he/him 
524 files, 40,000 lines of code, all linted in 3 seconds. Wall time.pic.twitter.com/uIUswozjRM
Show this thread
Cold cache. The time includes: - Node initialization overhead of 4 processes - A full crawl of the file system - Message passing overhead between processes - Parsing, linting, autofixing and formatting of EVERY file
-
-
That's still 23ms per file though, which isn't great. Hoping to drive that down even further.
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
How soon can we can a peek at what you’re working on???
-
No idea.
- 5 more replies
New conversation -
-
-
What sort of lint rules? Are some more intense than others, I'd imagine?
-
There are very few lint rules. But a lint rule is the exact same as a compiler transform. You have access to the same AST introspection, and can even return new AST nodes as a "fix". A majority of the overhead is already there. Should scale easily with more rules.
- 1 more reply
New conversation -
-
-
This is insane. Curious if you have some kind of faster format/data structure for IPC?
-
Depends on what process the communication is happening. For master <-> worker it's whatever IPC mechanism Node uses. For cli <-> master it's plain JSON with message length prefixes over a unix socket.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.