sqlite/test/analyze9.test

# 2013 August 3
#
# The author disclaims copyright to this source code.  In place of
# a legal notice, here is a blessing:
#
#    May you do good and not evil.
#    May you find forgiveness for yourself and forgive others.
#    May you share freely, never taking more than you give.
#
#***********************************************************************
#
# This file contains automated tests used to verify that the sqlite_stat4
# functionality is working.
#

set testdir [file dirname $argv0]
source $testdir/tester.tcl
set testprefix analyze9

ifcapable !stat4 {
  finish_test
  return
}

proc s {blob} {
  set ret ""
  binary scan $blob c* bytes
  foreach b $bytes {
    set t [binary format c $b]
    if {[string is print $t]} {
      append ret $t
    } else {
      append ret .
    }
  }
  return $ret
}
db function s s

do_execsql_test 1.0 {
  CREATE TABLE t1(a TEXT, b TEXT); 
  INSERT INTO t1 VALUES('(0)', '(0)');
  INSERT INTO t1 VALUES('(1)', '(1)');
  INSERT INTO t1 VALUES('(2)', '(2)');
  INSERT INTO t1 VALUES('(3)', '(3)');
  INSERT INTO t1 VALUES('(4)', '(4)');
  CREATE INDEX i1 ON t1(a, b);
} {}


do_execsql_test 1.1 {
  ANALYZE;
} {}

do_execsql_test 1.3 {
  SELECT tbl,idx,nEq,nLt,nDLt,test_decode(sample) FROM sqlite_stat4;
} {
  t1 i1 {1 1 1} {0 0 0} {0 0 0} {(0) (0) 1}
  t1 i1 {1 1 1} {1 1 1} {1 1 1} {(1) (1) 2}
  t1 i1 {1 1 1} {2 2 2} {2 2 2} {(2) (2) 3}
  t1 i1 {1 1 1} {3 3 3} {3 3 3} {(3) (3) 4}
  t1 i1 {1 1 1} {4 4 4} {4 4 4} {(4) (4) 5}
}

do_execsql_test 1.2 {
  SELECT tbl,idx,nEq,nLt,nDLt,s(sample) FROM sqlite_stat4;
} {
  t1 i1 {1 1 1} {0 0 0} {0 0 0} ....(0)(0)
  t1 i1 {1 1 1} {1 1 1} {1 1 1} ....(1)(1).
  t1 i1 {1 1 1} {2 2 2} {2 2 2} ....(2)(2).
  t1 i1 {1 1 1} {3 3 3} {3 3 3} ....(3)(3).
  t1 i1 {1 1 1} {4 4 4} {4 4 4} ....(4)(4).
}


#-------------------------------------------------------------------------
# This is really just to test SQL user function "test_decode".
#
reset_db
do_execsql_test 2.1 {
  CREATE TABLE t1(a, b, c);
  INSERT INTO t1 VALUES('some text', 14, NULL);
  INSERT INTO t1 VALUES(22.0, NULL, x'656667');
  CREATE INDEX i1 ON t1(a, b, c);
  ANALYZE;
  SELECT test_decode(sample) FROM sqlite_stat4;
} {
  {22.0 NULL x'656667' 2} 
  {{some text} 14 NULL 1}
}

#-------------------------------------------------------------------------
# 
reset_db
do_execsql_test 3.1 {
  CREATE TABLE t2(a, b);
  CREATE INDEX i2 ON t2(a, b);
  BEGIN;
}

do_test 3.2 {
  for {set i 0} {$i < 1000} {incr i} {
    set a [expr $i / 10]
    set b [expr int(rand() * 15.0)]
    execsql { INSERT INTO t2 VALUES($a, $b) }
  }
  execsql COMMIT
} {}

db func lindex lindex

# Each value of "a" occurs exactly 10 times in the table.
#
do_execsql_test 3.3.1 {
  SELECT count(*) FROM t2 GROUP BY a;
} [lrange [string repeat "10 " 100] 0 99]

# The first element in the "nEq" list of all samples should therefore be 10.
#
do_execsql_test 3.3.2 {
  ANALYZE;
  SELECT lindex(nEq, 0) FROM sqlite_stat4;
} [lrange [string repeat "10 " 100] 0 23]

#-------------------------------------------------------------------------
# 
do_execsql_test 3.4 {
  DROP TABLE IF EXISTS t1;
  CREATE TABLE t1(a INTEGER PRIMARY KEY, b, c);
  INSERT INTO t1 VALUES(1, 1, 'one-a');
  INSERT INTO t1 VALUES(11, 1, 'one-b');
  INSERT INTO t1 VALUES(21, 1, 'one-c');
  INSERT INTO t1 VALUES(31, 1, 'one-d');
  INSERT INTO t1 VALUES(41, 1, 'one-e');
  INSERT INTO t1 VALUES(51, 1, 'one-f');
  INSERT INTO t1 VALUES(61, 1, 'one-g');
  INSERT INTO t1 VALUES(71, 1, 'one-h');
  INSERT INTO t1 VALUES(81, 1, 'one-i');
  INSERT INTO t1 VALUES(91, 1, 'one-j');
  INSERT INTO t1 SELECT a+1,2,'two' || substr(c,4) FROM t1;
  INSERT INTO t1 SELECT a+2,3,'three'||substr(c,4) FROM t1 WHERE c GLOB 'one-*';
  INSERT INTO t1 SELECT a+3,4,'four'||substr(c,4) FROM t1 WHERE c GLOB 'one-*';
  INSERT INTO t1 SELECT a+4,5,'five'||substr(c,4) FROM t1 WHERE c GLOB 'one-*';
  INSERT INTO t1 SELECT a+5,6,'six'||substr(c,4) FROM t1 WHERE c GLOB 'one-*';	
  CREATE INDEX t1b ON t1(b);
  ANALYZE;
  SELECT c FROM t1 WHERE b=3 AND a BETWEEN 30 AND 60;
} {three-d three-e three-f}

#-------------------------------------------------------------------------
# These tests verify that the sample selection for stat4 appears to be 
# working as designed.
#

reset_db
db func lindex lindex
db func lrange lrange

do_execsql_test 4.0 {
  DROP TABLE IF EXISTS t1;
  CREATE TABLE t1(a, b, c);
  CREATE INDEX i1 ON t1(c, b, a);
}


proc insert_filler_rows_n {iStart args} {
  set A(-ncopy) 1
  set A(-nval) 1

  foreach {k v} $args {
    if {[info exists A($k)]==0} { error "no such option: $k" }
    set A($k) $v
  }
  if {[llength $args] % 2} {
    error "option requires an argument: [lindex $args end]"
  }

  for {set i 0} {$i < $A(-nval)} {incr i} {
    set iVal [expr $iStart+$i]
    for {set j 0} {$j < $A(-ncopy)} {incr j} {
      execsql { INSERT INTO t1 VALUES($iVal, $iVal, $iVal) }
    }
  }
}

do_test 4.1 {
  execsql { BEGIN }
  insert_filler_rows_n  0  -ncopy 10 -nval 19
  insert_filler_rows_n 20  -ncopy  1 -nval 100

  execsql {
    INSERT INTO t1(c, b, a) VALUES(200, 1, 'a');
    INSERT INTO t1(c, b, a) VALUES(200, 1, 'b');
    INSERT INTO t1(c, b, a) VALUES(200, 1, 'c');

    INSERT INTO t1(c, b, a) VALUES(200, 2, 'e');
    INSERT INTO t1(c, b, a) VALUES(200, 2, 'f');

    INSERT INTO t1(c, b, a) VALUES(201, 3, 'g');
    INSERT INTO t1(c, b, a) VALUES(201, 4, 'h');

    ANALYZE;
    SELECT count(*) FROM sqlite_stat4;
    SELECT count(*) FROM t1;
  }
} {24 297}

do_execsql_test 4.2 {
  SELECT 
    neq,
    lrange(nlt, 0, 2),
    lrange(ndlt, 0, 2),
    lrange(test_decode(sample), 0, 2)
    FROM sqlite_stat4
  ORDER BY rowid LIMIT 16;
} {
  {10 10 10 1} {0 0 0} {0 0 0} {0 0 0}
  {10 10 10 1} {10 10 10} {1 1 1} {1 1 1}
  {10 10 10 1} {20 20 20} {2 2 2} {2 2 2}
  {10 10 10 1} {30 30 30} {3 3 3} {3 3 3}
  {10 10 10 1} {40 40 40} {4 4 4} {4 4 4}
  {10 10 10 1} {50 50 50} {5 5 5} {5 5 5}
  {10 10 10 1} {60 60 60} {6 6 6} {6 6 6}
  {10 10 10 1} {70 70 70} {7 7 7} {7 7 7}
  {10 10 10 1} {80 80 80} {8 8 8} {8 8 8}
  {10 10 10 1} {90 90 90} {9 9 9} {9 9 9}
  {10 10 10 1} {100 100 100} {10 10 10} {10 10 10}
  {10 10 10 1} {110 110 110} {11 11 11} {11 11 11}
  {10 10 10 1} {120 120 120} {12 12 12} {12 12 12}
  {10 10 10 1} {130 130 130} {13 13 13} {13 13 13}
  {10 10 10 1} {140 140 140} {14 14 14} {14 14 14}
  {10 10 10 1} {150 150 150} {15 15 15} {15 15 15}
}

do_execsql_test 4.3 {
  SELECT 
    neq,
    lrange(nlt, 0, 2),
    lrange(ndlt, 0, 2),
    lrange(test_decode(sample), 0, 1)
    FROM sqlite_stat4
  ORDER BY rowid DESC LIMIT 2;
} {
  {2 1 1 1} {295 296 296} {120 122 125} {201 4} 
  {5 3 1 1} {290 290 292} {119 119 121} {200 1}
}

do_execsql_test 4.4 { SELECT count(DISTINCT c) FROM t1 WHERE c<201 } 120
do_execsql_test 4.5 { SELECT count(DISTINCT c) FROM t1 WHERE c<200 } 119

# Check that the perioidic samples are present.
do_execsql_test 4.6 {
  SELECT count(*) FROM sqlite_stat4
  WHERE lindex(test_decode(sample), 3) IN 
    ('34', '68', '102', '136', '170', '204', '238', '272')
} {8}


finish_test