Page MenuHomePhabricator

rhg: Replace subcommand boilerplate with a macro

Authored by SimonSapin on Mon, Feb 8, 5:44 PM.



This removes some repetition, and will avoid additional repetition
in the next commit.

Diff Detail

rHG Mercurial
Automatic diff as part of commit; lint not applicable.
Automatic diff as part of commit; unit tests not applicable.

Event Timeline

SimonSapin created this revision.Mon, Feb 8, 5:44 PM
Alphare accepted this revision.Tue, Feb 9, 4:03 AM
Alphare added a subscriber: Alphare.
Alphare added inline comments.

This should be a fallback to Python (if the "no fallback" config option is not set, when we're at that point) instead of a panic. Maybe this handling can be done in the same follow-up patch as the catch_unwind one from the previous patch?
Note that the previous behavior of exiting is not valid either, hence why I propose a follow-up.

SimonSapin added inline comments.Tue, Feb 9, 4:14 AM

My understanding of SubcommandRequired is that it makes get_matches_safe return an error (so this code is not even reached) in all cases that would otherwise return None here. Is that not how it works?

Alphare added inline comments.Tue, Feb 9, 4:18 AM

Ha, it does, but we also print an error to stderr at line 24 which confused me, when it should go to log::debug to grep for the fallback reason and not print anything to the user.

This revision was not accepted when it landed; it landed in state Needs Review.
This revision was automatically updated to reflect the committed changes.