Join on sq type fails

[DSLituiev]

Describe the bug
Join on sq type fails. I am trying to join two tables on sq column.

To Reproduce

tibble(sq=sq(c("CTGAATGCAGT", "ATGCCGT", "CAGACCATT"))) %>% inner_join(.,.)

->

Joining with `by = join_by(sq)`
Error: Unsupported type raw

Desktop :

• OS: MacOS 14.7.1 (23H222)
• R version 4.3.1 (2023-06-16)

[ErdaradunGaztea]

Hi, I remember trying to do the same (or maybe it was group_by()?) around the time we were developing {tidysq} - and it didn't work then either. Unfortunately, we've never figured out how to make {tibble} and related packages perform many such operations on raw vectors. Not sure if it's even possible without modifying {tibble} itself.

In retrospect, we could have decided to store sq objects as characters or integers instead of raw vectors, since the exact representation on R side is pretty much irrelevant - almost everything* happens in C++ anyways. Perhaps we were worried about extra storage overhead for characters/integers, perhaps we thought that raw vectors fit the intent the most.

As for the solution, I don't think we'll be able to make it work with {tidysq}; unfortunately, I can only reccomend casting these sequences to characters, which nullifies the key selling point of {tidysq}.