Adam Rodger

What you're saying makes sense if your workflow is a trunk based development one, where you only have a main branch and feature branches, and only have one version at...

Looking into the Rustls code, it appears it's essentially expected that some certs in the store will be invalid. The parsing code contains a method with the comments: /// Parse...

I've written a sample app to parse my OS cert store in two different ways: one the way that `rustls-native-certs` will do it and another via `x509-parser`. The `x509-parser` version...

It's hard to say from looking at the code/logs provided, but at first glance it looks like you probably have some consumer version selectors defined which only match the `main`...

I'm looking to do the same thing. My plan at the moment is to remove the check that makes sure there's only one active release branch and then whenever a...

@petervanderdoes it appears you've reimplemented my changes from this PR and not given me any credit for it which seems a bit unfair to the community. I did the hard...

This is something we need to support but it would be a really big breaking change. The correct solution is to allow the user to choose the serialiser via a...

So, to demonstrate the problem, the `PactNet.Abstractions` library currently has a reference to Newtonsoft because matchers need to be serialised in a special format recognised by the FFI: https://github.com/pact-foundation/pact-net/blob/085ce4eb59470d4697708dfb7285dbc50f868274/src/PactNet.Abstractions/Matchers/IMatcher.cs `PactConfig`...

I don't think it's all that outdated (the general API usage is correct and the flow is right) but it definitely needs some work. Some problems I can see: -...

This is probably a defect in the FFI not reporting the failure back to PactNet. I'll have to check the FFI source.