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
The hub README recommends aliasing hub as git. Since it's meant to be a replacement for git with added sugar, it might make sense to provide hub support by letting the user run all gitsh commands through hub instead of git.
At its simplest, this could be a git config setting (gitsh.gitBinary?) which could be pointed to the hub executable.
We could also get fancy and follow the lead of programs like vim by providing two executables: gitsh would run everything through git, hubsh would run everything through hub.
Indeed, I have such an alias and I like the gitsh/hubsh idea since the usage would be completely analogous to what I do with git/hub and I would just set up the same sort of alias.
My reasoning for suggesting we only call hub on unrecognized commands is that hub is slower (slightly, but noticeably so) than git. Making things conditional such that hub is only called when needed is annoying in shell, thus the always-hub alias is preferable; however, since we're in ruby-land now, I thought maybe we could have that logic more elegantly.
Hub's a pretty large part of my git workflow, should the script check for / use hub?
It could be smart about trying hub only on unrecognized commands.
The text was updated successfully, but these errors were encountered: