typer VS docopt

I have been using Typer on every one of my CLI projects which uses Click under the hood. The documentation is fantastic, the CLI app it produces looks great and lets you create things quickly. I high recommend it.

https://typer.tiangolo.com/

Adding CLI capabilities. My preferred library here is typer.

I just heard about this but it seems to be pretty much the type of thing you want and want fast.

Docstrings are for documenting how a function/ class/ method/ module works. Often you don't need to add a docstring to your main function because no one will be importing it to use elsewhere. And if you want it to run as a CLI, then there are better ways to document the available options. For example, typer does most of it for you, or in click you add the help text to the decorator.

Most computing tasks in airflow DAGs are KubernetesPodOperator containing a CLI (Python Typer). It allows us to pass arguments easily to run DAG manually if needed (the new UI to pass arguments to DAG in airflow 2.6 is really nice). Arguments allow us to replay DAG easily (change start / end dates for instance).

So it's basically the main .gitlab-ci.yml file plus a separate Python CI app using Typer for the AWS instrumentation.

Similarly for Typer, which is literally "the FastAPI of CLIs"[1]. Handy to type your `main` parameters and have CLI argument parsing. For more complicated cases, it's a wrapper around Click.

[1] https://typer.tiangolo.com/

interesting that you hate python, but love Click. Did you try Typer which uses Click underneath?

Symfony Console closely follows the well-established docopt conventions. Docopt, based on longstanding conventions from help messages and man pages, ensures a consistent and intuitive interface for describing a program's interface. Symfony Console's adherence to docopt conventions guarantees that your command line tools maintain a standardized and predictable user experience, simplifying development and user interaction.

You probably already know, but just in case you don't, you might read about http://docopt.org/ It seems to me a lot of your usage ideas could be refinements of / tooling around docopt-style interfaces.

http://docopt.org/

Not quite what you asked for, but close: type example invocations to generate the CLI, and just pull the arguments from a dictionary at runtime.

I've been using docopt to handle CLI arguments for years now.

http://docopt.org/

If you aren't averse to using a third party package, on my personal projects I always found https://github.com/docopt/docopt to be nice.

You can kill 2 birds with one stone by documenting your scripts while also providing the argument structure / parsing.

As for the implementation differences, complgen uses a trivial DSL that’s everybody is already familiar with more or less because it’s a slightly more rigorous version of what tools usually spit out when you do command --help (projects like docopt even use that for command line arguments parsing). Those happen to be regular languages and therefore can be represented as a Deterministic Finite Automata. complgen compiles the grammars to DFAs, minimizes the DFA and spits out shell-specific shell completions scripts that simply walk the DFA to match and complete the current input.

The current completion system has a list of rules of which completions to use at which time. It's purposely simple to make it as flexible as possible. The current things I'm planning is a derive macro like what clap has to generate these rules. I'm also considering introducing a plugin that let's you write rules in the format of docopt

For anyone unfamiliar, docopt is an established standard for specifying arguments in a script’s doc string. I use it for Python and it’s lovely. You’re going to write a docstring with examples anyway, why not make them functional?

http://docopt.org/

I like http://docopt.org/ a lot. You seem like someone who might have opinions on that.