velocity-python 0.0.109__py3-none-any.whl → 0.0.155__py3-none-any.whl

velocity/db/servers/sqlserver/sql.py

@@ -0,0 +1,805 @@
+import re
+import hashlib
+import decimal
+import datetime
+from typing import Any, Dict, List, Optional, Tuple, Union
+from collections.abc import Mapping, Sequence
+
+from velocity.db import exceptions
+from ..base.sql import BaseSQLDialect
+from .reserved import reserved_words
+from .types import TYPES
+from .operators import OPERATORS, SQLServerOperators
+from ..tablehelper import TableHelper
+
+
+# Configure TableHelper for SQL Server
+TableHelper.reserved = reserved_words
+TableHelper.operators = OPERATORS
+
+
+system_fields = [
+    "sys_id",
+    "sys_created",
+    "sys_modified",
+    "sys_modified_by",
+    "sys_dirty",
+    "sys_table",
+    "sys_modified_count",
+    "description",
+]
+
+
+def quote(data):
+    """Quote SQL Server identifiers."""
+    if isinstance(data, list):
+        return [quote(item) for item in data]
+    else:
+        parts = data.split(".")
+        new = []
+        for part in parts:
+            if "[" in part:
+                new.append(part)
+            elif part.upper() in reserved_words:
+                new.append("[" + part + "]")
+            elif re.findall("[/]", part):
+                new.append("[" + part + "]")
+            else:
+                new.append(part)
+        return ".".join(new)
+
+
+class SQL(BaseSQLDialect):
+    server = "SQL Server"
+    type_column_identifier = "DATA_TYPE"
+    is_nullable = "IS_NULLABLE"
+
+    default_schema = "dbo"
+
+    # SQL Server error numbers
+    ApplicationErrorCodes = []
+    DatabaseMissingErrorCodes = ["911"]  # Database not found
+    TableMissingErrorCodes = ["208"]     # Invalid object name
+    ColumnMissingErrorCodes = ["207"]    # Invalid column name
+    ForeignKeyMissingErrorCodes = ["1759"] # Foreign key error
+    ConnectionErrorCodes = ["2", "53", "1326"] # Connection errors
+    DuplicateKeyErrorCodes = ["2627", "2601"]  # Primary key / unique constraint
+    RetryTransactionCodes = ["1205"]     # Deadlock
+    TruncationErrorCodes = ["8152"]      # String truncation
+    LockTimeoutErrorCodes = ["1222"]     # Lock request timeout
+    DatabaseObjectExistsErrorCodes = ["2714"] # Object already exists
+    DataIntegrityErrorCodes = ["547", "515"] # Foreign key, null constraint
+
+    types = TYPES
+
+    @classmethod
+    def get_error(cls, e):
+        """Extract error information from SQL Server exception."""
+        # pytds exceptions have different attributes
+        error_number = getattr(e, "number", None) or getattr(e, "msgno", None)
+        error_message = getattr(e, "message", None) or str(e)
+        return error_number, error_message
+
+    @classmethod
+    def select(
+        cls,
+        tx,
+        columns=None,
+        table=None,
+        where=None,
+        orderby=None,
+        groupby=None,
+        having=None,
+        start=None,
+        qty=None,
+        lock=None,
+        skip_locked=None,
+    ):
+        """Generate a SQL Server SELECT statement."""
+        if not table:
+            raise ValueError("Table name is required")
+
+        sql_parts = []
+        vals = []
+
+        # SELECT clause with TOP (SQL Server pagination)
+        sql_parts.append("SELECT")
+        
+        # Handle TOP clause for SQL Server pagination
+        if qty is not None and start is None:
+            sql_parts.append(f"TOP {qty}")
+
+        # Column selection
+        if columns is None:
+            columns = ["*"]
+        elif isinstance(columns, str):
+            columns = [columns]
+        
+        sql_parts.append(", ".join(columns))
+
+        # FROM clause
+        sql_parts.append("FROM")
+        sql_parts.append(quote(table))
+
+        # WHERE clause
+        if where:
+            where_sql, where_vals = cls._build_where(where)
+            sql_parts.append("WHERE")
+            sql_parts.append(where_sql)
+            vals.extend(where_vals)
+
+        # GROUP BY clause
+        if groupby:
+            if isinstance(groupby, str):
+                groupby = [groupby]
+            sql_parts.append("GROUP BY")
+            sql_parts.append(", ".join(quote(col) for col in groupby))
+
+        # HAVING clause
+        if having:
+            having_sql, having_vals = cls._build_where(having)
+            sql_parts.append("HAVING")
+            sql_parts.append(having_sql)
+            vals.extend(having_vals)
+
+        # ORDER BY clause (required for OFFSET/FETCH)
+        if orderby:
+            if isinstance(orderby, str):
+                orderby = [orderby]
+            elif isinstance(orderby, dict):
+                orderby_list = []
+                for col, direction in orderby.items():
+                    orderby_list.append(f"{quote(col)} {direction.upper()}")
+                orderby = orderby_list
+            sql_parts.append("ORDER BY")
+            sql_parts.append(", ".join(orderby))
+        elif start is not None:
+            # ORDER BY is required for OFFSET/FETCH in SQL Server
+            sql_parts.append("ORDER BY")
+            sql_parts.append("(SELECT NULL)")
+
+        # OFFSET and FETCH (SQL Server 2012+)
+        if start is not None:
+            sql_parts.append(f"OFFSET {start} ROWS")
+            if qty is not None:
+                sql_parts.append(f"FETCH NEXT {qty} ROWS ONLY")
+
+        # Locking hints
+        if lock:
+            sql_parts.append("WITH (UPDLOCK)")
+            if skip_locked:
+                sql_parts.append("WITH (READPAST)")
+
+        return " ".join(sql_parts), vals
+
+    @classmethod
+    def _build_where(cls, where):
+        """Build WHERE clause for SQL Server."""
+        if isinstance(where, str):
+            return where, []
+        
+        if isinstance(where, dict):
+            where = list(where.items())
+
+        if not isinstance(where, (list, tuple)):
+            raise ValueError("WHERE clause must be string, dict, or list")
+
+        conditions = []
+        vals = []
+
+        for key, val in where:
+            if val is None:
+                if "!" in key:
+                    key = key.replace("!", "")
+                    conditions.append(f"{quote(key)} IS NOT NULL")
+                else:
+                    conditions.append(f"{quote(key)} IS NULL")
+            elif isinstance(val, (list, tuple)):
+                if "!" in key:
+                    key = key.replace("!", "")
+                    conditions.append(f"{quote(key)} NOT IN ({', '.join(['?'] * len(val))})")
+                else:
+                    conditions.append(f"{quote(key)} IN ({', '.join(['?'] * len(val))})")
+                vals.extend(val)
+            else:
+                # Handle operators
+                op = "="
+                if "<>" in key:
+                    key = key.replace("<>", "")
+                    op = "<>"
+                elif "!=" in key:
+                    key = key.replace("!=", "")
+                    op = "<>"
+                elif "%" in key:
+                    key = key.replace("%", "")
+                    op = "LIKE"
+                elif "!" in key:
+                    key = key.replace("!", "")
+                    op = "<>"
+
+                conditions.append(f"{quote(key)} {op} ?")
+                vals.append(val)
+
+        return " AND ".join(conditions), vals
+
+    @classmethod
+    def insert(cls, table, data):
+        """Generate an INSERT statement for SQL Server."""
+        if not data:
+            raise ValueError("Data cannot be empty")
+
+        columns = list(data.keys())
+        values = list(data.values())
+        
+        sql_parts = [
+            "INSERT INTO",
+            quote(table),
+            f"({', '.join(quote(col) for col in columns)})",
+            "VALUES",
+            f"({', '.join(['?'] * len(values))})"  # SQL Server uses ? placeholders
+        ]
+
+        return " ".join(sql_parts), values
+
+    @classmethod
+    def update(cls, tx, table, data, where=None, pk=None, excluded=False):
+        """Generate an UPDATE statement for SQL Server."""
+        if not data:
+            raise ValueError("Data cannot be empty")
+        
+        if not where and not pk:
+            raise ValueError("Either WHERE clause or primary key must be provided")
+
+        # Build SET clause
+        set_clauses = []
+        vals = []
+        
+        for col, val in data.items():
+            set_clauses.append(f"{quote(col)} = ?")
+            vals.append(val)
+
+        # Build WHERE clause
+        if pk:
+            if where:
+                # Merge pk into where
+                if isinstance(where, dict):
+                    where.update(pk)
+                else:
+                    # Convert to dict for merging
+                    where_dict = dict(where) if isinstance(where, (list, tuple)) else {}
+                    where_dict.update(pk)
+                    where = where_dict
+            else:
+                where = pk
+
+        where_sql, where_vals = cls._build_where(where) if where else ("", [])
+
+        sql_parts = [
+            "UPDATE",
+            quote(table),
+            "SET",
+            ", ".join(set_clauses)
+        ]
+
+        if where_sql:
+            sql_parts.extend(["WHERE", where_sql])
+            vals.extend(where_vals)
+
+        return " ".join(sql_parts), vals
+
+    @classmethod
+    def delete(cls, tx, table, where):
+        """Generate a DELETE statement for SQL Server."""
+        if not where:
+            raise ValueError("WHERE clause is required for DELETE")
+
+        where_sql, where_vals = cls._build_where(where)
+        
+        sql_parts = [
+            "DELETE FROM",
+            quote(table),
+            "WHERE",
+            where_sql
+        ]
+
+        return " ".join(sql_parts), where_vals
+
+    @classmethod
+    def merge(cls, tx, table, data, pk, on_conflict_do_nothing, on_conflict_update):
+        """Generate a MERGE statement for SQL Server."""
+        # SQL Server MERGE is complex - simplified version
+        if on_conflict_do_nothing:
+            # Use IF NOT EXISTS pattern
+            pk_conditions = " AND ".join([f"{quote(k)} = ?" for k in pk.keys()])
+            pk_values = list(pk.values())
+            
+            insert_sql, insert_vals = cls.insert(table, data)
+            wrapped_sql = f"""
+            IF NOT EXISTS (SELECT 1 FROM {quote(table)} WHERE {pk_conditions})
+            BEGIN
+                {insert_sql}
+            END
+            """
+            return wrapped_sql, pk_values + insert_vals
+        elif on_conflict_update:
+            # Use actual MERGE statement
+            pk_columns = list(pk.keys())
+            data_columns = [k for k in data.keys() if k not in pk_columns]
+            
+            # Build MERGE statement
+            merge_parts = [
+                f"MERGE {quote(table)} AS target",
+                f"USING (SELECT {', '.join(['?' for _ in data])} AS ({', '.join(quote(k) for k in data.keys())})) AS source",
+                f"ON ({' AND '.join([f'target.{quote(k)} = source.{quote(k)}' for k in pk_columns])})",
+                "WHEN MATCHED THEN",
+                f"UPDATE SET {', '.join([f'{quote(k)} = source.{quote(k)}' for k in data_columns])}",
+                "WHEN NOT MATCHED THEN",
+                f"INSERT ({', '.join(quote(k) for k in data.keys())})",
+                f"VALUES ({', '.join([f'source.{quote(k)}' for k in data.keys()])});",
+            ]
+            
+            return " ".join(merge_parts), list(data.values())
+        else:
+            return cls.insert(table, data)
+
+    # Metadata queries
+    @classmethod
+    def version(cls):
+        return "SELECT @@VERSION"
+
+    @classmethod
+    def timestamp(cls):
+        return "SELECT GETDATE()"
+
+    @classmethod
+    def user(cls):
+        return "SELECT SYSTEM_USER"
+
+    @classmethod
+    def databases(cls):
+        return "SELECT name FROM sys.databases WHERE database_id > 4"
+
+    @classmethod
+    def schemas(cls):
+        return "SELECT name FROM sys.schemas"
+
+    @classmethod
+    def current_schema(cls):
+        return "SELECT SCHEMA_NAME()"
+
+    @classmethod
+    def current_database(cls):
+        return "SELECT DB_NAME()"
+
+    @classmethod
+    def tables(cls, system=False):
+        if system:
+            return "SELECT name FROM sys.tables"
+        else:
+            return "SELECT name FROM sys.tables WHERE is_ms_shipped = 0"
+
+    @classmethod
+    def views(cls, system=False):
+        if system:
+            return "SELECT name FROM sys.views"
+        else:
+            return "SELECT name FROM sys.views WHERE is_ms_shipped = 0"
+
+    @classmethod
+    def create_database(cls, name):
+        return f"CREATE DATABASE {quote(name)}"
+
+    @classmethod
+    def drop_database(cls, name):
+        return f"DROP DATABASE {quote(name)}"
+
+    @classmethod
+    def create_table(cls, name, columns=None, drop=False):
+        if not name or not isinstance(name, str):
+            raise ValueError("Table name must be a non-empty string")
+
+        columns = columns or {}
+
+        if "." in name:
+            schema_part, table_part = name.split(".", 1)
+        else:
+            schema_part = cls.default_schema or "dbo"
+            table_part = name
+
+        schema_identifier = quote(schema_part)
+        table_identifier = quote(name if "." in name else f"{schema_part}.{table_part}")
+        base_name = table_part.replace("[", "").replace("]", "")
+        base_name_sql = base_name.replace("'", "''")
+        trigger_prefix = re.sub(r"[^0-9A-Za-z_]+", "_", f"CC_SYS_MOD_{base_name}")
+
+        statements = []
+        if drop:
+            statements.append(f"IF OBJECT_ID(N'{table_identifier}', N'U') IS NOT NULL DROP TABLE {table_identifier};")
+
+        statements.append(
+            f"""
+CREATE TABLE {table_identifier} (
+  [sys_id] BIGINT IDENTITY(1,1) PRIMARY KEY,
+  [sys_table] NVARCHAR(255),
+  [sys_created] DATETIME2 NOT NULL DEFAULT SYSDATETIME(),
+  [sys_modified] DATETIME2 NOT NULL DEFAULT SYSDATETIME(),
+  [sys_modified_by] NVARCHAR(255),
+  [sys_modified_count] INT NOT NULL DEFAULT 0,
+  [sys_dirty] BIT NOT NULL DEFAULT 0,
+  [description] NVARCHAR(MAX)
+);
+""".strip()
+        )
+
+        for key, val in columns.items():
+            clean_key = re.sub("<>!=%", "", key)
+            if clean_key in system_fields:
+                continue
+            col_type = TYPES.get_type(val)
+            statements.append(
+                f"ALTER TABLE {table_identifier} ADD {quote(clean_key)} {col_type};"
+            )
+
+        statements.extend(
+            [
+                f"IF OBJECT_ID(N'{schema_identifier}.{trigger_prefix}_insert', N'TR') IS NOT NULL DROP TRIGGER {schema_identifier}.{trigger_prefix}_insert;",
+                f"IF OBJECT_ID(N'{schema_identifier}.{trigger_prefix}_update', N'TR') IS NOT NULL DROP TRIGGER {schema_identifier}.{trigger_prefix}_update;",
+                f"""
+CREATE TRIGGER {schema_identifier}.{trigger_prefix}_insert
+ON {table_identifier}
+AFTER INSERT
+AS
+BEGIN
+    SET NOCOUNT ON;
+    UPDATE t
+    SET sys_created = ISNULL(i.sys_created, SYSDATETIME()),
+        sys_modified = SYSDATETIME(),
+        sys_modified_count = 0,
+        sys_dirty = ISNULL(i.sys_dirty, 0),
+        sys_table = '{base_name_sql}'
+    FROM {table_identifier} AS t
+    INNER JOIN inserted AS i ON t.sys_id = i.sys_id;
+END;
+""".strip(),
+                f"""
+CREATE TRIGGER {schema_identifier}.{trigger_prefix}_update
+ON {table_identifier}
+AFTER UPDATE
+AS
+BEGIN
+    SET NOCOUNT ON;
+    UPDATE t
+    SET sys_created = d.sys_created,
+        sys_modified = SYSDATETIME(),
+        sys_table = '{base_name_sql}',
+        sys_dirty = CASE WHEN d.sys_dirty = 1 AND i.sys_dirty = 0 THEN 0 ELSE 1 END,
+        sys_modified_count = CASE WHEN d.sys_dirty = 1 AND i.sys_dirty = 0 THEN ISNULL(d.sys_modified_count, 0) ELSE ISNULL(d.sys_modified_count, 0) + 1 END
+    FROM {table_identifier} AS t
+    INNER JOIN inserted AS i ON t.sys_id = i.sys_id
+    INNER JOIN deleted AS d ON d.sys_id = i.sys_id;
+END;
+""".strip(),
+            ]
+        )
+
+        return "\n".join(statements), tuple()
+
+    @classmethod
+    def ensure_system_columns(cls, name, existing_columns=None, force=False):
+        """Ensure SQL Server tables maintain Velocity system metadata."""
+        existing_columns = {col.lower() for col in existing_columns or []}
+
+        if "." in name:
+            schema, table_name = name.split(".", 1)
+        else:
+            schema = cls.default_schema or "dbo"
+            table_name = name
+
+        schema_identifier = quote(schema)
+        table_identifier = quote(name if "." in name else f"{schema}.{table_name}")
+        object_name = f"[{schema}].[{table_name}]"
+        table_name_sql = table_name.replace("'", "''")
+        trigger_prefix = re.sub(r"[^0-9A-Za-z_]+", "_", f"CC_SYS_MOD_{table_name}")
+
+        has_count = "sys_modified_count" in existing_columns
+
+        add_column = not has_count
+        recreate_triggers = force or add_column
+
+        if not recreate_triggers and not force:
+            return None
+
+        statements = []
+
+        if add_column:
+            statements.append(
+                f"IF COL_LENGTH(N'{object_name}', 'sys_modified_count') IS NULL BEGIN ALTER TABLE {table_identifier} ADD sys_modified_count INT NOT NULL CONSTRAINT DF_{trigger_prefix}_COUNT DEFAULT (0); END;"
+            )
+
+        statements.append(
+            f"UPDATE {table_identifier} SET sys_modified_count = 0 WHERE sys_modified_count IS NULL;"
+        )
+
+        statements.append(
+            f"IF OBJECT_ID(N'{schema_identifier}.{trigger_prefix}_insert', N'TR') IS NOT NULL DROP TRIGGER {schema_identifier}.{trigger_prefix}_insert;"
+        )
+        statements.append(
+            f"IF OBJECT_ID(N'{schema_identifier}.{trigger_prefix}_update', N'TR') IS NOT NULL DROP TRIGGER {schema_identifier}.{trigger_prefix}_update;"
+        )
+
+        statements.extend(
+            [
+                f"""
+CREATE TRIGGER {schema_identifier}.{trigger_prefix}_insert
+ON {table_identifier}
+AFTER INSERT
+AS
+BEGIN
+    SET NOCOUNT ON;
+    UPDATE t
+    SET sys_created = ISNULL(i.sys_created, SYSDATETIME()),
+        sys_modified = SYSDATETIME(),
+        sys_modified_count = 0,
+        sys_dirty = ISNULL(i.sys_dirty, 0),
+        sys_table = '{table_name_sql}'
+    FROM {table_identifier} AS t
+    INNER JOIN inserted AS i ON t.sys_id = i.sys_id;
+END;
+""".strip(),
+                f"""
+CREATE TRIGGER {schema_identifier}.{trigger_prefix}_update
+ON {table_identifier}
+AFTER UPDATE
+AS
+BEGIN
+    SET NOCOUNT ON;
+    UPDATE t
+    SET sys_created = d.sys_created,
+        sys_modified = SYSDATETIME(),
+        sys_table = '{table_name_sql}',
+        sys_dirty = CASE WHEN d.sys_dirty = 1 AND i.sys_dirty = 0 THEN 0 ELSE 1 END,
+        sys_modified_count = CASE WHEN d.sys_dirty = 1 AND i.sys_dirty = 0 THEN ISNULL(d.sys_modified_count, 0) ELSE ISNULL(d.sys_modified_count, 0) + 1 END
+    FROM {table_identifier} AS t
+    INNER JOIN inserted AS i ON t.sys_id = i.sys_id
+    INNER JOIN deleted AS d ON d.sys_id = i.sys_id;
+END;
+""".strip(),
+            ]
+        )
+
+        return "\n".join(statements), tuple()
+
+    @classmethod
+    def drop_table(cls, name):
+        return f"DROP TABLE {quote(name)}"
+
+    @classmethod
+    def truncate(cls, table):
+        return f"TRUNCATE TABLE {quote(table)}"
+
+    @classmethod
+    def columns(cls, name):
+        return f"""
+        SELECT 
+            COLUMN_NAME,
+            DATA_TYPE,
+            IS_NULLABLE,
+            COLUMN_DEFAULT,
+            CHARACTER_MAXIMUM_LENGTH
+        FROM INFORMATION_SCHEMA.COLUMNS 
+        WHERE TABLE_NAME = '{name}'
+        ORDER BY ORDINAL_POSITION
+        """
+
+    @classmethod
+    def column_info(cls, table, name):
+        return f"""
+        SELECT 
+            COLUMN_NAME,
+            DATA_TYPE,
+            IS_NULLABLE,
+            COLUMN_DEFAULT,
+            CHARACTER_MAXIMUM_LENGTH
+        FROM INFORMATION_SCHEMA.COLUMNS 
+        WHERE TABLE_NAME = '{table}' AND COLUMN_NAME = '{name}'
+        """
+
+    @classmethod
+    def drop_column(cls, table, name, cascade=True):
+        return f"ALTER TABLE {quote(table)} DROP COLUMN {quote(name)}"
+
+    @classmethod
+    def alter_add(cls, table, columns, null_allowed=True):
+        alter_parts = []
+        for col, col_type in columns.items():
+            null_clause = "NULL" if null_allowed else "NOT NULL"
+            alter_parts.append(f"ADD {quote(col)} {col_type} {null_clause}")
+        
+        return f"ALTER TABLE {quote(table)} {', '.join(alter_parts)}"
+
+    @classmethod
+    def alter_drop(cls, table, columns):
+        drop_parts = [f"DROP COLUMN {quote(col)}" for col in columns]
+        return f"ALTER TABLE {quote(table)} {', '.join(drop_parts)}"
+
+    @classmethod
+    def alter_column_by_type(cls, table, column, value, nullable=True):
+        null_clause = "NULL" if nullable else "NOT NULL"
+        return f"ALTER TABLE {quote(table)} ALTER COLUMN {quote(column)} {value} {null_clause}"
+
+    @classmethod
+    def alter_column_by_sql(cls, table, column, value):
+        return f"ALTER TABLE {quote(table)} ALTER COLUMN {quote(column)} {value}"
+
+    @classmethod
+    def rename_column(cls, table, orig, new):
+        return f"EXEC sp_rename '{table}.{orig}', '{new}', 'COLUMN'"
+
+    @classmethod
+    def rename_table(cls, table, new):
+        return f"EXEC sp_rename '{table}', '{new}'"
+
+    @classmethod
+    def primary_keys(cls, table):
+        return f"""
+        SELECT COLUMN_NAME
+        FROM INFORMATION_SCHEMA.KEY_COLUMN_USAGE
+        WHERE OBJECTPROPERTY(OBJECT_ID(CONSTRAINT_SCHEMA + '.' + CONSTRAINT_NAME), 'IsPrimaryKey') = 1
+        AND TABLE_NAME = '{table}'
+        """
+
+    @classmethod
+    def foreign_key_info(cls, table=None, column=None, schema=None):
+        sql = """
+        SELECT 
+            FK.TABLE_NAME,
+            CU.COLUMN_NAME,
+            PK.TABLE_NAME AS REFERENCED_TABLE_NAME,
+            PT.COLUMN_NAME AS REFERENCED_COLUMN_NAME,
+            C.CONSTRAINT_NAME
+        FROM INFORMATION_SCHEMA.REFERENTIAL_CONSTRAINTS C
+        INNER JOIN INFORMATION_SCHEMA.TABLE_CONSTRAINTS FK ON C.CONSTRAINT_NAME = FK.CONSTRAINT_NAME
+        INNER JOIN INFORMATION_SCHEMA.TABLE_CONSTRAINTS PK ON C.UNIQUE_CONSTRAINT_NAME = PK.CONSTRAINT_NAME
+        INNER JOIN INFORMATION_SCHEMA.KEY_COLUMN_USAGE CU ON C.CONSTRAINT_NAME = CU.CONSTRAINT_NAME
+        INNER JOIN INFORMATION_SCHEMA.KEY_COLUMN_USAGE PT ON PK.CONSTRAINT_NAME = PT.CONSTRAINT_NAME
+        """
+        if table:
+            sql += f" WHERE FK.TABLE_NAME = '{table}'"
+        if column:
+            conjunction = " AND" if table else " WHERE"
+            sql += f"{conjunction} CU.COLUMN_NAME = '{column}'"
+        return sql
+
+    @classmethod
+    def create_foreign_key(cls, table, columns, key_to_table, key_to_columns, name=None, schema=None):
+        if name is None:
+            name = f"FK_{table}_{'_'.join(columns)}"
+        
+        col_list = ", ".join(quote(col) for col in columns)
+        ref_col_list = ", ".join(quote(col) for col in key_to_columns)
+        
+        return f"""
+        ALTER TABLE {quote(table)} 
+        ADD CONSTRAINT {quote(name)} 
+        FOREIGN KEY ({col_list}) 
+        REFERENCES {quote(key_to_table)} ({ref_col_list})
+        """
+
+    @classmethod
+    def drop_foreign_key(cls, table, columns, key_to_table=None, key_to_columns=None, name=None, schema=None):
+        if name is None:
+            name = f"FK_{table}_{'_'.join(columns)}"
+        
+        return f"ALTER TABLE {quote(table)} DROP CONSTRAINT {quote(name)}"
+
+    @classmethod
+    def create_index(cls, tx, table=None, columns=None, unique=False, direction=None, where=None, name=None, schema=None, trigram=None, lower=None):
+        if name is None:
+            name = f"IX_{table}_{'_'.join(columns)}"
+        
+        index_type = "UNIQUE INDEX" if unique else "INDEX"
+        col_list = ", ".join(quote(col) for col in columns)
+        
+        sql = f"CREATE {index_type} {quote(name)} ON {quote(table)} ({col_list})"
+        
+        if where:
+            sql += f" WHERE {where}"
+        
+        return sql
+
+    @classmethod
+    def drop_index(cls, table=None, columns=None, name=None, schema=None, trigram=None):
+        if name is None:
+            name = f"IX_{table}_{'_'.join(columns)}"
+        
+        return f"DROP INDEX {quote(name)} ON {quote(table)}"
+
+    @classmethod
+    def indexes(cls, table):
+        return f"""
+        SELECT 
+            i.name AS index_name,
+            c.name AS column_name,
+            i.is_unique
+        FROM sys.indexes i
+        INNER JOIN sys.index_columns ic ON i.object_id = ic.object_id AND i.index_id = ic.index_id
+        INNER JOIN sys.columns c ON ic.object_id = c.object_id AND ic.column_id = c.column_id
+        WHERE i.object_id = OBJECT_ID('{table}')
+        ORDER BY i.name, ic.key_ordinal
+        """
+
+    @classmethod
+    def create_savepoint(cls, sp):
+        return f"SAVE TRANSACTION {sp}"
+
+    @classmethod
+    def release_savepoint(cls, sp):
+        return f"-- SQL Server doesn't support RELEASE SAVEPOINT {sp}"
+
+    @classmethod
+    def rollback_savepoint(cls, sp):
+        return f"ROLLBACK TRANSACTION {sp}"
+
+    @classmethod
+    def create_view(cls, name, query, temp=False, silent=True):
+        # SQL Server doesn't support temporary views in the same way
+        return f"CREATE VIEW {quote(name)} AS {query}"
+
+    @classmethod
+    def drop_view(cls, name, silent=True):
+        if silent:
+            return f"DROP VIEW IF EXISTS {quote(name)}"
+        else:
+            return f"DROP VIEW {quote(name)}"
+
+    @classmethod
+    def last_id(cls, table):
+        return "SELECT @@IDENTITY"
+
+    @classmethod
+    def current_id(cls, table):
+        return f"SELECT IDENT_CURRENT('{table}')"
+
+    @classmethod
+    def set_id(cls, table, start):
+        return f"DBCC CHECKIDENT('{table}', RESEED, {start})"
+
+    @classmethod
+    def set_sequence(cls, table, next_value):
+        return f"DBCC CHECKIDENT('{table}', RESEED, {next_value})"
+
+    @classmethod
+    def massage_data(cls, data):
+        """Massage data before insert/update operations."""
+        # SQL Server-specific data transformations
+        return data
+
+    @classmethod
+    def alter_trigger(cls, table, state="ENABLE", name="USER"):
+        state_cmd = "ENABLE" if state.upper() == "ENABLE" else "DISABLE"
+        return f"ALTER TABLE {quote(table)} {state_cmd} TRIGGER ALL"
+
+    @classmethod
+    def missing(cls, tx, table, list_values, column="SYS_ID", where=None):
+        """Generate query to find missing values from a list."""
+        # SQL Server version using VALUES clause
+        value_rows = ", ".join([f"(?)" for _ in list_values])
+        
+        sql = f"""
+        SELECT value_column FROM (
+            VALUES {value_rows}
+        ) AS input_values(value_column)
+        WHERE value_column NOT IN (
+            SELECT {quote(column)} FROM {quote(table)}
+        """
+        
+        vals = list_values
+        
+        if where:
+            where_sql, where_vals = cls._build_where(where)
+            sql += f" WHERE {where_sql}"
+            vals.extend(where_vals)
+        
+        sql += ")"
+        
+        return sql, vals