[Tarantool-patches] [PATCH v4 45/53] sql: introduce mem_explicit_cast()

[Mergen Imeev via Tarantool-patches <tarantool-patches@dev.tarantool.org>]

This patch introduces mem_explicit_cast(). This function is used to
convert a MEM to a given field type according to explicit cast rules.

Part of #5818
---
 src/box/sql/mem.c  | 302 +++++++++++++++++++--------------------------
 src/box/sql/mem.h  |  15 +--
 src/box/sql/vdbe.c |   4 +-
 3 files changed, 133 insertions(+), 188 deletions(-)

diff --git a/src/box/sql/mem.c b/src/box/sql/mem.c
index 262f48aca..559bf6121 100644
--- a/src/box/sql/mem.c
+++ b/src/box/sql/mem.c
@@ -943,6 +943,134 @@ mem_convert_to_string(struct Mem *mem)
 	return -1;
 }
 
+static inline int
+mem_convert_varstring_to_unsigned(struct Mem *mem)
+{
+	bool is_neg;
+	int64_t i;
+	if (sql_atoi64(mem->z, &i, &is_neg, mem->n) != 0)
+		return -1;
+	if (is_neg)
+		return -1;
+	mem_set_unsigned(mem, (uint64_t)i);
+	return 0;
+}
+
+static inline int
+mem_convert_boolean_to_unsigned(struct Mem *mem)
+{
+	mem_set_unsigned(mem, (uint64_t)mem->u.b);
+	return 0;
+}
+
+static inline int
+mem_convert_string_to_boolean(struct Mem *mem)
+{
+	char *str = mem->z;
+	bool b;
+	const char *str_true = "TRUE";
+	const char *str_false = "FALSE";
+	uint32_t len_true = strlen(str_true);
+	uint32_t len_false = strlen(str_false);
+
+	for (; str[0] == ' '; str++);
+	if (strncasecmp(str, str_true, len_true) == 0) {
+		b = true;
+		str += len_true;
+	} else if (strncasecmp(str, str_false, len_false) == 0) {
+		b = false;
+		str += len_false;
+	} else {
+		return -1;
+	}
+	for (; str[0] == ' '; str++);
+	if (str[0] != '\0')
+		return -1;
+	mem_set_boolean(mem, b);
+	return 0;
+}
+
+static inline int
+mem_convert_integer_to_boolean(struct Mem *mem)
+{
+	mem_set_boolean(mem, mem->u.u != 0);
+	return 0;
+}
+
+static inline int
+mem_convert_double_to_boolean(struct Mem *mem)
+{
+	mem_set_boolean(mem, mem->u.r != 0);
+	return 0;
+}
+
+static inline int
+mem_convert_string_to_varbinary(struct Mem *mem)
+{
+	mem->flags = (mem->flags & (MEM_Dyn | MEM_Static | MEM_Ephem)) |
+		     MEM_Blob;
+	mem->field_type = FIELD_TYPE_VARBINARY;
+	return 0;
+}
+
+int
+mem_explicit_cast(struct Mem *mem, enum field_type type)
+{
+	if ((mem->flags & MEM_Null) != 0) {
+		mem->field_type = type;
+		return 0;
+	}
+	switch (type) {
+	case FIELD_TYPE_UNSIGNED:
+		if ((mem->flags & MEM_UInt) != 0)
+			return 0;
+		if ((mem->flags & MEM_Int) != 0)
+			return -1;
+		if ((mem->flags & MEM_Blob) != 0 &&
+		    (mem->flags & MEM_Subtype) != 0)
+			return -1;
+		if ((mem->flags & (MEM_Str | MEM_Blob)) != 0)
+			return mem_convert_varstring_to_unsigned(mem);
+		if ((mem->flags & MEM_Real) != 0)
+			return mem_convert_double_to_integer(mem);
+		if ((mem->flags & MEM_Bool) != 0)
+			return mem_convert_boolean_to_unsigned(mem);
+		return -1;
+	case FIELD_TYPE_STRING:
+		return mem_convert_to_string(mem);
+	case FIELD_TYPE_DOUBLE:
+		return mem_convert_to_double(mem);
+	case FIELD_TYPE_INTEGER:
+		return mem_convert_to_integer(mem);
+	case FIELD_TYPE_BOOLEAN:
+		if ((mem->flags & MEM_Bool) != 0)
+			return 0;
+		if ((mem->flags & (MEM_UInt | MEM_Int)) != 0)
+			return mem_convert_integer_to_boolean(mem);
+		if ((mem->flags & MEM_Str) != 0)
+			return mem_convert_string_to_boolean(mem);
+		if ((mem->flags & MEM_Real) != 0)
+			return mem_convert_double_to_boolean(mem);
+		return -1;
+	case FIELD_TYPE_VARBINARY:
+		if (mem_is_binary(mem))
+			return 0;
+		if (mem_is_string(mem))
+			return mem_convert_string_to_varbinary(mem);
+		return -1;
+	case FIELD_TYPE_NUMBER:
+		return mem_convert_to_number(mem);
+	case FIELD_TYPE_SCALAR:
+		if ((mem->flags & MEM_Blob) != 0 &&
+		    (mem->flags & MEM_Subtype) != 0)
+			return -1;
+		return 0;
+	default:
+		break;
+	}
+	return -1;
+}
+
 int
 mem_copy(struct Mem *to, const struct Mem *from)
 {
@@ -1583,42 +1711,6 @@ valueToText(sql_value * pVal)
 	return pVal->z;
 }
 
-/**
- * According to ANSI SQL string value can be converted to boolean
- * type if string consists of literal "true" or "false" and
- * number of leading and trailing spaces.
- *
- * For instance, "   tRuE  " can be successfully converted to
- * boolean value true.
- *
- * @param str String to be converted to boolean. Assumed to be
- *        null terminated.
- * @param[out] result Resulting value of cast.
- * @retval 0 If string satisfies conditions above.
- * @retval -1 Otherwise.
- */
-static int
-str_cast_to_boolean(const char *str, bool *result)
-{
-	assert(str != NULL);
-	for (; *str == ' '; str++);
-	if (strncasecmp(str, SQL_TOKEN_TRUE, strlen(SQL_TOKEN_TRUE)) == 0) {
-		*result = true;
-		str += 4;
-	} else if (strncasecmp(str, SQL_TOKEN_FALSE,
-			       strlen(SQL_TOKEN_FALSE)) == 0) {
-		*result = false;
-		str += 5;
-	} else {
-		return -1;
-	}
-	for (; *str != '\0'; ++str) {
-		if (*str != ' ')
-			return -1;
-	}
-	return 0;
-}
-
 /*
  * Convert a 64-bit IEEE double into a 64-bit signed integer.
  * If the double is out of range of a 64-bit signed integer then
@@ -2000,109 +2092,6 @@ registerTrace(int iReg, Mem *p) {
 }
 #endif
 
-/*
- * Cast the datatype of the value in pMem according to the type
- * @type.  Casting is different from applying type in that a cast
- * is forced.  In other words, the value is converted into the desired
- * type even if that results in loss of data.  This routine is
- * used (for example) to implement the SQL "cast()" operator.
- */
-int
-sqlVdbeMemCast(Mem * pMem, enum field_type type)
-{
-	assert(type < field_type_MAX);
-	if (pMem->flags & MEM_Null)
-		return 0;
-	switch (type) {
-	case FIELD_TYPE_SCALAR:
-		return 0;
-	case FIELD_TYPE_BOOLEAN:
-		if ((pMem->flags & MEM_Int) != 0) {
-			mem_set_boolean(pMem, pMem->u.i);
-			return 0;
-		}
-		if ((pMem->flags & MEM_UInt) != 0) {
-			mem_set_boolean(pMem, pMem->u.u);
-			return 0;
-		}
-		if ((pMem->flags & MEM_Real) != 0) {
-			mem_set_boolean(pMem, pMem->u.r);
-			return 0;
-		}
-		if ((pMem->flags & MEM_Str) != 0) {
-			bool value;
-			if (str_cast_to_boolean(pMem->z, &value) != 0)
-				return -1;
-			mem_set_boolean(pMem, value);
-			return 0;
-		}
-		if ((pMem->flags & MEM_Bool) != 0)
-			return 0;
-		return -1;
-	case FIELD_TYPE_INTEGER:
-	case FIELD_TYPE_UNSIGNED:
-		if ((pMem->flags & (MEM_Blob | MEM_Str)) != 0) {
-			bool is_neg;
-			int64_t val;
-			if (sql_atoi64(pMem->z, &val, &is_neg, pMem->n) != 0)
-				return -1;
-			if (type == FIELD_TYPE_UNSIGNED && is_neg)
-				return -1;
-			mem_set_integer(pMem, val, is_neg);
-			return 0;
-		}
-		if ((pMem->flags & MEM_Bool) != 0) {
-			mem_set_integer(pMem, (int64_t)pMem->u.b, false);
-			return 0;
-		}
-		if ((pMem->flags & MEM_Real) != 0) {
-			double d;
-			if (sqlVdbeRealValue(pMem, &d) != 0)
-				return -1;
-			if (d < (double)INT64_MAX && d >= (double)INT64_MIN) {
-				mem_set_integer(pMem, d, d <= -1);
-				return 0;
-			}
-			if (d >= (double)INT64_MAX && d < (double)UINT64_MAX) {
-				mem_set_unsigned(pMem, d);
-				return 0;
-			}
-			return -1;
-		}
-		if (type == FIELD_TYPE_UNSIGNED &&
-		    (pMem->flags & MEM_UInt) == 0)
-			return -1;
-		return 0;
-	case FIELD_TYPE_DOUBLE:
-		return mem_convert_to_double(pMem);
-	case FIELD_TYPE_NUMBER:
-		return mem_convert_to_number(pMem);
-	case FIELD_TYPE_VARBINARY:
-		if ((pMem->flags & MEM_Blob) != 0)
-			return 0;
-		if ((pMem->flags & MEM_Str) != 0) {
-			MemSetTypeFlag(pMem, MEM_Str);
-			return 0;
-		}
-		return -1;
-	default:
-		assert(type == FIELD_TYPE_STRING);
-		assert(MEM_Str == (MEM_Blob >> 3));
-		if ((pMem->flags & MEM_Bool) != 0) {
-			const char *str_bool = SQL_TOKEN_BOOLEAN(pMem->u.b);
-			if (mem_copy_string0(pMem, str_bool) != 0)
-				return -1;
-			return 0;
-		}
-		pMem->flags |= (pMem->flags & MEM_Blob) >> 3;
-			sql_value_apply_type(pMem, FIELD_TYPE_STRING);
-		assert(pMem->flags & MEM_Str || pMem->db->mallocFailed);
-		pMem->flags &=
-			~(MEM_Int | MEM_UInt | MEM_Real | MEM_Blob | MEM_Zero);
-		return 0;
-	}
-}
-
 /*
  * Make sure the given Mem is \u0000 terminated.
  */
@@ -2255,41 +2244,6 @@ mem_apply_type(struct Mem *record, enum field_type type)
 	}
 }
 
-/**
- * Convert the numeric value contained in MEM to unsigned.
- *
- * @param mem The MEM that contains the numeric value.
- * @retval 0 if the conversion was successful, -1 otherwise.
- */
-static int
-mem_convert_to_unsigned(struct Mem *mem)
-{
-	if ((mem->flags & MEM_UInt) != 0)
-		return 0;
-	if ((mem->flags & MEM_Int) != 0)
-		return -1;
-	if ((mem->flags & MEM_Real) == 0)
-		return -1;
-	double d = mem->u.r;
-	if (d < 0.0 || d >= (double)UINT64_MAX)
-		return -1;
-	mem_set_unsigned(mem, (uint64_t) d);
-	return 0;
-}
-
-int
-mem_convert_to_numeric(struct Mem *mem, enum field_type type)
-{
-	assert(mem_is_number(mem) && sql_type_is_numeric(type));
-	assert(type != FIELD_TYPE_NUMBER);
-	if (type == FIELD_TYPE_DOUBLE)
-		return mem_convert_to_double(mem);
-	if (type == FIELD_TYPE_UNSIGNED)
-		return mem_convert_to_unsigned(mem);
-	assert(type == FIELD_TYPE_INTEGER);
-	return mem_convert_to_integer(mem);
-}
-
 /*
  * Make sure pMem->z points to a writable allocation of at least
  * min(n,32) bytes.
diff --git a/src/box/sql/mem.h b/src/box/sql/mem.h
index d370b448b..922dad272 100644
--- a/src/box/sql/mem.h
+++ b/src/box/sql/mem.h
@@ -336,6 +336,9 @@ mem_convert_to_string(struct Mem *mem);
 int
 mem_convert_to_string0(struct Mem *mem);
 
+int
+mem_explicit_cast(struct Mem *mem, enum field_type type);
+
 /**
  * Simple type to str convertor. It is used to simplify
  * error reporting.
@@ -386,7 +389,6 @@ registerTrace(int iReg, Mem *p);
 # define memAboutToChange(P,M)
 #endif
 
-int sqlVdbeMemCast(struct Mem *, enum field_type type);
 int sqlVdbeMemNulTerminate(struct Mem *);
 int sqlVdbeMemExpandBlob(struct Mem *);
 #define ExpandBlob(P) (mem_is_zeroblob(P)? sqlVdbeMemExpandBlob(P) : 0)
@@ -429,17 +431,6 @@ void sql_value_apply_type(struct Mem *val, enum field_type type);
 int
 mem_apply_type(struct Mem *record, enum field_type type);
 
-/**
- * Convert the numeric value contained in MEM to another numeric
- * type.
- *
- * @param mem The MEM that contains the numeric value.
- * @param type The type to convert to.
- * @retval 0 if the conversion was successful, -1 otherwise.
- */
-int
-mem_convert_to_numeric(struct Mem *mem, enum field_type type);
-
 /** Setters = Change MEM value. */
 
 int sqlVdbeMemGrow(struct Mem * pMem, int n, int preserve);
diff --git a/src/box/sql/vdbe.c b/src/box/sql/vdbe.c
index 6799cc9aa..a567f69bd 100644
--- a/src/box/sql/vdbe.c
+++ b/src/box/sql/vdbe.c
@@ -1389,7 +1389,7 @@ case OP_Cast: {                  /* in1 */
 	pIn1 = &aMem[pOp->p1];
 	if (ExpandBlob(pIn1) != 0)
 		goto abort_due_to_error;
-	rc = sqlVdbeMemCast(pIn1, pOp->p2);
+	rc = mem_explicit_cast(pIn1, pOp->p2);
 	/*
 	 * SCALAR is not type itself, but rather an aggregation
 	 * of types. Hence, cast to this type shouldn't change
@@ -2009,7 +2009,7 @@ case OP_ApplyType: {
 			if (!mem_is_number(pIn1))
 				goto type_mismatch;
 			/* Try to convert numeric-to-numeric. */
-			if (mem_convert_to_numeric(pIn1, type) != 0)
+			if (mem_explicit_cast(pIn1, type) != 0)
 				goto type_mismatch;
 		}
 		pIn1++;
-- 
2.25.1