Hello, I could see reparameterize for foreign path to work
effectively thanks to your advice. The key point was setting
use_remote_estimate to false and existence of another child to
get parameterized path in prior stages.
The overall patch was applied on HEAD and compiled cleanly except
for a warning.
> analyze.c: In function ‘acquire_inherited_sample_rows’:
> analyze.c:1461: warning: unused variable ‘saved_rel’
As for postgres-fdw, the point I felt uneasy in previous patch
was fixed already:) - which was coping with omission of
ReparameterizeForeignPath.
And for file-fdw, you made a change to re-create foreignscan node
instead of the previous copy-and-modify. Is the reason you did it
that you considered the cost of 're-checking whether to
selectively perform binary conversion' is low enough? Or other
reasons?
Finally, although I insist the necessity of the warning for child
foreign tables on alter table, if you belive it to be put off,
I'll compromise by putting a note to CF-app that last judgement
is left to committer.
regards,
--
Kyotaro Horiguchi
NTT Open Source Software Center