On 05/19/2015 02:22 PM, Tom Lane wrote:
> Mike Blackwell <mike.blackwell@rrd.com> writes:
>> See for example
>> http://docs.oracle.com/cd/B19306_01/text.102/b14218/cqoper.htm#i997330,
>> Table 3-1, third row, showing the precedence of '?'. Further down the
>> page, under "Fuzzy" see "Backward Compatibility Syntax".
> If I'm reading that right, that isn't a SQL-level operator but an operator
> in their text search query language, which would only appear in SQL
> queries within string literals (compare tsquery's query operators in PG).
> So it wouldn't be a hazard for ?-substitution, as long as the substituter
> was bright enough to not change string literals.
>
>
Yeah. What would be nice would be to have a functional notation
corresponding to the operators, so you would be able to write
something."?>"(a,b)
and it would mean exactly the same thing, including indexability, as
a ?> b
I presume that wouldn't give the drivers a headache.
cheers
andrew