| Summary: | Use @NonNull and @Nullable annotations for IWorkbench API | ||
|---|---|---|---|
| Product: | [Eclipse Project] Platform | Reporter: | Lars Vogel <Lars.Vogel> |
| Component: | UI | Assignee: | Brian de Alwis <bsd> |
| Status: | RESOLVED WONTFIX | QA Contact: | |
| Severity: | enhancement | ||
| Priority: | P3 | CC: | daniel_megert, Lars.Vogel, loskutov, stephan.herrmann |
| Version: | 4.5 | ||
| Target Milestone: | --- | ||
| Hardware: | All | ||
| OS: | All | ||
| See Also: |
https://git.eclipse.org/r/62344 https://bugs.eclipse.org/bugs/show_bug.cgi?id=472631 https://bugs.eclipse.org/bugs/show_bug.cgi?id=540433 |
||
| Whiteboard: | |||
|
Description
Lars Vogel
(In reply to Lars Vogel from comment #0) > We should use @NotNullable for this API. @NotNullable ? Are you proposing new API to replace org.eclipse.jdt.annotation.NonNull, see https://wiki.eclipse.org/JDT_Core/Null_Analysis? Why not just use existing one? The (hidden) reason is: FindBugs also understands org.eclipse.jdt.annotation.NonNull but it will not understand some custom NotNullable annotation. (In reply to Andrey Loskutov from comment #1) > (In reply to Lars Vogel from comment #0) > > We should use @NotNullable for this API. @NotNullable > > ? Are you proposing new API to replace org.eclipse.jdt.annotation.NonNull No plan to introduce new API. I was planning to use the JDT annotations, just mistyped them as I did not remember there they are documented. Thanks for the pointer to the wiki. (In reply to Lars Vogel from comment #2) > [...] I did not remember there they are documented. Thanks > for the pointer to the wiki. See also bug 472631 comment 12 (official documentation is in the help) New Gerrit change created: https://git.eclipse.org/r/62344 I think we should first enable the Null analysis in our plug-ins before merging this change. See Bug 472631. Mass move to M6 Can this be done for M3? Yes! I'll announce on platform-ui-dev after M2. Mass move to M4. This was more difficult than I imagined as we have some inconsistent code ourselves. I'm going to take a different approach and start at JFace and move up with a series of commits. (In reply to Brian de Alwis from comment #10) > This was more difficult than I imagined as we have some inconsistent code > ourselves. > > I'm going to take a different approach and start at JFace and move up with a > series of commits. First usage was done via Bug 540433. Please reopen if you plan to work on this. |