bullseye-sample VS Marten

We will take a simple sample .NET project available in GitHub which has the following setup and then attempt to add a build script for it:

Check out Marten for a fully fleshed out implementation https://github.com/JasperFx/marten

I always loved Marten, it is so simple to use and yet powerful. If you are unfamiliar with it, it is a data access library (like an ORM) that is using JSON serialization and LINQ to store and query data from/to Postgres. It basically turns Postgres into document DB. Comparing it to EF, Marten doesn't require migrations since it stores documents.

For that part, the samples use EventStoreDB (https://www.eventstore.com/), which is the only mature event store I know in Node.js land. Event Sourcing allows using any database as backing storage. I'm co-maintainer of the Marten (https://martendb.io/), which is a .NET library that allows using Postgres as event store and document db.

Then you should check out Marten (https://martendb.io/). Our intention is to remove the boilerplate, we're using Postgres e having the built-in projections.

Not to be confused with the C# document database built on Postgres.

https://martendb.io/

Haven't tried it myself yet, but they were in .NET foundation, them left after some drama: https://martendb.io/

Thanks for the feedback and the link! Indeed that looks similar. I think that too many NodeJS tooling is made by C# or Java devs bringing their OOP habits. In my opinion, TypeScript and NodeJS development shines the most when you're cutting all the boilerplate and keeping things simple. Functional composition helps in that. I'm also coming from the .NET world and am co-maintainer of the Postgres-based library: https://martendb.io/, but I try to keep my OOP habits on a leash ;) Out of curiosity, how are you dealing with the eventual consistency and the read-your-own-writes guarantee on MongoDB in your library?

I personally advocate Bullseye in combination with SimpleExec as a build tool for .NET projects due to its sheer simplicity by providing a bare bones API to define the tasks and their dependent tasks in plain C# code. Both these libraries are build by Adam Ralph. It does not enforce any specific model or structure to write your build tasks. When someone runs through the build script, it is fairly straight forward to understand whats going on. We use this build tool setup for Marten for both local dev and CI build tasks and has worked well so far. Also many OSS projects have adopted it.

I have been a co-maintainer of .NET OSS library Marten and largely involved in taking care of the DevOps stuff pertaining to CI pipeline, build and docs system. In the early days, the build was driven using a RAKE build script written by Jeremy Miller, our lead dev and creator of Marten. NuGet packages were published from the local dev machine after confirming that we had bumped up the versions of the packages in preparation for the release. Apparently, this being a manual step, we publish the NuGet packages but sometimes tend to forget tagging and creation of release in GitHub (you know how an OSS author life is around managing day job and doing open source work).