Re: Query Discrepancy in Postgres HLL Test

On Wed, May 1, 2024 at 1:10 PM Ayush Vatsa <ayushvatsa1810@gmail.com> wrote:
> I'm currently delving into Postgres HLL (HyperLogLog) functionality and have encountered an unexpected behavior while
executingqueries from the "cumulative_add_sparse_edge.sql" regress test. This particular test data file involves three
columns,with the last column representing an HLL (HyperLogLog) value derived from the previous HLL value and the
currentraw value. 
>
> Upon manual inspection of the query responsible for deriving the last row's HLL value, I noticed a discrepancy. When
executingthe query: 
> """
> -- '\x148B481002....' is second last rows hll value
> SELECT hll_add('\x148B481002.....', hll_hashval(2561));
> """
> instead of obtaining the expected value (''\x148B481002....''), I received a different output which is
('\x138b48000200410061008100a1........'). 

PostgreSQL has no function called hll_add or hll_hashval, and no
regression test file called cumulative_add_sparse_edge.sql. A quick
Google search suggests that these things are part of citusdata's fork
of PostgreSQL, so you might want to contact them.

--
Robert Haas
EDB: http://www.enterprisedb.com