Werkzeug VS attrs

The angle brackets on their own should be a sign that this is not the case, because that's not regular expression syntax. Flask uses werkzeug.routing instead. In fact, the module docs specifically call out the difference:

Flask doesn’t support ASGI; neither does Werkzeug. The official guidance for running Flask with an ASGI server is to wrap it in middleware. The open GitHub issue for Werkzeug to support ASGI is #1322.

But this may be a starting point: https://github.com/pallets/werkzeug/blob/main/src/werkzeug/middleware/http_proxy.py

> In this case, he initially presumed that the problem would be with Flask, Bottle, and Tornado still honoring a semi-colon as a separator in a uri

Flask (or rather Werkzeug what it's based on) removed implicit semicolon support for URL 13 years ago: https://github.com/pallets/werkzeug/commit/0ea28bbc6f5f05eef...

Werkzeug (library used by Flask) had a repr for its sentinel long before getting type annotations.

Flask getting async support is a huge step for the asyncio web ecosystem in Python. I wonder how Flask's ASGI support will evolve in parallel with Quart[1] which bills itself as the ASGI version of Flask and is maintained by a member of the Pallets team. One of the maintainers of Werkzeug even recommended using Quart[2]

[1] https://pgjones.gitlab.io/quart/

[2] https://github.com/pallets/werkzeug/issues/1322#issuecomment...

I’m pretty sure you need to establish a ‘werkzeug’ logger in a logging configuration and make sure that it has a handler for its effective level. But that’s just what I got from reading the source code. I could very well be wrong

You're not able to change the loglevel with werkzeug. It's hardcoded to info: https://github.com/pallets/werkzeug/blob/72b2e48e7d44927b1b7d6b2f940d0691230de893/src/werkzeug/serving.py#L388

The PRs are this initial one and this update. Also note this issue - almost 5 years to close :o.

Full support for validation and serialisation of attrs classes and msgspec Structs. Where previously only Pydantic models and types where supported, you can now mix and match any of these three libraries. In addition to this, adding support for another modelling library has been greatly simplified with the new plugin architecture

Pydantic is by far not the only library of its kind, with prominent members of the same class being attrs, cattrs or even plain dataclasses for some use cases.

Anything requiring e.g. setattr, getattr, delattr? Without looking far,

https://github.com/python-attrs/attrs/blob/main/src/attr/_ma...

I was looking for an example of using locals() to "fill a data class from kwargs" or something similar to that. The example here doesn't use locals().

That aside, I generally wouldn't use the kwargs approach shown in this example either. I'd use [dataclasses](https://docs.python.org/3/library/dataclasses.html ) or [attrs](https://www.attrs.org/) instead.

hynek: developer of attrs

I continue to be surprised how easy it can be to consume a SOAP API with the right client libraries. Such as https://docs.python-zeep.org/en/master/ for Python. Now that's not to say it will always work, you can design a terrible API with any mechanism, no SOAP or REST client will help you if the other end has desided to succumb to madness and done something like turn their entire API into just "two endpoints" and driven by the payload content you post to the inbound endpoint, and you have to sit there polling the outbound endpoint with the inbound endpoints response ID because to find out what the eventual response is...

But horror story aside, consuming a decent SOAP endpoint with a good client library can be practically magical.

Between attrs (https://www.attrs.org/), cattrs (https://cattrs.readthedocs.io/), and the aforementioned zeep soap client I've got a serialisation pipeline from soap endpoint into an attrs dataclass with type hints and basic type validation down to a snippet so small it fits right here (type hints removed to minimise size).

  from zeep import helpers

attrs has at least two.