Thread: Suggestion for psql command interpretation
I just made a typo in psql, and it resulted in something I wouldn't have expected. js=# \timimng Showing only tuples. js=# \t Tuples only is off. I also wouldn't have expected \timimng to have been interpreted as \timing, which I tried to type in the first place, but perhaps a typo like this should throw an error? Colin
Colin Wetherbee wrote: > I just made a typo in psql, and it resulted in something I wouldn't have > expected. Oh, um, this was in the 8.3.1 psql. Colin
Colin Wetherbee <cww@denterprises.org> writes:
> I just made a typo in psql, and it resulted in something I wouldn't have
> expected.
> js=# \timimng
> Showing only tuples.
This is related to the discussion about whether to require a space
between a backslash command name and its arguments ...
regards, tom lane
Tom Lane wrote: > Colin Wetherbee <cww@denterprises.org> writes: >> I just made a typo in psql, and it resulted in something I >> wouldn't have expected. > >> js=# \timimng Showing only tuples. > > This is related to the discussion about whether to require a space > between a backslash command name and its arguments ... I see. So, this was interpreted by psql in the same way "\t imimng" would have been? Not that \t does anything with arguments, though... I just thought I'd report it here in case it wasn't supposed to happen, but from what you say, it seems like it's a "feature". Colin
Colin Wetherbee <cww@denterprises.org> writes:
> I just thought I'd report it here in case it wasn't supposed to
> happen, but from what you say, it seems like it's a "feature".
Well, it's more of a historical hangover. Personally I'd not have
much problem with breaking backward compatibility on this point.
regards, tom lane
Tom Lane wrote: > Colin Wetherbee <cww@denterprises.org> writes: >> I just thought I'd report it here in case it wasn't supposed to >> happen, but from what you say, it seems like it's a "feature". > > Well, it's more of a historical hangover. Personally I'd not have > much problem with breaking backward compatibility on this point. I'm not generally in the habit of making typos, so this doesn't affect me too much. However, IMHO and as I mentioned previously, I don't think "\timimng" should succeed. I'll leave the rest of the discussion up to you and the other developers, though. :) Colin
At 04:46 AM 4/16/2008, Colin Wetherbee wrote: >Tom Lane wrote: >>Colin Wetherbee <cww@denterprises.org> writes: >>>I just thought I'd report it here in case it wasn't supposed to >>>happen, but from what you say, it seems like it's a "feature". >>Well, it's more of a historical hangover. Personally I'd not have >>much problem with breaking backward compatibility on this point. > >I'm not generally in the habit of making typos, so this doesn't >affect me too much. > >However, IMHO and as I mentioned previously, I don't think >"\timimng" should succeed. I'll leave the rest of the discussion up >to you and the other developers, though. :) I too agree that \timimng should not work. Maybe someone can think of an undesirable surprise that springs from this "feature" ;). \i \e Are possible candidates (probably others could clobber files given a typo). I think requiring a space between command and arguments would be a good idea. And not doing stuff if there are syntax errors... Link.