Bug 2491

Summary: SQL: use UNIQUE instead of GROUP BY
Product: SQL Reporter: Stefan de Konink <skinkie>
Component: allAssignee: SQL devs <bugs-sql>
Status: NEW ---    
Severity: enhancement Keywords: NONEEDTOTEST
Priority: ---    
Version: -- development   
Hardware: x86 (i?86)   
OS: Linux   
URL: http://sourceforge.net/support/tracker.php?aid=2778533

Description Stefan de Konink 2009-04-22 13:56:58 CEST
#update(sys.node_tags,
#| project (
#| | left outer join (
#| | | table(sys.node_tags) [ node_tags.node, node_tags.k, node_tags.v, node_tags.vlen, node_tags.%TID% NOT NULL ],
#| | | group by (
#| | | | table(sys.node_tags) [ node_tags.node as y.node, node_tags.k as y.k, node_tags.v as y.v, node_tags.vlen as y.vlen, node_tags.%TID% NOT NULL as y.%TID% ]
#| | | ) [ y.node as L1.node, y.k as L1.k, y.v as L1.v ] [ y.node as L1.node, y.k as L1.k, y.v as L1.v, zero_or_one(length(y.v)) NOT NULL as L2 ]
#| | ) [ node_tags.node = L1.node, node_tags.k = L1.k, node_tags.v = L1.v ]
#| ) [ node_tags.%TID% NOT NULL, L2 NOT NULL as L3.L3 ]
#) [ node_tags.vlen NOT NULL ]

This could use the unique property instead of group by.
Comment 1 Fabian Groffen 2009-12-01 10:05:35 CET
does it make a notable difference?
Comment 2 Pseudo user for Sourceforge import 2010-05-04 09:32:09 CEST
This bug was previously known as tracker item 2778533 at http://sourceforge.net/support/tracker.php?aid=2778533