class QualifyClauseSegment(BaseSegment):
"""A `QUALIFY` clause like in `SELECT`.
https://docs.snowflake.com/en/sql-reference/constructs/qualify.html
"""
type = "having_clause"
match_grammar = StartsWith(
"QUALIFY",
terminator=OneOf(
"ORDER",
"LIMIT",
),
)
parse_grammar = Sequence(
"QUALIFY",
Indent,
OneOf(
Bracketed(Ref("ExpressionSegment"), ),
Ref("ExpressionSegment"),
),
Dedent,
)
class AlterTableConstraintClauses(BaseSegment):
"""ALTER TABLE `constraint_clauses` per defined in Oracle's grammar.
https://docs.oracle.com/en/database/oracle/oracle-database/21/sqlrf/ALTER-TABLE.html
If possible, please match the order of this sequence with what's defined in
Oracle's constraint_clauses grammar.
"""
type = "alter_table_constraint_clauses"
match_grammar = OneOf(
Sequence(
"ADD",
Ref("TableConstraintSegment"),
),
# @TODO MODIFY
# @TODO RENAME
# @TODO DROP
# drop_constraint_clause
Sequence(
"DROP",
OneOf(
Sequence(
"PRIMARY",
"KEY",
),
Sequence(
"UNIQUE",
Bracketed(Ref("ColumnReferenceSegment")),
),
Sequence("CONSTRAINT", Ref("ObjectReferenceSegment")),
),
Ref.keyword("CASCADE", optional=True),
Sequence(
OneOf(
"KEEP",
"DROP",
),
"INDEX",
optional=True,
),
Ref.keyword("ONLINE", optional=True),
),
)
class InsertStatementSegment(BaseSegment):
"""An `INSERT` statement.
Full Apache Hive `INSERT` reference here:
https://cwiki.apache.org/confluence/display/Hive/LanguageManual+DML
"""
type = "insert_statement"
match_grammar = StartsWith("INSERT")
parse_grammar = Sequence(
"INSERT",
OneOf(
Sequence(
"OVERWRITE",
OneOf(
Sequence(
"TABLE",
Ref("TableReferenceSegment"),
Ref("PartitionSpecGrammar", optional=True),
Ref("IfNotExistsGrammar", optional=True),
Ref("SelectableGrammar"),
),
Sequence(
Sequence("LOCAL", optional=True),
"DIRECTORY",
Ref("QuotedLiteralSegment"),
Ref("RowFormatClauseSegment", optional=True),
Ref("StoredAsGrammar", optional=True),
Ref("SelectableGrammar"),
),
),
),
Sequence(
"INTO",
"TABLE",
Ref("TableReferenceSegment"),
Ref("PartitionSpecGrammar", optional=True),
OneOf(
Ref("SelectableGrammar"),
Ref("ValuesClauseSegment"),
),
),
),
)
class FunctionDefinitionGrammar(BaseSegment):
"""This is the body of a `CREATE FUNCTION AS` statement."""
match_grammar = Sequence(
AnyNumberOf(
Sequence(
OneOf("DETERMINISTIC", Sequence("NOT", "DETERMINISTIC")),
optional=True,
),
Sequence(
"LANGUAGE",
# Not really a parameter, but best fit for now.
Ref("ParameterNameSegment"),
Sequence(
"OPTIONS",
Bracketed(
Delimited(
Sequence(
Ref("ParameterNameSegment"),
Ref("EqualsSegment"),
Anything(),
),
delimiter=Ref("CommaSegment"),
)),
optional=True,
),
),
# There is some syntax not implemented here,
Sequence(
"AS",
OneOf(
Ref("DoubleQuotedUDFBody"),
Ref("SingleQuotedUDFBody"),
Bracketed(
OneOf(Ref("ExpressionSegment"),
Ref("SelectStatementSegment"))),
),
),
))
class CreateTableStatementSegment(BaseSegment):
"""A `CREATE [MULTISET| SET] TABLE` statement."""
type = "create_table_statement"
match_grammar = Sequence(
"CREATE",
Sequence("OR", "REPLACE", optional=True),
# Adding Teradata specific [MULTISET| SET]
OneOf("SET", "MULTISET", optional=True),
OneOf(Sequence("GLOBAL", "TEMPORARY"), "VOLATILE", optional=True),
"TABLE",
Sequence("IF", "NOT", "EXISTS", optional=True),
Ref("TableReferenceSegment"),
# , NO FALLBACK, NO BEFORE JOURNAL, NO AFTER JOURNAL
OneOf(Ref("TdCreateTableOptions"), optional=True),
OneOf(
# Columns and comment syntax:
Sequence(
Bracketed(
Delimited(
OneOf(
Ref("ColumnDefinitionSegment"),
Ref("TableConstraintSegment"),
),
delimiter=Ref("CommaSegment"),
)
),
Ref("CommentClauseSegment", optional=True),
),
# Create AS syntax:
Sequence("AS", Ref("SelectableGrammar")),
# Create like syntax
Sequence("LIKE", Ref("TableReferenceSegment")),
),
# PRIMARY INDEX( COD_TARJETA, COD_EST, IND_TIPO_TARJETA, FEC_ANIO_MES )
OneOf(Ref("TdTableConstraints"), optional=True),
)
class CreateSchemaStatementSegment(BaseSegment):
"""A `CREATE SCHEMA` statement.
https://docs.aws.amazon.com/redshift/latest/dg/r_CREATE_SCHEMA.html
TODO: support optional SCHEMA_ELEMENT
"""
type = "create_schema_statement"
match_grammar = Sequence(
"CREATE",
"SCHEMA",
OneOf(
Sequence(
Ref("IfNotExistsGrammar", optional=True),
Ref("SchemaReferenceSegment"),
Sequence(
"AUTHORIZATION",
Ref("ObjectReferenceSegment"),
optional=True,
),
),
Sequence(
"AUTHORIZATION",
Ref("ObjectReferenceSegment"),
),
),
Sequence(
"QUOTA",
OneOf(
Sequence(
Ref("NumericLiteralSegment"),
OneOf(
"MB",
"GB",
"TB",
),
),
"UNLIMITED",
),
optional=True,
),
)
class AlterUserStatementSegment(BaseSegment):
"""`ALTER USER` statement.
https://docs.aws.amazon.com/redshift/latest/dg/r_ALTER_USER.html
"""
type = "alter_user"
match_grammar = Sequence(
"ALTER",
"USER",
Ref("ObjectReferenceSegment"),
Ref.keyword("WITH", optional=True),
AnySetOf(
OneOf(
"CREATEDB",
"NOCREATEDB",
),
OneOf(
"CREATEUSER",
"NOCREATEUSER",
),
Sequence(
"SYSLOG",
"ACCESS",
OneOf(
"RESTRICTED",
"UNRESTRICTED",
),
),
Sequence(
"PASSWORD",
OneOf(
Ref("QuotedLiteralSegment"),
"DISABLE",
),
Sequence("VALID",
"UNTIL",
Ref("QuotedLiteralSegment"),
optional=True),
),
Sequence(
"RENAME",
"TO",
Ref("ObjectReferenceSegment"),
),
Sequence(
"CONNECTION",
"LIMIT",
OneOf(
Ref("NumericLiteralSegment"),
"UNLIMITED",
),
),
OneOf(
Sequence(
"SESSION",
"TIMEOUT",
Ref("NumericLiteralSegment"),
),
Sequence(
"RESET",
"SESSION",
"TIMEOUT",
),
),
OneOf(
Sequence(
"SET",
Ref("ObjectReferenceSegment"),
OneOf(
"TO",
Ref("EqualsSegment"),
),
OneOf(
"DEFAULT",
Ref("LiteralGrammar"),
),
),
Sequence(
"RESET",
Ref("ObjectReferenceSegment"),
),
),
min_times=1,
),
)
class DatatypeSegment(BaseSegment):
"""A data type segment.
Indicates a data type.
https://docs.aws.amazon.com/redshift/latest/dg/c_Supported_data_types.html
"""
type = "data_type"
match_grammar = OneOf(
# numeric types
"SMALLINT",
"INT2",
"INTEGER",
"INT",
"INT4",
"BIGINT",
"INT8",
"REAL",
"FLOAT4",
Sequence("DOUBLE", "PRECISION"),
"FLOAT8",
"FLOAT",
# numeric types [precision ["," scale])]
Sequence(
OneOf("DECIMAL", "NUMERIC"),
Bracketed(
Delimited(Ref("NumericLiteralSegment")),
optional=True,
),
),
# character types
OneOf(
Sequence(
OneOf(
"CHAR",
"CHARACTER",
"NCHAR",
"VARCHAR",
Sequence("CHARACTER", "VARYING"),
"NVARCHAR",
),
Bracketed(
OneOf(
Ref("NumericLiteralSegment"),
"MAX",
),
optional=True,
),
),
"BPCHAR",
"TEXT",
),
Ref("DateTimeTypeIdentifier"),
# INTERVAL is a data type *only* for conversion operations
"INTERVAL",
# boolean types
OneOf("BOOLEAN", "BOOL"),
# hllsketch type
"HLLSKETCH",
# super type
"SUPER",
# spatial data
"GEOMETRY",
"GEOGRAPHY",
# binary type
Sequence(
OneOf(
"VARBYTE",
"VARBINARY",
Sequence("BINARY", "VARYING"),
),
Bracketed(
Ref("NumericLiteralSegment"),
optional=True,
),
),
)
class TdCommentStatementSegment(BaseSegment):
"""A `COMMENT` statement.
COMMENT [ON] (object_kind_1|object_kind_2) name [[AS|IS] comment]
object_kind_1: (COLUMN|FUNCTION|GLOP SET|MACRO|MAP|METHOD|PROCEDURE|PROFILE|ROLE|TRIGGER|TYPE|VIEW)
object_kind_2: (DATABASE|FILE|TABLE|USER)
"""
type = "comment_clause"
is_ddl = True
is_dml = False
is_dql = False
is_dcl = False
match_grammar = Sequence(
"COMMENT",
OneOf("ON", optional=True),
OneOf(
Sequence("COLUMN", Ref("ColumnReferenceSegment")),
Sequence("FUNCTION", Ref("ObjectReferenceSegment")),
Sequence("MACRO", Ref("ObjectReferenceSegment")),
Sequence("MAP", Ref("ObjectReferenceSegment")),
Sequence("METHOD", Ref("ObjectReferenceSegment")),
Sequence("PROCEDURE", Ref("ObjectReferenceSegment")),
Sequence("PROFILE", Ref("ObjectReferenceSegment")),
Sequence("ROLE", Ref("ObjectReferenceSegment")),
Sequence("TRIGGER", Ref("ObjectReferenceSegment")),
Sequence("TYPE", Ref("ObjectReferenceSegment")),
Sequence("VIEW", Ref("TableReferenceSegment")),
Sequence("DATABASE", Ref("DatabaseReferenceSegment")),
Sequence("FILE", Ref("ObjectReferenceSegment")),
Sequence("TABLE", Ref("TableReferenceSegment")),
Sequence("USER", Ref("ObjectReferenceSegment")),
),
Sequence(
OneOf("AS", "IS", optional=True),
Ref("QuotedLiteralSegment"),
optional=True,
),
)
class TdOrderByStatClauseSegment(BaseSegment):
"""An `ORDER BY (VALUES|HASH) (column_name)` clause in COLLECT STATS."""
type = "stat_orderby_clause"
match_grammar = Sequence("ORDER", "BY", OneOf("VALUES", "HASH"),
Bracketed(Ref("ColumnReferenceSegment")))
)
# Reserve WITHIN (required for the WithinGroupClauseSegment)
postgres_dialect.sets("unreserved_keywords").remove("WITHIN")
postgres_dialect.sets("reserved_keywords").add("WITHIN")
# Add the EPOCH datetime unit
postgres_dialect.sets("datetime_units").update(["EPOCH"])
postgres_dialect.add(JsonOperatorSegment=NamedSegment.make(
"json_operator", name="json_operator", type="binary_operator"), )
postgres_dialect.replace(
PostFunctionGrammar=OneOf(
Ref("WithinGroupClauseSegment"),
Sequence(
Sequence(OneOf("IGNORE", "RESPECT"), "NULLS", optional=True),
Ref("OverClauseSegment"),
),
),
BinaryOperatorGrammar=OneOf(
Ref("ArithmeticBinaryOperatorGrammar"),
Ref("StringBinaryOperatorGrammar"),
Ref("BooleanBinaryOperatorGrammar"),
Ref("ComparisonOperatorGrammar"),
# Add JSON operators
Ref("JsonOperatorSegment"),
),
)
@postgres_dialect.segment(replace=True)
class SelectClauseModifierSegment(BaseSegment):
class UnloadStatementSegment(BaseSegment):
"""A `UNLOAD` statement.
https://docs.aws.amazon.com/redshift/latest/dg/r_UNLOAD.html
"""
type = "unload_statement"
match_grammar = Sequence(
"UNLOAD",
Bracketed(Ref("QuotedLiteralSegment")),
"TO",
Ref("QuotedLiteralSegment"),
AnySetOf(
Ref("AuthorizationSegment", optional=False),
Sequence(
"REGION",
Ref.keyword("AS", optional=True),
Ref("QuotedLiteralSegment"),
optional=True,
),
Ref("CompressionTypeGrammar", optional=True),
Sequence(
Sequence(
"FORMAT",
Ref.keyword("AS", optional=True),
optional=True,
),
OneOf(
"CSV",
"JSON",
"PARQUET",
),
optional=True,
),
Sequence(
"PARTITION",
"BY",
Ref("BracketedColumnReferenceListGrammar"),
Ref.keyword("INCLUDE", optional=True),
),
Sequence(
"PARALLEL",
OneOf(
"PRESET",
"ON",
"OFF",
"TRUE",
"FALSE",
optional=True,
),
optional=True,
),
OneOf(
Sequence(
"DELIMITER",
Ref.keyword("AS", optional=True),
Ref("QuotedLiteralSegment"),
),
Sequence(
"FIXEDWIDTH",
Ref.keyword("AS", optional=True),
Ref("QuotedLiteralSegment"),
),
optional=True,
),
Sequence(
"MANIFEST",
Ref.keyword("VERBOSE", optional=True),
optional=True,
),
Sequence(
"NULL",
"AS",
Ref("QuotedLiteralSegment"),
optional=True,
),
Sequence(
"NULL",
"AS",
Ref("QuotedLiteralSegment"),
optional=True,
),
AnySetOf(
OneOf(
"MAXFILESIZE",
"ROWGROUPSIZE",
),
Ref.keyword("AS", optional=True),
Ref("NumericLiteralSegment"),
OneOf(
"MB",
"GB",
),
optional=True,
),
Sequence(
"ENCRYPTED",
Ref.keyword("AUTO", optional=True),
optional=True,
),
Ref.keyword("ALLOWOVERWRITE", optional=True),
Ref.keyword("CLEANPATH", optional=True),
Ref.keyword("ESCAPE", optional=True),
Ref.keyword("ADDQUOTES", optional=True),
Ref.keyword("HEADER", optional=True),
),
)
class FunctionSegment(BaseSegment):
"""A scalar or aggregate function.
Extended version of `ansi` to add support of row typecasting
https://prestodb.io/docs/current/language/types.html#row
```
cast(row(val1, val2) as row(a integer, b integer))
```
"""
type = "function"
match_grammar = OneOf(
Sequence(
# Treat functions which take date parts separately
# So those functions parse date parts as DatetimeUnitSegment
# rather than identifiers.
Sequence(
Ref("DatePartFunctionNameSegment"),
Bracketed(
Delimited(
Ref("DatetimeUnitSegment"),
Ref(
"FunctionContentsGrammar",
# The brackets might be empty for some functions...
optional=True,
ephemeral_name="FunctionContentsGrammar",
),
)
),
),
),
Sequence(
# This unusual syntax is used to cast the Keyword ROW to
# to the function_name to avoid rule linting exceptions
StringParser("ROW", KeywordSegment, type="function_name"),
Bracketed(
Delimited(
Sequence(
Ref("BaseExpressionElementGrammar"),
),
),
),
"AS",
"ROW",
Bracketed(
Delimited(
Sequence(
Ref("SingleIdentifierGrammar"),
Ref("DatatypeSegment", optional=True),
),
),
),
),
Sequence(
Sequence(
Ref(
"FunctionNameSegment",
exclude=OneOf(
Ref("DatePartFunctionNameSegment"),
Ref("ValuesClauseSegment"),
),
),
Bracketed(
Ref(
"FunctionContentsGrammar",
# The brackets might be empty for some functions...
optional=True,
ephemeral_name="FunctionContentsGrammar",
)
),
),
Ref("PostFunctionGrammar", optional=True),
),
)
class DatatypeSegment(BaseSegment):
"""Data types."""
type = "data_type"
match_grammar = OneOf(
Ref("PrimitiveTypeSegment"),
Sequence(
"ARRAY",
Bracketed(
Ref("DatatypeSegment"),
bracket_pairs_set="angle_bracket_pairs",
bracket_type="angle",
),
),
Sequence(
"MAP",
Bracketed(
Sequence(
Ref("PrimitiveTypeSegment"),
Ref("CommaSegment"),
Ref("DatatypeSegment"),
),
bracket_pairs_set="angle_bracket_pairs",
bracket_type="angle",
),
),
Sequence(
"STRUCT",
Bracketed(
Delimited(
Sequence(
Ref("NakedIdentifierSegment"),
Ref("ColonSegment"),
Ref("DatatypeSegment"),
Ref("CommentGrammar", optional=True),
),
bracket_pairs_set="angle_bracket_pairs",
),
bracket_pairs_set="angle_bracket_pairs",
bracket_type="angle",
),
),
Sequence(
"UNIONTYPE",
Bracketed(
Delimited(
Ref("DatatypeSegment"), bracket_pairs_set="angle_bracket_pairs"
),
bracket_pairs_set="angle_bracket_pairs",
bracket_type="angle",
),
),
# array types
OneOf(
AnyNumberOf(
Bracketed(
Ref("ExpressionSegment", optional=True), bracket_type="square"
)
),
Ref("SimpleArrayTypeGrammar"),
Sequence(Ref("SimpleArrayTypeGrammar"), Ref("ArrayLiteralSegment")),
optional=True,
),
)
StructKeywordSegment=StringParser("struct", KeywordSegment, name="struct"),
StartAngleBracketSegment=StringParser("<",
SymbolSegment,
name="start_angle_bracket",
type="start_angle_bracket"),
EndAngleBracketSegment=StringParser(">",
SymbolSegment,
name="end_angle_bracket",
type="end_angle_bracket"),
)
bigquery_dialect.replace(
FunctionContentsExpressionGrammar=OneOf(
Ref("DatetimeUnitSegment"),
Sequence(
Ref("ExpressionSegment"),
Sequence(OneOf("IGNORE", "RESPECT"), "NULLS", optional=True),
),
),
SimpleArrayTypeGrammar=Sequence(
"ARRAY",
Bracketed(
Ref("DatatypeIdentifierSegment"),
bracket_type="angle",
bracket_pairs_set="angle_bracket_pairs",
),
),
# BigQuery also supports the special "Struct" construct.
BaseExpressionElementGrammar=ansi_dialect.get_grammar(
"BaseExpressionElementGrammar").copy(
insert=[Ref("TypelessStructSegment")]),
FunctionContentsGrammar=ansi_dialect.get_grammar(
Ref("TdCollectStatisticsStatementSegment"),
Ref("BteqStatementSegment"),
Ref("TdRenameStatementSegment"),
Ref("QualifyClauseSegment"),
Ref("TdCommentStatementSegment"),
Ref("DatabaseStatementSegment"),
Ref("SetSessionStatementSegment"),
],
)
match_grammar = ansi.StatementSegment.match_grammar.copy()
teradata_dialect.add(
TdCastIdentifierGrammar=Sequence(
OneOf("DATE", "TIMESTAMP"), Ref("ExpressionSegment")
),
)
teradata_dialect.replace(
SingleIdentifierGrammar=OneOf(
Ref("NakedIdentifierSegment"),
Ref("QuotedIdentifierSegment"),
Ref("TdCastIdentifierGrammar"),
),
SelectClauseSegmentGrammar=Sequence(
OneOf("SELECT", "SEL"),
Ref("SelectClauseModifierSegment", optional=True),
Indent,
Delimited(
Ref("SelectClauseElementSegment"),
class AlterUserSegment(BaseSegment):
"""`ALTER USER` statement.
https://docs.snowflake.com/en/sql-reference/sql/alter-user.html
All user parameters can be found here
https://docs.snowflake.com/en/sql-reference/parameters.html
"""
type = "alter_user"
match_grammar = StartsWith(Sequence("ALTER", "USER"), )
parse_grammar = Sequence(
"ALTER",
"USER",
Sequence("IF", "EXISTS", optional=True),
Ref("ObjectReferenceSegment"),
OneOf(
Sequence("RENAME", "TO", Ref("ObjectReferenceSegment")),
Sequence("RESET", "PASSWORD"),
Sequence("ABORT", "ALL", "QUERIES"),
Sequence(
"ADD",
"DELEGATED",
"AUTHORIZATION",
"OF",
"ROLE",
Ref("ObjectReferenceSegment"),
"TO",
"SECURITY",
"INTEGRATION",
Ref("ObjectReferenceSegment"),
),
Sequence(
"REMOVE",
"DELEGATED",
OneOf(
Sequence("AUTHORIZATION", "OF", "ROLE",
Ref("ObjectReferenceSegment")),
"AUTHORIZATIONS",
),
"FROM",
"SECURITY",
"INTEGRATION",
Ref("ObjectReferenceSegment"),
),
# Snowflake supports the SET command with space delimitted parameters, but it also supports
# using commas which is better supported by `Delimited`, so we will just use that.
Sequence(
"SET",
Delimited(
Sequence(
Ref("ParameterNameSegment"),
Ref("EqualsSegment"),
OneOf(Ref("LiteralGrammar"),
Ref("ObjectReferenceSegment")),
), ),
),
Sequence("UNSET", Delimited(Ref("ParameterNameSegment"))),
),
)
class ShowStatementSegment(BaseSegment):
"""A snowflake `SHOW` statement.
https://docs.snowflake.com/en/sql-reference/sql/show.html
"""
_object_types_plural = OneOf(
"PARAMETERS",
Sequence("GLOBAL", "ACCOUNTS"),
"REGIONS",
Sequence("REPLICATION", "ACCOUNTS"),
Sequence("REPLICATION", "DATABASES"),
"PARAMETERS",
"VARIABLES",
"TRANSACTIONS",
"LOCKS",
"PARAMETERS",
"FUNCTIONS",
Sequence("NETWORK", "POLICIES"),
"SHARES",
"ROLES",
"GRANTS",
"USERS",
"WAREHOUSES",
"DATABASES",
Sequence(
OneOf("API", "NOTIFICATION", "SECURITY", "STORAGE", optional=True),
"INTEGRATIONS",
),
"SCHEMAS",
"OBJECTS",
"TABLES",
Sequence("EXTERNAL", "TABLES"),
"VIEWS",
Sequence("MATERIALIZED", "VIEWS"),
Sequence("MASKING", "POLICIES"),
"COLUMNS",
Sequence("FILE", "FORMATS"),
"SEQUENCES",
"STAGES",
"PIPES",
"STREAMS",
"TASKS",
Sequence("USER", "FUNCTIONS"),
Sequence("EXTERNAL", "FUNCTIONS"),
"PROCEDURES",
Sequence("FUTURE", "GRANTS"),
)
_object_scope_types = OneOf(
"ACCOUNT",
"SESSION",
Sequence(
OneOf(
"DATABASE",
"SCHEMA",
"SHARE",
"ROLE",
"TABLE",
"TASK",
"USER",
"WAREHOUSE",
"VIEW",
),
Ref("ObjectReferenceSegment", optional=True),
),
)
type = "show_statement"
match_grammar = Sequence(
"SHOW",
OneOf("TERSE", optional=True),
_object_types_plural,
OneOf("HISTORY", optional=True),
Sequence("LIKE", Ref("QuotedLiteralSegment"), optional=True),
Sequence(
OneOf("ON", "TO", "OF", "IN"),
OneOf(
Sequence(_object_scope_types),
Ref("ObjectReferenceSegment"),
),
optional=True,
),
Sequence("STARTS", "WITH", Ref("QuotedLiteralSegment"), optional=True),
Sequence("WITH",
"PRIMARY",
Ref("ObjectReferenceSegment"),
optional=True),
Sequence(
Ref("LimitClauseSegment"),
Sequence("FROM", Ref("QuotedLiteralSegment"), optional=True),
optional=True,
),
)
class CreateStatementSegment(BaseSegment):
"""A snowflake `CREATE` statement.
https://docs.snowflake.com/en/sql-reference/sql/create.html
"""
type = "create_statement"
match_grammar = Sequence(
"CREATE",
Sequence("OR", "REPLACE", optional=True),
OneOf(
Sequence("NETWORK", "POLICY"),
Sequence("RESOURCE", "MONITOR"),
"SHARE",
"ROLE",
"USER",
"WAREHOUSE",
Sequence("NOTIFICATION", "INTEGRATION"),
Sequence("SECURITY", "INTEGRATION"),
Sequence("STORAGE", "INTEGRATION"),
Sequence("EXTERNAL", "TABLE"),
"VIEW",
Sequence("MATERIALIZED", "VIEW"),
Sequence("SECURE", "VIEW"),
Sequence("MASKING", "POLICY"),
"PIPE",
"FUNCTION",
Sequence("EXTERNAL", "FUNCTION"),
"PROCEDURE",
# Objects that also support clone
"DATABASE",
"SCHEMA",
"TABLE",
"SEQUENCE",
Sequence("FILE", "FORMAT"),
"STAGE",
"STREAM",
"TASK",
),
Sequence("IF", "NOT", "EXISTS", optional=True),
Ref("ObjectReferenceSegment"),
Ref("CreateStatementCommentSegment", optional=True),
Ref.keyword("AS", optional=True),
Ref("SelectStatementSegment", optional=True),
)
class CreateModelStatementSegment(BaseSegment):
"""A `CREATE MODEL` statement.
https://docs.aws.amazon.com/redshift/latest/dg/r_CREATE_MODEL.html
NB: order of keywords matter
"""
type = "create_model_statement"
match_grammar = Sequence(
"CREATE",
"MODEL",
Ref("ObjectReferenceSegment"),
Sequence(
"FROM",
OneOf(
Ref("QuotedLiteralSegment"),
Bracketed(Ref("SelectableGrammar")),
Ref("ObjectReferenceSegment"),
),
optional=True,
),
Sequence(
"TARGET",
Ref("ColumnReferenceSegment"),
optional=True,
),
Sequence(
"FUNCTION",
Ref("ObjectReferenceSegment"),
Bracketed(
Delimited(Ref("DatatypeSegment")),
optional=True,
),
),
Sequence(
"RETURNS",
Ref("DatatypeSegment"),
optional=True,
),
Sequence(
"SAGEMAKER",
Ref("QuotedLiteralSegment"),
optional=True,
),
Sequence(
"IAM_ROLE",
OneOf(
"DEFAULT",
Ref("QuotedLiteralSegment"),
),
),
Sequence(
"AUTO",
OneOf(
"ON",
"OFF",
),
optional=True,
),
Sequence(
"MODEL_TYPE",
OneOf(
"XGBOOST",
"MLP",
"KMEANS",
),
optional=True,
),
Sequence(
"PROBLEM_TYPE",
OneOf(
"REGRESSION",
"BINARY_CLASSIFICATION",
"MULTICLASS_CLASSIFICATION",
),
optional=True,
),
Sequence(
"OBJECTIVE",
Ref("QuotedLiteralSegment"),
optional=True,
),
Sequence(
"PREPROCESSORS",
Ref("QuotedLiteralSegment"),
optional=True,
),
Sequence(
"HYPERPARAMETERS",
"DEFAULT",
Sequence(
"EXCEPT",
Bracketed(Delimited(Anything(), ), ),
optional=True,
),
optional=True,
),
Sequence(
"SETTINGS",
Bracketed(
Sequence(
"S3_BUCKET",
Ref("QuotedLiteralSegment"),
Sequence(
"KMS_KEY_ID",
Ref("QuotedLiteralSegment"),
optional=True,
),
Sequence(
"S3_GARBAGE_COLLECT",
OneOf(
"ON",
"OFF",
),
optional=True,
),
Sequence(
"MAX_CELLS",
Ref("NumericLiteralSegment"),
optional=True,
),
Sequence(
"MAX_RUNTIME",
Ref("NumericLiteralSegment"),
optional=True,
),
), ),
optional=True,
),
)
class CreateExternalTableStatementSegment(BaseSegment):
"""A `CREATE EXTERNAL TABLE` statement.
https://docs.aws.amazon.com/redshift/latest/dg/r_CREATE_EXTERNAL_TABLE.html
"""
type = "create_external_table_statement"
match_grammar = Sequence(
"CREATE",
"EXTERNAL",
"TABLE",
Ref("TableReferenceSegment"),
Bracketed(
# Columns and comment syntax:
Delimited(
Sequence(
Ref("ColumnReferenceSegment"),
Ref("DatatypeSegment"),
), ), ),
Ref("PartitionedBySegment", optional=True),
Sequence(
"ROW",
"FORMAT",
OneOf(
Sequence(
"DELIMITED",
Ref("RowFormatDelimitedSegment"),
),
Sequence(
"SERDE",
Ref("QuotedLiteralSegment"),
Sequence(
"WITH",
"SERDEPROPERTIES",
Bracketed(
Delimited(
Sequence(
Ref("QuotedLiteralSegment"),
Ref("EqualsSegment"),
Ref("QuotedLiteralSegment"),
), ), ),
optional=True,
),
),
),
optional=True,
),
"STORED",
"AS",
OneOf(
"PARQUET",
"RCFILE",
"SEQUENCEFILE",
"TEXTFILE",
"ORC",
"AVRO",
Sequence(
"INPUTFORMAT",
Ref("QuotedLiteralSegment"),
"OUTPUTFORMAT",
Ref("QuotedLiteralSegment"),
),
),
"LOCATION",
Ref("QuotedLiteralSegment"),
Sequence(
"TABLE",
"PROPERTIES",
Bracketed(
Delimited(
Sequence(
Ref("QuotedLiteralSegment"),
Ref("EqualsSegment"),
Ref("QuotedLiteralSegment"),
), ), ),
optional=True,
),
)
KeywordSegment,
name="json_file",
type="file_format"),
RcfileKeywordSegment=StringParser("RCFILE",
KeywordSegment,
name="rc_file",
type="file_format"),
SequencefileKeywordSegment=StringParser("SEQUENCEFILE",
KeywordSegment,
name="sequence_file",
type="file_format"),
TextfileKeywordSegment=StringParser("TEXTFILE",
KeywordSegment,
name="text_file",
type="file_format"),
LocationGrammar=Sequence("LOCATION", Ref("QuotedLiteralSegment")),
PropertyGrammar=Sequence(
Ref("QuotedLiteralSegment"),
Ref("EqualsSegment"),
Ref("QuotedLiteralSegment"),
),
BracketedPropertyListGrammar=Bracketed(Delimited(Ref("PropertyGrammar"))),
TablePropertiesGrammar=Sequence("TBLPROPERTIES",
Ref("BracketedPropertyListGrammar")),
SerdePropertiesGrammar=Sequence("WITH", "SERDEPROPERTIES",
Ref("BracketedPropertyListGrammar")),
TerminatedByGrammar=Sequence("TERMINATED", "BY",
Ref("QuotedLiteralSegment")),
FileFormatGrammar=OneOf(
"SEQUENCEFILE",
"TEXTFILE",
class CopyStatementSegment(BaseSegment):
"""A `COPY` statement.
:
- https://docs.aws.amazon.com/redshift/latest/dg/r_COPY.html
- https://docs.aws.amazon.com/redshift/latest/dg/r_COPY-parameters.html
"""
type = "copy_statement"
match_grammar = Sequence(
"COPY",
Ref("TableReferenceSegment"),
Ref("BracketedColumnReferenceListGrammar", optional=True),
"FROM",
Ref("QuotedLiteralSegment"),
AnySetOf(
Ref("AuthorizationSegment", optional=False),
Sequence(
"REGION",
Ref.keyword("AS", optional=True),
Ref("QuotedLiteralSegment"),
optional=True,
),
Ref("CompressionTypeGrammar", optional=True),
Ref("DataFormatSegment", optional=True),
OneOf(
Sequence(
"DELIMITER",
Ref.keyword("AS", optional=True),
Ref("QuotedLiteralSegment"),
),
Sequence(
"FIXEDWIDTH",
Ref.keyword("AS", optional=True),
Ref("QuotedLiteralSegment"),
),
optional=True,
),
Sequence(
"ENCRYPTED",
Ref.keyword("AUTO", optional=True),
optional=True,
),
Ref.keyword("MANIFEST", optional=True),
Sequence(
"COMPROWS",
Ref("NumericLiteralSegment"),
optional=True,
),
Sequence(
"MAXERROR",
Ref.keyword("AS", optional=True),
Ref("NumericLiteralSegment"),
optional=True,
),
Sequence(
"COMPUPDATE",
OneOf(
"PRESET",
"ON",
"OFF",
"TRUE",
"FALSE",
optional=True,
),
optional=True,
),
Sequence(
"STATUPDATE",
OneOf(
"ON",
"OFF",
"TRUE",
"FALSE",
optional=True,
),
optional=True,
),
Ref.keyword("NOLOAD", optional=True),
Ref.keyword("ACCEPTANYDATE", optional=True),
Sequence(
"ACCEPTINVCHARS",
Ref.keyword("AS", optional=True),
Ref("QuotedLiteralSegment"),
optional=True,
),
Ref.keyword("BLANKSASNULL", optional=True),
Sequence(
"DATEFORMAT",
Ref.keyword("AS", optional=True),
OneOf(
"AUTO",
Ref("QuotedLiteralSegment"),
),
optional=True,
),
Ref.keyword("EMPTYASNULL", optional=True),
Sequence(
"ENCODING",
Ref.keyword("AS", optional=True),
OneOf(
"UTF8",
"UTF16",
"UTF16BE",
"UTF16LE",
),
optional=True,
),
Ref.keyword("ESCAPE", optional=True),
Ref.keyword("EXPLICIT_IDS", optional=True),
Ref.keyword("FILLRECORD", optional=True),
Ref.keyword("IGNOREBLANKLINES", optional=True),
Sequence(
"IGNOREHEADER",
Ref.keyword("AS", optional=True),
Ref("QuotedLiteralSegment"),
optional=True,
),
Sequence(
"NULL",
"AS",
Ref("QuotedLiteralSegment"),
optional=True,
),
Sequence(
"READRATIO",
Ref("NumericLiteralSegment"),
optional=True,
),
Ref.keyword("REMOVEQUOTES", optional=True),
Ref.keyword("ROUNDEC", optional=True),
Sequence(
"TIMEFORMAT",
Ref.keyword("AS", optional=True),
OneOf(
"AUTO",
"EPOCHSECS",
"EPOCHMILLISECS",
Ref("QuotedLiteralSegment"),
),
optional=True,
),
Ref.keyword("TRIMBLANKS", optional=True),
Ref.keyword("TRUNCATECOLUMNS", optional=True),
),
)
class CreateTableStatementSegment(BaseSegment):
"""A `CREATE TABLE` statement.
Full Apache Hive `CREATE TABLE` reference here:
https://cwiki.apache.org/confluence/display/hive/languagemanual+ddl#LanguageManualDDL-CreateTable
"""
type = "create_table_statement"
match_grammar = StartsWith(
Sequence(
"CREATE",
Ref.keyword("TEMPORARY", optional=True),
Ref.keyword("EXTERNAL", optional=True),
"TABLE",
))
parse_grammar = Sequence(
"CREATE",
Ref.keyword("TEMPORARY", optional=True),
Ref.keyword("EXTERNAL", optional=True),
"TABLE",
Ref("IfNotExistsGrammar", optional=True),
Ref("TableReferenceSegment"),
OneOf(
# Columns and comment syntax:
Sequence(
Bracketed(
Delimited(
OneOf(
# TODO: support all constraints
Ref("TableConstraintSegment", optional=True),
Sequence(
Ref("ColumnDefinitionSegment"),
Ref("CommentGrammar", optional=True),
),
),
bracket_pairs_set="angle_bracket_pairs",
),
optional=True,
),
Ref("CommentGrammar", optional=True),
# `STORED AS` can be called before or after the additional table
# properties below
Ref("StoredAsGrammar", optional=True),
Sequence(
"PARTITIONED",
"BY",
Bracketed(
Delimited(
Sequence(
Ref("ColumnDefinitionSegment"),
Ref("CommentGrammar", optional=True),
), ), ),
optional=True,
),
Sequence(
"CLUSTERED",
"BY",
Ref("BracketedColumnReferenceListGrammar"),
Sequence(
"SORTED",
"BY",
Bracketed(
Delimited(
Sequence(
Ref("ColumnReferenceSegment"),
OneOf("ASC", "DESC", optional=True),
))),
optional=True,
),
"INTO",
Ref("NumericLiteralSegment"),
"BUCKETS",
optional=True,
),
# Second call of `STORED AS` to match when appears after
Ref("StoredAsGrammar", optional=True),
Ref("SkewedByClauseSegment", optional=True),
Ref("StorageFormatGrammar", optional=True),
Ref("LocationGrammar", optional=True),
Ref("TablePropertiesGrammar", optional=True),
Ref("CommentGrammar", optional=True),
Sequence(
"AS",
OptionallyBracketed(Ref("SelectableGrammar")),
optional=True,
),
),
# Create like syntax
Sequence(
"LIKE",
Ref("TableReferenceSegment"),
Ref("LocationGrammar", optional=True),
Ref("TablePropertiesGrammar", optional=True),
),
),
)
class TdCollectStatUsingOptionClauseSegment(BaseSegment):
"""'using_option' for COLLECT STAT clause."""
type = "collect_stat_using_option_clause"
match_grammar = Sequence(
OneOf(
Sequence("SAMPLE", Ref("NumericLiteralSegment"), "PERCENT"),
Sequence("SYSTEM", "THRESHOLD",
OneOf("PERCENT", "DAYS", optional=True)),
Sequence("SYSTEM", "SAMPLE"),
Sequence(
"THRESHOLD",
Ref("NumericLiteralSegment"),
OneOf("PERCENT", "DAYS"),
),
Sequence("NO", "THRESHOLD", OneOf("PERCENT", "DAYS",
optional=True)),
Sequence("NO", "SAMPLE"),
Sequence("MAXINTERVALS", Ref("NumericLiteralSegment")),
Sequence("SYSTEM", "MAXINTERVALS"),
Sequence("MAXVALUELENGTH", Ref("NumericLiteralSegment")),
Sequence("SYSTEM", "MAXVALUELENGTH"),
"SAMPLE",
),
Sequence("FOR", "CURRENT", optional=True),
)
Ref("SelectableGrammar"),
Ref("InsertStatementSegment"),
Ref("TransactionStatementSegment"),
Ref("DropStatementSegment"),
Ref("AccessStatementSegment"),
Ref("CreateTableStatementSegment"),
Ref("CreateViewStatementSegment"),
Ref("DeleteStatementSegment"),
Ref("UpdateStatementSegment"),
# Teradata specific statements
Ref("TdCollectStatisticsStatementSegment"),
Ref("BteqStatementSegment"),
Ref("TdRenameStatementSegment"),
)
match_grammar = GreedyUntil(Ref("SemicolonSegment"))
teradata_dialect.add(
TdCastIdentifierSegment=Sequence(
OneOf("DATE", "TIMESTAMP"), Ref("ExpressionSegment")
),
)
teradata_dialect.replace(
SingleIdentifierGrammar=OneOf(
Ref("NakedIdentifierSegment"),
Ref("QuotedIdentifierSegment"),
Ref("TdCastIdentifierSegment"),
)
)
class TdCollectStatisticsStatementSegment(BaseSegment):
"""A `COLLECT STATISTICS (Optimizer Form)` statement.
# TODO: add expression
COLLECT [SUMMARY] (STATISTICS|STAT) [[COLUMN| [UNIQUE] INDEX] (expression (, expression ...)] ON TABLENAME
"""
type = "collect_statistics_statement"
match_grammar = Sequence(
"COLLECT",
Ref.keyword("SUMMARY", optional=True),
OneOf("STATISTICS", "STAT"),
Sequence(
"USING",
Delimited(
Ref("TdCollectStatUsingOptionClauseSegment"),
delimiter="AND",
),
optional=True,
),
Delimited(
OneOf(
# UNIQUE INDEX index_name ALL (column_name, ...) ORDER BY VALUES|HASH (column_name)
Sequence(
Ref.keyword("UNIQUE", optional=True),
"INDEX",
Ref("IndexReferenceSegment", optional=True),
Ref.keyword("ALL", optional=True),
Bracketed(
Delimited(Ref("ColumnReferenceSegment"),
delimiter=Ref("CommaSegment"))),
Ref("TdOrderByStatClauseSegment", optional=True),
),
# UNIQUE INDEX index_name
Sequence(
Ref.keyword("UNIQUE", optional=True),
"INDEX",
Ref("IndexReferenceSegment"),
),
# COLUMN ...
Sequence(
"COLUMN",
Bracketed(
Delimited(
OneOf(
Ref("ColumnReferenceSegment"),
Ref.keyword("PARTITION"),
# TODO: expression
),
delimiter=Ref("CommaSegment"),
), ),
Sequence(
Ref.keyword("AS", optional=True),
Ref("ObjectReferenceSegment"), # statistics_name
optional=True,
),
),
),
delimiter=Ref("CommaSegment"),
optional=True,
),
"ON",
Ref.keyword("TEMPORARY", optional=True),
Ref("TableReferenceSegment"),
)
class AltereDatashareStatementSegment(BaseSegment):
"""An `ALTER DATASHARE` statement.
https://docs.aws.amazon.com/redshift/latest/dg/r_ALTER_DATASHARE.html
"""
type = "create_datashare_statement"
match_grammar = Sequence(
"ALTER",
"DATASHARE",
Ref("ObjectReferenceSegment"),
OneOf(
# add or remove objects to the datashare
Sequence(
OneOf(
"ADD",
"REMOVE",
),
OneOf(
Sequence(
"TABLE",
Delimited(Ref("TableReferenceSegment")),
),
Sequence(
"SCHEMA",
Delimited(Ref("SchemaReferenceSegment")),
),
Sequence(
"FUNCTION",
Delimited(Ref("FunctionNameSegment")),
),
Sequence(
"ALL",
OneOf("TABLES", "FUNCTIONS"),
"IN",
"SCHEMA",
Delimited(Ref("SchemaReferenceSegment")),
),
),
),
# configure the properties of the datashare
Sequence(
"SET",
OneOf(
Sequence(
"PUBLICACCESSIBLE",
Ref("EqualsSegment", optional=True),
Ref("BooleanLiteralGrammar"),
),
Sequence(
"INCLUDENEW",
Ref("EqualsSegment", optional=True),
Ref("BooleanLiteralGrammar"),
"FOR",
"SCHEMA",
Ref("SchemaReferenceSegment"),
),
),
),
),
)
class StatementSegment(BaseSegment):
"""A generic segment, to any of its child subsegments."""
type = "statement"
parse_grammar = OneOf(
Ref("SelectableGrammar"),
Ref("InsertStatementSegment"),
Ref("TransactionStatementSegment"),
Ref("DropStatementSegment"),
Ref("AccessStatementSegment"),
Ref("CreateTableStatementSegment"),
Ref("CreateViewStatementSegment"),
Ref("DeleteStatementSegment"),
Ref("UpdateStatementSegment"),
# Teradata specific statements
Ref("TdCollectStatisticsStatementSegment"),
Ref("BteqStatementSegment"),
Ref("TdRenameStatementSegment"),
)
match_grammar = GreedyUntil(Ref("SemicolonSegment"))
teradata_dialect.add(TdCastIdentifierSegment=Sequence(
OneOf("DATE", "TIMESTAMP"), Ref("ExpressionSegment")), )
teradata_dialect.replace(SingleIdentifierGrammar=OneOf(
Ref("NakedIdentifierSegment"),
Ref("QuotedIdentifierSegment"),
Ref("TdCastIdentifierSegment"),
))
type="identifier",
),
)
snowflake_dialect.replace(
Accessor_Grammar=AnyNumberOf(
Ref("ArrayAccessorSegment"),
# Add in semi structured expressions
Ref("SemiStructuredAccessorSegment"),
),
PreTableFunctionKeywordsGrammar=OneOf(Ref("LateralKeywordSegment")),
FunctionContentsExpressionGrammar=OneOf(
Ref("DatetimeUnitSegment"),
Ref("NamedParameterExpressionSegment"),
Sequence(
Ref("ExpressionSegment"),
Sequence(OneOf("IGNORE", "RESPECT"), "NULLS", optional=True),
),
),
JoinLikeClauseGrammar=Sequence(
AnyNumberOf(
Ref("FromAtExpressionSegment"),
Ref("FromBeforeExpressionSegment"),
Ref("FromPivotExpressionSegment"),
Ref("FromUnpivotExpressionSegment"),
Ref("SamplingExpressionSegment"),
min_times=1,
),
Ref("TableAliasExpressionSegment", optional=True),
),
SingleIdentifierGrammar=OneOf(
Ref("NakedIdentifierSegment"),
