You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Apr 23, 2019. It is now read-only.
One feature of SublimeJedi which is very nice is that when you reopen a project, it seems that the autocompletes are cached from the previous time. For a big project this is very important. Currently when I launch atom I am waiting about 30 seconds to a minute at 100% CPU load before I am able to autocomplete. I'm not sure where this cache is being kept in SublimeJedi, I guess it is also possible that it is indexing several orders of magnitude faster.
The text was updated successfully, but these errors were encountered:
I think that if we want to speed it up -- we should have some manual action which will scan the whole project and cache it. Right now the cache is populated "on-demand" so every time when you make a first call for some bog modules -- you will need to wait a lot. But after it works pretty fast.
The problem is that it's hard to determine what the 'project' even is. Atom does not have projects the way sublime has them. I thought about ascending directories until I cannot find any more init.py's, but that feels kinda hacky.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
One feature of SublimeJedi which is very nice is that when you reopen a project, it seems that the autocompletes are cached from the previous time. For a big project this is very important. Currently when I launch atom I am waiting about 30 seconds to a minute at 100% CPU load before I am able to autocomplete. I'm not sure where this cache is being kept in SublimeJedi, I guess it is also possible that it is indexing several orders of magnitude faster.
The text was updated successfully, but these errors were encountered: