On Thu, Sep 8, 2016 at 5:21 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Stephen Frost <sfrost@snowman.net> writes:
>> * Alvaro Herrera (alvherre@2ndquadrant.com) wrote:
>>> Can't you keep those words as Sconst or something (DefElems?) until the
>>> execution phase, so that they don't need to be keywords at all?
>
>> Seems like we could do that, though I'm not convinced that it really
>> gains us all that much. These are only unreserved keywords, of course,
>> so they don't impact users the way reserved keywords (of any kind) can.
>> While there may be some places where we use a string to represent a set
>> of defined options, I don't believe that's typical
>
> -1 for having to write them as string literals; but I think what Alvaro
> really means is to arrange for the words to just be identifiers in the
> grammar, which you strcmp against at execution. See for example
> reloption_list. (Whether you use DefElem as the internal representation
> is a minor detail, though it might help for making the parsetree
> copyObject-friendly.)
>
> vacuum_option_elem shows another way to avoid making a word into a
> keyword, although to me that one is more of an antipattern; it'd be better
> to leave the strcmp to execution, since there's so much other code that
> does things that way.
There are other cases like that, too, e.g. AlterOptRoleElem; I don't
think it's a bad pattern. Regardless of the specifics, I do think
that it would be better not to bloat the keyword table with things
that don't really need to be keywords.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company