Kyle Lanclos writes:
> Tom Lane wrote:
>> Well, that would depend on the data type being indexed, which you did
>> not specify; and if it's a variable-length type then it's really hard to
>> give a concrete answer.
> In my specific case, I have an index on (text, double). There are individual
> i
Tom Lane wrote:
> Well, that would depend on the data type being indexed, which you did
> not specify; and if it's a variable-length type then it's really hard to
> give a concrete answer.
Thanks for the quick reply; I did not appreciate that the Knuth order
would vary according to the data being
Kyle Lanclos writes:
> I spent some time last week staring at the code for the PostgreSQL
> B+-tree implementation. What I hoped to find, and was not immediately
> able to determine, was the Knuth order for the PostgreSQL B+-tree
> implementation. It is entirely possible that I simply got lost in
I spent some time last week staring at the code for the PostgreSQL
B+-tree implementation. What I hoped to find, and was not immediately
able to determine, was the Knuth order for the PostgreSQL B+-tree
implementation. It is entirely possible that I simply got lost in the
wrong C file.
My goal is
