std.format: Make formatting 0.0 100% @nogc. #8000
Conversation
|
Thanks for your pull request and interest in making D better, @berni44! We are looking forward to reviewing it, and you should be hearing from a maintainer soon.
Please see CONTRIBUTING.md for more information. If you have addressed all reviews or aren't sure how to proceed, don't hesitate to ping us with a simple comment. Bugzilla references
Testing this PR locallyIf you don't have a local development environment setup, you can use Digger to test this PR: dub run digger -- build "master + phobos#8000" |
| @@ -925,6 +925,12 @@ if (is(FloatingPointTypeOf!T) && !is(T == enum) && !hasToString!(T, Char)) | |||
| assert(format("%a", r) == "0x1p-20"); | |||
| } | |||
|
|
|||
| // https://issues.dlang.org/show_bug.cgi?id=21836 | |||
| @safe pure unittest | |||
There was a problem hiding this comment.
Not @nogc since some code paths in the called function allocate, even though those paths are not taken when called with the argument in the unit test.
There was a problem hiding this comment.
do we have an assertGCNotCalled test helper?
There was a problem hiding this comment.
There is a @nogc-test in std.format.internal.floats. I think, it can be made @nogc once the three followups on this are merged and formatting floats is completely @nogc, not only zeros, nans and infs.
|
|
||
| return result; | ||
| } | ||
|
|
||
| // check no allocations | ||
| @system unittest |
Followup on #7971; see there for the roadmap.
The calls to
printFloat0are replaced by a calls towriteAlignedwhich does the same more general and without allocating.writeAlignedneeded another parameter: fractional digits are yet another stretchable element due to trailing zeros. The interpretation ofprecisionfor floating point values differs from the interpretation of it for integral values. Therefore I added anenum PrecisionTypeto distinguish the various meanings ofprecision. Eventually it will replace the boolean parameterinteger_precision- for the time being I added an additional wrapper.