On Thu, 8 Jan 2026 at 03:07, Greg Sabino Mullane <htamfids@gmail.com> wrote:
> No, it's not yet in the SQL standard, but why not be more liberal in what we accept,
I believe the concern is that it could hinder us from being able to
implement new SQL standard features. Back in 8.4, "AS" became optional
in the SELECT list. I recall that it was a bit tricky to get around
some grammar ambiguities with that. There's some detail on that in the
comment added in e67867b26.
> and not have people invent various workaround hacks (e.g. dummy columns)? Because the parser eats the comma, there
areno behavioral changes post-parsing.
Why do people have to add dummy columns? Is this because these people
don't know how to program a check to see if they're on the first
iteration of a loop or not?
> Unsurprisingly, I also don't agree that new SQL committee changes will be that hard to support around some trailing
commasupport.
To me, that seems like a risky thing to disagree with. I expect you
don't have a vision of what the standard committee might dream up
infinitely far into the future.
Personally, I don't want this feature. I'd rather receive an error
message when I make silly mistakes in code that I write. If I have
code to dynamically build an SQL statement and I have a bug that
causes a variable that stores a column name to sometimes be an empty
string, with your patch, I'll get an error message in all cases unless
the column's position is last. I want the error at all times, not
sometimes.
David
