On 04/16/2013 08:23 PM, Tom Lane wrote:
> Lee Hachadoorian <Lee.Hachadoorian+L@gmail.com> writes:
>> SQL seems to be behaving in a case-sensitive manner:
>> universe=# select 1;
>> ?column?
>> ----------
>> 1
>> (1 row)
>> universe=# SELECT 1;
>> ERROR: syntax error at or near "SELECT 1"
>> LINE 1: SELECT 1;
>> ^
> That's really bizarre, but I don't think it's a case sensitivity problem
> as such. Watch what I get from a syntax error on a normally-functioning
> system:
>
> $ psql
> psql (9.1.9)
> Type "help" for help.
>
> regression=# select 1;
> ?column?
> ----------
> 1
> (1 row)
>
> regression=# SELECT 1;
> ?column?
> ----------
> 1
> (1 row)
>
> regression=# xELECT 1;
> ERROR: syntax error at or near "xELECT"
> LINE 1: xELECT 1;
> ^
>
> See the differences? The error message indicates that your parser saw
> "SELECT 1" as all one token. Also, if you transcribed this accurately,
> it looks like the error cursor is pointing to the second character of
> the "SELECT" not the first (could you confirm that?). Which is even
> more bizarre. I'm not sure what's going on, but I think it's more
> likely to be something to do with whitespace not being really whitespace
> than it is with case as such. Consider the possibility that you're
> somehow typing a non-breaking space or some such character. One thing
> that might be useful is to examine the error report in the postmaster
> log using an editor that will show you any non-printing characters.
>
> regards, tom lane
Tom,
Yes, nonbreaking spaces was the problem. Sorry for the red herring re:
case sensitivity, and thanks for figuring it out. Adrian, thanks for
your input as well.
Best,
--Lee
--
Lee Hachadoorian
Assistant Professor in Geography, Dartmouth College
http://freecity.commons.gc.cuny.edu
