refactor(hardhat): switch to ethers v6 #142
Conversation
Rubilmax
commented
Jul 18, 2023
Rubilmax
commented
- Fixes Fix never ending hardhat tests #132
- Fixes hardhat tests are all in the same block #127
- Fixes Fix hardhat "uninitialized provider" issue #19
Rubilmax
requested review from
QGarchery,
MerlinEgalite,
pakim249CAL,
MathisGD,
Jean-Grimal and
makcandrov
|
The CI is failing though |
Fixed in #141 |
|
#141 has been merged, the CI should pass with a rebase |
| @@ -70,24 +72,25 @@ describe("Blue", () => { | |||
| borrowableOracle = await OracleMockFactory.deploy(); | |||
| collateralOracle = await OracleMockFactory.deploy(); | |||
|
|
|||
| await borrowableOracle.connect(admin).setPrice(BigNumber.WAD); | |||
| await collateralOracle.connect(admin).setPrice(BigNumber.WAD); | |||
| await borrowableOracle.connect(admin).setPrice(0); // Make health check always pass. | |||
There was a problem hiding this comment.
Why do we need that now ? I tried reverting this change in local and it seems to pass
There was a problem hiding this comment.
indeed, people supply some collateral. I would revert the change as it might change the cost of the health check
There was a problem hiding this comment.
Sometimes, the test run fails because of not enough collateral. I'd say it happens to me 1/3 of the time.
There was a problem hiding this comment.
but isn't the test supposed to be deterministic ?
There was a problem hiding this comment.
Let's open a new issue and merge this PR ?
There was a problem hiding this comment.
I like this frictionless, productivity-oriented development process 😁😁
I'll dig this up though, as I believe it is in the scope of this PR and I'm just curious about the 2 points raised in this comment.
|
Unfortunately, local tests show that ethers v6 is on average 6x slower than ethers v5. The CI duration is thus increased by roughly 5x. So I'm against ethers v6 for now. |
|
For now, this work is replaced by #146. |
|
Can be closed no ? |
|
Closed but please not delete |
