Version 1.0 - Released

[dabeaz]

I have released a version 1.0 of Curio. The overall core of the library is basically the same as before, but if you were using any its more experimental features, you might find them gone (sorry). This release also represents a general change in focus for the project. Early versions of Curio were very much about having a playground for trying out different ideas related to async programming. However, that experiment has now ended. Instead, the 1.0 release represents a desire to simply have a no-nonsense async library that's small, fast, and understandable. I'd also like it to be something stable that I can use in other projects.

I do not anticipate further major features being added to Curio. That said, I do intend to fix bugs and to maintain it so that it continues to work with new versions of Python. Thanks for your support.

[Tronic]

Is there a reason why you wouldn't just use Trio? It seems to have similar goals and everyone could benefit of a joint development effort in the already fragmented Python async world.

[nskalis]

thank you :) I am in love with curio

[dabeaz]

Despite a similar name and the use of "await", Trio is not an embodiment of what I was trying to build with Curio. So no, I'm not especially interested in using it. I wrote Curio to be something that I would want to use as a programmer. If other people want to use it, that's fine. If not, that's also fine. The world is a big place.

[imrn]

Is there a reason why you wouldn't just use Trio?

It's complicated. ;)

It seems to have similar goals and everyone could benefit of a joint development effort in the already fragmented Python async world.

Did you ever mentioned this on trio issues or lists?

Some history: Trio project owner once had some 'sophisticated' async ideas that could not find appropriate place in curio. So, he started his own project. This is what you call fragmentation I guess.

It's still unclear whether those ideas created any value that worth the hassle and complexity.