Charvi Mendiratta [off-list ref] writes:
Earlier, I thought to implement the UI of amend! commit as :
"git commit --fixup=a/amend:<commit>"
So users can either use 'a' as abbreviation or 'amend'. But I want to once
confirm if I got this right ? As I am doubtful about, what does allowing to
accept any prefix of "amend" and "reword" means ?
It means that all of these do the same thing ...
git commit --fixup=amend:<commit>
git commit --fixup=amen:<commit>
git commit --fixup=ame:<commit>
git commit --fixup=am:<commit>
git commit --fixup=a:<commit>
... until somebody more brilliant than either of us comes along and
invents another operation that sits next to 'amend' and 'reword',
say, 'annoy'. At that point,
git commit --fixup=a:<commit>
no longer uniquely identifies a choice among three ('amend',
'annoy', or 'reword') and you'd give an error message, but
git commit --fixup=am:<commit>
will keep working as an abbreviation for "amend".
Hi Junio,
On Sun, 17 Jan 2021 at 10:03, Junio C Hamano [off-list ref] wrote:
Charvi Mendiratta [off-list ref] writes:
quoted
Earlier, I thought to implement the UI of amend! commit as :
"git commit --fixup=a/amend:<commit>"
So users can either use 'a' as abbreviation or 'amend'. But I want to once
confirm if I got this right ? As I am doubtful about, what does allowing to
accept any prefix of "amend" and "reword" means ?
It means that all of these do the same thing ...
git commit --fixup=amend:<commit>
git commit --fixup=amen:<commit>
git commit --fixup=ame:<commit>
git commit --fixup=am:<commit>
git commit --fixup=a:<commit>
... until somebody more brilliant than either of us comes along and
invents another operation that sits next to 'amend' and 'reword',
say, 'annoy'. At that point,
git commit --fixup=a:<commit>
no longer uniquely identifies a choice among three ('amend',
'annoy', or 'reword') and you'd give an error message, but
git commit --fixup=am:<commit>
will keep working as an abbreviation for "amend".
Thanks for explaining, now I can relate with the previous discussions and
will look around for the implementation in the above way. I think may be we
can do this while parsing the options and checking for the list of prefixes,
I am not sure but will look into it.
Thanks and Regards,
Charvi