> I think I'd like to quibble with some of the names a bit, though.
That is a good idea, because I'm not a native English speaker and I was
not so sure for many options.
> The patch adds --fill-factor, but I think we should spell it
> without the hyphen: --fillfactor.
Fine with me.
> I think --quiet-log should be spelled --quiet.
ISTM that --quiet usually means "not verbose on stdout", so I added log
because this was specific to the log output, and that there may be a need
for a --quiet option for stdout at some time.
> I think --connect for each connection is not very descriptive; maybe
> --connection-per-transaction or something, although that's kind of long.
Yes, I think that it is too long. You have to type them!
What about '--reconnect'?
> I think -M should be aliased to --protocol, not --query-mode.
Fine with me.
> --skip-some-update is incorrectly pluralized; if that's
> what we're going to use, it should be --skip-some-updates.
Indeed.
> Alternatively, we could use --update-large-tables-only, which might
> make the intent more clear.
Yep, but quite long.
> On another note, it doesn't look like this updates the output of
> pgbench --help, which seems important.
Indeed, it should.
Please find attached a v4 which takes into account most of your comments,
but "very very long" option names.
--
Fabien.