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
If there's duplicate arguments (such as -m provided more than once) the argument parser currently uses the last argument provided without warning or error that the previous definition is ignored.
This cause me to think that multiple mocks were supported in a chain with priority order. They're not. But why not?
But if chaining of mocks was supported, one could avoid running ./tunnel twice on the client system. This should also cause fewer transitions in/out of kernel space
If there's duplicate arguments (such as
-m
provided more than once) the argument parser currently uses the last argument provided without warning or error that the previous definition is ignored.This cause me to think that multiple mocks were supported in a chain with priority order. They're not. But why not?
Currently it's possible to do:
which makes a chain:
But if chaining of mocks was supported, one could avoid running
./tunnel
twice on the client system. This should also cause fewer transitions in/out of kernel spaceOrder of mocks
I think there's a natural ordering of what makes sense, and this order could be used regardless of the order of things on the commandline.
Alternatively, argument ordering could be used to specify the order of the pipeline, such as putting socks5 inside of a https_ws_client
The text was updated successfully, but these errors were encountered: