Bugzilla – Full Text Bug Listing
| Summary: | [hotbug] @regression: DefaultInferenceProvider is always enabled | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [WebTools] JSDT | Reporter: | Victor Rubezhny <vrubezhny> | ||||
| Component: | General | Assignee: | Victor Rubezhny <vrubezhny> | ||||
| Status: | RESOLVED FIXED | QA Contact: | Chris Jaun <cmjaun> | ||||
| Severity: | blocker | ||||||
| Priority: | P3 | CC: | cbridgha, cmjaun, david_williams, kaloyan, naci.dai, neil.hauge, paul.verest, raghunathan.srinivasan, thatnitind, vrubezhny, zulus | ||||
| Version: | 3.5.1 | Flags: | vrubezhny:
pmc_approved?
(david_williams) vrubezhny: pmc_approved? (raghunathan.srinivasan) vrubezhny: pmc_approved? (naci.dai) vrubezhny: pmc_approved? (neil.hauge) cbridgha: pmc_approved+ vrubezhny: pmc_approved? (kaloyan) cmjaun: review+ |
||||
| Target Milestone: | 3.5.2 | ||||||
| Hardware: | PC | ||||||
| OS: | Mac OS X | ||||||
| Whiteboard: | PMC_Approved | ||||||
| Bug Depends on: | 419871 | ||||||
| Bug Blocks: | |||||||
| Attachments: |
|
||||||
|
Description
Victor Rubezhny
The issue is the regression made by fixing bug [384319] (CPU utilization to 100% when this is appended to new object definition) in commit ec87a3b8e0a51fc2baec20756c00429d9e68a587 This regression was found (see [419871] - DefaultInferenceProvider is always enabled) and fixed in master in commit 802e0ab714f7e3f9bf852ce72689628b218e9e62 The fix for Luna WTP 3.6 was provided and tested by Jason Peterson <jasonpet@us.ibm.com> 2013-07-15 23:11:57. Backport to WTP 3.5.2 for the fix is requested by the adopters. Explain why you believe this is a stop-ship defect: * It's not possible for adopters to remove InferredType from CompilationResult, when new InferrenceProvider is added using InferrenceProvider.ONLY_THIS option. Is there a work-around? If so, why do you believe the work-around is insufficient? * No, DefaultInferrenceProvider to InferrenceManager is is always added to InferrenceManager in addition to any Inferrence Provider implemented by an adopter that is added using InferrenceProvider.ONLY_THIS option. How has the fix been tested? Is there a test case attached to the bugzilla record? Has a JUnit Test been added? * Tested with JSDT JUnit tests on R3_5_maintenance branch. All tests are passed with no errors. Give a brief technical overview. Who has reviewed this fix? * The adding of DefaultInferrenceProvider to InferrenceManager is hardcoded in line 112 of InferrenceManager, so it's always added to the new Inferrence Provider (from adopters). What is the risk associated with this fix? * Low. All JUnit Tests are good after applying the patch in R3_5_maintenance branch. Smoke testing shows that Content Assist/Validation/Outline View works as expected when DefaultInferrenceManager is used. Created attachment 239354 [details] Patch that backports the fix for Bug 419871 to R3_5_maintenance branch Thanks for pushing this forward, Victor. You can mark for PMC approval. Thanks for the fix... Pushed to 3.5.2: http://git.eclipse.org/c/jsdt/webtools.jsdt.core.git/commit/?h=R3_5_maintenance&id=cdef9cb517ebb5befcdee683ced8d38f69b17d0d Thanks for the fix! |