use raw serializer for e2e output snapshots

[SimenB]

Summary

Noticed we had some inconsistencies in newlines on CI. Added a trim, but that looked pretty bad due to the " all over.

So I've added jest-snapshot-serializer-raw, which really cleans up the snapshots, IMO (should probably be default behavior in Jest... (#5660)).

An advantage is that now the snapshots actually match what the user sees, as we get rid of a lot of escaping

Test plan

Updated a lot of snapshots, hopefully green CI

[codecov-io]

Codecov Report

Merging #7651 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master    #7651   +/-   ##
=======================================
  Coverage   68.23%   68.23%           
=======================================
  Files         251      251           
  Lines        9639     9639           
  Branches        5        6    +1     
=======================================
  Hits         6577     6577           
  Misses       3060     3060           
  Partials        2        2

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 44054af...e076ea2. Read the comment docs.

[jeysal]

Wow, this looks like a huge DX improvement to me.
Would it be an option to make this automatic by having our e2e utils (runJest, extractSummary, ...) set a symbol property on the output strings and adding a serializer that detects those?

[SimenB]

I had that at the start, but we do quite a lot of .split and .replace here and there in the tests.

Adding a property doesn't work, as pretty-format itself will handle typeof 'string'

[jeysal]

I had that at the start, but we do quite a lot of .split and .replace here and there in the tests.

Implement String methods ourselves? :D Just kidding, if there's no simple way to automate it then it's probably not worth it as the extra effort required in each test seems manageable.
Just a bit worried that the e2e tests will inevitably become inconsistent with some using wrap and some forgetting it.

[SimenB]

Just a bit worried that the e2e tests will inevitably become inconsistent with some using wrap and some forgetting it.

For sure, I agree! Personally, I'd like some option in Jest that does not escape strings in snapshots (beyond what must be escaped in a template string for it to still be valid JS)

[github-actions]

This pull request has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.
Please note this issue tracker is not a help forum. We recommend using StackOverflow or our discord channel for questions.