Open helloworld922 opened 8 years ago
So, was this closed because the fix was merged? Closed because it won't be fixed? @gennadiycivil -- can you please elaborate?
I have to say that I dont know. I have been doing a lot of cleanup, this was back in August and it is possible that I simply clicked the wrong button. Sorry about that Thanks G
On Thu, Jan 24, 2019 at 1:38 PM Chip Hogg notifications@github.com wrote:
So, was this closed because the fix was merged? Closed because it won't be fixed? @gennadiycivil https://github.com/gennadiycivil -- can you please elaborate?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/google/googletest/issues/890#issuecomment-457308244, or mute the thread https://github.com/notifications/unsubscribe-auth/AMJSMnXKSsCQBhNLqeWKda6ZQ2P-DSn5ks5vGf2OgaJpZM4KJGnT .
It seems there is an open pull request. Perhaps this issue should be reopened?
Re-Opening. Please read https://github.com/google/googletest/pull/2077#issuecomment-460729350.
Even though it looks like I shouldn't expect this to close any time soon, I'm still happy the issue is in the correct state. :slightly_smiling_face: Thanks!
Perhaps a good way to move this forward would be:
What do you think?
I am encountering similar issues with EXPECT_NEAR
. Looks like there was some recent discussion in https://github.com/google/googletest/pull/2077#issuecomment-783465831 about a possible fix -- might this be a way forward?
The EXPECT_NEAR and ASSERT_NEAR macros expect double parameters, which potentially means loss of precision or for types with greater accuracy than double, or compile errors for types not implicitly convertible to double (gcc/intel quad precision long double, Boost.MultiPrecision, etc.).
I've attached a patch which demonstrates a working example fix which generalizes this to work with arbitrary types as long as the required basic operations are available (can find an abs function, subtraction operator, comparison operator).
fp_fix.txt