Horton

Re: Cache lookup errors with functions manipulation object addresses

Daniel Gustafsson <daniel@yesql.se>

View thread View on postgr.es
From: Daniel Gustafsson <daniel@yesql.se>
To: Michael Paquier <michael@paquier.xyz>
Cc: Alvaro Herrera <alvherre@2ndquadrant.com>, Dmitry Dolgov <9erthalion6@gmail.com>, Andrew Dunstan <andrew.dunstan@2ndquadrant.com>, PostgreSQL Developers <pgsql-hackers@lists.postgresql.org>
Date: 2020-07-03T10:34:39Z
Lists: pgsql-hackers 
> On 2 Jul 2020, at 07:35, Michael Paquier <michael@paquier.xyz> wrote:
> 
> On Thu, Mar 19, 2020 at 10:30:11PM +0900, Michael Paquier wrote:
>> The new FORMAT_TYPE_* flag still makes sense to me while reading this
>> code once again, as well as the extensibility of the new APIs for
>> operators and procedures.  One doubt I still have is if we should
>> really change the signature of getObjectDescription() to include the
>> new missing_ok argument or if a new function should be introduced.
>> I'd rather keep only one function as, even if this is called in many
>> places in the backend, I cannot track down an extension using it, but
>> I won't go against Alvaro's will either if he thinks something
>> different as this is his original design and commit as of f8348ea3.
> 
> Attached is an updated patch set, because of conflicts in the docs.
> Daniel, you are still registered as a reviewer of this patch, and it
> is marked as ready for committer?  Do you have more comments?  Would
> anybody have objections if I move on with 0001 and 0002 that extend
> the APIs to get the procedure, operator and type names?

No objections from me after skimming over the updated patchset.  I still
maintain that the format_operator_extended comment should be amended to include
that it can return NULL and not alwatys palloced string, but that's it.

cheers ./daniel

Commits

  1. Eliminate cache lookup errors in SQL functions for object addresses

  2. Refactor routines for name lookups of procedures and operators

  3. Add new flag to format_type_extended() to get NULL for undefined type

  4. Introduce new extended routines for FDW and foreign server lookups

  5. Refactor routines for subscription and publication lookups

  6. get_relid_attribute_name is dead, long live get_attname

  7. Tweak parser so that there is a defined representation for datatypes