Eric Blake <ebl...@redhat.com> writes:
> On 02/27/2017 07:36 AM, Markus Armbruster wrote:
>
>>
>>> Maybe just 'foo.array[]' (without any =) for an empty
>>> array or something like that.
>>
>> Yes, that should do. Likewise foo.object{} for empty object.
>>
>> {} doesn't even need quoting. [] may.
>
> [contents] needs quoting, but [] does NOT need shell quoting (no shells
> treat it as a glob), for the same reason that 'if [ -e "$file" ];' needs
> no quoting around the [ or ] (the shell only requires quotes for [ if
> the rest of the shell word can look like a valid glob, but globs require
> intermediate content before the ]).
Bash shell options failglob and nullglob can still mess it up.
>>> Before we introduce anything like this, do we actually need it?
>>
>> I don't know whether anything needs optional, present and empty. But
>> even if the answer is "no" today, it need not remain "no".
>>
>> Anyone running into a case of "yes", will have to fall back to the JSON
>> form of -blockdev. Strengthens my belief that providing JSON there is a
>> good idea.
>>
>> The insufficient generality of dotted keys bugs me a bit. Not sure
>> whether it justifies more syntax now. But we should document it.
>
> I agree that documenting it as a shortcoming of dotted form and pointing
> to JSON form is okay. I also like that we are leaving the door open for
> future expansion, if needed, and think that is better than inventing the
> syntax now, especially for what we are trying to get into 2.9.
Okay, I'll work on improving the documentation.
