thoth-dbmanager 0.4.0__py3-none-any.whl → 0.4.2__py3-none-any.whl

thoth_dbmanager/helpers/multi_db_generator.py

@@ -0,0 +1,508 @@
+import logging
+import random
+import re
+from typing import Dict, List, Optional
+from pathlib import Path
+
+from dbmanager.helpers.schema import DatabaseSchema
+
+
+class MultiDbGenerator:
+    """
+    Class for multi-database schema generation.
+    Works with any ThothDbManager implementation.
+    """
+
+    CACHED_DB_SCHEMA = {}
+
+    def __init__(
+        self,
+        dbmanager,
+        tentative_schema: Optional[DatabaseSchema] = None,
+        schema_with_examples: Optional[DatabaseSchema] = None,
+        schema_with_descriptions: Optional[DatabaseSchema] = None,
+        add_examples: bool = True):
+        self.db_manager = dbmanager
+        self.db_id = dbmanager.db_id if dbmanager else None
+        self.add_examples = add_examples
+        self.schema_structure = tentative_schema or DatabaseSchema()
+        self.schema_with_examples = schema_with_examples or DatabaseSchema()
+        self.schema_with_descriptions = schema_with_descriptions or DatabaseSchema()
+
+        if dbmanager and self.db_id not in MultiDbGenerator.CACHED_DB_SCHEMA:
+            self._load_schema_into_cache()
+        self._initialize_schema_structure()
+
+    def _load_schema_into_cache(self) -> None:
+        """Load database schema into cache using manager methods"""
+        schema_dict = self.db_manager.get_schema_dict()
+        db_schema = DatabaseSchema.from_schema_dict(schema_dict)
+        MultiDbGenerator.CACHED_DB_SCHEMA[self.db_id] = db_schema
+
+        # Set primary keys
+        primary_keys = self.db_manager.get_primary_keys()
+        schema_with_primary_keys = {
+            table_name: {col: {"primary_key": True} for col in cols}
+            for table_name, cols in primary_keys.items()
+        }
+        db_schema.set_columns_info(schema_with_primary_keys)
+
+        # Set foreign keys
+        foreign_keys = self.db_manager.get_foreign_keys()
+        schema_with_references = {
+            table_name: {
+                column_name: {
+                    "foreign_keys": info.get("foreign_keys", []),
+                    "referenced_by": info.get("referenced_by", []),
+                }
+                for column_name, info in columns.items()
+            }
+            for table_name, columns in foreign_keys.items()
+        }
+        db_schema.set_columns_info(schema_with_references)
+
+    def _set_primary_keys(self, database_schema: DatabaseSchema) -> None:
+        """Set primary keys using manager method"""
+        primary_keys = self.db_manager.get_primary_keys()
+        schema_with_primary_keys = {
+            table_name: {col: {"primary_key": True} for col in cols}
+            for table_name, cols in primary_keys.items()
+        }
+        database_schema.set_columns_info(schema_with_primary_keys)
+
+    def _set_foreign_keys(self, database_schema: DatabaseSchema) -> None:
+        """Set foreign keys using manager method"""
+        foreign_keys = self.db_manager.get_foreign_keys()
+        schema_with_references = {
+            table_name: {
+                column_name: {
+                    "foreign_keys": info.get("foreign_keys", []),
+                    "referenced_by": info.get("referenced_by", []),
+                }
+                for column_name, info in columns.items()
+            }
+            for table_name, columns in foreign_keys.items()
+        }
+        database_schema.set_columns_info(schema_with_references)
+
+    def _initialize_schema_structure(self) -> None:
+        """Initialize the schema structure with table and column info"""
+        self._load_table_and_column_info()
+        self._load_column_examples()
+        self._load_column_descriptions()
+
+    def _load_table_and_column_info(self) -> None:
+        """Load table and column information from cached schema"""
+        if self.db_id in MultiDbGenerator.CACHED_DB_SCHEMA:
+            self.schema_structure = MultiDbGenerator.CACHED_DB_SCHEMA[
+                self.db_id
+            ].subselect_schema(self.schema_structure)
+            self.schema_structure.add_info_from_schema(
+                schema=MultiDbGenerator.CACHED_DB_SCHEMA[self.db_id],
+                field_names=["type", "primary_key", "foreign_keys", "referenced_by"],
+            )
+
+    def _load_column_examples(self) -> None:
+        """Load examples for columns in the schema"""
+        self.schema_structure.add_info_from_schema(
+            schema=self.schema_with_examples, field_names=["examples"]
+        )
+
+    def _load_column_descriptions(self) -> None:
+        """Load descriptions for columns in the schema"""
+        self.schema_structure.add_info_from_schema(
+            schema=self.schema_with_descriptions,
+            field_names=[
+                "original_column_name",
+                "column_name",
+                "column_description",
+                "data_format",
+                "value_description",
+            ],
+        )
+
+    def _extract_create_ddl_commands(self) -> Dict[str, str]:
+        """Extract CREATE TABLE DDL commands for all tables
+
+        Returns:
+            Dict[str, str]: Dictionary mapping table names to their CREATE TABLE DDL statements
+        """
+        if hasattr(self.db_manager, 'extract_create_ddl_commands'):
+            return self.db_manager.extract_create_ddl_commands()
+        else:
+            raise NotImplementedError("extract_create_ddl_commands method not implemented for this database manager")
+
+    def generate_schema_string(
+        self,
+        include_value_description: bool = True,
+        shuffle_cols: bool = True,
+        shuffle_tables: bool = True,
+    ) -> str:
+        """
+        Generates a schema string with descriptions and examples.
+
+        Args:
+            include_value_description (bool): Flag to include value descriptions.
+            shuffle_cols (bool): Flag to shuffle columns within tables.
+            shuffle_tables (bool): Flag to shuffle tables in the output.
+
+        Returns:
+            str: The generated schema string.
+        """
+        ddl_commands = self._extract_create_ddl_commands()
+        schema_strings = []
+        
+        if shuffle_tables:
+            ddl_tables = list(ddl_commands.keys())
+            random.shuffle(ddl_tables)
+            ddl_commands = {
+                table_name: ddl_commands[table_name] for table_name in ddl_tables
+            }
+        
+        for table_name, ddl_command in ddl_commands.items():
+            ddl_command = re.sub(r"\s+", " ", ddl_command.strip())
+            create_table_match = re.match(
+                r'CREATE TABLE "?`?([\w -]+)`?"?\s*\((.*)\)', ddl_command, re.DOTALL
+            )
+            
+            if not create_table_match:
+                logging.warning(f"Could not parse DDL command for table {table_name}")
+                continue
+            
+            table = create_table_match.group(1).strip()
+            if table != table_name:
+                logging.warning(f"Table name mismatch: {table} != {table_name}")
+            
+            column_definitions = create_table_match.group(2).strip()
+            
+            if table_name in self.schema_structure.tables:
+                table_schema = self.schema_structure.tables[table_name]
+                
+                # Start building the new CREATE TABLE statement
+                schema_lines = [f"CREATE TABLE {table_name}", "("]
+                
+                # Process column definitions
+                definitions = self._separate_column_definitions(column_definitions)
+                column_defs = []
+                constraint_defs = []
+                
+                # Extract column definitions and constraints
+                for definition in definitions:
+                    if definition.lower().startswith("foreign key") or definition.lower().startswith("constraint"):
+                        constraint_defs.append(definition)
+                    else:
+                        column_match = re.match(r'"?`?([\w_]+)`?"?\s+(.*)', definition)
+                        if column_match:
+                            column_name = column_match.group(1)
+                            column_type = column_match.group(2)
+                            
+                            # Remove NULL/NOT NULL constraints as requested
+                            column_type = re.sub(r'\s+(?:NOT\s+)?NULL', '', column_type, flags=re.IGNORECASE)
+                            
+                            # Check if this is a primary key
+                            is_primary_key = "primary key" in column_type.lower()
+                            
+                            # Format the column definition
+                            column_def = f"\t{column_name} {column_type}"
+                            
+                            # Add comments with examples and descriptions
+                            if column_name in table_schema.columns:
+                                column_info = table_schema.columns[column_name]
+                                comment_parts = []
+                                
+                                # Add examples if available
+                                if hasattr(column_info, 'examples') and column_info.examples:
+                                    examples = [f"`{ex}`" for ex in column_info.examples[:3]]  # Limit to 3 examples
+                                    comment_parts.append(f"examples: {', '.join(examples)}")
+                                
+                                # Add column name if available
+                                if hasattr(column_info, 'column_name') and column_info.column_name:
+                                    comment_parts.append(f"| `{column_info.column_name}`")
+                                
+                                # Add column description if available
+                                if hasattr(column_info, 'column_description') and column_info.column_description:
+                                    comment_parts.append(f"description: {column_info.column_description}")
+                                
+                                # Add value description if available and requested
+                                if include_value_description and hasattr(column_info, 'value_description') and column_info.value_description:
+                                    comment_parts.append(f"values: {column_info.value_description}")
+                                
+                                # Add the comment to the column definition
+                                if comment_parts:
+                                    column_def += f" -- {' '.join(comment_parts)}"
+                            
+                            column_defs.append(column_def)
+        
+                # Process foreign key constraints with references
+                for column_name, column_info in table_schema.columns.items():
+                    if hasattr(column_info, 'foreign_keys') and column_info.foreign_keys:
+                        for ref_table, ref_column in column_info.foreign_keys:
+                            # Fixed: Properly access tuple elements instead of using dictionary access
+                            fk_constraint = f"\tforeign key ({column_name}) references {ref_table} ({ref_column}) on update cascade on delete cascade"
+                            constraint_defs.append(fk_constraint)
+        
+                # Combine column definitions and constraints
+                all_defs = column_defs + constraint_defs
+                schema_lines.extend(all_defs)
+                schema_lines.append(");")
+                
+                # Join all lines to form the complete CREATE TABLE statement
+                schema_strings.append("\n".join(schema_lines))
+        
+        return "\n\n".join(schema_strings)
+
+    @staticmethod
+    def _separate_column_definitions(column_definitions: str) -> List[str]:
+        """Separate column definitions from a CREATE TABLE statement
+
+        Args:
+            column_definitions (str): The column definitions part of a CREATE TABLE statement
+
+        Returns:
+            List[str]: List of individual column definitions
+        """
+        definitions = []
+        current_def = ""
+        paren_count = 0
+        
+        for char in column_definitions:
+            if char == '(' and not current_def.strip().lower().startswith("constraint"):
+                paren_count += 1
+            elif char == ')' and not current_def.strip().lower().startswith("constraint"):
+                paren_count -= 1
+            
+            current_def += char
+            
+            if char == ',' and paren_count == 0:
+                definitions.append(current_def[:-1].strip())
+                current_def = ""
+        
+        if current_def.strip():
+            definitions.append(current_def.strip())
+        
+        return definitions
+
+    def _is_connection(self, table_name: str, column_name: str) -> bool:
+        """
+        Checks if a column is a connection (primary key or foreign key).
+
+        Args:
+            table_name (str): The name of the table.
+            column_name (str): The name of the column.
+
+        Returns:
+            bool: True if the column is a connection, False otherwise.
+        """
+        column_info = self.CACHED_DB_SCHEMA[self.db_id].get_column_info(
+            table_name, column_name
+        )
+        if column_info is None:
+            return False
+        if column_info.primary_key:
+            return True
+        for target_table, _ in column_info.foreign_keys:
+            if self.schema_structure.get_table_info(target_table):
+                return True
+        for target_table, _ in column_info.referenced_by:
+            if self.schema_structure.get_table_info(target_table):
+                return True
+        for target_table_name, table_schema in self.schema_structure.tables.items():
+            if table_name.lower() == target_table_name.lower():
+                continue
+            for target_column_name, target_column_info in table_schema.columns.items():
+                if (
+                    target_column_name.lower() == column_name.lower()
+                    and target_column_info.primary_key
+                ):
+                    return True
+        return False
+
+    def _get_connections(self) -> Dict[str, List[str]]:
+        """
+        Retrieves connections between tables in the schema.
+
+        Returns:
+            Dict[str, List[str]]: A dictionary mapping table names to lists of connected columns.
+        """
+        connections = {}
+        for table_name, table_schema in self.schema_structure.tables.items():
+            connections[table_name] = []
+            for column_name, column_info in (
+                self.CACHED_DB_SCHEMA[self.db_id].tables[table_name].columns.items()
+            ):
+                if self._is_connection(table_name, column_name):
+                    connections[table_name].append(column_name)
+        return connections
+
+    def get_schema_with_connections(self) -> Dict[str, List[str]]:
+        """
+        Gets schema with connections included.
+
+        Returns:
+            Dict[str, List[str]]: The schema with connections included.
+        """
+        schema_structure_dict = self.schema_structure.to_dict()
+        connections = self._get_connections()
+        for table_name, connected_columns in connections.items():
+            for column_name in connected_columns:
+                if column_name.lower() not in [
+                    col.lower() for col in schema_structure_dict[table_name]
+                ]:
+                    schema_structure_dict[table_name].append(column_name)
+        return schema_structure_dict
+
+    def _get_example_column_name_description(
+        self, table_name: str, column_name: str, include_value_description: bool = True
+    ) -> str:
+        """
+        Retrieves example values and descriptions for a column.
+
+        Args:
+            table_name (str): The name of the table.
+            column_name (str): The name of the column.
+            include_value_description (bool): Flag to include value description.
+
+        Returns:
+            str: The example values and descriptions for the column.
+        """
+        example_part = ""
+        name_string = ""
+        description_string = ""
+        value_statics_string = ""
+        value_description_string = ""
+
+        column_info = self.schema_structure.get_column_info(table_name, column_name)
+        if column_info:
+            if column_info.examples:
+                example_part = f" Example Values: {', '.join([f'`{str(x)}`' for x in column_info.examples])}"
+            if column_info.value_statics:
+                value_statics_string = f" Value Statics: {column_info.value_statics}"
+            if column_info.column_name:
+                if (column_info.column_name.lower() != column_name.lower()) and (
+                    column_info.column_name.strip() != ""
+                ):
+                    name_string = f"| Column Name Meaning: {column_info.column_name}"
+            if column_info.column_description:
+                description_string = (
+                    f"| Column Description: {column_info.column_description}"
+                )
+            if column_info.value_description and include_value_description:
+                value_description_string = (
+                    f"| Value Description: {column_info.value_description}"
+                )
+
+        description_part = (
+            f"{name_string} {description_string} {value_description_string}"
+        )
+        joint_string = (
+            f" --{example_part} |{value_statics_string} {description_part}"
+            if example_part and description_part
+            else f" --{example_part or description_part or value_statics_string}"
+        )
+        if joint_string == " --":
+            joint_string = ""
+        return joint_string.replace("\n", " ") if joint_string else ""
+
+    def get_column_profiles(
+        self, with_keys: bool = False, with_references: bool = False
+    ) -> Dict[str, Dict[str, str]]:
+        """
+        Retrieves profiles for columns in the schema.
+        The output is a dictionary with table names as keys mapping to dictionaries with column names as keys and column profiles as values.
+
+        Args:
+            with_keys (bool): Flag to include primary keys and foreign keys.
+            with_references (bool): Flag to include referenced columns.
+
+        Returns:
+            Dict[str, Dict[str, str]]: The column profiles.
+        """
+        column_profiles = {}
+        for table_name, table_schema in self.schema_structure.tables.items():
+            column_profiles[table_name] = {}
+            for column_name, column_info in table_schema.columns.items():
+                if with_keys or not (
+                    column_info.primary_key
+                    or column_info.foreign_keys
+                    or column_info.referenced_by
+                ):
+                    column_profile = f"Table name: `{table_name}`\nOriginal column name: `{column_name}`\n"
+                    if (
+                        column_info.column_name.lower().strip()
+                        != column_name.lower().strip()
+                    ) and (column_info.column_name.strip() != ""):
+                        column_profile += (
+                            f"Expanded column name: `{column_info.column_name}`\n"
+                        )
+                    if column_info.type:
+                        column_profile += f"Data type: {column_info.type}\n"
+                    if column_info.column_description:
+                        column_profile += (
+                            f"Description: {column_info.column_description}\n"
+                        )
+                    if column_info.value_description:
+                        column_profile += (
+                            f"Value description: {column_info.value_description}\n"
+                        )
+                    if column_info.examples:
+                        column_profile += f"Example of values in the column: {', '.join([f'`{str(x)}`' for x in column_info.examples])}\n"
+                    if column_info.primary_key:
+                        column_profile += "This column is a primary key.\n"
+                    if with_references:
+                        if column_info.foreign_keys:
+                            column_profile += (
+                                "This column references the following columns:\n"
+                            )
+                            for target_table, target_column in column_info.foreign_keys:
+                                column_profile += f"    Table: `{target_table}`, Column: `{target_column}`\n"
+                        if column_info.referenced_by:
+                            column_profile += (
+                                "This column is referenced by the following columns:\n"
+                            )
+                            for (
+                                source_table,
+                                source_column,
+                            ) in column_info.referenced_by:
+                                column_profile += f"    Table: `{source_table}`, Column: `{source_column}`\n"
+                    column_profiles[table_name][column_name] = column_profile
+        return column_profiles
+
+    def validate_schema_consistency(self) -> List[str]:
+        """
+        Validates the consistency between the schema in the generator and the database manager.
+        
+        Returns:
+            List[str]: A list of validation error messages, empty if no errors.
+        """
+        errors = []
+        
+        # Skip validation if no database manager is provided
+        if not self.db_manager:
+            return ["No database manager provided for validation"]
+        
+        # Compare schema tables with database manager tables
+        db_schema_dict = self.db_manager.get_schema_dict()
+        for table_name in self.schema_structure.tables:
+            if table_name not in db_schema_dict:
+                errors.append(f"Table '{table_name}' exists in schema but not in database manager")
+        
+        # Validate foreign key references
+        for table_name, table_schema in self.schema_structure.tables.items():
+            for column_name, column_info in table_schema.columns.items():
+                if column_info.foreign_keys:
+                    for ref_table, ref_column in column_info.foreign_keys:
+                        # Check if referenced table exists
+                        if ref_table not in self.schema_structure.tables:
+                            errors.append(f"Foreign key in {table_name}.{column_name} references non-existent table {ref_table}")
+                            continue
+                        
+                        # Check if referenced column exists
+                        if ref_column not in self.schema_structure.tables[ref_table].columns:
+                            errors.append(f"Foreign key in {table_name}.{column_name} references non-existent column {ref_table}.{ref_column}")
+        
+        # Use the DatabaseSchema's validate_schema method if available
+        if hasattr(self.schema_structure, 'validate_schema'):
+            schema_errors = self.schema_structure.validate_schema()
+            errors.extend(schema_errors)
+        
+        return errors

thoth_dbmanager/helpers/preprocess_values.py

@@ -0,0 +1,159 @@
+import logging
+import pickle
+from pathlib import Path
+from typing import Dict, List, Tuple
+
+from datasketch import MinHash, MinHashLSH
+from tqdm import tqdm
+
+
+def _create_minhash(signature_size: int, string: str, n_gram: int) -> MinHash:
+    """
+    Creates a MinHash object for a given string.
+
+    Args:
+        signature_size (int): The size of the MinHash signature.
+        string (str): The input string to create the MinHash for.
+        n_gram (int): The n-gram size for the MinHash.
+
+    Returns:
+        MinHash: The MinHash object for the input string.
+    """
+    m = MinHash(num_perm=signature_size)
+    for d in [string[i : i + n_gram] for i in range(len(string) - n_gram + 1)]:
+        m.update(d.encode("utf8"))
+    return m
+
+
+def skip_column(column_name: str, column_values: List[str]) -> bool:
+    """
+    Determines whether to skip processing a column based on its values.
+
+    Args:
+        column_name (str): The name of the column.
+        column_values (List[str]): The list of values in the column.
+
+    Returns:
+        bool: True if the column should be skipped, False otherwise.
+    """
+    if "name" in column_name.lower():
+        return False
+    sum_of_lengths = sum(len(value) for value in column_values)
+    average_length = sum_of_lengths / len(column_values)
+    return (sum_of_lengths > 50000) and (average_length > 20)
+
+
+def make_lsh(
+    unique_values: Dict[str, Dict[str, List[str]]],
+    signature_size: int,
+    n_gram: int,
+    threshold: float,
+    verbose: bool = True,
+) -> Tuple[MinHashLSH, Dict[str, Tuple[MinHash, str, str, str]]]:
+    """
+    Creates a MinHash Locality-Sensitive Hashing (LSH) index from unique values in a database.
+
+    This function processes unique values from database tables and columns, creates MinHash
+    signatures for each value, and builds an LSH index for efficient similarity search.
+
+    Args:
+        unique_values (Dict[str, Dict[str, List[str]]]): A nested dictionary containing unique values
+            from the database. The structure is {table_name: {column_name: [values]}}.
+        signature_size (int): The number of permutations to use in the MinHash signatures.
+        n_gram (int): The size of n-grams to use when creating MinHash signatures.
+        threshold (float): The similarity threshold for the LSH index. Values closer to 1 require
+            higher similarity for matches.
+        verbose (bool, optional): If True, displays a progress bar during processing. Defaults to True.
+
+    Returns:
+        Tuple[MinHashLSH, Dict[str, Tuple[MinHash, str, str, str]]]: A tuple containing:
+            - MinHashLSH: The constructed LSH index.
+            - Dict[str, Tuple[MinHash, str, str, str]]: A dictionary mapping unique keys to tuples
+              containing (MinHash object, table name, column name, original value).
+
+    Raises:
+        Exception: If an error occurs during LSH creation, it's logged but not raised.
+
+    Note:
+        This function uses the datasketch library for MinHash and LSH operations.
+    """
+    lsh = MinHashLSH(threshold=threshold, num_perm=signature_size)
+    minhashes: Dict[str, Tuple[MinHash, str, str, str]] = {}
+    try:
+        total_unique_values = sum(
+            len(column_values)
+            for table_values in unique_values.values()
+            for column_values in table_values.values()
+        )
+        logging.info(f"Total unique values: {total_unique_values}")
+
+        progress_bar = (
+            tqdm(total=total_unique_values, desc="Creating LSH") if verbose else None
+        )
+
+        for table_name, table_values in unique_values.items():
+            for column_name, column_values in table_values.items():
+                if column_name.lower() == "doctype":
+                    print("=" * 20)
+                    print("Doctype found")
+                    print("=" * 20)
+                logging.info(
+                    f"Processing {table_name} - {column_name} - {len(column_values)}"
+                )
+
+                for id, value in enumerate(column_values):
+                    minhash = _create_minhash(signature_size, value, n_gram)
+                    minhash_key = f"{table_name}_{column_name}_{id}"
+                    minhashes[minhash_key] = (minhash, table_name, column_name, value)
+                    lsh.insert(minhash_key, minhash)
+
+                    if verbose:
+                        progress_bar.update(1)
+
+        if verbose:
+            progress_bar.close()
+    except Exception as e:
+        logging.error(f"Error creating LSH: {e}")
+
+    return lsh, minhashes
+
+
+def make_db_lsh(db, db_directory_path, db_name, **kwargs) -> None:
+    """
+    Creates a MinHash LSH for the database and saves the results.
+    
+    This function maintains backward compatibility while using the new LSH architecture.
+
+    Args:
+        db: Database manager instance
+        db_directory_path (str): The path to the database directory.
+        db_name (str): Name of the database
+        **kwargs (Any): Additional arguments for the LSH creation.
+    """
+    # Use the new LSH factory for database-independent creation
+    from ..lsh.factory import LshFactory
+    
+    try:
+        # Try using the new architecture
+        LshFactory.create_lsh_from_db(db, **kwargs)
+    except Exception as e:
+        logging.warning(f"New LSH creation failed, falling back to old method: {e}")
+        
+        # Fallback to old method for backward compatibility
+        preprocessed_path = Path(db_directory_path) / "preprocessed"
+        logging.info(f"Preprocessed directory: {preprocessed_path}")
+        preprocessed_path.mkdir(parents=True, exist_ok=True)
+
+        unique_values = db.get_unique_values()
+        logging.info("Unique values obtained")
+
+        with open(preprocessed_path / f"{db_name}_unique_values.pkl", "wb") as file:
+            pickle.dump(unique_values, file)
+        logging.info("Saved unique values")
+
+        lsh, minhashes = make_lsh(unique_values, **kwargs)
+
+        with open(preprocessed_path / f"{db_name}_lsh.pkl", "wb") as file:
+            pickle.dump(lsh, file)
+        with open(preprocessed_path / f"{db_name}_minhashes.pkl", "wb") as file:
+            pickle.dump(minhashes, file)