new aegir #1
new aegir #1
Conversation
|
things were passing for me last time I checked |
|
@dignifiedquire is this something you know what's going on or that requires more debugging? |
|
@diasdavid need probably the old timeouts of |
|
@dignifiedquire still failing |
|
Those failures look like issues with the tests/code rather than aegir |
|
@dignifiedquire what surprises me is that aegir 12.0.0 got merged and released without having it fully passing in this repo. This might mean that JEST is not suited yet to our type of testing. If there are bugs with the tests lets fix it, however, we should not lightly release aegir without testing it extensively in this repo. |
|
I was not under the impression that releasing a version of aegir has the requirement of this repo passing on that version. For me that was simply a nice way to test certain things. The current version working well on a number of repos, the test on js-ipfs and js-ipfs-api have been very brittle for some time, so while it is very possible there is an issue in aegir and its components I am not convinced that should have held up the release making it harder and more problematic to test. |
Ok. It is a great test to ensure that new versions of aegir work well for our set up of tests. That is why this repo is a clone of js-ipfs-api because it has one of the most complicated and complete test setups we have built. Let's get everything working :) Happy to help. |
|
The latest version of this branch together with ipfs-inactive/interface-js-ipfs-core#151 passes all node test on my local machine |
|
update all browser tests just passed locally as well |
| "test": "aegir test", | ||
| "test:node": "aegir test --target node", | ||
| "test:browser": "aegir test --target browser webworker", | ||
| "test": "aegir test --no-parallel --timeout 80000", |
There was a problem hiding this comment.
Which tests still require increased timeout?
There was a problem hiding this comment.
not sure..ci was just being slow and I didn't want to spend an hour on going test by test on ci
|
and webworker tests passed as well |
|
As part of testing aegir with this repo, every release must get every single feature tested, from testing, to coverage to release. Right now ipfs-inactive/js-ipfs-http-client#585 is being super painful and there is incoming changes to interface-ipfs-core that I'll have to release in patch versions of the previous minor so that it works with old aegir if we don't get new aegir to work. |
|
solved with #3 |
@dignifiedquire I'm seeing the same errors I'm getting in ipfs-inactive/js-ipfs-http-client#585 (comment) on this test repo with the latest aegir.
Did you get to test the new aegir before merge and release? This repo should always be tested before shipping a new version as it is a very solid way to ensure that it works with our projects.