From: | Melanie Plageman <melanieplageman(at)gmail(dot)com> |
---|---|
To: | Pg Hackers <pgsql-hackers(at)postgresql(dot)org> |
Cc: | Andres Freund <andres(at)anarazel(dot)de> |
Subject: | make tuplestore helper function |
Date: | 2021-11-02 16:34:30 |
Message-ID: | CAAKRu_azyd1Z3W_r7Ou4sorTjRCs+PxeHw1CWJeXKofkE6TuZg@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Attached is a patch to create a helper function which creates a
tuplestore to be used by FUNCAPI-callable functions.
It was suggested in [1] that I start a separate thread for this patch.
A few notes:
There are a few places with very similar code to the new helper
(MakeTuplestore()) but which, for example, don't expect the return type
to be a row (pg_ls_dir(), pg_tablespace_databases(), text_to_table()). I
wasn't sure if it was worth modifying it for their benefit.
I wasn't sure if the elog() check for call result type should be changed
to an ereport().
All callers of MakeTuplestore() pass work_mem as the third parameter, so
I could just omit it and hard-code it in, but I wasn't sure that felt
right in a utility/helper function.
I inlined deflist_to_tuplestore() in foreign.c since it was billed as a
helper function but wasn't used elsewhere and it made it harder to use
MakeTuplestore() in this location.
-- melanie
[1] https://www.postgresql.org/message-id/flat/20200124195226.lth52iydq2n2uilq%40alap3.anarazel.de
Attachment | Content-Type | Size |
---|---|---|
v1-0001-Add-helper-to-make-tuplestore.patch | text/x-patch | 42.8 KB |
From | Date | Subject | |
---|---|---|---|
Next Message | Julien Rouhaud | 2021-11-02 16:35:45 | Re: should we enable log_checkpoints out of the box? |
Previous Message | Robert Haas | 2021-11-02 16:34:05 | Re: refactoring basebackup.c |