summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorDaniel Baumann <daniel.baumann@progress-linux.org>2022-09-23 17:07:13 +0000
committerDaniel Baumann <daniel.baumann@progress-linux.org>2022-09-23 17:07:13 +0000
commita3a22f545e149b714b6f2b17cd04a5d6e0bd5294 (patch)
treec85091fdb1158b1f0edaf40267d812197c0bc76c
parentAdding upstream version 6.1.1. (diff)
downloadsqlglot-a3a22f545e149b714b6f2b17cd04a5d6e0bd5294.tar.xz
sqlglot-a3a22f545e149b714b6f2b17cd04a5d6e0bd5294.zip
Adding upstream version 6.2.0.upstream/6.2.0
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to '')
-rw-r--r--CHANGELOG.md17
-rw-r--r--sqlglot/__init__.py2
-rw-r--r--sqlglot/dialects/__init__.py1
-rw-r--r--sqlglot/dialects/dialect.py19
-rw-r--r--sqlglot/dialects/duckdb.py2
-rw-r--r--sqlglot/dialects/snowflake.py20
-rw-r--r--sqlglot/dialects/spark.py1
-rw-r--r--sqlglot/dialects/tsql.py38
-rw-r--r--sqlglot/expressions.py74
-rw-r--r--sqlglot/generator.py5
-rw-r--r--sqlglot/parser.py26
-rw-r--r--sqlglot/tokens.py14
-rw-r--r--tests/dialects/test_dialect.py5
-rw-r--r--tests/dialects/test_duckdb.py3
-rw-r--r--tests/dialects/test_hive.py2
-rw-r--r--tests/dialects/test_oracle.py6
-rw-r--r--tests/dialects/test_presto.py4
-rw-r--r--tests/dialects/test_snowflake.py45
-rw-r--r--tests/dialects/test_spark.py12
-rw-r--r--tests/dialects/test_tsql.py26
-rw-r--r--tests/test_expressions.py32
-rw-r--r--tests/test_parser.py3
22 files changed, 312 insertions, 45 deletions
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 0eba6cc..f16fc70 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,6 +1,23 @@
Changelog
=========
+v6.2.0
+------
+
+Changes:
+
+- New: TSQL support
+
+- Breaking: Removed $ from tokenizer, added @ placeholders
+
+- Improvement: Nodes can now be removed in transform and replace [8cd81c3](https://github.com/tobymao/sqlglot/commit/8cd81c36561463b9849a8e0c2d70248c5b1feb62)
+
+- Improvement: Snowflake timestamp support
+
+- Improvement: Property conversion for CTAS Builder
+
+- Improvement: Tokenizers are now unique per dialect instance
+
v6.1.0
------
diff --git a/sqlglot/__init__.py b/sqlglot/__init__.py
index 3fa40ce..44d349b 100644
--- a/sqlglot/__init__.py
+++ b/sqlglot/__init__.py
@@ -20,7 +20,7 @@ from sqlglot.generator import Generator
from sqlglot.parser import Parser
from sqlglot.tokens import Tokenizer, TokenType
-__version__ = "6.1.1"
+__version__ = "6.2.0"
pretty = False
diff --git a/sqlglot/dialects/__init__.py b/sqlglot/dialects/__init__.py
index f7d03ad..0f80723 100644
--- a/sqlglot/dialects/__init__.py
+++ b/sqlglot/dialects/__init__.py
@@ -14,3 +14,4 @@ from sqlglot.dialects.sqlite import SQLite
from sqlglot.dialects.starrocks import StarRocks
from sqlglot.dialects.tableau import Tableau
from sqlglot.dialects.trino import Trino
+from sqlglot.dialects.tsql import TSQL
diff --git a/sqlglot/dialects/dialect.py b/sqlglot/dialects/dialect.py
index f338c81..0120e71 100644
--- a/sqlglot/dialects/dialect.py
+++ b/sqlglot/dialects/dialect.py
@@ -27,6 +27,7 @@ class Dialects(str, Enum):
STARROCKS = "starrocks"
TABLEAU = "tableau"
TRINO = "trino"
+ TSQL = "tsql"
class _Dialect(type):
@@ -53,7 +54,6 @@ class _Dialect(type):
klass.parser_class = getattr(klass, "Parser", Parser)
klass.generator_class = getattr(klass, "Generator", Generator)
- klass.tokenizer = klass.tokenizer_class()
klass.quote_start, klass.quote_end = list(klass.tokenizer_class._QUOTES.items())[0]
klass.identifier_start, klass.identifier_end = list(klass.tokenizer_class._IDENTIFIERS.items())[0]
@@ -95,7 +95,6 @@ class Dialect(metaclass=_Dialect):
tokenizer_class = None
parser_class = None
generator_class = None
- tokenizer = None
@classmethod
def get_or_raise(cls, dialect):
@@ -138,6 +137,12 @@ class Dialect(metaclass=_Dialect):
def transpile(self, code, **opts):
return self.generate(self.parse(code), **opts)
+ @property
+ def tokenizer(self):
+ if not hasattr(self, "_tokenizer"):
+ self._tokenizer = self.tokenizer_class()
+ return self._tokenizer
+
def parser(self, **opts):
return self.parser_class(
**{
@@ -170,7 +175,15 @@ class Dialect(metaclass=_Dialect):
def rename_func(name):
- return lambda self, expression: f"{name}({csv(*[self.sql(e) for e in expression.args.values()])})"
+ def _rename(self, expression):
+ args = (
+ self.expressions(expression, flat=True)
+ if isinstance(expression, exp.Func) and expression.is_var_len_args
+ else csv(*[self.sql(e) for e in expression.args.values()])
+ )
+ return f"{name}({args})"
+
+ return _rename
def approx_count_distinct_sql(self, expression):
diff --git a/sqlglot/dialects/duckdb.py b/sqlglot/dialects/duckdb.py
index ff3a8b1..4ca9e84 100644
--- a/sqlglot/dialects/duckdb.py
+++ b/sqlglot/dialects/duckdb.py
@@ -108,7 +108,7 @@ class DuckDB(Dialect):
TRANSFORMS = {
**Generator.TRANSFORMS,
exp.ApproxDistinct: approx_count_distinct_sql,
- exp.Array: lambda self, e: f"LIST_VALUE({self.expressions(e, flat=True)})",
+ exp.Array: rename_func("LIST_VALUE"),
exp.ArraySize: rename_func("ARRAY_LENGTH"),
exp.ArraySort: _array_sort_sql,
exp.ArraySum: rename_func("LIST_SUM"),
diff --git a/sqlglot/dialects/snowflake.py b/sqlglot/dialects/snowflake.py
index 8d6ee78..b5d4f0a 100644
--- a/sqlglot/dialects/snowflake.py
+++ b/sqlglot/dialects/snowflake.py
@@ -106,6 +106,11 @@ class Snowflake(Dialect):
"TO_TIMESTAMP": _snowflake_to_timestamp,
}
+ FUNCTION_PARSERS = {
+ **Parser.FUNCTION_PARSERS,
+ "DATE_PART": lambda self: self._parse_extract(),
+ }
+
COLUMN_OPERATORS = {
**Parser.COLUMN_OPERATORS,
TokenType.COLON: lambda self, this, path: self.expression(
@@ -118,10 +123,20 @@ class Snowflake(Dialect):
class Tokenizer(Tokenizer):
QUOTES = ["'", "$$"]
ESCAPE = "\\"
+
+ SINGLE_TOKENS = {
+ **Tokenizer.SINGLE_TOKENS,
+ "$": TokenType.DOLLAR, # needed to break for quotes
+ }
+
KEYWORDS = {
**Tokenizer.KEYWORDS,
"QUALIFY": TokenType.QUALIFY,
"DOUBLE PRECISION": TokenType.DOUBLE,
+ "TIMESTAMP_LTZ": TokenType.TIMESTAMPLTZ,
+ "TIMESTAMP_NTZ": TokenType.TIMESTAMP,
+ "TIMESTAMP_TZ": TokenType.TIMESTAMPTZ,
+ "TIMESTAMPNTZ": TokenType.TIMESTAMP,
}
class Generator(Generator):
@@ -132,6 +147,11 @@ class Snowflake(Dialect):
exp.UnixToTime: _unix_to_time,
}
+ TYPE_MAPPING = {
+ **Generator.TYPE_MAPPING,
+ exp.DataType.Type.TIMESTAMP: "TIMESTAMPNTZ",
+ }
+
def except_op(self, expression):
if not expression.args.get("distinct", False):
self.unsupported("EXCEPT with All is not supported in Snowflake")
diff --git a/sqlglot/dialects/spark.py b/sqlglot/dialects/spark.py
index a331191..c051178 100644
--- a/sqlglot/dialects/spark.py
+++ b/sqlglot/dialects/spark.py
@@ -82,6 +82,7 @@ class Spark(Hive):
TRANSFORMS = {
**{k: v for k, v in Hive.Generator.TRANSFORMS.items() if k not in {exp.ArraySort}},
+ exp.FileFormatProperty: lambda self, e: f"USING {e.text('value').upper()}",
exp.ArraySum: lambda self, e: f"AGGREGATE({self.sql(e, 'this')}, 0, (acc, x) -> acc + x, acc -> acc)",
exp.BitwiseLeftShift: rename_func("SHIFTLEFT"),
exp.BitwiseRightShift: rename_func("SHIFTRIGHT"),
diff --git a/sqlglot/dialects/tsql.py b/sqlglot/dialects/tsql.py
new file mode 100644
index 0000000..68bb9bd
--- /dev/null
+++ b/sqlglot/dialects/tsql.py
@@ -0,0 +1,38 @@
+from sqlglot import exp
+from sqlglot.dialects.dialect import Dialect
+from sqlglot.generator import Generator
+from sqlglot.tokens import Tokenizer, TokenType
+
+
+class TSQL(Dialect):
+ null_ordering = "nulls_are_small"
+ time_format = "'yyyy-mm-dd hh:mm:ss'"
+
+ class Tokenizer(Tokenizer):
+ IDENTIFIERS = ['"', ("[", "]")]
+
+ KEYWORDS = {
+ **Tokenizer.KEYWORDS,
+ "BIT": TokenType.BOOLEAN,
+ "REAL": TokenType.FLOAT,
+ "NTEXT": TokenType.TEXT,
+ "SMALLDATETIME": TokenType.DATETIME,
+ "DATETIMEOFFSET": TokenType.TIMESTAMPTZ,
+ "TIME": TokenType.TIMESTAMP,
+ "VARBINARY": TokenType.BINARY,
+ "IMAGE": TokenType.IMAGE,
+ "MONEY": TokenType.MONEY,
+ "SMALLMONEY": TokenType.SMALLMONEY,
+ "ROWVERSION": TokenType.ROWVERSION,
+ "SQL_VARIANT": TokenType.SQL_VARIANT,
+ "UNIQUEIDENTIFIER": TokenType.UNIQUEIDENTIFIER,
+ "XML": TokenType.XML,
+ }
+
+ class Generator(Generator):
+ TYPE_MAPPING = {
+ **Generator.TYPE_MAPPING,
+ exp.DataType.Type.BOOLEAN: "BIT",
+ exp.DataType.Type.INT: "INTEGER",
+ exp.DataType.Type.DECIMAL: "NUMERIC",
+ }
diff --git a/sqlglot/expressions.py b/sqlglot/expressions.py
index b983bf9..9299132 100644
--- a/sqlglot/expressions.py
+++ b/sqlglot/expressions.py
@@ -1,4 +1,5 @@
import inspect
+import numbers
import re
import sys
from collections import deque
@@ -6,7 +7,7 @@ from copy import deepcopy
from enum import auto
from sqlglot.errors import ParseError
-from sqlglot.helper import AutoName, camel_to_snake_case, ensure_list
+from sqlglot.helper import AutoName, camel_to_snake_case, ensure_list, list_get
class _Expression(type):
@@ -350,7 +351,8 @@ class Expression(metaclass=_Expression):
Args:
fun (function): a function which takes a node as an argument and returns a
- new transformed node or the same node without modifications.
+ new transformed node or the same node without modifications. If the function
+ returns None, then the corresponding node will be removed from the syntax tree.
copy (bool): if set to True a new tree instance is constructed, otherwise the tree is
modified in place.
@@ -360,9 +362,7 @@ class Expression(metaclass=_Expression):
node = self.copy() if copy else self
new_node = fun(node, *args, **kwargs)
- if new_node is None:
- raise ValueError("A transformed node cannot be None")
- if not isinstance(new_node, Expression):
+ if new_node is None or not isinstance(new_node, Expression):
return new_node
if new_node is not node:
new_node.parent = node.parent
@@ -843,10 +843,6 @@ class Ordered(Expression):
arg_types = {"this": True, "desc": True, "nulls_first": True}
-class Properties(Expression):
- arg_types = {"expressions": True}
-
-
class Property(Expression):
arg_types = {"this": True, "value": True}
@@ -891,6 +887,42 @@ class AnonymousProperty(Property):
pass
+class Properties(Expression):
+ arg_types = {"expressions": True}
+
+ PROPERTY_KEY_MAPPING = {
+ "AUTO_INCREMENT": AutoIncrementProperty,
+ "CHARACTER_SET": CharacterSetProperty,
+ "COLLATE": CollateProperty,
+ "COMMENT": SchemaCommentProperty,
+ "ENGINE": EngineProperty,
+ "FORMAT": FileFormatProperty,
+ "LOCATION": LocationProperty,
+ "PARTITIONED_BY": PartitionedByProperty,
+ "TABLE_FORMAT": TableFormatProperty,
+ }
+
+ @classmethod
+ def from_dict(cls, properties_dict):
+ expressions = []
+ for key, value in properties_dict.items():
+ property_cls = cls.PROPERTY_KEY_MAPPING.get(key.upper(), AnonymousProperty)
+ expressions.append(property_cls(this=Literal.string(key), value=cls._convert_value(value)))
+ return cls(expressions=expressions)
+
+ @staticmethod
+ def _convert_value(value):
+ if isinstance(value, Expression):
+ return value
+ if isinstance(value, str):
+ return Literal.string(value)
+ if isinstance(value, numbers.Number):
+ return Literal.number(value)
+ if isinstance(value, list):
+ return Tuple(expressions=[_convert_value(v) for v in value])
+ raise ValueError(f"Unsupported type '{type(value)}' for value '{value}'")
+
+
class Qualify(Expression):
pass
@@ -1562,15 +1594,7 @@ class Select(Subqueryable, Expression):
)
properties_expression = None
if properties:
- properties_str = " ".join(
- [f"{k} = '{v}'" if isinstance(v, str) else f"{k} = {v}" for k, v in properties.items()]
- )
- properties_expression = maybe_parse(
- properties_str,
- into=Properties,
- dialect=dialect,
- **opts,
- )
+ properties_expression = Properties.from_dict(properties)
return Create(
this=table_expression,
@@ -1650,6 +1674,10 @@ class Star(Expression):
return "*"
+class Parameter(Expression):
+ pass
+
+
class Placeholder(Expression):
arg_types = {}
@@ -1688,6 +1716,7 @@ class DataType(Expression):
INTERVAL = auto()
TIMESTAMP = auto()
TIMESTAMPTZ = auto()
+ TIMESTAMPLTZ = auto()
DATE = auto()
DATETIME = auto()
ARRAY = auto()
@@ -1702,6 +1731,13 @@ class DataType(Expression):
SERIAL = auto()
SMALLSERIAL = auto()
BIGSERIAL = auto()
+ XML = auto()
+ UNIQUEIDENTIFIER = auto()
+ MONEY = auto()
+ SMALLMONEY = auto()
+ ROWVERSION = auto()
+ IMAGE = auto()
+ SQL_VARIANT = auto()
@classmethod
def build(cls, dtype, **kwargs):
@@ -2976,7 +3012,7 @@ def replace_children(expression, fun):
else:
new_child_nodes.append(cn)
- expression.args[k] = new_child_nodes if is_list_arg else new_child_nodes[0]
+ expression.args[k] = new_child_nodes if is_list_arg else list_get(new_child_nodes, 0)
def column_table_names(expression):
diff --git a/sqlglot/generator.py b/sqlglot/generator.py
index a445178..d264e59 100644
--- a/sqlglot/generator.py
+++ b/sqlglot/generator.py
@@ -748,6 +748,9 @@ class Generator:
def structkwarg_sql(self, expression):
return f"{self.sql(expression, 'this')} {self.sql(expression, 'expression')}"
+ def parameter_sql(self, expression):
+ return f"@{self.sql(expression, 'this')}"
+
def placeholder_sql(self, *_):
return "?"
@@ -903,7 +906,7 @@ class Generator:
return f"UNIQUE ({columns})"
def if_sql(self, expression):
- return self.case_sql(exp.Case(ifs=[expression], default=expression.args.get("false")))
+ return self.case_sql(exp.Case(ifs=[expression.copy()], default=expression.args.get("false")))
def in_sql(self, expression):
query = expression.args.get("query")
diff --git a/sqlglot/parser.py b/sqlglot/parser.py
index f46bafe..6ad6391 100644
--- a/sqlglot/parser.py
+++ b/sqlglot/parser.py
@@ -81,6 +81,7 @@ class Parser:
TokenType.INTERVAL,
TokenType.TIMESTAMP,
TokenType.TIMESTAMPTZ,
+ TokenType.TIMESTAMPLTZ,
TokenType.DATETIME,
TokenType.DATE,
TokenType.DECIMAL,
@@ -92,6 +93,13 @@ class Parser:
TokenType.SERIAL,
TokenType.SMALLSERIAL,
TokenType.BIGSERIAL,
+ TokenType.XML,
+ TokenType.UNIQUEIDENTIFIER,
+ TokenType.MONEY,
+ TokenType.SMALLMONEY,
+ TokenType.ROWVERSION,
+ TokenType.IMAGE,
+ TokenType.SQL_VARIANT,
*NESTED_TYPE_TOKENS,
}
@@ -233,6 +241,7 @@ class Parser:
TIMESTAMPS = {
TokenType.TIMESTAMP,
TokenType.TIMESTAMPTZ,
+ TokenType.TIMESTAMPLTZ,
}
SET_OPERATIONS = {
@@ -315,6 +324,7 @@ class Parser:
TokenType.TRUE: lambda *_: exp.Boolean(this=True),
TokenType.FALSE: lambda *_: exp.Boolean(this=False),
TokenType.PLACEHOLDER: lambda *_: exp.Placeholder(),
+ TokenType.PARAMETER: lambda self, _: exp.Parameter(this=self._parse_var() or self._parse_primary()),
TokenType.BIT_STRING: lambda _, token: exp.BitString(this=token.text),
TokenType.HEX_STRING: lambda _, token: exp.HexString(this=token.text),
TokenType.INTRODUCER: lambda self, token: self.expression(
@@ -1497,12 +1507,19 @@ class Parser:
if type_token in self.TIMESTAMPS:
tz = self._match(TokenType.WITH_TIME_ZONE) or type_token == TokenType.TIMESTAMPTZ
- self._match(TokenType.WITHOUT_TIME_ZONE)
if tz:
return exp.DataType(
this=exp.DataType.Type.TIMESTAMPTZ,
expressions=expressions,
)
+ ltz = self._match(TokenType.WITH_LOCAL_TIME_ZONE) or type_token == TokenType.TIMESTAMPLTZ
+ if ltz:
+ return exp.DataType(
+ this=exp.DataType.Type.TIMESTAMPLTZ,
+ expressions=expressions,
+ )
+ self._match(TokenType.WITHOUT_TIME_ZONE)
+
return exp.DataType(
this=exp.DataType.Type.TIMESTAMP,
expressions=expressions,
@@ -1845,8 +1862,11 @@ class Parser:
def _parse_extract(self):
this = self._parse_var() or self._parse_type()
- if not self._match(TokenType.FROM):
- self.raise_error("Expected FROM after EXTRACT", self._prev)
+ if self._match(TokenType.FROM):
+ return self.expression(exp.Extract, this=this, expression=self._parse_bitwise())
+
+ if not self._match(TokenType.COMMA):
+ self.raise_error("Expected FROM or comma after EXTRACT", self._prev)
return self.expression(exp.Extract, this=this, expression=self._parse_bitwise())
diff --git a/sqlglot/tokens.py b/sqlglot/tokens.py
index bd95bc7..7a50fc3 100644
--- a/sqlglot/tokens.py
+++ b/sqlglot/tokens.py
@@ -41,6 +41,7 @@ class TokenType(AutoName):
LR_ARROW = auto()
ANNOTATION = auto()
DOLLAR = auto()
+ PARAMETER = auto()
SPACE = auto()
BREAK = auto()
@@ -75,6 +76,7 @@ class TokenType(AutoName):
JSON = auto()
TIMESTAMP = auto()
TIMESTAMPTZ = auto()
+ TIMESTAMPLTZ = auto()
DATETIME = auto()
DATE = auto()
UUID = auto()
@@ -86,6 +88,13 @@ class TokenType(AutoName):
SERIAL = auto()
SMALLSERIAL = auto()
BIGSERIAL = auto()
+ XML = auto()
+ UNIQUEIDENTIFIER = auto()
+ MONEY = auto()
+ SMALLMONEY = auto()
+ ROWVERSION = auto()
+ IMAGE = auto()
+ SQL_VARIANT = auto()
# keywords
ADD_FILE = auto()
@@ -247,6 +256,7 @@ class TokenType(AutoName):
WINDOW = auto()
WITH = auto()
WITH_TIME_ZONE = auto()
+ WITH_LOCAL_TIME_ZONE = auto()
WITHIN_GROUP = auto()
WITHOUT_TIME_ZONE = auto()
UNIQUE = auto()
@@ -340,7 +350,7 @@ class Tokenizer(metaclass=_Tokenizer):
"~": TokenType.TILDA,
"?": TokenType.PLACEHOLDER,
"#": TokenType.ANNOTATION,
- "$": TokenType.DOLLAR,
+ "@": TokenType.PARAMETER,
# used for breaking a var like x'y' but nothing else
# the token type doesn't matter
"'": TokenType.QUOTE,
@@ -520,6 +530,7 @@ class Tokenizer(metaclass=_Tokenizer):
"WHERE": TokenType.WHERE,
"WITH": TokenType.WITH,
"WITH TIME ZONE": TokenType.WITH_TIME_ZONE,
+ "WITH LOCAL TIME ZONE": TokenType.WITH_LOCAL_TIME_ZONE,
"WITHIN GROUP": TokenType.WITHIN_GROUP,
"WITHOUT TIME ZONE": TokenType.WITHOUT_TIME_ZONE,
"ARRAY": TokenType.ARRAY,
@@ -561,6 +572,7 @@ class Tokenizer(metaclass=_Tokenizer):
"BYTEA": TokenType.BINARY,
"TIMESTAMP": TokenType.TIMESTAMP,
"TIMESTAMPTZ": TokenType.TIMESTAMPTZ,
+ "TIMESTAMPLTZ": TokenType.TIMESTAMPLTZ,
"DATE": TokenType.DATE,
"DATETIME": TokenType.DATETIME,
"UNIQUE": TokenType.UNIQUE,
diff --git a/tests/dialects/test_dialect.py b/tests/dialects/test_dialect.py
index 6b7bfd3..4e0a3c6 100644
--- a/tests/dialects/test_dialect.py
+++ b/tests/dialects/test_dialect.py
@@ -228,6 +228,7 @@ class TestDialect(Validator):
"duckdb": "STRPTIME(x, '%Y-%m-%dT%H:%M:%S')",
"hive": "CAST(FROM_UNIXTIME(UNIX_TIMESTAMP(x, 'yyyy-MM-ddTHH:mm:ss')) AS TIMESTAMP)",
"presto": "DATE_PARSE(x, '%Y-%m-%dT%H:%i:%S')",
+ "redshift": "TO_TIMESTAMP(x, 'YYYY-MM-DDTHH:MI:SS')",
"spark": "TO_TIMESTAMP(x, 'yyyy-MM-ddTHH:mm:ss')",
},
)
@@ -237,6 +238,7 @@ class TestDialect(Validator):
"duckdb": "STRPTIME('2020-01-01', '%Y-%m-%d')",
"hive": "CAST('2020-01-01' AS TIMESTAMP)",
"presto": "DATE_PARSE('2020-01-01', '%Y-%m-%d')",
+ "redshift": "TO_TIMESTAMP('2020-01-01', 'YYYY-MM-DD')",
"spark": "TO_TIMESTAMP('2020-01-01', 'yyyy-MM-dd')",
},
)
@@ -246,6 +248,7 @@ class TestDialect(Validator):
"duckdb": "STRPTIME(x, '%y')",
"hive": "CAST(FROM_UNIXTIME(UNIX_TIMESTAMP(x, 'yy')) AS TIMESTAMP)",
"presto": "DATE_PARSE(x, '%y')",
+ "redshift": "TO_TIMESTAMP(x, 'YY')",
"spark": "TO_TIMESTAMP(x, 'yy')",
},
)
@@ -287,6 +290,7 @@ class TestDialect(Validator):
"duckdb": "STRFTIME(x, '%Y-%m-%d')",
"hive": "DATE_FORMAT(x, 'yyyy-MM-dd')",
"presto": "DATE_FORMAT(x, '%Y-%m-%d')",
+ "redshift": "TO_CHAR(x, 'YYYY-MM-DD')",
},
)
self.validate_all(
@@ -295,6 +299,7 @@ class TestDialect(Validator):
"duckdb": "CAST(x AS TEXT)",
"hive": "CAST(x AS STRING)",
"presto": "CAST(x AS VARCHAR)",
+ "redshift": "CAST(x AS TEXT)",
},
)
self.validate_all(
diff --git a/tests/dialects/test_duckdb.py b/tests/dialects/test_duckdb.py
index 501301f..f52decb 100644
--- a/tests/dialects/test_duckdb.py
+++ b/tests/dialects/test_duckdb.py
@@ -66,6 +66,9 @@ class TestDuckDB(Validator):
def test_duckdb(self):
self.validate_all(
"LIST_VALUE(0, 1, 2)",
+ read={
+ "spark": "ARRAY(0, 1, 2)",
+ },
write={
"bigquery": "[0, 1, 2]",
"duckdb": "LIST_VALUE(0, 1, 2)",
diff --git a/tests/dialects/test_hive.py b/tests/dialects/test_hive.py
index 55086e3..a9b5168 100644
--- a/tests/dialects/test_hive.py
+++ b/tests/dialects/test_hive.py
@@ -131,7 +131,7 @@ class TestHive(Validator):
write={
"presto": "CREATE TABLE test WITH (FORMAT = 'parquet', x = '1', Z = '2') AS SELECT 1",
"hive": "CREATE TABLE test STORED AS PARQUET TBLPROPERTIES ('x' = '1', 'Z' = '2') AS SELECT 1",
- "spark": "CREATE TABLE test STORED AS PARQUET TBLPROPERTIES ('x' = '1', 'Z' = '2') AS SELECT 1",
+ "spark": "CREATE TABLE test USING PARQUET TBLPROPERTIES ('x' = '1', 'Z' = '2') AS SELECT 1",
},
)
self.validate_all(
diff --git a/tests/dialects/test_oracle.py b/tests/dialects/test_oracle.py
new file mode 100644
index 0000000..1fadb84
--- /dev/null
+++ b/tests/dialects/test_oracle.py
@@ -0,0 +1,6 @@
+from tests.dialects.test_dialect import Validator
+
+
+class TestOracle(Validator):
+ def test_oracle(self):
+ self.validate_identity("SELECT * FROM V$SESSION")
diff --git a/tests/dialects/test_presto.py b/tests/dialects/test_presto.py
index eb9aa5c..96c299d 100644
--- a/tests/dialects/test_presto.py
+++ b/tests/dialects/test_presto.py
@@ -173,7 +173,7 @@ class TestPresto(Validator):
write={
"presto": "CREATE TABLE test WITH (FORMAT = 'PARQUET') AS SELECT 1",
"hive": "CREATE TABLE test STORED AS PARQUET AS SELECT 1",
- "spark": "CREATE TABLE test STORED AS PARQUET AS SELECT 1",
+ "spark": "CREATE TABLE test USING PARQUET AS SELECT 1",
},
)
self.validate_all(
@@ -181,7 +181,7 @@ class TestPresto(Validator):
write={
"presto": "CREATE TABLE test WITH (FORMAT = 'PARQUET', X = '1', Z = '2') AS SELECT 1",
"hive": "CREATE TABLE test STORED AS PARQUET TBLPROPERTIES ('X' = '1', 'Z' = '2') AS SELECT 1",
- "spark": "CREATE TABLE test STORED AS PARQUET TBLPROPERTIES ('X' = '1', 'Z' = '2') AS SELECT 1",
+ "spark": "CREATE TABLE test USING PARQUET TBLPROPERTIES ('X' = '1', 'Z' = '2') AS SELECT 1",
},
)
self.validate_all(
diff --git a/tests/dialects/test_snowflake.py b/tests/dialects/test_snowflake.py
index 2eeff52..165f8e2 100644
--- a/tests/dialects/test_snowflake.py
+++ b/tests/dialects/test_snowflake.py
@@ -175,3 +175,48 @@ class TestSnowflake(Validator):
"snowflake": r"SELECT FIRST_VALUE(TABLE1.COLUMN1) IGNORE NULLS OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1"
},
)
+
+ def test_timestamps(self):
+ self.validate_all(
+ "SELECT CAST(a AS TIMESTAMP)",
+ write={
+ "snowflake": "SELECT CAST(a AS TIMESTAMPNTZ)",
+ },
+ )
+ self.validate_all(
+ "SELECT a::TIMESTAMP_LTZ(9)",
+ write={
+ "snowflake": "SELECT CAST(a AS TIMESTAMPLTZ(9))",
+ },
+ )
+ self.validate_all(
+ "SELECT a::TIMESTAMPLTZ",
+ write={
+ "snowflake": "SELECT CAST(a AS TIMESTAMPLTZ)",
+ },
+ )
+ self.validate_all(
+ "SELECT a::TIMESTAMP WITH LOCAL TIME ZONE",
+ write={
+ "snowflake": "SELECT CAST(a AS TIMESTAMPLTZ)",
+ },
+ )
+ self.validate_identity("SELECT EXTRACT(month FROM a)")
+ self.validate_all(
+ "SELECT EXTRACT('month', a)",
+ write={
+ "snowflake": "SELECT EXTRACT('month' FROM a)",
+ },
+ )
+ self.validate_all(
+ "SELECT DATE_PART('month', a)",
+ write={
+ "snowflake": "SELECT EXTRACT('month' FROM a)",
+ },
+ )
+ self.validate_all(
+ "SELECT DATE_PART(month FROM a::DATETIME)",
+ write={
+ "snowflake": "SELECT EXTRACT(month FROM CAST(a AS DATETIME))",
+ },
+ )
diff --git a/tests/dialects/test_spark.py b/tests/dialects/test_spark.py
index 8794fed..22f6947 100644
--- a/tests/dialects/test_spark.py
+++ b/tests/dialects/test_spark.py
@@ -44,15 +44,7 @@ class TestSpark(Validator):
write={
"presto": "CREATE TABLE test WITH (FORMAT = 'PARQUET') AS SELECT 1",
"hive": "CREATE TABLE test STORED AS PARQUET AS SELECT 1",
- "spark": "CREATE TABLE test STORED AS PARQUET AS SELECT 1",
- },
- )
- self.validate_all(
- "CREATE TABLE test USING ICEBERG STORED AS PARQUET AS SELECT 1",
- write={
- "presto": "CREATE TABLE test WITH (TABLE_FORMAT = 'ICEBERG', FORMAT = 'PARQUET') AS SELECT 1",
- "hive": "CREATE TABLE test USING ICEBERG STORED AS PARQUET AS SELECT 1",
- "spark": "CREATE TABLE test USING ICEBERG STORED AS PARQUET AS SELECT 1",
+ "spark": "CREATE TABLE test USING PARQUET AS SELECT 1",
},
)
self.validate_all(
@@ -86,7 +78,7 @@ COMMENT 'Test comment: blah'
PARTITIONED BY (
date STRING
)
-STORED AS ICEBERG
+USING ICEBERG
TBLPROPERTIES (
'x' = '1'
)""",
diff --git a/tests/dialects/test_tsql.py b/tests/dialects/test_tsql.py
new file mode 100644
index 0000000..0619eaa
--- /dev/null
+++ b/tests/dialects/test_tsql.py
@@ -0,0 +1,26 @@
+from tests.dialects.test_dialect import Validator
+
+
+class TestTSQL(Validator):
+ dialect = "tsql"
+
+ def test_tsql(self):
+ self.validate_identity('SELECT "x"."y" FROM foo')
+
+ self.validate_all(
+ "SELECT CAST([a].[b] AS SMALLINT) FROM foo",
+ write={
+ "tsql": 'SELECT CAST("a"."b" AS SMALLINT) FROM foo',
+ "spark": "SELECT CAST(`a`.`b` AS SHORT) FROM foo",
+ },
+ )
+
+ def test_types(self):
+ self.validate_identity("CAST(x AS XML)")
+ self.validate_identity("CAST(x AS UNIQUEIDENTIFIER)")
+ self.validate_identity("CAST(x AS MONEY)")
+ self.validate_identity("CAST(x AS SMALLMONEY)")
+ self.validate_identity("CAST(x AS ROWVERSION)")
+ self.validate_identity("CAST(x AS IMAGE)")
+ self.validate_identity("CAST(x AS SQL_VARIANT)")
+ self.validate_identity("CAST(x AS BIT)")
diff --git a/tests/test_expressions.py b/tests/test_expressions.py
index 716e457..59d584c 100644
--- a/tests/test_expressions.py
+++ b/tests/test_expressions.py
@@ -224,9 +224,6 @@ class TestExpressions(unittest.TestCase):
self.assertEqual(actual_expression_2.sql(dialect="presto"), "IF(c - 2 > 0, c - 2, b)")
self.assertIs(actual_expression_2, expression)
- with self.assertRaises(ValueError):
- parse_one("a").transform(lambda n: None)
-
def test_transform_no_infinite_recursion(self):
expression = parse_one("a")
@@ -247,6 +244,35 @@ class TestExpressions(unittest.TestCase):
self.assertEqual(expression.transform(fun).sql(), "SELECT a, b FROM x")
+ def test_transform_node_removal(self):
+ expression = parse_one("SELECT a, b FROM x")
+
+ def remove_column_b(node):
+ if isinstance(node, exp.Column) and node.name == "b":
+ return None
+ return node
+
+ self.assertEqual(expression.transform(remove_column_b).sql(), "SELECT a FROM x")
+ self.assertEqual(expression.transform(lambda _: None), None)
+
+ expression = parse_one("CAST(x AS FLOAT)")
+
+ def remove_non_list_arg(node):
+ if isinstance(node, exp.DataType):
+ return None
+ return node
+
+ self.assertEqual(expression.transform(remove_non_list_arg).sql(), "CAST(x AS )")
+
+ expression = parse_one("SELECT a, b FROM x")
+
+ def remove_all_columns(node):
+ if isinstance(node, exp.Column):
+ return None
+ return node
+
+ self.assertEqual(expression.transform(remove_all_columns).sql(), "SELECT FROM x")
+
def test_replace(self):
expression = parse_one("SELECT a, b FROM x")
expression.find(exp.Column).replace(parse_one("c"))
diff --git a/tests/test_parser.py b/tests/test_parser.py
index 1054103..9e430e2 100644
--- a/tests/test_parser.py
+++ b/tests/test_parser.py
@@ -114,6 +114,9 @@ class TestParser(unittest.TestCase):
with self.assertRaises(ParseError):
parse_one("SELECT FROM x ORDER BY")
+ def test_parameter(self):
+ self.assertEqual(parse_one("SELECT @x, @@x, @1").sql(), "SELECT @x, @@x, @1")
+
def test_annotations(self):
expression = parse_one(
"""