Filter subsequent payloads when parent field is null

[robrichard]

Implementing what was discussed in graphql/defer-stream-wg#45

[netlify]

✅ Deploy Preview for compassionate-pike-271cb3 ready!

Name	Link
🔨 Latest commit	e0a539c
🔍 Latest deploy log	https://app.netlify.com/sites/compassionate-pike-271cb3/deploys/631b2bb4f150260008fcaed9
😎 Deploy Preview	https://deploy-preview-3720--compassionate-pike-271cb3.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site settings.

[github-actions]

Hi @robrichard, I'm @github-actions bot happy to help you with this PR 👋

Supported commands

Please post this commands in separate comments and only one per comment:

• @github-actions run-benchmark - Run benchmark comparing base and merge commits for this PR
• @github-actions publish-pr-on-npm - Build package from this PR and publish it on NPM

[benjie]

I don't think this is the approach I would have taken. Instead:

1. When a nullable position becomes null because of an error that has bubbled (no need to handle when the field itself throws an error), add this path to a new set on exeContext (let's call the new set errorNulledPaths since it's a list of the paths that were nulled out due to an error caught by a nullable type)
2. In filterSubsequentPayloads, find and cancel the subsequentPayloads whose path starts with or matches one of the paths in errorNulledPaths - no need to traverse the response data

Assuming that the length of paths is negligible, I think this basic approach would scale at O(N*M) where N is the number of errors and M is the number of subsequent payloads, but I'm sure there are clever ways to optimize it better.

[robrichard]

@benjie I significantly rewrote this. Now, as discussed in the WG today, I've added the logic to filter subsequent payloads at the point where the bubbled null is ultimately set. This no longer requires inspecting the payload response and should not add any overhead in the non-error case.

[benjie]

Definitely looks better; thanks @robrichard!