On 19.11.2024 00:38, Tomas Vondra wrote:
> On 11/18/24 22:15, Tomas Vondra wrote:
>> ...
>>
>> So I think the correct solution is to not pass any expressions with
>> RestrictInfo to deparse_expression(). Either by stripping the nodes, or
>> by not adding them at all.
>>
>> The patch tries to do the stripping by maybe_extract_actual_clauses(),
>> but that only looks at the top node, and that is not sufficient here.
>> Maybe it would be possible to walk the whole tree, and remove all the
>> RestrictInfos nodes - including intermediate ones, not just the top. But
>> I'm not quite sure it wouldn't cause issues elsewhere (assuming it
>> modifies the existing nodes). It still feels a bit like fixing a problem
>> we shouldn't really have ...
>>
> To make this idea a bit more concrete, here's a patch removing all
> RestrictInfo nodes in show_stat_qual(). But still, it feels wrong.
>
>
> regards
>
Yes, removing all 'RestrictInfos' during deparsing using
'expression_tree_mutator()' is not optimal. However, I don't see an
alternative. Perhaps it could be done this earlier in extended_stats.c
to avoid the need for cleanup later ...
--
Best regards,
Ilia Evdokimov,
Tantor Labs LLC.
