| From: | Noah Misch <noah(at)leadboat(dot)com> |
|---|---|
| To: | pgsql-hackers(at)postgresql(dot)org |
| Subject: | pg_dump vs. GRANT OPTION among initial privileges |
| Date: | 2021-01-09 10:24:23 |
| Message-ID: | 20210109102423.GA160022@rfd.leadboat.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
pg_dump can generate invalid SQL syntax if an aclitem in pg_init_privs
contains a GRANT OPTION (an asterisk in the aclitemout() representation).
Also, pg_dump uses REVOKE GRANT OPTION FOR where it needs plain REVOKE. For
more details, see the log message in the attached patch. I recommend
reviewing "diff -w" output first. I'm not certain the test suite will always
find those REVOKE statements in the same order; if that order varies, one
regex will need more help.
| Attachment | Content-Type | Size |
|---|---|---|
| grant-option-dump-v1.patch | text/plain | 7.3 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Simon Riggs | 2021-01-09 10:39:15 | Re: WIP: System Versioned Temporal Table |
| Previous Message | Noah Misch | 2021-01-09 10:17:59 | Re: Spurious "apparent wraparound" via SimpleLruTruncate() rounding |