diff options
author | Helmut Grohne <helmut@subdivi.de> | 2016-06-12 07:52:28 +0200 |
---|---|---|
committer | Helmut Grohne <helmut@subdivi.de> | 2016-06-12 07:52:28 +0200 |
commit | 90efeacbc079bb87d44f4ed4bca4d9069ed06eeb (patch) | |
tree | e41c19f385fb74c3f9173d3e4f5ecddc874cfcbe /multiarchanalyze.sql | |
parent | e123f857a74e2ec8742ceebddebd7d296dcaca3c (diff) | |
download | debian-dedup-90efeacbc079bb87d44f4ed4bca4d9069ed06eeb.tar.gz |
multiarchanalyze: speed up on sqlite3 3.8.7.1
Since all users of archdepcandidate run the results through "exists()"
or "group by", "union" vs "union all" does not make any difference to
the results.
On the performance side however, it avoids a b-tree merge getting the
maforeign_candidate query down from hours to seconds.
Diffstat (limited to 'multiarchanalyze.sql')
-rw-r--r-- | multiarchanalyze.sql | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/multiarchanalyze.sql b/multiarchanalyze.sql index 6b26ecd..95ebd7d 100644 --- a/multiarchanalyze.sql +++ b/multiarchanalyze.sql @@ -45,7 +45,7 @@ CREATE VIEW archdepcandidate AS OR p1.architecture = 'all' OR p2.architecture = 'all' OR p2.architecture = d.archqual) - UNION + UNION ALL SELECT p1.id AS dependerid, p2.id AS dependeeid FROM package AS p1, depends AS d, provides AS r, package AS p2 WHERE p1.id = d.pid |