[patches] [PATCH 4/4] Allow to do not specify tail nullable index columns

Konstantin Osipov kostja at tarantool.org
Sat Feb 17 23:51:46 MSK 2018


* v.shpilevoy at tarantool.org <v.shpilevoy at tarantool.org> [18/02/17 23:41]:
> > AFAIU is_optional should be updated for all indexes in alter_space_do()
> > or in AlterSpaceOps, not here.
> 
> I disagree.

Vlad, a code review process exists so that people can exchange
their opinions, not disagreements. With this remarks you've just
deadlocked the review process - it's not the reviewer's job to
convince you you're wrong, the burden of proof lays on you.

> >> struct key_def;
> >> @@ -126,6 +128,11 @@ struct key_def {
> >>    uint32_t unique_part_count;
> >>    /** True, if at least one part can store NULL. */
> >>    bool is_nullable;
> >> +    /**
> >> +     * True, if some key parts can absense in a tuple. These
> >> +     * fields assumed to be MP_NIL.
> >> +     */
> >> +    bool is_optional;
> > 
> > You add is_optional to both key_def and key_part, but I fail to
> > understand why you need them. You can detect if a field is missing
> > in a tuple without any flags, right? So why do you need them?
> 
> It is Kostja’s decision.

I never asked you to add is_optional to both key_def and key_part
- I simply asked you to split two concepts - is_optional and
is_nullable. 

Buy saying it's my decision you misrepresent me. I *never* "decide"
on matters which can be resolved by a technical argument, this is
not my job.

Could you please begin by explaining what you mean was my decision here? 

Please note once again that by referring to me you stall the
review process. 


-- 
Konstantin Osipov, Moscow, Russia, +7 903 626 22 32
http://tarantool.org - www.twitter.com/kostja_osipov



More information about the Tarantool-patches mailing list