Re: Regress tests reveal *serious* psql bug

On 2000-01-12, Tom Lane mentioned:

> I think this raises the bar to the point where we must
> have a transparent backward-compatible approach to psql variables.
> I was not all that thrilled about blowing off colon as an operator,
> and blowing off array subscripts *too* is just too far above my
> threshold of pain.

To clear something out here: I'm with you all the way. I didn't make up
that syntax, and I too was forgetful about the array issue. If y'all think
that a variable must be defined to be substituted and that that will fix
things to a reasonable state, thus it shall be. My concern was more that
this would only work around this particular problem, while being short
sighted. Just want to make sure we have a consensus.

> > If I'm going to hack around in that code, one related question: what
> > should the deal be regarding variable interpolation into quoted
> > strings? Yes/No/Maybe?
>
> No bloody way, IMHO --- that increases the odds of unwanted

I'll take that as a No. ;)

--
Peter Eisentraut Sernanders väg 10:115
peter_e(at)gmx(dot)net 75262 Uppsala
http://yi.org/peter-e/ Sweden