a one   a one      a one   b two    \
   a one   c three    b two   a one    \
   b two   b two      b two   c three  \
   c three a one      c three b two    \
   c three c three                     \
]














































#-------------------------------------------------------------------------
# The following tests focus on FROM clause (join) processing.
#
# EVIDENCE-OF: R-26491-65072 If the join-op is a comma (","), then the
# composite dataset is the cartesian product of the sets of records from
# the left and right sides of the join-op.
................................................................................
      {one I two II three III}                                             \
   10 { SELECT t1.b, t2.b FROM t1 %JOIN% t2 ON (t1.a = 'a') }              \
      {one I one II one III}                                               \
   11 { SELECT t1.b, t2.b 
        FROM t1 %JOIN% t2 ON (CASE WHEN t1.a = 'a' THEN NULL ELSE 1 END) } \
      {two I two II two III three I three II three III}                    \
] {
  foreach {tn2 joinop} [list    1 ,    2 "CROSS JOIN"    3 "INNER JOIN"] {
    set S [string map [list %JOIN% $joinop] $select]
    do_execsql_test e_select-1.3.$tn.$tn2 $S $res
  }
}

# EVIDENCE-OF: R-63358-54862 If there is a USING clause specified as
# part of the join-constraint, then each of the column names specified
# must exist in the datasets to both the left and right of the join-op.
#
foreach {tn select col} {
................................................................................
foreach {tn select res} {
  1 { SELECT * FROM t5 %JOIN% t6 USING (a) } {AA cc cc BB dd DD}
  2 { SELECT * FROM t6 %JOIN% t5 USING (a) } {}
  3 { SELECT * FROM (SELECT a COLLATE nocase, b FROM t6) %JOIN% t5 USING (a) } 
    {aa cc cc bb DD dd}
  4 { SELECT * FROM t5 %JOIN% t6 USING (a,b) } {AA cc}
  5 { SELECT * FROM t6 %JOIN% t5 USING (a,b) } {}































} {
  foreach {tn2 joinop} [list    1 ,    2 "CROSS JOIN"    3 "INNER JOIN"] {
    set S [string map [list %JOIN% $joinop] $select]










    do_execsql_test e_select-1.6.$tn.$tn2 $S $res


  }


}























































finish_test

   a one   a one      a one   b two    \
   a one   c three    b two   a one    \
   b two   b two      b two   c three  \
   c three a one      c three b two    \
   c three c three                     \
]


# This proc is a specialized version of [do_execsql_test].
#
# The second argument to this proc must be a SELECT statement that 
# features a cross join of some time. Instead of the usual ",", 
# "CROSS JOIN" or "INNER JOIN" join-op, the string %JOIN% must be 
# substituted.
#
# This test runs the SELECT three times - once with:
#
#   * s/%JOIN%/,/
#   * s/%JOIN%/INNER JOIN/
#   * s/%JOIN%/CROSS JOIN/
#
# and checks that each time the results of the SELECT are $res.
#
proc do_join_test {tn select res} {
  foreach {tn2 joinop} [list    1 ,    2 "CROSS JOIN"    3 "INNER JOIN"] {
    set S [string map [list %JOIN% $joinop] $select]
    uplevel do_execsql_test $tn.$tn2 [list $S] [list $res]
  }
}

#-------------------------------------------------------------------------
# The following tests check that all paths on the syntax diagrams on
# the lang_select.html page may be taken.
#
# EVIDENCE-OF: R-18428-22111 -- syntax diagram join-constraint
#
do_join_test e_select-0.1.1 {
  SELECT count(*) FROM t1 %JOIN% t2 ON (t1.a=t2.a)
} {3}
do_join_test e_select-0.1.2 {
  SELECT count(*) FROM t1 %JOIN% t2 USING (a)
} {3}
do_join_test e_select-0.1.3 {
  SELECT count(*) FROM t1 %JOIN% t2
} {9}
do_catchsql_test e_select-0.1.4 {
  SELECT count(*) FROM t1, t2 ON (t1.a=t2.a) USING (a)
} {1 {cannot have both ON and USING clauses in the same join}}
do_catchsql_test e_select-0.1.5 {
  SELECT count(*) FROM t1, t2 USING (a) ON (t1.a=t2.a)
} {1 {near "ON": syntax error}}

#-------------------------------------------------------------------------
# The following tests focus on FROM clause (join) processing.
#
# EVIDENCE-OF: R-26491-65072 If the join-op is a comma (","), then the
# composite dataset is the cartesian product of the sets of records from
# the left and right sides of the join-op.
................................................................................
      {one I two II three III}                                             \
   10 { SELECT t1.b, t2.b FROM t1 %JOIN% t2 ON (t1.a = 'a') }              \
      {one I one II one III}                                               \
   11 { SELECT t1.b, t2.b 
        FROM t1 %JOIN% t2 ON (CASE WHEN t1.a = 'a' THEN NULL ELSE 1 END) } \
      {two I two II two III three I three II three III}                    \
] {


  do_join_test e_select-1.3.$tn $select $res

}

# EVIDENCE-OF: R-63358-54862 If there is a USING clause specified as
# part of the join-constraint, then each of the column names specified
# must exist in the datasets to both the left and right of the join-op.
#
foreach {tn select col} {
................................................................................
foreach {tn select res} {
  1 { SELECT * FROM t5 %JOIN% t6 USING (a) } {AA cc cc BB dd DD}
  2 { SELECT * FROM t6 %JOIN% t5 USING (a) } {}
  3 { SELECT * FROM (SELECT a COLLATE nocase, b FROM t6) %JOIN% t5 USING (a) } 
    {aa cc cc bb DD dd}
  4 { SELECT * FROM t5 %JOIN% t6 USING (a,b) } {AA cc}
  5 { SELECT * FROM t6 %JOIN% t5 USING (a,b) } {}
} {
  do_join_test e_select-1.6.$tn $select $res
}

# EVIDENCE-OF: R-57047-10461 For each pair of columns identified by a
# USING clause, the column from the right-hand dataset is omitted from
# the joined dataset.
#
# EVIDENCE-OF: R-56132-15700 This is the only difference between a USING
# clause and its equivalent ON constraint.
#
foreach {tn select res} {
  1a { SELECT * FROM t1 %JOIN% t2 USING (a)      } 
     {a one I b two II c three III}
  1b { SELECT * FROM t1 %JOIN% t2 ON (t1.a=t2.a) }
     {a one a I b two b II c three c III}

  2a { SELECT * FROM t3 %JOIN% t4 USING (a)      }  
     {a 1 {} b 2 2}
  2b { SELECT * FROM t3 %JOIN% t4 ON (t3.a=t4.a) } 
     {a 1 a {} b 2 b 2}

  3a { SELECT * FROM t3 %JOIN% t4 USING (a,c)                  } {b 2}
  3b { SELECT * FROM t3 %JOIN% t4 ON (t3.a=t4.a AND t3.c=t4.c) } {b 2 b 2}

  4a { SELECT * FROM (SELECT a COLLATE nocase, b FROM t6) AS x 
       %JOIN% t5 USING (a) } 
     {aa cc cc bb DD dd}
  4b { SELECT * FROM (SELECT a COLLATE nocase, b FROM t6) AS x
       %JOIN% t5 ON (x.a=t5.a) } 
     {aa cc AA cc bb DD BB dd}
} {


  do_join_test e_select-1.7.$tn $select $res
}

# EVIDENCE-OF: R-04095-00676 If the join-op is a "LEFT JOIN" or "LEFT
# OUTER JOIN", then the composite dataset is created as for an "INNER
# JOIN". Except, after the ON or USING filtering clauses have been
# applied, an extra row is added to the output for each row in the
# original left-hand input dataset (if any) that corresponds to no rows
# at all in the composite dataset.
#
do_execsql_test e_select-1.8.0 {
  CREATE TABLE t7(a, b, c);
  CREATE TABLE t8(a, d, e);

  INSERT INTO t7 VALUES('x', 'ex',  24);
  INSERT INTO t7 VALUES('y', 'why', 25);

  INSERT INTO t8 VALUES('x', 'abc', 24);
  INSERT INTO t8 VALUES('z', 'ghi', 26);
} {}

do_execsql_test e_select-1.8.1a {
  SELECT count(*) FROM t7 JOIN t8 ON (t7.a=t8.a)
} {1}
do_execsql_test e_select-1.8.1b {
  SELECT count(*) FROM t7 LEFT JOIN t8 ON (t7.a=t8.a)
} {2}

do_execsql_test e_select-1.8.2a {
  SELECT count(*) FROM t7 JOIN t8 USING (a)
} {1}
do_execsql_test e_select-1.8.2b {
  SELECT count(*) FROM t7 LEFT JOIN t8 USING (a)
} {2}

# EVIDENCE-OF: R-15607-52988 The added rows contain NULL values in the
# columns that would normally contain values copied from the right-hand
# input dataset.
#
do_execsql_test e_select-1.9.1a {
  SELECT * FROM t7 JOIN t8 ON (t7.a=t8.a)
} {x ex 24 x abc 24}
do_execsql_test e_select-1.9.1b {
  SELECT * FROM t7 LEFT JOIN t8 ON (t7.a=t8.a)
} {x ex 24 x abc 24 y why 25 {} {} {}}

do_execsql_test e_select-1.9.2a {
  SELECT * FROM t7 JOIN t8 USING (a)
} {x ex 24 abc 24}
do_execsql_test e_select-1.9.2b {
  SELECT * FROM t7 LEFT JOIN t8 USING (a)
} {x ex 24 abc 24 y why 25 {} {}}

# EVIDENCE-OF: R-01809-52134 If the NATURAL keyword is added to any of
# the join-ops, then an implicit USING clause is added to the
# join-constraints. The implicit USING clause contains each of the
# column names that appear in both the left and right-hand input
# datasets.
#
foreach {tn s1 s2 res} {
  1 { SELECT * FROM t7 JOIN t8 USING (a) }
    { SELECT * FROM t7 NATURAL JOIN t8 }
    {x ex 24 abc 24}

  2 { SELECT * FROM t8 JOIN t7 USING (a) }
    { SELECT * FROM t8 NATURAL JOIN t7 }
    {x abc 24 ex 24}

  3 { SELECT * FROM t7 LEFT JOIN t8 USING (a) }
    { SELECT * FROM t7 NATURAL LEFT JOIN t8 }
    {x ex 24 abc 24 y why 25 {} {}}


  4 { SELECT * FROM t8 LEFT JOIN t7 USING (a) }
    { SELECT * FROM t8 NATURAL LEFT JOIN t7 }
    {x abc 24 ex 24 z ghi 26 {} {}}

  5 { SELECT * FROM t3 JOIN t4 USING (a,c) }
    { SELECT * FROM t3 NATURAL JOIN t4 }
    {b 2}

  6 { SELECT * FROM t3 LEFT JOIN t4 USING (a,c) }
    { SELECT * FROM t3 NATURAL LEFT JOIN t4 }
    {a 1 b 2}
} {
  do_execsql_test e_select-1.10.${tn}a $s1 $res
  do_execsql_test e_select-1.10.${tn}b $s2 $res
}

# EVIDENCE-OF: R-49566-01570 If the left and right-hand input datasets
# feature no common column names, then the NATURAL keyword has no effect
# on the results of the join.
#
do_execsql_test e_select-1.11.0 {
  CREATE TABLE t10(x, y);
  INSERT INTO t10 VALUES(1, 'true');
  INSERT INTO t10 VALUES(0, 'false');
} {}
foreach {tn s1 s2 res} {
  1 { SELECT a, x FROM t1 CROSS JOIN t10 }
    { SELECT a, x FROM t1 NATURAL CROSS JOIN t10 }
    {a 1 a 0 b 1 b 0 c 1 c 0}
} {
  do_execsql_test e_select-1.11.${tn}a $s1 $res
  do_execsql_test e_select-1.11.${tn}b $s2 $res
}

# EVIDENCE-OF: R-39625-59133 A USING or ON clause may not be added to a
# join that specifies the NATURAL keyword.
#
foreach {tn sql} {
  1 {SELECT * FROM t1 NATURAL LEFT JOIN t2 USING (a)}
  2 {SELECT * FROM t1 NATURAL LEFT JOIN t2 ON (t1.a=t2.a)}
  3 {SELECT * FROM t1 NATURAL LEFT JOIN t2 ON (45)}
} {
  do_catchsql_test e_select-1.12.$tn "
    $sql
  " {1 {a NATURAL join may not have an ON or USING clause}}
}

finish_test