| Summary: | REGRESSION (r182121): SVG animation macros cause warnings in MSVC | ||||||
|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Alexey Proskuryakov <ap> | ||||
| Component: | SVG | Assignee: | Alexey Proskuryakov <ap> | ||||
| Status: | RESOLVED FIXED | ||||||
| Severity: | Normal | CC: | darin, sabouhallawa, zimmermann | ||||
| Priority: | P2 | ||||||
| Version: | 528+ (Nightly build) | ||||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Attachments: |
|
||||||
|
Description
Alexey Proskuryakov
2015-04-01 12:00:39 PDT
This seems to be almost certainly a regression from <http://trac.webkit.org/changeset/182121>. I'm not sure how to observe this, or even whether there was any change in behavior with either clang or MSVC. Created attachment 249937 [details]
proposed fix
Comment on attachment 249937 [details]
proposed fix
Oops! We really need to add the “override” to the overrides of this function. That would have caused this to be a compilation failure rather than a subtle cause of failing tests.
Wait, no this is a big misunderstanding. The warning is telling you about a behavior change where the Microsoft tools development team fixed a bug in the old version of Visual Studio. The warning simply indicates that this code would have had different behavior *in an old version of MSVC*. I don’t care about this passionately one way or another, but there’s nothing wrong here at all. We could just turn off this warning instead. I think the title of this bug is misleading. I don’t know quite what to title it. It’s more appropriate to say “MSVC started warning, but there is no problem other than the warning”. Unless there is some symptom here other than the warning that I’m unaware of? Can’t imagine there is given my understanding of the situation. |