edge VS stripe-python

I totally respect that, and Clojure could invest more in offering frameworks or even no-code platforms or such features, but the truth is it doesn't. The language very much targets the software/information engineer category in my opinion, where by that I mean, the people who are interested in not just the functional requirements, but also the non-functional requirements of performance, scale, architectural runway, future extensibility, operations, maintainability, correctness, re-usability, etc. Especially, Clojure targets those who believe a balance between all these and functional requirements is the holy grail. That's why it won't be the most correct, the most performant, the most productive, but a pragmatic balance of all these in almost equal parts.

Maybe it should also embrace the people looking to get a product out by simply using a framework, and I'd say there's more of that in Clojure today than ever before, but the community I think is more composed of the former people that I describe, which is why you don't see any attempted framework take hold in the community, because most current members are not in the group that "just want to build the product using an established framework".

I think the community has settled, ounce again, on a bit of a balanced approach, Kit (https://github.com/kit-clj/kit) and Edge (https://github.com/juxt/edge) are such hybrids. And some more direct viable frameworks have come along like Biff (https://biffweb.com/) and Fulcro (https://fulcro.fulcrologic.com/).

That said, since the community is more composed of people like me, you don't see a mass move of every Clojurian switching to one of those.

So it creates some questions?

1. Is it a problem that the language targets engineers more interested in a balance between non-functional and functional?

2. Should it be mutually exclusive, or can Clojure equally serve both niche? And if so, should it, why?

3. Is the claim that you can be as productive and it is just as easy to build a product without using a framework in Clojure true? Does this apply to everyone, or only certain personalities or people with certain amount of lower level knowledge?

4. Is Clojure's marketing misleading? Are people looking to just "build the product using an established framework" mislead in thinking Clojure will offer them salvation?

5. Where do most developer fall in, if they don't fall in the category Clojure currently targets, than does that mean Clojure cannot become mainstream? To go mainstream does it mean you have to target frameworks because there are more developers looking to just make a product using a framework?

I don't have answers to these, I'm just trying to define the current state and what the problem with it might be, or if it even is a problem.

Make async requests in the beta release of stripe-python: The stripe-python beta channel now has built-in support for async control flow, backed by the httpx or aiohttp http libraries.

Instantiate StripeClient in the Python SDK: Stripe-python v8.0.0 introduced StripeClient, upgrading from a single global configuration to enable multiple thread-safe instances.

Let's start by installing the Python library for Stripe, which is available on Github.

Use types with the Python SDK: Strengthen Python integrations with type annotations using the stripe-python library starting in v6.8.0b3 through the beta channel.

You should use the python package that stripe provides: https://github.com/stripe/stripe-python

The api_key is being passed down to every object. You can see it being fed to the constructor for Customer because it is a child of StripeObject.

I've been maintaining async-stripe for over a year. It is the only async client for Python. It fixes a long standing feature request (issue #327).

New API version released: Version 2022-08-01 of the Stripe API has been released along with major version upgrades to all official client libraries, namely Dotnet, Go, Java, PHP, Node, Python, and Ruby. Read more about breaking changes in the API upgrades section and how to upgrade.

> Every time a new language comes out we have to recreate millions of baseline libraries and it just sucks. As a dev I want to be able to make use of great libraries oblivious to what they are created with.

Technically this tool mostly does exist already with the OpenAPI specification if we're talking about REST APIs. If you put in the leg work to create a very detailed specification which is a YAML file, you can generate programming language specific SDKs out of it.

Stripe has publicly mentioned[0] they mostly use this spec to generate their SDKs (even as of a few years ago), even so auto-generated code still requires developer time and there's a level of "hey we're dedicated to internally supporting this". It's a huge deal having a provider internally support your language's SDK.

[0]: https://github.com/stripe/stripe-python/issues/694#issuecomm...

The stripe package is actually maintained by Stripe.