Mailspring/internal_packages/participant-profile/package.json

{
  "name": "participant-profile",
  "version": "0.1.0",
  "title": "Participant Profile",
  "description": "Information about a participant",
  "isOptional": true,
  "main": "lib/main",
  "windowTypes": {
    "default": true
  },
  "dependencies": {
    "clearbit": "^1.2"
  },
  "license": "GPL-3.0"
}
feat(sidebar): Add thread list of currently selected participants Summary: WIP. I added a collection index to make displaying the threads of a currently selected participant on the sidebar easy and fast. The problem is that the `participants` of a thread, while a collection of `Contact` objects, have no "ids" for those contact objects. One idea was to create the join table but access contacts by email instead of id. This required a minor change to the way the data is entered in the join table. This means the sidebar can now simply do: `DatabaseStore.findAll(Thread).where(Thread.attributes.participants.contains('foo@bar.com'))` While I didn't for this initial test, we could also/instead create the `Message-Contact` join table. The trick about a Message-Contact table is that I believe we'd have to create additional columns further specifying which field we're interested in. The following two queries: `DatabaseStore.findAll(Message).where(Message.attributes.to.contains('foo@bar.com'))` `DatabaseStore.findAll(Message).where(Message.attributes.from.contains('foo@bar.com'))` would require additional columns in the `Message-Contact` join table because currently the only columns are `id` and `value`. In the case of the sidebar use case, I think the Thread participants is what you want to see anyway. Unfortunately an email-centric scheme can't distinguish between `noreply@phab.com <Evan>` and `noreply@phab.com <Juan>`. I actually think this may be a good thing since I think most people think in terms of email address as the unique key anyway and for the use case of showing related emails in the sidebar I'd rather overshow than undershow. This solution seems to be working pretty well in initial testing, but I want to see if you guys can think of anything this may subtly screw up down the line, or if you can think of a simpler way to do this. Test Plan: todo Reviewers: juan, bengotow Reviewed By: bengotow Differential Revision: https://phab.nylas.com/D2687 2016-03-10 03:33:31 +08:00			`{`
			`"name": "participant-profile",`
			`"version": "0.1.0",`
			`"title": "Participant Profile",`
			`"description": "Information about a participant",`
			`"isOptional": true,`
			`"main": "lib/main",`
			`"windowTypes": {`
			`"default": true`
			`},`
			`"dependencies": {`
			`"clearbit": "^1.2"`
			`},`
			`"license": "GPL-3.0"`
			`}`