Hello everybody. This is a short summary of OpenSim development for the past week up until the end of last Saturday. The letters in brackets after bug and feature points are the Git revision identifiers in which the work appears. We only give the first 6 characters of each revision but this should be enough to identify it.
The changes described here appear only in OpenSim’s cutting edge Git ‘master’ branch. Please do not attempt to use code taken directly from ‘master’ in any production environment. At particular points in time this in-development code may not work or may even cause data corruption. If you are not a developer or tester, please treat this summary instead as a preview of what will be coming up in the next release. The latest official release is OpenSim 0.6.6 (branched at old Subversion revision r9962).
This week in OpenSim ( 04aca6 – ddac88 )
General news
- None this week.
Features & bug fixes
- Melanie removed the script sponsor logic introduced in ( ddac88 ) because this was causing some scripts to timeout. Unfortunately, this reinstates a memory leak in regions that see a lot of script starting and stopping (e.g. script attachments or script development) ( 72a2e6 ).
Infrastructure
- Diva started working on overhauling the inventory service. At the moment, when a user enters a region (whether by login, teleport or direct crossing) their entire inventory is downloaded and cached on the region simulator. In the new model, region simulators will instead simply relay inventory requests between the client and the inventory service without caching. This should improve inventory stability at the cost of increasing the number of requests made to the inventory service. This work is not yet active.
- I did some more work on Inventory Archives (IARs).
- Teravus added viewer-simulator circuit tests.
- Teravus filled in method and class documentation for quite a few classes – hopefully this should go a small way towards making the codebase more understandable for third-party developers.
- Teravus worked on trying to track down why automated tests run in our Panda continuous integration server occasionally fail with ThreadAbortExceptions.