Merging upstream version 18.3.0.

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

6 files changed, 70 insertions, 3 deletions

diff --git a/sqlglot/dialects/doris.py b/sqlglot/dialects/doris.py
index 4b8919c..bd7e0f2 100644
--- a/sqlglot/dialects/doris.py
+++ b/sqlglot/dialects/doris.py
@@ -33,6 +33,8 @@ class Doris(MySQL):
             exp.DataType.Type.TIMESTAMPTZ: "DATETIME",
         }
 
+        TIMESTAMP_FUNC_TYPES = set()
+
         TRANSFORMS = {
             **MySQL.Generator.TRANSFORMS,
             exp.ApproxDistinct: approx_count_distinct_sql,
diff --git a/sqlglot/dialects/mysql.py b/sqlglot/dialects/mysql.py
index f9249eb..6327796 100644
--- a/sqlglot/dialects/mysql.py
+++ b/sqlglot/dialects/mysql.py
@@ -555,7 +555,26 @@ class MySQL(Dialect):
             exp.WeekOfYear: rename_func("WEEKOFYEAR"),
         }
 
-        TYPE_MAPPING = generator.Generator.TYPE_MAPPING.copy()
+        UNSIGNED_TYPE_MAPPING = {
+            exp.DataType.Type.UBIGINT: "BIGINT",
+            exp.DataType.Type.UINT: "INT",
+            exp.DataType.Type.UMEDIUMINT: "MEDIUMINT",
+            exp.DataType.Type.USMALLINT: "SMALLINT",
+            exp.DataType.Type.UTINYINT: "TINYINT",
+        }
+
+        TIMESTAMP_TYPE_MAPPING = {
+            exp.DataType.Type.TIMESTAMP: "DATETIME",
+            exp.DataType.Type.TIMESTAMPTZ: "TIMESTAMP",
+            exp.DataType.Type.TIMESTAMPLTZ: "TIMESTAMP",
+        }
+
+        TYPE_MAPPING = {
+            **generator.Generator.TYPE_MAPPING,
+            **UNSIGNED_TYPE_MAPPING,
+            **TIMESTAMP_TYPE_MAPPING,
+        }
+
         TYPE_MAPPING.pop(exp.DataType.Type.MEDIUMTEXT)
         TYPE_MAPPING.pop(exp.DataType.Type.LONGTEXT)
         TYPE_MAPPING.pop(exp.DataType.Type.MEDIUMBLOB)
@@ -580,6 +599,18 @@ class MySQL(Dialect):
             exp.DataType.Type.VARCHAR: "CHAR",
         }
 
+        TIMESTAMP_FUNC_TYPES = {
+            exp.DataType.Type.TIMESTAMPTZ,
+            exp.DataType.Type.TIMESTAMPLTZ,
+        }
+
+        def datatype_sql(self, expression: exp.DataType) -> str:
+            # https://dev.mysql.com/doc/refman/8.0/en/numeric-type-syntax.html
+            result = super().datatype_sql(expression)
+            if expression.this in self.UNSIGNED_TYPE_MAPPING:
+                result = f"{result} UNSIGNED"
+            return result
+
         def limit_sql(self, expression: exp.Limit, top: bool = False) -> str:
             # MySQL requires simple literal values for its LIMIT clause.
             expression = simplify_literal(expression.copy())
@@ -599,6 +630,9 @@ class MySQL(Dialect):
             return f"{self.sql(expression, 'this')} MEMBER OF({self.sql(expression, 'expression')})"
 
         def cast_sql(self, expression: exp.Cast, safe_prefix: t.Optional[str] = None) -> str:
+            if expression.to.this in self.TIMESTAMP_FUNC_TYPES:
+                return self.func("TIMESTAMP", expression.this)
+
             to = self.CAST_MAPPING.get(expression.to.this)
 
             if to:
diff --git a/sqlglot/dialects/postgres.py b/sqlglot/dialects/postgres.py
index c26e121..5027013 100644
--- a/sqlglot/dialects/postgres.py
+++ b/sqlglot/dialects/postgres.py
@@ -190,7 +190,11 @@ def _remove_target_from_merge(expression: exp.Expression) -> exp.Expression:
     if isinstance(expression, exp.Merge):
         alias = expression.this.args.get("alias")
 
-        normalize = lambda identifier: Postgres.normalize_identifier(identifier).name
+        normalize = (
+            lambda identifier: Postgres.normalize_identifier(identifier).name
+            if identifier
+            else None
+        )
 
         targets = {normalize(expression.this.this)}
 
diff --git a/sqlglot/dialects/spark.py b/sqlglot/dialects/spark.py
index a4435f6..9d4a1ab 100644
--- a/sqlglot/dialects/spark.py
+++ b/sqlglot/dialects/spark.py
@@ -35,6 +35,13 @@ def _parse_datediff(args: t.List) -> exp.Expression:
 
 
 class Spark(Spark2):
+    class Tokenizer(Spark2.Tokenizer):
+        RAW_STRINGS = [
+            (prefix + q, q)
+            for q in t.cast(t.List[str], Spark2.Tokenizer.QUOTES)
+            for prefix in ("r", "R")
+        ]
+
     class Parser(Spark2.Parser):
         FUNCTIONS = {
             **Spark2.Parser.FUNCTIONS,
diff --git a/sqlglot/dialects/teradata.py b/sqlglot/dialects/teradata.py
index 163cc13..d9de968 100644
--- a/sqlglot/dialects/teradata.py
+++ b/sqlglot/dialects/teradata.py
@@ -45,6 +45,7 @@ class Teradata(Dialect):
             "MOD": TokenType.MOD,
             "NE": TokenType.NEQ,
             "NOT=": TokenType.NEQ,
+            "SAMPLE": TokenType.TABLE_SAMPLE,
             "SEL": TokenType.SELECT,
             "ST_GEOMETRY": TokenType.GEOMETRY,
             "TOP": TokenType.TOP,
@@ -55,6 +56,8 @@ class Teradata(Dialect):
         SINGLE_TOKENS.pop("%")
 
     class Parser(parser.Parser):
+        TABLESAMPLE_CSV = True
+
         CHARSET_TRANSLATORS = {
             "GRAPHIC_TO_KANJISJIS",
             "GRAPHIC_TO_LATIN",
@@ -171,6 +174,11 @@ class Teradata(Dialect):
             exp.Use: lambda self, e: f"DATABASE {self.sql(e, 'this')}",
         }
 
+        def tablesample_sql(
+            self, expression: exp.TableSample, seed_prefix: str = "SEED", sep=" AS "
+        ) -> str:
+            return f"{self.sql(expression, 'this')} SAMPLE {self.expressions(expression)}"
+
         def partitionedbyproperty_sql(self, expression: exp.PartitionedByProperty) -> str:
             return f"PARTITION BY {self.sql(expression, 'this')}"
 
diff --git a/sqlglot/dialects/tsql.py b/sqlglot/dialects/tsql.py
index b26f499..19c586e 100644
--- a/sqlglot/dialects/tsql.py
+++ b/sqlglot/dialects/tsql.py
@@ -57,6 +57,8 @@ TRANSPILE_SAFE_NUMBER_FMT = {"N", "C"}
 
 DEFAULT_START_DATE = datetime.date(1900, 1, 1)
 
+BIT_TYPES = {exp.EQ, exp.NEQ, exp.Is, exp.In, exp.Select, exp.Alias}
+
 
 def _format_time_lambda(
     exp_class: t.Type[E], full_format_mapping: t.Optional[bool] = None
@@ -584,6 +586,7 @@ class TSQL(Dialect):
         RETURNING_END = False
         NVL2_SUPPORTED = False
         ALTER_TABLE_ADD_COLUMN_KEYWORD = False
+        LIMIT_FETCH = "FETCH"
 
         TYPE_MAPPING = {
             **generator.Generator.TYPE_MAPPING,
@@ -630,7 +633,16 @@ class TSQL(Dialect):
             exp.VolatileProperty: exp.Properties.Location.UNSUPPORTED,
         }
 
-        LIMIT_FETCH = "FETCH"
+        def boolean_sql(self, expression: exp.Boolean) -> str:
+            if type(expression.parent) in BIT_TYPES:
+                return "1" if expression.this else "0"
+
+            return "(1 = 1)" if expression.this else "(1 = 0)"
+
+        def is_sql(self, expression: exp.Is) -> str:
+            if isinstance(expression.expression, exp.Boolean):
+                return self.binary(expression, "=")
+            return self.binary(expression, "IS")
 
         def createable_sql(self, expression: exp.Create, locations: t.DefaultDict) -> str:
             sql = self.sql(expression, "this")