GraphQL queries running in parallel and in their own context

[joelgriffith]

Really like this idea. Some thoughts I have:

• No clear way to "clone" a tab in Chrome...
• Would probably need to have some limitation on the amount of parallelism in requests.
• Technically might be challenging to orchestrate all of it due to the async nature of the API and how all of these pieces interact.

[mute]

Totally agreed with those concerns -- here's my hand wavy explanation of how I think this might be possible, but it's hand wavy because I'm not a GraphQL guru:

1) Parse GraphQL query into an array of Promise chains which resolve to the sequence of actual navalia / chrome API functions. This should result in n Promises where n is the number of leaf nodes in the query. 2) For each of these Promises, spawn a tab up to whatever the maximum specified number of tabs is. 3) Resolve each promise chain in its corresponding tab. 4) Continue until tabs have been spawned / resolved for each leaf node.

So, we shouldn't need to clone tabs, each tab will just execute the same series of steps in parallel up until the query branches.

[joelgriffith]

Ah, that definitely removes some of the challenges related to having branching work as expected. I know this is entirely possible (things like join-monster do this), but would definitely take some meaty work to do so. I'll dig into how to do this, as it's a pretty interesting idea!