Re: BUG #17148: About --no-strict-names option and --quiet option of pg_amcheck command

On 17.08.21 19:00, Euler Taveira wrote:
>> Well, problem is that it’s plain not true.  If you pass --quiet 
>> --verbose you
>> will get a lot of output, albeit less than if not using --quiet.  
>> Consistency
>> with other tools is obviously good, but only when it’s correct IMO.
> Indeed, it is not a good design. It should be one option --verbose that
> increases the verbosity according to a number or an enum value. --verbose=0
> means "quiet". However, that ship has sailed.

I was confused by this the other day as well.  Having all of

   -q, --quiet                     don't write any messages
   -P, --progress                  show progress information
   -v, --verbose                   write a lot of output

is surely a lot.

If you look at what --quiet does, it

1) disables logging warnings if there are no matches for object patterns 
and --no-strict-names is given, and

2) sets PQsetErrorVerbosity(free_slot->connection, PQERRORS_TERSE).

I think this both of these things could be deleted and we could get rid 
of the --quiet option, to simplify all this.  Neither of these behaviors 
is in common with any other PostgreSQL tool.