Tarantool development patches archive
 help / color / mirror / Atom feed
From: Nikita Pettik <korablev@tarantool.org>
To: tarantool-patches@freelists.org
Cc: v.shpilevoy@tarantool.org, Nikita Pettik <korablev@tarantool.org>
Subject: [tarantool-patches] [PATCH] sql: make IN operator stop ignoring type and collation
Date: Wed, 16 Jan 2019 16:34:01 +0300	[thread overview]
Message-ID: <20190116133401.5844-1-korablev@tarantool.org> (raw)

SQLite discards type and collation of IN operator when it comes with
only one operand. This leads to different results of straight comparison
using '=' operator and IN:

SELECT x FROM t1 WHERE x IN (1.0);
-- Result is empty set
SELECT x FROM t1 WHERE x = 1.0;
- - ['1']

Lets remove this strange ignorance and always take into consideration
types and collations of operands.

Closes #3934
---
Branch: https://github.com/tarantool/tarantool/tree/np/gh-3934-IN-operator-fix
Issue: https://github.com/tarantool/tarantool/issues/3934

 src/box/sql/expr.c                   |  4 ----
 src/box/sql/parse.y                  | 11 -----------
 src/box/sql/sqliteInt.h              |  1 -
 test/sql-tap/in4.test.lua            |  8 ++++----
 test/sql-tap/tkt-9a8b09f8e6.test.lua |  7 ++++---
 5 files changed, 8 insertions(+), 23 deletions(-)

diff --git a/src/box/sql/expr.c b/src/box/sql/expr.c
index b67b22c23..d83be5101 100644
--- a/src/box/sql/expr.c
+++ b/src/box/sql/expr.c
@@ -73,8 +73,6 @@ char
 sqlite3ExprAffinity(Expr * pExpr)
 {
 	pExpr = sqlite3ExprSkipCollate(pExpr);
-	if (pExpr->flags & EP_Generic)
-		return 0;
 	uint8_t op = pExpr->op;
 	struct ExprList *el;
 	if (op == TK_REGISTER)
@@ -197,8 +195,6 @@ sql_expr_coll(Parse *parse, Expr *p, bool *is_explicit_coll, uint32_t *coll_id)
 	*coll_id = COLL_NONE;
 	while (p != NULL) {
 		int op = p->op;
-		if (p->flags & EP_Generic)
-			break;
 		if (op == TK_CAST || op == TK_UPLUS) {
 			p = p->pLeft;
 			continue;
diff --git a/src/box/sql/parse.y b/src/box/sql/parse.y
index 50bb2ba01..0bcf41594 100644
--- a/src/box/sql/parse.y
+++ b/src/box/sql/parse.y
@@ -1101,21 +1101,10 @@ expr(A) ::= expr(A) in_op(N) LP exprlist(Y) RP(E). [IN] {
     **
     **      expr1 == ?1
     **      expr1 <> ?2
-    **
-    ** But, the RHS of the == or <> is marked with the EP_Generic flag
-    ** so that it may not contribute to the computation of comparison
-    ** affinity or the collating sequence to use for comparison.  Otherwise,
-    ** the semantics would be subtly different from IN or NOT IN.
     */
     Expr *pRHS = Y->a[0].pExpr;
     Y->a[0].pExpr = 0;
     sql_expr_list_delete(pParse->db, Y);
-    /* pRHS cannot be NULL because a malloc error would have been detected
-    ** before now and control would have never reached this point */
-    if( ALWAYS(pRHS) ){
-      pRHS->flags &= ~EP_Collate;
-      pRHS->flags |= EP_Generic;
-    }
     A.pExpr = sqlite3PExpr(pParse, N ? TK_NE : TK_EQ, A.pExpr, pRHS);
   }else{
     A.pExpr = sqlite3PExpr(pParse, TK_IN, A.pExpr, 0);
diff --git a/src/box/sql/sqliteInt.h b/src/box/sql/sqliteInt.h
index 7e16edc9a..ee24e0337 100644
--- a/src/box/sql/sqliteInt.h
+++ b/src/box/sql/sqliteInt.h
@@ -2158,7 +2158,6 @@ struct Expr {
 #define EP_DblQuoted 0x000040	/* token.z was originally in "..." */
 #define EP_InfixFunc 0x000080	/* True for an infix function: LIKE, etc */
 #define EP_Collate   0x000100	/* Tree contains a TK_COLLATE operator */
-#define EP_Generic   0x000200	/* Ignore COLLATE or affinity on this tree */
 #define EP_IntValue  0x000400	/* Integer value contained in u.iValue */
 #define EP_xIsSelect 0x000800	/* x.pSelect is valid (otherwise x.pList is) */
 #define EP_Skip      0x001000	/* COLLATE, AS, or UNLIKELY */
diff --git a/test/sql-tap/in4.test.lua b/test/sql-tap/in4.test.lua
index b029535af..c488cde45 100755
--- a/test/sql-tap/in4.test.lua
+++ b/test/sql-tap/in4.test.lua
@@ -518,8 +518,8 @@ test:do_execsql_test(
     })
 
 -- MUST_WORK_TEST
--- Make sure that when "x IN (?)" is converted into "x==?" that collating
--- sequence and affinity computations do not get messed up.
+-- Make sure that when "x IN (?)" is converted into "x==?":
+-- type and collation sequence should be taken into consideration.
 --
 test:do_execsql_test(
     "in4-4.1",
@@ -571,7 +571,7 @@ test:do_execsql_test(
         SELECT c FROM t4a WHERE a IN (b) ORDER BY c;
     ]], {
         -- <in4-4.5>
-        3
+        1, 3
         -- </in4-4.5>
     })
 
@@ -581,7 +581,7 @@ test:do_execsql_test(
         SELECT c FROM t4a WHERE (a||'') IN (b) ORDER BY c;
     ]], {
         -- <in4-4.6>
-        3
+        1, 3
         -- </in4-4.6>
     })
 
diff --git a/test/sql-tap/tkt-9a8b09f8e6.test.lua b/test/sql-tap/tkt-9a8b09f8e6.test.lua
index 4d08cd657..996809afe 100755
--- a/test/sql-tap/tkt-9a8b09f8e6.test.lua
+++ b/test/sql-tap/tkt-9a8b09f8e6.test.lua
@@ -99,7 +99,7 @@ test:do_execsql_test(
         SELECT x FROM t1 WHERE x IN (1.0);
     ]], {
         -- <2.2>
-        
+        "1"
         -- </2.2>
     })
 
@@ -239,7 +239,7 @@ test:do_execsql_test(
         SELECT x FROM t2 WHERE '1.0' IN (x);
     ]], {
         -- <3.8>
-
+        1
         -- </3.8>
     })
 
@@ -309,7 +309,7 @@ test:do_execsql_test(
         SELECT x FROM t3 WHERE '1' IN (x);
     ]], {
         -- <4.7>
-
+        1
         -- </4.7>
     })
 
@@ -519,6 +519,7 @@ test:do_execsql_test(
         SELECT x, y FROM t5 WHERE '1.0' IN (x);
     ]], {
         -- <6.8>
+        1, "one", 1, "two", 1, "three", 1, "four"
         -- </6.8>
     })
 
-- 
2.15.1

             reply	other threads:[~2019-01-16 13:34 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-01-16 13:34 Nikita Pettik [this message]
2019-01-17 19:53 ` [tarantool-patches] " Vladislav Shpilevoy
2019-01-24 13:56 ` Kirill Yukhin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190116133401.5844-1-korablev@tarantool.org \
    --to=korablev@tarantool.org \
    --cc=tarantool-patches@freelists.org \
    --cc=v.shpilevoy@tarantool.org \
    --subject='Re: [tarantool-patches] [PATCH] sql: make IN operator stop ignoring type and collation' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox