# Casting struct into an apparently identical type
**URL:** https://discourse.julialang.org/t/casting-struct-into-an-apparently-identical-type/108980
**Category:** General Usage
**Created:** [January 18, 2024, 6:10pm UTC](https://discourse.julialang.org/t/casting-struct-into-an-apparently-identical-type/108980 "2024-01-18T18:10:07Z")
**Posts on this page:** 3
**Page:** 1
### Author:  [@mike.ingold](https://discourse.julialang.org/u/mike.ingold)
#### Post date: [January 18, 2024, 6:10pm UTC](https://discourse.julialang.org/t/casting-struct-into-an-apparently-identical-type/108980/1 "2024-01-18T18:10:07Z")
Is it possible (even if hacky and not advisable) to do something like the following? Forewarning: this is a contrived example that just happens to work as an MWE exercise for my use case.
Let `Foo` be some type whose definition I can’t control and with a fundamentally-incompatible inner constructor.
```julia
struct Foo{T}
x::T
Foo(x) = error()
end
```
Is it possible to declare a shadow implementation a la
```julia
struct _ShadowFoo{T}
x::T
end
```
and then cast an instance of this type to the type being shadowed?
```julia
reinterpret(Foo{Float64}, _ShadowFoo(1.0))
```
This use of `reinterpret` doesn’t seem to work for me and I haven’t been able to find a different function that achieves this effect.
---
### Author:  [@mkitti](https://discourse.julialang.org/u/mkitti)
#### Post date: [January 18, 2024, 6:50pm UTC](https://discourse.julialang.org/t/casting-struct-into-an-apparently-identical-type/108980/2 "2024-01-18T18:50:40Z")
Reinterpret usually works on mutable arrays and may be implemented for some other mutable types.
We can accomplish what you want as follows.
```julia
julia> shadows = [_ShadowFoo(1.0)]
1-element Vector{_ShadowFoo{Float64}}:
_ShadowFoo{Float64}(1.0)
julia> reinterpret(Foo{Float64}, shadows)[1]
Foo{Float64}(1.0)
```
Here we managed to construct `Foo` despite having no constructor.
---
### Author:  [@mikmoore](https://discourse.julialang.org/u/mikmoore)
#### Post date: [January 18, 2024, 6:55pm UTC](https://discourse.julialang.org/t/casting-struct-into-an-apparently-identical-type/108980/3 "2024-01-18T18:55:23Z")
I think v1.10 added support for reinterpreting between `isbitstype` types. On v1.10, the original post’s code all works for me, concluding with
```julia-repl
julia> reinterpret(Foo{Float64}, _ShadowFoo(1.0))
Foo{Float64}(1.0)
```
I think it mostly uses the procedure that **mkitti** suggests, but it’s builtin and maybe does it in a way that is slightly more sophisticated and possibly more performant. See `@less Base._reinterpret(ComplexF32, 1.0)` for the source.