# Accessing elements of a ComponentArray based on a variable

**URL:** <https://discourse.julialang.org/t/accessing-elements-of-a-componentarray-based-on-a-variable/119066>\
**Category:** General Usage\
**Created:** [September 5, 2024, 7:23am UTC](https://discourse.julialang.org/t/accessing-elements-of-a-componentarray-based-on-a-variable/119066 "2024-09-05T07:23:59Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![julien-arino](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/julien-arino/32/211710_2.png) [@julien-arino](https://discourse.julialang.org/u/julien-arino)\
**Post date:** [September 5, 2024, 7:23am UTC](https://discourse.julialang.org/t/accessing-elements-of-a-componentarray-based-on-a-variable/119066/1 "2024-09-05T07:23:59Z")

</div>

Warning: newbie here. Suppose I have something like

```julia
using ComponentArrays
p = ComponentArray()
p = ComponentArray(p; thing = 1)
tmp = ComponentArray{Int32}()
tmp = ComponentArray(tmp; a = 2)
tmp = ComponentArray(tmp; b = 3:4)
p = ComponentArray(p; other = tmp)

```

Is there a way to access `p.other` fields (or “top level” ones, for that matter) using something like `str="b"`, then `p.other.str`?

I am thinking of the analogy in `R`, where I would, in a list, access entries by using `p$other[[str]]` as opposed to `p$other$b` if I know that I want to access `b`.

I have a lookup table from which I am getting the field I want to use. I could work in the other direction (iterate on the keys of `p.other` and go find the info in my lookup table), but there are instances where my lookup table has 10 entries while my `p.other` has millions (my guess is that the keys are well indexed and quickly searched, so finding a key is fast).

So and also out of curiosity, I was wondering if there’s a way to do the equivalent of my `R` statement.

I have so far worked out that I could find the index of `b` in the keys of `p.other` using `findfirst`, but for that, I need to figure out how to convert `str` to a symbol… That will be my next question if there is no way to directly do some direct equivalent of the `R` statement.

---

<div class="post-metadata">

**Author:** ![dawbarton](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/dawbarton/32/215461_2.png) [@dawbarton](https://discourse.julialang.org/u/dawbarton)\
**Post date:** [September 5, 2024, 8:48am UTC](https://discourse.julialang.org/t/accessing-elements-of-a-componentarray-based-on-a-variable/119066/2 "2024-09-05T08:48:36Z")

</div>

The easiest way to do this is to reuse the machinery that Julia uses under the hood. Whenever you do `p.other`, Julia rewrites this as `getproperty(p, :other)` (note that `:other` is a symbol).

So for what you want, you should be able to do

```julia
sym = :b
getproperty(p.other, sym)

```

or if you want to work with strings (this will be a bit more inefficient because of the extra conversion)

```julia
str = "b"
getproperty(p.other, Symbol(str))

```

Alternatively, if you are often iterating over various properties, a (potentially - I haven’t timed this) faster approach to use the `label2index` function in `ComponentArrays` and store the indices you need for direct use.

```julia
other_labels = ["a", "b"] 
other_indices = [label2index(p.other, label) for label in other_labels]
@show p.other[other_indices[2]] == p.other.b # true

```

(Note rather than manually specifying all the labels, you can get all the labels using the `labels` function, but unfortunately it turns vectors into a list of their component scalars.)