Reproducer for S1905 FP: Floating point arithmetic #9520
Conversation
| void Sample() | ||
| { | ||
| float x = 3.14f; | ||
| float res = 1 / (float)(x * 2); // Fixed |
There was a problem hiding this comment.
We do not fix casts, only .Cast<T> and as T expressions.
| float x = 3.14f; | ||
| float res = 1 / (float)(x * 2); // Noncompliant FP |
There was a problem hiding this comment.
Can you find an example where the precision change is visible?
There was a problem hiding this comment.
I don't think it matters, as the solution would probably be what Jared suggested on the post.
Which is to be conservative and raise only in very few circumstances, when the RHS is a variable.
Even more conservative, we could also not raise on floating point arithmetic at all.
What do you think?
There was a problem hiding this comment.
I think we should document a case where this is actually an FP and think about a concrete solution later (or consider it a niche case and ignore it).
github-actions
bot
assigned gregory-paidis-sonarsource and unassigned Tim-Pohlmann and gregory-paidis-sonarsource
github-actions
bot
assigned gregory-paidis-sonarsource and unassigned gregory-paidis-sonarsource
|
|
Repro for #9498