Push from aleph -> concat #147
Conversation
|
okay, I think this is in decent shape, although maybe we should merge #146 first, since this branches off of it. I could also do a bit of squashing to sand off some of the edges in the commit history 😄 The last commit adds an The node's publisher id is currently optional, which isn't ideal... it needs to be generated asynchronously, so you can't make one in the constructor. And I didn't want to make it mandatory since we'll need to update existing usages. The ingest command will fail with an error if it's not set. |
| .then(() => concatNodeClient()) | ||
| .then(concat => concat.authorize(alephPeerIdB58, ['scratch.*'])) | ||
| .catch(err => { | ||
| console.error('error during push setup:', err) |
There was a problem hiding this comment.
do we really want to catch this?
There was a problem hiding this comment.
ah, sorry bout that! I stuck that in when things were failing during setup so I could print a stack trace - need to remove it. well spotted :)
| this.db = Levelup(levelOpts) | ||
| const location = (options.location == null || options.location === '') | ||
| ? '/aleph/data-' + uuid.v4() | ||
| : options.location |
There was a problem hiding this comment.
const location = options.location || '/aleph/data-' + uuid.v4()
Unless you want number 0, NaN and false to be valid locations?
(there's a moment of wondering if js messes up the precedence in this but it's actually OK)
yusefnapora
force-pushed
the
yn-aleph-push
branch
from
69b0763
to
eb1aa80
Compare
|
OK is this ready for final review? |
|
I think so; although I realized I don't have a test for a partially-successful push (some valid results + error). I think I'll add that real quick; hopefully it'll work as expected :) |
| function seedStatementsToAleph (alephNode: AlephNode): Promise<Array<string>> { | ||
| return Promise.all( | ||
| seedObjects.map(obj => | ||
| alephNode.ingestSimpleStatement('scratch.test', obj, { refs: [obj.id] }) |
There was a problem hiding this comment.
one comment here would be towards separating the test flow and the normal flow, i.e. not using ingest* functions to seed test data, but this is probably a separate discussion around testing methodology
quick summary: