I’m just wondering if the julia compiler can treat a expression like x in (y,) the same as x == y, assuming that the right argument for in is always a 1-sized tuple. Moreover, what would be the performance difference between these two pieces of code? Edit: typo in the title

Sure looks like it! julia> @code_native debuginfo=:none 1 in (1,) pushq %rbp .cfi_def_cfa_offset 16 .cfi_offset %rbp, -16 movq %rsp, %rbp .cfi_def_cfa_register %rbp cmpq %rcx, (%rdx) sete %al popq %rbp retq juli…

Is the julia compiler smart enough to treat `x in (y,)` the same as `x == y`?

Internals & Design

dylanxyz June 15, 2022, 7:36pm 3

Oh that’s cool! i guess there’s no performance difference then

2 Likes

Topic		Replies	Views
A metaprogramming question (sorry, can't find a more informative title) New to Julia	9	905	February 2, 2017
How to determine whether two variables refer to the same object New to Julia question	2	1561	August 16, 2017
Future-proof elementwise array comparison General Usage question	1	747	May 28, 2018
Benchmark showing 2x slower result between nearly identical functions on Julia 0.6 General Usage	4	638	January 27, 2017
Julia's in.() seems slow compared to R's %in% Performance	40	3160	May 9, 2019

Is the julia compiler smart enough to treat `x in (y,)` the same as `x == y`?

Related topics