Prioritize errors in response

[nicholascioli]

Description here

Previously, the router would always respond with data first, and then show any errors in the request. The GraphQL spec suggests having the errors show up before data, so this commit makes errors output first before data.

As expected, this change updates a LOT of snapshot tests to move the errors up. Care was taken to make sure that the only change was the order of the output, but please double check during review.

Fixes #4375

---

Checklist

Complete the checklist (and note appropriate exceptions) before the PR is marked ready-for-review.

• [ ] Changes are compatible[^1]
• [ ] Documentation[^2] completed
• [ ] Performance impact assessed and acceptable
• Tests added and passing[^3]
  • [X] Unit Tests
  • [X] Integration Tests
  • [ ] Manual Tests

Exceptions

Note any exceptions here

Notes

[^1]: It may be appropriate to bring upcoming changes to the attention of other (impacted) groups. Please endeavour to do this before seeking PR approval. The mechanism for doing this will vary considerably, so use your judgement as to how and when to do this. [^2]: Configuration is an important part of many changes. Where applicable please try to document configuration examples. [^3]: Tick whichever testing boxes are applicable. If you are adding Manual Tests, please document the manual testing (extensively) in the Exceptions.

[github-actions[bot]]

@nicholascioli, please consider creating a changeset entry in /.changesets/. These instructions describe the process and tooling.

[router-perf[bot]]

CI performance tests

• [ ] reload - Reload test over a long period of time at a constant rate of users
• [ ] events_big_cap_high_rate_callback - Stress test for events with a lot of users, deduplication enabled and high rate event with a big queue capacity using callback mode
• [ ] events_without_dedup_callback - Stress test for events with a lot of users and deduplication DISABLED using callback mode
• [ ] large-request - Stress test with a 1 MB request payload
• [x] const - Basic stress test that runs with a constant number of users
• [ ] no-graphos - Basic stress test, no GraphOS.
• [ ] step-jemalloc-tuning - Clone of the basic stress test for jemalloc tuning
• [ ] events - Stress test for events with a lot of users and deduplication ENABLED
• [ ] events_callback - Stress test for events with a lot of users and deduplication ENABLED in callback mode
• [ ] events_big_cap_high_rate - Stress test for events with a lot of users, deduplication enabled and high rate event with a big queue capacity
• [ ] events_without_dedup - Stress test for events with a lot of users and deduplication DISABLED
• [ ] xxlarge-request - Stress test with 100 MB request payload
• [ ] xlarge-request - Stress test with 10 MB request payload
• [x] step - Basic stress test that steps up the number of users over time

[abernix]

I'd be interested in talking about this with the client teams a bit to get a full understanding. I feel like if we had to update all our snapshots then all users are also going to have to update their snapshots which could just raise eyebrows and require scrutiny. Some existing clients could also be making parsing assumptions about where items in the data structures will appear in the streams.

As I recall, the entire original motivation for the spec recommendation (and bringing it to the working group) was to make it so that errors appeared at the top of the "Human User interface showing JSON" which defaults to being scrolled to the top of the frame. I can appreciate this convenience.

However, it wasn't considered how this would impact users who are running terminal commands where the errors would no longer be at the bottom (that is at the bottom of the, for example, curl command output).

Combined with the potential impacts I mentioned above, how about we talk next week with some folks and figure out where this fits? For example, maybe content type negotiation is the place to have our eyes.

Wdyt?