Make gix-transport compile with all features #2241
Conversation
djc
force-pushed
the
transport-all
branch
4 times, most recently
from
a336dfa to
564e885
Compare
Byron
left a comment
Byron
left a comment
There was a problem hiding this comment.
Thanks so much, and sorry for the delay!
The most notable change that I did was the renaming of Outcome::from_lines…() to Handshake, which seemed like a better fit.
Now I don't know if this is a mistake because it introduced ambiguity elsewhere, but if you encounter it, I am sure you'd be most qualified to find better names.
|
|
||
| ## If set, blocking implementations of the typical git transports become available in `crate::client::blocking_io` | ||
| ## | ||
| ## If used in conjunction with an async implementation, this one takes precedence. |
There was a problem hiding this comment.
This might also be something to double-check @djc, I hope it's correct.
There was a problem hiding this comment.
I think the async and blocking implementations are fully parallel -- I don't remember seeing any code where either one is prioritized, and I don't intend to introduce any such code.
There was a problem hiding this comment.
Thanks! I fixed that in the other PR which is about to merge.
The |
Yes, there are everywhere, and typically they really are nothing more than the outcome of some function. In the handshake case though, it made sense to name it, and in the other PR I also changed it, I think for the better. Please feel entitled to review these common names as well. |
2 participants
Some miscellaneous stuff unrelated to transport implementations, and a bunch of stuff I missed because some async code was previously configured out if blocking code was enabled. Test that all features work in CI.