2008/9/5 Heikki Linnakangas <[EMAIL PROTECTED]>: > Heikki Linnakangas wrote: >> >> I'll review the parser/planner changes from the current patch. > > Looks pretty sane to me. Few issues: > > Is it always OK to share a window between two separate window function > invocations, if they both happen to have identical OVER clause? It seems OK > for stable functions, but I'm not sure that's correct for expressions > involving volatile functions. I wonder if the SQL spec has anything to say > about that.
It may be here: ---quote--- In general, two <window function>s are computed independently, each one performing its own sort of its data, even if they use the same data and the same <sort specification list>. Since sorts may specify partial orderings, the computation of <window function>s is inevitably non-deterministic to the extent that the ordering is not total. Nevertheless, the user may desire that two <window function>s be computed using the same ordering, so that, for example, two moving aggregates move through the rows of a partition in precisely the same order. Two <window function>s are computed using the same (possibly non-deterministic) window ordering of the rows if any of the following are true: ― The <window function>s identify the same window structure descriptor. ― The <window function>s' window structure descriptors have window partitioning clauses that enumerate the same number of column references, and those column references are pairwise equivalent in their order of occurrence; and their window structure descriptors have window ordering clauses with the same number of <sort key>s, and those <sort key>s are all column references, and those column references are pairwise equivalent in their order of occurrence, and the <sort specification>s pairwise specify or imply <collate clause>s that specify equivalent <collation name>s, the same <ordering specification> (ASC or DESC), and the same <null ordering> (NULLS FIRST or NULLS LAST). ― The window structure descriptor of one <window function> is the ordering window of the other <window function>, or both window structure descriptors identify the same ordering window. /---quote--- But it doesn't say anything about volatile functions. Do you have example that is bad with the current design? The other issuses are OK. I missed those cases. will fix them. Regards, -- Hitoshi Harada -- Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-hackers
