Create 2018-09-10--js-core-dev-team-weekly.md #695
Conversation
ghost
assigned 
|
|
||
| @alanshaw | ||
| - Done: | ||
| - Published release candidate for `js-ipfs` 0.32.0-rc.1 and 0.32.0-rc.2 |
There was a problem hiding this comment.
@alanshaw, releasing RC is an aegir step or is it done manually?
There was a problem hiding this comment.
Manual step at the moment until aegir supports it.
Details: ipfs/js-ipfs#1533
PR adding step to release doc: ipfs/js-ipfs#1536
Tracking issue in AEgir: ipfs/aegir#261
| @alanshaw | ||
| - Done: | ||
| - Published release candidate for `js-ipfs` 0.32.0-rc.1 and 0.32.0-rc.2 | ||
| - Test stabilisation work [#1541](https://github.com/ipfs/js-ipfs/pull/1541) |
There was a problem hiding this comment.
Nice! We should avoid the global flags though, they will hide tests that are slow when they shouldn't be.
| @achingbrain | ||
| - Done: | ||
| - Spun up npm registry at https://registry.js.ipfs.io | ||
| - Play along at home with `npm i --registry=https://registry.js.ipfs.io` |
There was a problem hiding this comment.
Wooot!! :D This is nice :D
@achingbrain, Does it have the whole registry? What's the repo size?
There was a problem hiding this comment.
At the moment it's configured to only fetch individual tarballs when they've been requested by a client and aren't present locally as the EC2 instance it's running on doesn't have enough storage to eagerly fetch the whole thing.
This issue is to sort out a more scalable storage solution, once that's in place we can hold whole the whole repo: ipfs/infra#432
There was a problem hiding this comment.
Did you have the chance to download the entire registry locally to see the size it took?
| - Done: | ||
| - Spun up npm registry at https://registry.js.ipfs.io | ||
| - Play along at home with `npm i --registry=https://registry.js.ipfs.io` | ||
| - Fixed flaky pubsub tests https://github.com/ipfs/interop/pull/33 |
| - Done: | ||
| - IPNS over Pubsub OKR | ||
| - Created `js-datastore-pubsub` module | ||
| - Initial implementation of [ipfs/js-datastore-pubsub#1](https://github.com/ipfs/js-datastore-pubsub/pull/1) |
There was a problem hiding this comment.
@vasco-santos, This design is incorrect. We don't want a datastore to publish every key that gets stored.
Do you have a diagram of how things get plugged together? The IPNS system (module) should use PubSub to publish and subscribe to the new keys and then make the decision of storing them in the DataStore.
@vasco-santos do you have a diagram of your architecture? We can workshop this together if you need.
| - Blocked: | ||
| - IPNS | ||
| - [interop#26](https://github.com/ipfs/interop/pull/26) | ||
| - Needs js-ipfs new release with IPNS, and to be reviewed |
| - [#1540](https://github.com/ipfs/js-ipfs/pull/1540) | ||
| - ipfs-api chunked add | ||
| - [#851](https://github.com/ipfs/js-ipfs-api/pull/851) | ||
| - [stream-http/tree/fix/body-handling](https://github.com/hugomrdias/stream-http/tree/fix/body-handling) |
There was a problem hiding this comment.
@hugomrdias Does this break/change interop with go-ipfs?
|
|
||
| @travisperson | ||
| - Done: | ||
| - iptb [go/js/browser demo](https://asciinema.org/a/zn07AUFgZ5RWzAwit7mjpdqvW) |
|
|
||
| @vmx | ||
| - Done: | ||
| - Conference (FOSS4G) and travels in Tanzania |
There was a problem hiding this comment.
It's not easy to link to it, here's the best one I could find: https://fkih7q.attendify.io/schedule.html#9YYK6haD49GaQsORLc
jacobheun
added
⚡️ⒿⓈ Core Dev
Notes for the Sep 10 js core team weekly!