snowflake-sqlalchemy 1.7.3__py3-none-any.whl

snowflake/sqlalchemy/base.py

@@ -0,0 +1,1188 @@
+#
+# Copyright (c) 2012-2023 Snowflake Computing Inc. All rights reserved.
+#
+
+import itertools
+import operator
+import re
+import string
+import warnings
+from typing import List
+
+from sqlalchemy import exc as sa_exc
+from sqlalchemy import inspect, sql
+from sqlalchemy import util as sa_util
+from sqlalchemy.engine import default
+from sqlalchemy.orm import context
+from sqlalchemy.orm.context import _MapperEntity
+from sqlalchemy.schema import Sequence, Table
+from sqlalchemy.sql import compiler, expression, functions
+from sqlalchemy.sql.base import CompileState
+from sqlalchemy.sql.elements import BindParameter, quoted_name
+from sqlalchemy.sql.expression import Executable
+from sqlalchemy.sql.selectable import Lateral, SelectState
+
+from snowflake.sqlalchemy._constants import DIALECT_NAME
+from snowflake.sqlalchemy.compat import IS_VERSION_20, args_reducer, string_types
+from snowflake.sqlalchemy.custom_commands import (
+    AWSBucket,
+    AzureContainer,
+    ExternalStage,
+)
+
+from ._constants import NOT_NULL
+from .exc import (
+    CustomOptionsAreOnlySupportedOnSnowflakeTables,
+    UnexpectedOptionTypeError,
+)
+from .functions import flatten
+from .sql.custom_schema.custom_table_base import CustomTableBase
+from .sql.custom_schema.options.table_option import TableOption
+from .util import (
+    _find_left_clause_to_join_from,
+    _set_connection_interpolate_empty_sequences,
+    _Snowflake_ORMJoin,
+    _Snowflake_Selectable_Join,
+)
+
+RESERVED_WORDS = frozenset(
+    [
+        "ALL",  # ANSI Reserved words
+        "ALTER",
+        "AND",
+        "ANY",
+        "AS",
+        "BETWEEN",
+        "BY",
+        "CHECK",
+        "COLUMN",
+        "CONNECT",
+        "COPY",
+        "CREATE",
+        "CURRENT",
+        "DELETE",
+        "DISTINCT",
+        "DROP",
+        "ELSE",
+        "EXISTS",
+        "FOR",
+        "FROM",
+        "GRANT",
+        "GROUP",
+        "HAVING",
+        "IN",
+        "INSERT",
+        "INTERSECT",
+        "INTO",
+        "IS",
+        "LIKE",
+        "NOT",
+        "NULL",
+        "OF",
+        "ON",
+        "OR",
+        "ORDER",
+        "REVOKE",
+        "ROW",
+        "ROWS",
+        "SAMPLE",
+        "SELECT",
+        "SET",
+        "START",
+        "TABLE",
+        "THEN",
+        "TO",
+        "TRIGGER",
+        "UNION",
+        "UNIQUE",
+        "UPDATE",
+        "VALUES",
+        "WHENEVER",
+        "WHERE",
+        "WITH",
+        "REGEXP",
+        "RLIKE",
+        "SOME",  # Snowflake Reserved words
+        "MINUS",
+        "INCREMENT",  # Oracle reserved words
+    ]
+)
+
+# Snowflake DML:
+# - UPDATE
+# - INSERT
+# - DELETE
+# - MERGE
+AUTOCOMMIT_REGEXP = re.compile(
+    r"\s*(?:UPDATE|INSERT|DELETE|MERGE|COPY)", re.I | re.UNICODE
+)
+# used for quoting identifiers ie. table names, column names, etc.
+ILLEGAL_INITIAL_CHARACTERS = frozenset({d for d in string.digits}.union({"$"}))
+
+
+# used for quoting identifiers ie. table names, column names, etc.
+ILLEGAL_IDENTIFIERS = frozenset({d for d in string.digits}.union({"_"}))
+
+"""
+Overwrite methods to handle Snowflake BCR change:
+https://docs.snowflake.com/en/release-notes/bcr-bundles/2023_04/bcr-1057
+- _join_determine_implicit_left_side
+- _join_left_to_right
+"""
+
+
+# handle Snowflake BCR bcr-1057
+@CompileState.plugin_for("default", "select")
+class SnowflakeSelectState(SelectState):
+    def _setup_joins(self, args, raw_columns):
+        for right, onclause, left, flags in args:
+            isouter = flags["isouter"]
+            full = flags["full"]
+
+            if left is None:
+                (
+                    left,
+                    replace_from_obj_index,
+                ) = self._join_determine_implicit_left_side(
+                    raw_columns, left, right, onclause
+                )
+            else:
+                (replace_from_obj_index) = self._join_place_explicit_left_side(left)
+
+            if replace_from_obj_index is not None:
+                # splice into an existing element in the
+                # self._from_obj list
+                left_clause = self.from_clauses[replace_from_obj_index]
+
+                self.from_clauses = (
+                    self.from_clauses[:replace_from_obj_index]
+                    + (
+                        _Snowflake_Selectable_Join(  # handle Snowflake BCR bcr-1057
+                            left_clause,
+                            right,
+                            onclause,
+                            isouter=isouter,
+                            full=full,
+                        ),
+                    )
+                    + self.from_clauses[replace_from_obj_index + 1 :]
+                )
+            else:
+                self.from_clauses = self.from_clauses + (
+                    # handle Snowflake BCR bcr-1057
+                    _Snowflake_Selectable_Join(
+                        left, right, onclause, isouter=isouter, full=full
+                    ),
+                )
+
+    @sa_util.preload_module("sqlalchemy.sql.util")
+    def _join_determine_implicit_left_side(self, raw_columns, left, right, onclause):
+        """When join conditions don't express the left side explicitly,
+        determine if an existing FROM or entity in this query
+        can serve as the left hand side.
+
+        """
+
+        replace_from_obj_index = None
+
+        from_clauses = self.from_clauses
+
+        if from_clauses:
+            # handle Snowflake BCR bcr-1057
+            indexes = _find_left_clause_to_join_from(from_clauses, right, onclause)
+
+            if len(indexes) == 1:
+                replace_from_obj_index = indexes[0]
+                left = from_clauses[replace_from_obj_index]
+        else:
+            potential = {}
+            statement = self.statement
+
+            for from_clause in itertools.chain(
+                itertools.chain.from_iterable(
+                    [element._from_objects for element in raw_columns]
+                ),
+                itertools.chain.from_iterable(
+                    [element._from_objects for element in statement._where_criteria]
+                ),
+            ):
+                potential[from_clause] = ()
+
+            all_clauses = list(potential.keys())
+            # handle Snowflake BCR bcr-1057
+            indexes = _find_left_clause_to_join_from(all_clauses, right, onclause)
+
+            if len(indexes) == 1:
+                left = all_clauses[indexes[0]]
+
+        if len(indexes) > 1:
+            raise sa_exc.InvalidRequestError(
+                "Can't determine which FROM clause to join "
+                "from, there are multiple FROMS which can "
+                "join to this entity. Please use the .select_from() "
+                "method to establish an explicit left side, as well as "
+                "providing an explicit ON clause if not present already to "
+                "help resolve the ambiguity."
+            )
+        elif not indexes:
+            raise sa_exc.InvalidRequestError(
+                "Don't know how to join to %r. "
+                "Please use the .select_from() "
+                "method to establish an explicit left side, as well as "
+                "providing an explicit ON clause if not present already to "
+                "help resolve the ambiguity." % (right,)
+            )
+        return left, replace_from_obj_index
+
+
+# handle Snowflake BCR bcr-1057
+@sql.base.CompileState.plugin_for("orm", "select")
+class SnowflakeORMSelectCompileState(context.ORMSelectCompileState):
+    def _join_determine_implicit_left_side(
+        self, entities_collection, left, right, onclause
+    ):
+        """When join conditions don't express the left side explicitly,
+        determine if an existing FROM or entity in this query
+        can serve as the left hand side.
+
+        """
+
+        # when we are here, it means join() was called without an ORM-
+        # specific way of telling us what the "left" side is, e.g.:
+        #
+        # join(RightEntity)
+        #
+        # or
+        #
+        # join(RightEntity, RightEntity.foo == LeftEntity.bar)
+        #
+
+        r_info = inspect(right)
+
+        replace_from_obj_index = use_entity_index = None
+
+        if self.from_clauses:
+            # we have a list of FROMs already.  So by definition this
+            # join has to connect to one of those FROMs.
+
+            # handle Snowflake BCR bcr-1057
+            indexes = _find_left_clause_to_join_from(
+                self.from_clauses, r_info.selectable, onclause
+            )
+
+            if len(indexes) == 1:
+                replace_from_obj_index = indexes[0]
+                left = self.from_clauses[replace_from_obj_index]
+            elif len(indexes) > 1:
+                raise sa_exc.InvalidRequestError(
+                    "Can't determine which FROM clause to join "
+                    "from, there are multiple FROMS which can "
+                    "join to this entity. Please use the .select_from() "
+                    "method to establish an explicit left side, as well as "
+                    "providing an explicit ON clause if not present already "
+                    "to help resolve the ambiguity."
+                )
+            else:
+                raise sa_exc.InvalidRequestError(
+                    "Don't know how to join to %r. "
+                    "Please use the .select_from() "
+                    "method to establish an explicit left side, as well as "
+                    "providing an explicit ON clause if not present already "
+                    "to help resolve the ambiguity." % (right,)
+                )
+
+        elif entities_collection:
+            # we have no explicit FROMs, so the implicit left has to
+            # come from our list of entities.
+
+            potential = {}
+            for entity_index, ent in enumerate(entities_collection):
+                entity = ent.entity_zero_or_selectable
+                if entity is None:
+                    continue
+                ent_info = inspect(entity)
+                if ent_info is r_info:  # left and right are the same, skip
+                    continue
+
+                # by using a dictionary with the selectables as keys this
+                # de-duplicates those selectables as occurs when the query is
+                # against a series of columns from the same selectable
+                if isinstance(ent, context._MapperEntity):
+                    potential[ent.selectable] = (entity_index, entity)
+                else:
+                    potential[ent_info.selectable] = (None, entity)
+
+            all_clauses = list(potential.keys())
+            # handle Snowflake BCR bcr-1057
+            indexes = _find_left_clause_to_join_from(
+                all_clauses, r_info.selectable, onclause
+            )
+
+            if len(indexes) == 1:
+                use_entity_index, left = potential[all_clauses[indexes[0]]]
+            elif len(indexes) > 1:
+                raise sa_exc.InvalidRequestError(
+                    "Can't determine which FROM clause to join "
+                    "from, there are multiple FROMS which can "
+                    "join to this entity. Please use the .select_from() "
+                    "method to establish an explicit left side, as well as "
+                    "providing an explicit ON clause if not present already "
+                    "to help resolve the ambiguity."
+                )
+            else:
+                raise sa_exc.InvalidRequestError(
+                    "Don't know how to join to %r. "
+                    "Please use the .select_from() "
+                    "method to establish an explicit left side, as well as "
+                    "providing an explicit ON clause if not present already "
+                    "to help resolve the ambiguity." % (right,)
+                )
+        else:
+            raise sa_exc.InvalidRequestError(
+                "No entities to join from; please use "
+                "select_from() to establish the left "
+                "entity/selectable of this join"
+            )
+
+        return left, replace_from_obj_index, use_entity_index
+
+    @args_reducer(positions_to_drop=(6, 7))
+    def _join_left_to_right(
+        self, entities_collection, left, right, onclause, prop, outerjoin, full
+    ):
+        """given raw "left", "right", "onclause" parameters consumed from
+        a particular key within _join(), add a real ORMJoin object to
+        our _from_obj list (or augment an existing one)
+
+        """
+
+        if left is None:
+            # left not given (e.g. no relationship object/name specified)
+            # figure out the best "left" side based on our existing froms /
+            # entities
+            assert prop is None
+            (
+                left,
+                replace_from_obj_index,
+                use_entity_index,
+            ) = self._join_determine_implicit_left_side(
+                entities_collection, left, right, onclause
+            )
+        else:
+            # left is given via a relationship/name, or as explicit left side.
+            # Determine where in our
+            # "froms" list it should be spliced/appended as well as what
+            # existing entity it corresponds to.
+            (
+                replace_from_obj_index,
+                use_entity_index,
+            ) = self._join_place_explicit_left_side(entities_collection, left)
+
+        if left is right:
+            raise sa_exc.InvalidRequestError(
+                "Can't construct a join from %s to %s, they "
+                "are the same entity" % (left, right)
+            )
+
+        # the right side as given often needs to be adapted.  additionally
+        # a lot of things can be wrong with it.  handle all that and
+        # get back the new effective "right" side
+
+        if IS_VERSION_20:
+            r_info, right, onclause = self._join_check_and_adapt_right_side(
+                left, right, onclause, prop
+            )
+        else:
+            r_info, right, onclause = self._join_check_and_adapt_right_side(
+                left, right, onclause, prop, False, False
+            )
+
+        if not r_info.is_selectable:
+            extra_criteria = self._get_extra_criteria(r_info)
+        else:
+            extra_criteria = ()
+
+        if replace_from_obj_index is not None:
+            # splice into an existing element in the
+            # self._from_obj list
+            left_clause = self.from_clauses[replace_from_obj_index]
+
+            self.from_clauses = (
+                self.from_clauses[:replace_from_obj_index]
+                + [
+                    _Snowflake_ORMJoin(  # handle Snowflake BCR bcr-1057
+                        left_clause,
+                        right,
+                        onclause,
+                        isouter=outerjoin,
+                        full=full,
+                        _extra_criteria=extra_criteria,
+                    )
+                ]
+                + self.from_clauses[replace_from_obj_index + 1 :]
+            )
+        else:
+            # add a new element to the self._from_obj list
+            if use_entity_index is not None:
+                # make use of _MapperEntity selectable, which is usually
+                # entity_zero.selectable, but if with_polymorphic() were used
+                # might be distinct
+                assert isinstance(entities_collection[use_entity_index], _MapperEntity)
+                left_clause = entities_collection[use_entity_index].selectable
+            else:
+                left_clause = left
+
+            self.from_clauses = self.from_clauses + [
+                _Snowflake_ORMJoin(  # handle Snowflake BCR bcr-1057
+                    left_clause,
+                    r_info,
+                    onclause,
+                    isouter=outerjoin,
+                    full=full,
+                    _extra_criteria=extra_criteria,
+                )
+            ]
+
+
+class SnowflakeIdentifierPreparer(compiler.IdentifierPreparer):
+    reserved_words = {x.lower() for x in RESERVED_WORDS}
+    illegal_initial_characters = ILLEGAL_INITIAL_CHARACTERS
+    illegal_identifiers = ILLEGAL_IDENTIFIERS
+
+    def __init__(self, dialect, **kw):
+        quote = '"'
+
+        super().__init__(dialect, initial_quote=quote, escape_quote=quote)
+
+    def _quote_free_identifiers(self, *ids):
+        """
+        Unilaterally identifier-quote any number of strings.
+        """
+        return tuple(self.quote(i) for i in ids if i is not None)
+
+    def quote_schema(self, schema, force=None):
+        """
+        Split schema by a dot and merge with required quotes
+        """
+        idents = self._split_schema_by_dot(schema)
+        return ".".join(self._quote_free_identifiers(*idents))
+
+    def format_label(self, label, name=None):
+        n = name or label.name
+        s = n.replace(self.escape_quote, "")
+
+        if not isinstance(n, quoted_name) or n.quote is None:
+            return self.quote(s)
+
+        return self.quote_identifier(s) if n.quote else s
+
+    def _requires_quotes(self, value: str) -> bool:
+        """Return True if the given identifier requires quoting."""
+        lc_value = value.lower()
+        return (
+            lc_value in self.reserved_words
+            or lc_value in self.illegal_identifiers
+            or value[0] in self.illegal_initial_characters
+            or not self.legal_characters.match(str(value))
+            or (lc_value != value)
+        )
+
+    def _split_schema_by_dot(self, schema):
+        ret = []
+        idx = 0
+        pre_idx = 0
+        in_quote = False
+        while idx < len(schema):
+            if not in_quote:
+                if schema[idx] == "." and pre_idx < idx:
+                    ret.append(schema[pre_idx:idx])
+                    pre_idx = idx + 1
+                elif schema[idx] == '"':
+                    in_quote = True
+                    pre_idx = idx + 1
+            else:
+                if schema[idx] == '"' and pre_idx < idx:
+                    ret.append(schema[pre_idx:idx])
+                    in_quote = False
+                    pre_idx = idx + 1
+            idx += 1
+            if pre_idx < len(schema) and schema[pre_idx] == ".":
+                pre_idx += 1
+        if pre_idx < idx:
+            ret.append(schema[pre_idx:idx])
+
+        # convert the returning strings back to quoted_name types, and assign the original 'quote' attribute on it
+        quoted_ret = [
+            quoted_name(value, quote=getattr(schema, "quote", None)) for value in ret
+        ]
+
+        return quoted_ret
+
+
+class SnowflakeCompiler(compiler.SQLCompiler):
+    def visit_sequence(self, sequence, **kw):
+        return self.dialect.identifier_preparer.format_sequence(sequence) + ".nextval"
+
+    def visit_now_func(self, now, **kw):
+        return "CURRENT_TIMESTAMP"
+
+    def visit_merge_into(self, merge_into, **kw):
+        clauses = " ".join(
+            clause._compiler_dispatch(self, **kw) for clause in merge_into.clauses
+        )
+        return (
+            f"MERGE INTO {merge_into.target} USING {merge_into.source} ON {merge_into.on}"
+            + (" " + clauses if clauses else "")
+        )
+
+    def visit_merge_into_clause(self, merge_into_clause, **kw):
+        case_predicate = (
+            f" AND {str(merge_into_clause.predicate._compiler_dispatch(self, **kw))}"
+            if merge_into_clause.predicate is not None
+            else ""
+        )
+        if merge_into_clause.command == "INSERT":
+            sets, sets_tos = zip(*merge_into_clause.set.items())
+            sets, sets_tos = list(sets), list(sets_tos)
+            if kw.get("deterministic", False):
+                sets, sets_tos = zip(
+                    *sorted(merge_into_clause.set.items(), key=operator.itemgetter(0))
+                )
+            return "WHEN NOT MATCHED{} THEN {} ({}) VALUES ({})".format(
+                case_predicate,
+                merge_into_clause.command,
+                ", ".join(sets),
+                ", ".join(map(lambda e: e._compiler_dispatch(self, **kw), sets_tos)),
+            )
+        else:
+            set_list = list(merge_into_clause.set.items())
+            if kw.get("deterministic", False):
+                set_list.sort(key=operator.itemgetter(0))
+            sets = (
+                ", ".join(
+                    [
+                        f"{set[0]} = {set[1]._compiler_dispatch(self, **kw)}"
+                        for set in set_list
+                    ]
+                )
+                if merge_into_clause.set
+                else ""
+            )
+            return "WHEN MATCHED{} THEN {}{}".format(
+                case_predicate,
+                merge_into_clause.command,
+                " SET %s" % sets if merge_into_clause.set else "",
+            )
+
+    def visit_copy_into(self, copy_into, **kw):
+        if hasattr(copy_into, "formatter") and copy_into.formatter is not None:
+            formatter = copy_into.formatter._compiler_dispatch(self, **kw)
+        else:
+            formatter = ""
+        into = (
+            copy_into.into
+            if isinstance(copy_into.into, Table)
+            else copy_into.into._compiler_dispatch(self, **kw)
+        )
+        if isinstance(copy_into.from_, Table):
+            from_ = copy_into.from_.name
+        # this is intended to catch AWSBucket and AzureContainer
+        elif (
+            isinstance(copy_into.from_, AWSBucket)
+            or isinstance(copy_into.from_, AzureContainer)
+            or isinstance(copy_into.from_, ExternalStage)
+        ):
+            from_ = copy_into.from_._compiler_dispatch(self, **kw)
+        # everything else (selects, etc.)
+        else:
+            from_ = f"({copy_into.from_._compiler_dispatch(self, **kw)})"
+
+        partition_by_value = None
+        if isinstance(copy_into.partition_by, (BindParameter, Executable)):
+            partition_by_value = copy_into.partition_by.compile(
+                compile_kwargs={"literal_binds": True}
+            )
+        elif copy_into.partition_by is not None:
+            partition_by_value = copy_into.partition_by
+
+        partition_by = (
+            f"PARTITION BY {partition_by_value}"
+            if partition_by_value is not None and partition_by_value != ""
+            else ""
+        )
+
+        credentials, encryption = "", ""
+        if isinstance(into, tuple):
+            into, credentials, encryption = into
+        elif isinstance(from_, tuple):
+            from_, credentials, encryption = from_
+        options_list = list(copy_into.copy_options.items())
+        if kw.get("deterministic", False):
+            options_list.sort(key=operator.itemgetter(0))
+        options = (
+            (
+                " ".join(
+                    [
+                        "{} = {}".format(
+                            n,
+                            (
+                                v._compiler_dispatch(self, **kw)
+                                if getattr(v, "compiler_dispatch", False)
+                                else str(v)
+                            ),
+                        )
+                        for n, v in options_list
+                    ]
+                )
+            )
+            if copy_into.copy_options
+            else ""
+        )
+        if credentials:
+            options += f" {credentials}"
+        if encryption:
+            options += f" {encryption}"
+        return f"COPY INTO {into} FROM {' '.join([from_, partition_by, formatter, options])}"
+
+    def visit_copy_formatter(self, formatter, **kw):
+        options_list = list(formatter.options.items())
+        if kw.get("deterministic", False):
+            options_list.sort(key=operator.itemgetter(0))
+        if "format_name" in formatter.options:
+            return f"FILE_FORMAT=(format_name = {formatter.options['format_name']})"
+        return "FILE_FORMAT=(TYPE={}{})".format(
+            formatter.file_format,
+            (
+                " "
+                + " ".join(
+                    [
+                        "{}={}".format(
+                            name,
+                            (
+                                value._compiler_dispatch(self, **kw)
+                                if hasattr(value, "_compiler_dispatch")
+                                else formatter.value_repr(name, value)
+                            ),
+                        )
+                        for name, value in options_list
+                    ]
+                )
+                if formatter.options
+                else ""
+            ),
+        )
+
+    def visit_aws_bucket(self, aws_bucket, **kw):
+        credentials_list = list(aws_bucket.credentials_used.items())
+        if kw.get("deterministic", False):
+            credentials_list.sort(key=operator.itemgetter(0))
+        credentials = "CREDENTIALS=({})".format(
+            " ".join(f"{n}='{v}'" for n, v in credentials_list)
+        )
+        encryption_list = list(aws_bucket.encryption_used.items())
+        if kw.get("deterministic", False):
+            encryption_list.sort(key=operator.itemgetter(0))
+        encryption = "ENCRYPTION=({})".format(
+            " ".join(
+                ("{}='{}'" if isinstance(v, string_types) else "{}={}").format(n, v)
+                for n, v in encryption_list
+            )
+        )
+        uri = "'s3://{}{}'".format(
+            aws_bucket.bucket, f"/{aws_bucket.path}" if aws_bucket.path else ""
+        )
+        return (
+            uri,
+            credentials if aws_bucket.credentials_used else "",
+            encryption if aws_bucket.encryption_used else "",
+        )
+
+    def visit_azure_container(self, azure_container, **kw):
+        credentials_list = list(azure_container.credentials_used.items())
+        if kw.get("deterministic", False):
+            credentials_list.sort(key=operator.itemgetter(0))
+        credentials = "CREDENTIALS=({})".format(
+            " ".join(f"{n}='{v}'" for n, v in credentials_list)
+        )
+        encryption_list = list(azure_container.encryption_used.items())
+        if kw.get("deterministic", False):
+            encryption_list.sort(key=operator.itemgetter(0))
+        encryption = "ENCRYPTION=({})".format(
+            " ".join(
+                f"{n}='{v}'" if isinstance(v, string_types) else f"{n}={v}"
+                for n, v in encryption_list
+            )
+        )
+        uri = "'azure://{}.blob.core.windows.net/{}{}'".format(
+            azure_container.account,
+            azure_container.container,
+            f"/{azure_container.path}" if azure_container.path else "",
+        )
+        return (
+            uri,
+            credentials if azure_container.credentials_used else "",
+            encryption if azure_container.encryption_used else "",
+        )
+
+    def visit_external_stage(self, external_stage, **kw):
+        if external_stage.file_format is None:
+            return (
+                f"@{external_stage.namespace}{external_stage.name}{external_stage.path}"
+            )
+        return f"@{external_stage.namespace}{external_stage.name}{external_stage.path} (file_format => {external_stage.file_format})"
+
+    def delete_extra_from_clause(
+        self, delete_stmt, from_table, extra_froms, from_hints, **kw
+    ):
+        return "USING " + ", ".join(
+            t._compiler_dispatch(self, asfrom=True, fromhints=from_hints, **kw)
+            for t in extra_froms
+        )
+
+    def update_from_clause(
+        self, update_stmt, from_table, extra_froms, from_hints, **kw
+    ):
+        return "FROM " + ", ".join(
+            t._compiler_dispatch(self, asfrom=True, fromhints=from_hints, **kw)
+            for t in extra_froms
+        )
+
+    def _get_regexp_args(self, binary, kw):
+        string = self.process(binary.left, **kw)
+        pattern = self.process(binary.right, **kw)
+        flags = binary.modifiers["flags"]
+        if flags is not None:
+            flags = self.process(flags, **kw)
+        return string, pattern, flags
+
+    def visit_regexp_match_op_binary(self, binary, operator, **kw):
+        string, pattern, flags = self._get_regexp_args(binary, kw)
+        if flags is None:
+            return f"REGEXP_LIKE({string}, {pattern})"
+        else:
+            return f"REGEXP_LIKE({string}, {pattern}, {flags})"
+
+    def visit_regexp_replace_op_binary(self, binary, operator, **kw):
+        string, pattern, flags = self._get_regexp_args(binary, kw)
+        try:
+            replacement = self.process(binary.modifiers["replacement"], **kw)
+        except KeyError:
+            # in sqlalchemy 1.4.49, the internal structure of the expression is changed
+            # that binary.modifiers doesn't have "replacement":
+            # https://docs.sqlalchemy.org/en/20/changelog/changelog_14.html#change-1.4.49
+            return f"REGEXP_REPLACE({string}, {pattern}{'' if flags is None else f', {flags}'})"
+
+        if flags is None:
+            return f"REGEXP_REPLACE({string}, {pattern}, {replacement})"
+        else:
+            return f"REGEXP_REPLACE({string}, {pattern}, {replacement}, {flags})"
+
+    def visit_not_regexp_match_op_binary(self, binary, operator, **kw):
+        return f"NOT {self.visit_regexp_match_op_binary(binary, operator, **kw)}"
+
+    def visit_join(self, join, asfrom=False, from_linter=None, **kwargs):
+        if from_linter:
+            from_linter.edges.update(
+                itertools.product(join.left._from_objects, join.right._from_objects)
+            )
+
+        if join.full:
+            join_type = " FULL OUTER JOIN "
+        elif join.isouter:
+            join_type = " LEFT OUTER JOIN "
+        else:
+            join_type = " JOIN "
+
+        join_statement = (
+            join.left._compiler_dispatch(
+                self, asfrom=True, from_linter=from_linter, **kwargs
+            )
+            + join_type
+            + join.right._compiler_dispatch(
+                self, asfrom=True, from_linter=from_linter, **kwargs
+            )
+        )
+
+        if join.onclause is None and isinstance(join.right, Lateral):
+            # in snowflake, onclause is not accepted for lateral due to BCR change:
+            # https://docs.snowflake.com/en/release-notes/bcr-bundles/2023_04/bcr-1057
+            # sqlalchemy only allows join with on condition.
+            # to adapt to snowflake syntax change,
+            # we make the change such that when oncaluse is None and the right part is
+            # Lateral, we do not append the on condition
+            return join_statement
+
+        return (
+            join_statement
+            + " ON "
+            # TODO: likely need asfrom=True here?
+            + join.onclause._compiler_dispatch(self, from_linter=from_linter, **kwargs)
+        )
+
+    def visit_truediv_binary(self, binary, operator, **kw):
+        if self.dialect.div_is_floordiv:
+            warnings.warn(
+                "div_is_floordiv value will be changed to False in a future release. This will generate a behavior change on true and floor division. Please review https://docs.sqlalchemy.org/en/20/changelog/whatsnew_20.html#python-division-operator-performs-true-division-for-all-backends-added-floor-division",
+                PendingDeprecationWarning,
+                stacklevel=2,
+            )
+        return (
+            self.process(binary.left, **kw) + " / " + self.process(binary.right, **kw)
+        )
+
+    def visit_floordiv_binary(self, binary, operator, **kw):
+        if self.dialect.div_is_floordiv and IS_VERSION_20:
+            warnings.warn(
+                "div_is_floordiv value will be changed to False in a future release. This will generate a behavior change on true and floor division. Please review https://docs.sqlalchemy.org/en/20/changelog/whatsnew_20.html#python-division-operator-performs-true-division-for-all-backends-added-floor-division",
+                PendingDeprecationWarning,
+                stacklevel=2,
+            )
+        return super().visit_floordiv_binary(binary, operator, **kw)
+
+    def render_literal_value(self, value, type_):
+        # escape backslash
+        return super().render_literal_value(value, type_).replace("\\", "\\\\")
+
+
+class SnowflakeExecutionContext(default.DefaultExecutionContext):
+    INSERT_SQL_RE = re.compile(r"^insert\s+into", flags=re.IGNORECASE)
+
+    def fire_sequence(self, seq, type_):
+        return self._execute_scalar(
+            f"SELECT {self.identifier_preparer.format_sequence(seq)}.nextval",
+            type_,
+        )
+
+    def should_autocommit_text(self, statement):
+        return AUTOCOMMIT_REGEXP.match(statement)
+
+    @sa_util.memoized_property
+    def should_autocommit(self):
+        autocommit = self.execution_options.get(
+            "autocommit",
+            not self.compiled
+            and self.statement
+            and expression.PARSE_AUTOCOMMIT
+            or False,
+        )
+
+        if autocommit is expression.PARSE_AUTOCOMMIT:
+            return self.should_autocommit_text(self.unicode_statement)
+        else:
+            return autocommit and not self.isddl
+
+    def pre_exec(self):
+        if self.compiled and self.identifier_preparer._double_percents:
+            # for compiled statements, percent is doubled for escape, we turn on _interpolate_empty_sequences
+            _set_connection_interpolate_empty_sequences(self._dbapi_connection, True)
+
+            # if the statement is executemany insert, setting _interpolate_empty_sequences to True is not enough,
+            # because executemany pre-processes the param binding and then pass None params to execute so
+            # _interpolate_empty_sequences condition not getting met for the command.
+            # Therefore, we manually revert the escape percent in the command here
+            if self.executemany and self.INSERT_SQL_RE.match(self.statement):
+                self.statement = self.statement.replace("%%", "%")
+        else:
+            # for other cases, do no interpolate empty sequences as "%" is not double escaped
+            _set_connection_interpolate_empty_sequences(self._dbapi_connection, False)
+
+    def post_exec(self):
+        if self.compiled and self.identifier_preparer._double_percents:
+            # for compiled statements, percent is doubled for escapeafter execution
+            # we reset _interpolate_empty_sequences to false which is turned on in pre_exec
+            _set_connection_interpolate_empty_sequences(self._dbapi_connection, False)
+
+    @property
+    def rowcount(self):
+        return self.cursor.rowcount
+
+
+class SnowflakeDDLCompiler(compiler.DDLCompiler):
+    def denormalize_column_name(self, name):
+        if name is None:
+            return None
+        elif name.lower() == name and not self.preparer._requires_quotes(name.lower()):
+            # no quote as case insensitive
+            return name
+        return self.preparer.quote(name)
+
+    def get_column_specification(self, column, **kwargs):
+        """
+        Gets Column specifications
+        """
+        colspec = [
+            self.preparer.format_column(column),
+            self.dialect.type_compiler.process(column.type, type_expression=column),
+        ]
+
+        has_identity = (
+            column.identity is not None and self.dialect.supports_identity_columns
+        )
+
+        if not column.nullable:
+            colspec.append("NOT NULL")
+
+        default = self.get_column_default_string(column)
+        if default is not None:
+            colspec.append("DEFAULT " + default)
+
+        # TODO: This makes the first INTEGER column AUTOINCREMENT.
+        # But the column is not really considered so unless
+        # postfetch_lastrowid is enabled. But it is very unlikely to happen...
+        if (
+            column.table is not None
+            and column is column.table._autoincrement_column
+            and column.server_default is None
+        ):
+            if isinstance(column.default, Sequence):
+                colspec.append(
+                    f"DEFAULT {self.dialect.identifier_preparer.format_sequence(column.default)}.nextval"
+                )
+            else:
+                colspec.append("AUTOINCREMENT")
+
+        if has_identity:
+            colspec.append(self.process(column.identity))
+
+        return " ".join(colspec)
+
+    def handle_cluster_by(self, table):
+        """
+        Handles snowflake-specific ``CREATE TABLE ... CLUSTER BY`` syntax.
+
+        Users can specify the `clusterby` property per table
+        using the dialect specific syntax.
+        For example, to specify a cluster by key you apply the following:
+
+        >>> import sqlalchemy as sa
+        >>> from sqlalchemy.schema import CreateTable
+        >>> engine = sa.create_engine('snowflake://om1')
+        >>> metadata = sa.MetaData()
+        >>> user = sa.Table(
+        ...     'user',
+        ...     metadata,
+        ...     sa.Column('id', sa.Integer, primary_key=True),
+        ...     sa.Column('name', sa.String),
+        ...     snowflake_clusterby=['id', 'name', text("id > 5")]
+        ... )
+        >>> print(CreateTable(user).compile(engine))
+        <BLANKLINE>
+        CREATE TABLE "user" (
+            id INTEGER NOT NULL AUTOINCREMENT,
+            name VARCHAR,
+            PRIMARY KEY (id)
+        ) CLUSTER BY (id, name, id > 5)
+        <BLANKLINE>
+        <BLANKLINE>
+        """
+        text = ""
+        info = table.dialect_options[DIALECT_NAME]
+        cluster = info.get("clusterby")
+        if cluster:
+            text += " CLUSTER BY ({})".format(
+                ", ".join(
+                    (
+                        self.denormalize_column_name(key)
+                        if isinstance(key, str)
+                        else str(key)
+                    )
+                    for key in cluster
+                )
+            )
+        return text
+
+    def post_create_table(self, table):
+        text = self.handle_cluster_by(table)
+        options = []
+        invalid_options: List[str] = []
+
+        for key, option in table.dialect_options[DIALECT_NAME].items():
+            if isinstance(option, TableOption):
+                options.append(option)
+            elif key not in ["clusterby", "*"]:
+                invalid_options.append(key)
+
+        if len(invalid_options) > 0:
+            raise UnexpectedOptionTypeError(sorted(invalid_options))
+
+        if isinstance(table, CustomTableBase):
+            options.sort(key=lambda x: (x.priority.value, x.option_name), reverse=True)
+            for option in options:
+                text += "\t" + option.render_option(self)
+        elif len(options) > 0:
+            raise CustomOptionsAreOnlySupportedOnSnowflakeTables()
+
+        return text
+
+    def visit_create_stage(self, create_stage, **kw):
+        """
+        This visitor will create the SQL representation for a CREATE STAGE command.
+        """
+        return "CREATE {or_replace}{temporary}STAGE {}{} URL={}".format(
+            create_stage.stage.namespace,
+            create_stage.stage.name,
+            repr(create_stage.container),
+            or_replace="OR REPLACE " if create_stage.replace_if_exists else "",
+            temporary="TEMPORARY " if create_stage.temporary else "",
+        )
+
+    def visit_create_file_format(self, file_format, **kw):
+        """
+        This visitor will create the SQL representation for a CREATE FILE FORMAT
+        command.
+        """
+        return "CREATE {}FILE FORMAT {} TYPE='{}' {}".format(
+            "OR REPLACE " if file_format.replace_if_exists else "",
+            file_format.format_name,
+            file_format.formatter.file_format,
+            " ".join(
+                [
+                    f"{name} = {file_format.formatter.value_repr(name, value)}"
+                    for name, value in file_format.formatter.options.items()
+                ]
+            ),
+        )
+
+    def visit_drop_table_comment(self, drop, **kw):
+        """Snowflake does not support setting table comments as NULL.
+
+        Reflection has to account for this and convert any empty comments to NULL.
+        """
+        table_name = self.preparer.format_table(drop.element)
+        return f"COMMENT ON TABLE {table_name} IS ''"
+
+    def visit_drop_column_comment(self, drop, **kw):
+        """Snowflake does not support directly setting column comments as NULL.
+
+        Instead we are forced to use the ALTER COLUMN ... UNSET COMMENT instead.
+        """
+        return "ALTER TABLE {} ALTER COLUMN {} UNSET COMMENT".format(
+            self.preparer.format_table(drop.element.table),
+            self.preparer.format_column(drop.element),
+        )
+
+    def visit_identity_column(self, identity, **kw):
+        text = "IDENTITY"
+        if identity.start is not None or identity.increment is not None:
+            start = 1 if identity.start is None else identity.start
+            increment = 1 if identity.increment is None else identity.increment
+            text += f"({start},{increment})"
+        if identity.order is not None:
+            order = "ORDER" if identity.order else "NOORDER"
+            text += f" {order}"
+        return text
+
+    def get_identity_options(self, identity_options):
+        text = []
+        if identity_options.increment is not None:
+            text.append("INCREMENT BY %d" % identity_options.increment)
+        if identity_options.start is not None:
+            text.append("START WITH %d" % identity_options.start)
+        if identity_options.minvalue is not None:
+            text.append("MINVALUE %d" % identity_options.minvalue)
+        if identity_options.maxvalue is not None:
+            text.append("MAXVALUE %d" % identity_options.maxvalue)
+        if identity_options.nominvalue is not None:
+            text.append("NO MINVALUE")
+        if identity_options.nomaxvalue is not None:
+            text.append("NO MAXVALUE")
+        if identity_options.cache is not None:
+            text.append("CACHE %d" % identity_options.cache)
+        if identity_options.cycle is not None:
+            text.append("CYCLE" if identity_options.cycle else "NO CYCLE")
+        if identity_options.order is not None:
+            text.append("ORDER" if identity_options.order else "NOORDER")
+        return " ".join(text)
+
+
+class SnowflakeTypeCompiler(compiler.GenericTypeCompiler):
+    def visit_BYTEINT(self, type_, **kw):
+        return "BYTEINT"
+
+    def visit_CHARACTER(self, type_, **kw):
+        return "CHARACTER"
+
+    def visit_DEC(self, type_, **kw):
+        return "DEC"
+
+    def visit_DOUBLE(self, type_, **kw):
+        return "DOUBLE"
+
+    def visit_FIXED(self, type_, **kw):
+        return "FIXED"
+
+    def visit_INT(self, type_, **kw):
+        return "INT"
+
+    def visit_NUMBER(self, type_, **kw):
+        return "NUMBER"
+
+    def visit_STRING(self, type_, **kw):
+        return "STRING"
+
+    def visit_TINYINT(self, type_, **kw):
+        return "TINYINT"
+
+    def visit_VARIANT(self, type_, **kw):
+        return "VARIANT"
+
+    def visit_MAP(self, type_, **kw):
+        not_null = f" {NOT_NULL}" if type_.not_null else ""
+        return (
+            f"MAP({type_.key_type.compile()}, {type_.value_type.compile()}{not_null})"
+        )
+
+    def visit_ARRAY(self, type_, **kw):
+        return "ARRAY"
+
+    def visit_SNOWFLAKE_ARRAY(self, type_, **kw):
+        if type_.is_semi_structured:
+            return "ARRAY"
+        not_null = f" {NOT_NULL}" if type_.not_null else ""
+        return f"ARRAY({type_.value_type.compile()}{not_null})"
+
+    def visit_OBJECT(self, type_, **kw):
+        if type_.is_semi_structured:
+            return "OBJECT"
+        else:
+            contents = []
+            for key in type_.items_types:
+
+                row_text = f"{key} {type_.items_types[key][0].compile()}"
+                # Type and not null is specified
+                if len(type_.items_types[key]) > 1:
+                    row_text += f"{' NOT NULL' if type_.items_types[key][1] else ''}"
+                contents.append(row_text)
+            return "OBJECT" if contents == [] else f"OBJECT({', '.join(contents)})"
+
+    def visit_BLOB(self, type_, **kw):
+        return "BINARY"
+
+    def visit_datetime(self, type_, **kw):
+        return "datetime"
+
+    def visit_DATETIME(self, type_, **kw):
+        return "DATETIME"
+
+    def visit_TIMESTAMP_NTZ(self, type_, **kw):
+        return "TIMESTAMP_NTZ"
+
+    def visit_TIMESTAMP_TZ(self, type_, **kw):
+        return "TIMESTAMP_TZ"
+
+    def visit_TIMESTAMP_LTZ(self, type_, **kw):
+        return "TIMESTAMP_LTZ"
+
+    def visit_TIMESTAMP(self, type_, **kw):
+        return "TIMESTAMP"
+
+    def visit_GEOGRAPHY(self, type_, **kw):
+        return "GEOGRAPHY"
+
+    def visit_GEOMETRY(self, type_, **kw):
+        return "GEOMETRY"
+
+
+construct_arguments = [(Table, {"clusterby": None})]
+
+functions.register_function("flatten", flatten, "snowflake")