I have already rolled back to 0.42 in my corporate machine and the world is beautiful again however I do have 0.43 on my personal machine so I would like to continue to give feedback.
With that being said - I do not see the long context option ( gemini 500k, sonnet 200k, gpt 128k etc ) in the AI chat in version 0.43.5 in my personal machine. Like I do not even see it. Is it somehow included in the regular options?
maybe it was too expensive for them?
Also chat with codebase is gone!
That was also often very missleading because you can chat with codebase tihe normal sonnet or with long context sonnet. But chat with codebase without long context - does that make sense?
maybe everything is used automatically now depending on how big your project is?
Anyway as cursor is a coding tool it is very important for us to know about such things.
I wish for more information/documentation about long context, chat with codebase, how cursorrules is implemented (only at the beginning of a chat, right?) and how notepads (on every submission it is injected, right?) are implemented.
We need to know such things.
I can imagine the cursor people beeing very very busy. But how can we really use cursor well, if we dont know how it works
I also noticed the option to always enable @web is gone as well. Dunno if that was intentional or maybe just a bug on my end as the docs still mention it being available as a setting
Ah I see. Well I have my Ubuntu programmed to revert back to 0.42.5 upon boot so I’m not too worried about it so I have it at least in my corporate machine.