Hi @ozonep how are you?
Thanks for the interest in closing #528 and for the great effort!

I checked out to your branch and it seems that there is much more work to do, so I was wondering if you were aware of it/are running predator locally while working on the feature to see this? Also you can run the tests locally to see the passing/failing ones.

As I said from my perspective I feel like the work left to do on this MR is substantial and just wanted to know if you had a specific motive of upgrading express to fastify or just for the overall improvement of predator?

Thanks again!

Hi @ozonep how are you?
Thanks for the interest in closing #528 and for the great effort!

I checked out to your branch and it seems that there is much more work to do, so I was wondering if you were aware of it/are running predator locally while working on the feature to see this? Also you can run the tests locally to see the passing/failing ones.

As I said from my perspective I feel like the work left to do on this MR is substantial and just wanted to know if you had a specific motive of upgrading express to fastify or just for the overall improvement of predator?

Thanks again!

Hello! )
I liked this tool & I believe it should gain more attraction and contributors!
it makes more sense to use Fastify/"got" to future proof Predator & reduce number of critical vulnerabilities.
I am open for negative feedback :-D
Right now I am working on updating tests.

I guess this should be released as v2.0.0 because of all the changes...
Anyway, I can keep this as just my fork of the project for some time.

@NivLipetz any ideas why test "call kafka produce with StreamingMessage when streaming manager initialized as kafka manager" fails sometimes because of date not being equal?
Example:
expected: "published_at":"2021-02-09T11:04:30.359Z"
actual: "published_at":"2021-02-09T11:04:30.360Z"

This happens ~1 time out of 10...

@NivLipetz any ideas why test "call kafka produce with StreamingMessage when streaming manager initialized as kafka manager" fails sometimes because of date not being equal?
Example:
expected: "published_at":"2021-02-09T11:04:30.359Z"
actual: "published_at":"2021-02-09T11:04:30.360Z"

This happens ~1 time out of 10...

yea it's an unstable test, i stubbed the prototype of Date constructor in order for it to always equal in the expectation but i guess it didn't work as i expected it to, I would ignore that failure in the unit-tests.

You can also fix the assertion to check a delta in the timerange where a 10ms difference can be "acceptable" and pass in order to make the test more stable. Here i see the idfference is 1ms