Making std.math.isIdentical work for real NaN when real is more than 64 bits requires either changing `is` for floats to be a binary comparison (like https://github.com/dlang/dmd/pull/7568) or adding support for CTFE reinterpretation of reals as byte arrays. This is a subissue of Issue #17007.
Actually it doesn't require this since in CTFE there is no way to construct a real nan with a payload that would be lost when casting to double.
@n8sh created dlang/phobos pull request #7515 "Make std.math.isIdentical work in CTFE" fixing this issue: - Fix Issue 20197 - Make std.math.isIdentical work in CTFE https://github.com/dlang/phobos/pull/7515
THIS ISSUE HAS BEEN MOVED TO GITHUB https://github.com/dlang/phobos/issues/10386 DO NOT COMMENT HERE ANYMORE, NOBODY WILL SEE IT, THIS ISSUE HAS BEEN MOVED TO GITHUB
@runiixx updated dlang/phobos pull request #10576 "Make std.math.isIdentical work in CTFE with coverage tests" fixing this issue: - Fix Bugzilla Issue 20197 Signed-off-by: runiixx <murgua03@gmail.com> https://github.com/dlang/phobos/pull/10576