵Justin Clark-Casey (justincc)2010-10-162-7/+11
| | | | | | serialization format can actually be loaded by older OpenSims after all This bumps the OAR version to 0.4 instead, signalling some change but with backwards compatability... for now.
* change --old-guids switch on the save oar command line to --version=<x>Justin Clark-Casey (justincc)2010-10-165-9/+31
| | | | | | if x is 0, then an old version 0.3 archive is saved. If it is anything else or missing, then a version 1.0 archive is saved version 1.0 archives cannot be loaded on OpenSim 0.7.0.2 and earlier also add various informational notices about what version we've saving/loading
* Have OpenSim throw a strop if it tries to load an OAR with a major version ↵Justin Clark-Casey (justincc)2010-10-162-1/+23
| | | | that is too high for it to handle
* save oar control file first rather than in the middle so that it's the first ↵Justin Clark-Casey (justincc)2010-10-163-38/+41
| | | | | | thing we come accross on load this exposes a weekness with using tar where we don't control the order in which files are loaded. can't be helped for now
* Made OARs use the new serialization procedure. (TPs/crossings still on the ↵Diva Canto2010-10-153-14/+25
| | | | old one) Added an options argument down the pipeline. For the time being it takes --old-guids as an option to produce <Guid> instead of <UUID>.
* Changed SceneObjectGroup to store parts with the fast and thread-safe ↵John Hurliman2010-09-161-5/+1
| | | | MapAndArray collection
* Formatting cleanup.Jeff Ames2010-09-121-4/+4
|
* First pass at cleaning up thread safety in EntityManager and SceneGraphJohn Hurliman2010-09-101-1/+1
|
* Merge branch 'master' of ssh://opensimulator.org/var/git/opensimJustin Clark-Casey (justincc)2010-09-081-1/+4
|\
| * * Cache null account responses in the SimianUserAccountServiceConnector to ↵John Hurliman2010-09-071-1/+4
| | | | | | | | | | | | avoid repeated requests for missing avatar IDs * Updated to OpenMetaverse r3442 to fix a timezone issue with ExpiringCache
* | If a scene object part UUID is changed (only possible when not in a scene), ↵Justin Clark-Casey (justincc)2010-09-071-1/+1
|/ | | | then adjust the inventory items to point to the new uuid as well
* Improve liveness by operating on list copies of SOG.Children where appropriateJustin Clark-Casey (justincc)2010-08-281-28/+29
|
* Improve consistency of locking for SOG.m_parts in order to avoid race ↵Justin Clark-Casey (justincc)2010-08-261-26/+29
| | | | conditions in linking and unlinking
* Revert "don't report the null uuid as a missing asset on saving oars/iars"Justin Clark-Casey (justincc)2010-06-291-5/+3
| | | | | | | This reverts commit 5ce119174e03715bfc61f7864bd3e7182c5d21ea. This change led to infinite timeout since the expected number of assets no longer matched requests. Rather than introduce yet fresh bugs with another fix, just revert this since the inaccurate report is just inconvenient
* don't report the null uuid as a missing asset on saving oars/iarsJustin Clark-Casey (justincc)2010-06-251-3/+5
|
* remove pointless mock user account service since the real one can now be ↵Justin Clark-Casey (justincc)2010-06-041-1/+1
| | | | easily configured for test purposes
* refactor: move GetStream and URI methods from ArchiveReadRequest -> ↵Justin Clark-Casey (justincc)2010-05-282-63/+65
| | | | ArchiveHelpers
* Fix a problem where SceneGraph.AddSceneObject() would return false on ↵Justin Clark-Casey (justincc)2010-05-211-1/+0
| | | | | | successfully adding an object rather than true, in defiance of its method documentation This meant that the returns were inconsistent - false would be returned both for various scene object failure conditions (e.g. root part was null) and if the object was successfully added.
* make oar object filename/pathname creation a helper methodJustin Clark-Casey (justincc)2010-05-214-38/+75
| | | | | reused in both tests and oar code reduction in checking is outweighed by greater test clarity
* add prim item and test asset save in save oar unit testJustin Clark-Casey (justincc)2010-05-212-3/+31
|
* All scripts are now created suspended and are only unsuspended when the objectMelanie2010-04-191-0/+1
| | | | | | is fully rezzed and all scripts in it are instantiated. This ensures that link messages will not be lost on rez/region crossing and makes heavily scripted objects reliable.
* minor: use the static ascii and utf8 encodings instead of instantiating our ownJustin Clark-Casey (justincc)2010-04-161-6/+3
|
* Add --skip-assets option to load oar.Justin Clark-Casey (justincc)2010-04-163-24/+37
| | | | | This allows you to load an oar without loading its assets. This is useful if you know that the required assets are already in the asset service, since loading without assets is quicker. This option will become more useful when the ability to save oars without assets is added, which should happen fairly soon. At this point there will also be better documentation.
* * UuidGatherer now tracks asset types for assets it discovers. The asset ↵John Hurliman2010-03-152-12/+25
| | | | | | types are inferred from context * OAR saving will attempt to correct unknown asset types before writing broken assets to the OAR file
* Merge branch 'master' into presence-refactorMelanie2010-03-011-0/+2
|\
| * fix bug where region settings loaded via an oar (water height, terrain ↵Justin Clark-Casey (justincc)2010-02-261-0/+2
| | | | | | | | | | | | textures, etc.) were not being persisted to the database this meant that the loaded settings would disappear after a restart
* | Changed asset CreatorID to a stringJohn Hurliman2010-02-222-2/+2
| |
* | Merge branch 'presence-refactor' of ssh://opensimulator.org/var/git/opensim ↵John Hurliman2010-02-223-15/+39
|\ \ | | | | | | | | | into presence-refactor
| * \ Merge branch 'master' into presence-refactorMelanie2010-02-223-14/+38
| |\ \ | | |/ | | | | | | This brings presence-refactor up to master again
| | * forestall bug reports about the Command error: ↵Justin Clark-Casey (justincc)2010-02-192-2/+26
| | | | | | | | | | | | System.EntryPointNotFoundException: CreateZStream save/load oar/iar issue by telling the user what the likely problem is
| | * Formatting cleanup.Jeff Ames2010-02-151-12/+12
| | |
| * | One more test running.Diva Canto2010-02-211-1/+1
| | |
generated by cgit v1.1 at 2024-07-20 03:28:11 +0000