Sigurd Spieckermann

> Note that when updating, the `copy` message might be displayed before the `update` one, and not _instead_ of, though that's coming from tasks, so maybe a tighter integration can...

Thanks for the clarification! :+1: Yes, a tighter integration should be able to control the printing better. There's actually another point about message printing. Should the `-q, --quiet` flag suppress...

I've submitted a PR that should fix this problem, @mhoeger @pQraus. See #1069.

Perhaps a little late since the v2 roadmap has been set, but possibly still relevant/helpful for the future development of Pydantic: There's a state management library implemented in TypeScript called...

@HappyDingning I think the `context` attribute should be a [private model attribute](https://docs.pydantic.dev/usage/models/#private-model-attributes), so its value doesn't get (de)serialized and it doesn't appear in the schema. WDYT?

Or maybe not ... :thinking: Private model attributes cannot be set upon model instantiation. Still, the problem remains how to omit the context from (de)serialization and the schema.

I wonder whether we actually need to require a Copier template to contain a file `{{ _copier_conf.answers_file }}.jinja`. It's always the same anyway, so why burden the template creator with...

I think neither `_updatable: true` nor the answers file template are needed, instead a new CLI flag could be introduced which is less intrusive IMO. The different ways of generating...

Opt-in vs. opt-out is a fair point. I'm not opinionated at all. The current behavior is fine for me.

I still find the `{{ _copier_conf.answers_file }}.jinja` file redundant and have thought of a backwards compatible opt-in approach to remove it: * If a template has the `{{ _copier_conf.answers_file }}.jinja`...