Ask HN: What are some tools / libraries you built yourself?

Not sure if it counts, but I made two Roslyn analyzers for C# code. A Roslyn analyzer is basically a C# compiler plugin which can emit additional diagnostics if it encounters a certain pattern in the code being compiled.

One of them lets a developer mark a C# method such that the method must be invoked with named args [1], calling it with positional args is a compile time error then.

And the other makes it a compile time error to discard a marked method's return value [2]. Kind of like it is in functional languages.

Was learning/practicing F# along the way, so the code is probably not so good. But it works :)

[1]: https://github.com/mykolav/require-named-args-fs

[2]: https://github.com/mykolav/must-use-ret-val-fs

I built Backtest[0], the backtesting tool for European index investors, as a way to analyze the historical performance of my portfolio of index funds. US investors can use a tool called Portfolio Visualizer[1], which is really powerful and has tons of different analyses you can run on portfolios. But the funds and indexes that Europeans such as myself have access to are different. And then there's the currency you hold your investments in (USD vs EUR), which has a non-negligible effect on the historical performance.

So I built Backtest to scratch my own itch. It's been growing more or less organically for the past 1.5 years though and currently sees about 500 visitors per day.

I'm thinking about open-sourcing it but I've yet to thoroughly analyze the pros and cons, and decide on important things such as the license.

[0]: https://backtest.curvo.eu [1]: https://www.portfoliovisualizer.com/

i often am annoyed with the generality and complexity of existing solutions, and seek something simpler. additionally, that which i cannot build, i do not understand. to summarize: building things, for fun and profit.

agr: https://github.com/nathants/agr

like ag or ack, but with the ability to replace matches or delete lines.

tinysnitch: https://github.com/nathants/tinysnitch

a minimal system firewall for inbound/outbound connections with visual prompts on connections without an existing rule. fails closed.

cli-aws: https://github.com/nathants/cli-aws

a cli for working with aws, particularly ec2 and lambda, that dramatically simplify operations. currently exploring a go port.

bsv: https://github.com/nathants/bsv

a binary data format and composable tools for maximum performance data processing. several tools have triplicate implementations in go and rust for performance experiments. generative testing with small and varying buffer sizes ensure correctness. prototype with pipes, optimize with a single binary (see related posts in readme).

s4: https://github.com/nathants/s4

maximum performance distributed object storage with adhoc distributed computation and data shuffle. duplicate implementations in go and python, for fine grained objects prefer go, for medium and coarse grained, use either. use with ephemeral data and instance local storage (i3en). prefer vendor object storage for durability.

Some tools I built at my last job:

1. Pronto [0]: An "ORM" that mapped a protobuf to an object in MongoDB. The idea was to extend it to other storage backends (RocksDB, sstable, sqlite, etc) to make it easy for a service owner to decide what access pattern they needed and easily and efficiently map a proto to it. It wasn't too slow and was very intuitive for people to use since it only exposed a LevelDB-like API (Get(key), Set(key, value), List(filter)) and it made use of some of the newer features of MongoDB's libraries to do sorts and streaming on the client side. Really lowered latency and DB size requirements.

2. bazel_compose [1]: A tool to wrap Bazel, rules_docker, ibazel, and docker-compose to give developers a fast, hermetic, and automatically rebuilding/restarting local environment for development. You could essentially go to `http://frontent-a.localhost` in your browser and get routed to the right frontend.

3. branchpoke [2]: When I joined we had ~300 open branches per developer. People would hack something, push it to git, and forget about it. Some branches had critical fixes that were deployed before getting into master. Lovely startup things ;). I wrote this to send everyone a slack message once a week about 10 branches they had open that were over a month old in our monorepo. Over a few months we deleted most of the old branches.

Stuff I built at college:

1. Parking Analysis [3]: My college had an awful parking situation. I setup a cron to hit an undocumented API grab the number of spots open in the parking lots. I graphed those with a heatmap to find when our parking lot filled up around different holidays.

2. Butcher School [4]: A website used to "study" (memorize) the ARRL exam question set. It used some regex to parse the .doc file that is published by ARRL into it's json so anyone can create an updated set of questions. It also ran in local storage so you could study without internet on your laptop (like in the car on your way to Dayton). I built this for students who were having a tough time joining the Amateur Radio Club at NJIT (I was the treasurer). It's not the best tool by any means. It's just the most free (price and libre).

[0] - https://github.com/CaperAi/pronto/

[1] - https://github.com/CaperAi/bazel_compose

[2] - https://github.com/CaperAi/branchpoke

[3] - https://github.com/gravypod/NJITParking/

[4] - https://github.com/gravypod/ButcherSchool

Here I'm showing something I did some time ago https://meimporta.eu/htmleditor/ it's a crude WYSIWYG html editor for people that already knows HTML/CSS and uses basically the browser contenteditable capabilities. It was inpired by Webflow (https://webflow.com/) but very simple and basic, it has 4 main sections: editor shortcuts, wysiwyg editor, html editor, tree view. Some things are easy done in the wysiwyg editor but sometimes the raw html editor is better, we can edit in both of then and get instant result.

Don't bother comment on the UI/UX it's an personal naive tool straight to the point I needed. Anyway any feedback to improve is welcome.