| Summary: | Fix -Wformat= warnings when passing enum to integer specifiers | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Lauro Moura <lmoura> | ||||||
| Component: | New Bugs | Assignee: | Lauro Moura <lmoura> | ||||||
| Status: | RESOLVED FIXED | ||||||||
| Severity: | Normal | CC: | calvaris, darin, eric.carlson, ews-watchlist, glenn, hta, jer.noble, philipj, pnormand, sergio, tommyw, webkit-bug-importer, youennf | ||||||
| Priority: | P2 | Keywords: | InRadar | ||||||
| Version: | WebKit Nightly Build | ||||||||
| Hardware: | Unspecified | ||||||||
| OS: | Unspecified | ||||||||
| Attachments: |
|
||||||||
|
Description
Lauro Moura
2020-07-16 20:00:16 PDT
Created attachment 404524 [details]
Patch
Comment on attachment 404524 [details]
Patch
Those all look good. But I wonder if we could use a specific integer type as underlying type for the enumeration instead.
Created attachment 404535 [details]
Patch using uint8_t as underlying type
The fast/mediastream tests worked as expected here with both enums (WebCore::MediaConstraintType and WebCore::MediaConstraint::DataType) using uint8_t as underlying type, but I'm not sure if there is any uncovered case where it could be an issue.
Comment on attachment 404535 [details]
Patch using uint8_t as underlying type
Seems OK. Awkward to have to write static_cast<int> out each time. Would have thought that an underlying type would help, and there would be some simple format string we could use, but not really sure how that all works with integer type promotion and all the PRIxx stuff.
Comment on attachment 404535 [details]
Patch using uint8_t as underlying type
I also thought it would end up not needing the casts but both clang and GCC complained.
Thanks for the review.
(Regarding the red bubbles, other builds from the same time also failed in the same tests)
Committed r264534: <https://trac.webkit.org/changeset/264534> All reviewed patches have been landed. Closing bug and clearing flags on attachment 404535 [details]. |