Story #2002
Create test for use of LocalCache in libclient
0%
Description
could be integration test or unit test.
History
#1 Updated by Dave Vieglais over 12 years ago
- Target version set to Sprint-2011.47-Block.6
- Position set to 1
- Position deleted (
21)
#2 Updated by Dave Vieglais over 12 years ago
- Position deleted (
15) - Position changed from 1 to 322
- Position set to 1
- Target version changed from Sprint-2011.47-Block.6 to Sprint-2011.48-Block.6
#3 Updated by Matthew Jones over 12 years ago
- Position deleted (
325) - Target version changed from Sprint-2011.48-Block.6 to Sprint-2011.49-Block.6
- Position set to 18
#4 Updated by Dave Vieglais over 12 years ago
- Target version changed from Sprint-2011.49-Block.6 to Sprint-2012.03-Block.1.2
- Position set to 1
- Position deleted (
31)
#5 Updated by Dave Vieglais over 12 years ago
- Position deleted (
1) - Position changed from 1 to 329
- Target version deleted (
Sprint-2012.03-Block.1.2) - Position set to 1
#6 Updated by Dave Vieglais over 9 years ago
- translation missing: en.field_remaining_hours set to 0.0
- Due date set to 2014-10-02
- Target version set to Maintenance Backlog
- Start date set to 2014-10-02
#7 Updated by Rob Nahf over 9 years ago
- Description updated (diff)
- Subject changed from Create integration test for use of LocalCache in libclient to Create test for use of LocalCache in libclient
- Milestone changed from CCI-0.6.4 to None
- Assignee set to Rob Nahf
- Due date changed from 2014-10-02 to 2014-11-19
#8 Updated by Rob Nahf about 9 years ago
- Category set to d1_libclient_java
#9 Updated by Rob Nahf about 8 years ago
- Status changed from New to Rejected
JCS / LocalCache is slated for removal from libclient due to the advent of series identifiers. (we can't make the same assertions that the object bytes behind an identifier are the same).
#10 Updated by Matthew Jones about 8 years ago
Note that we could still be caching based on the PID in the sysmeta, which will still exist, even for objects with SIDs. So technically this could still be useful.
But still, I think its fine as long as we check and fix all apps that depend on that caching behavior -- someone will need to review Metacat and other stacks.
This would be a breaking change for anyone external to DataONE who relied on caching, so we should bump the major version number in libclient to 3.x, and make sure that we've documented the change. In general, I would prefer that a major change in behavior like this first be deprecated through a round of versions so people have time to update their codebases. So, keep it but deprecate it now, then remove it after a suitable time period and with notice. Generally good behavior for software that you want people to use and build upon.
