Horton

Re: Improving isolationtester's data output

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Noah Misch <noah@leadboat.com>
Cc: Andres Freund <andres@anarazel.de>, Alvaro Herrera <alvherre@alvh.no-ip.org>, pgsql-hackers@lists.postgresql.org
Date: 2021-06-16T14:03:52Z
Lists: pgsql-hackers 
Noah Misch <noah@leadboat.com> writes:
> I'm +1 for back-patching this class of change.  I've wasted time adapting a
> back-patch's test case to account for non-back-patched test infrastructure
> changes.  Every back-patch of test infrastructure has been a strict win from
> my perspective.

Hearing few objections, I'll plan on back-patching.  I'm thinking that the
best thing to do is apply these changes after beta2 wraps, but before we
branch v14.  Waiting till after the branch would just create duplicate
work.

BTW, as long as we're thinking of back-patching nontrivial specfile
changes, I have another modest proposal.  What do people think of
removing the requirement for step/session names to be double-quoted,
and instead letting them work like SQL identifiers?  A quick grep
shows that practically all the existing names are plain identifiers,
so we could just drop their quotes for a useful notational savings.
While I haven't actually tried yet, I doubt it'd be hard to adopt
scan.l's identifier rules into specscanner.l.  (Probably wouldn't
bother with auto case-folding, though.)

			regards, tom lane

Commits

  1. Improve display of query results in isolation tests.

  2. Detect unused steps in isolation specs and do some cleanup

  3. Remove dry-run mode from isolationtester