Spaces:
Paused
Paused
File size: 11,271 Bytes
a584f85 85de883 a584f85 4bb196e a584f85 4bb196e a584f85 85de883 a584f85 85de883 a584f85 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 |
"""Data models for Schema Translator using Pydantic."""
from datetime import datetime, timezone
from enum import Enum
from typing import Any, Dict, List, Optional
from pydantic import BaseModel, Field
# Enums
class SemanticType(str, Enum):
"""Semantic types for values."""
LIFETIME_TOTAL = "lifetime_total"
ANNUAL_RECURRING_REVENUE = "annual_recurring_revenue"
DATE = "date"
DAYS_REMAINING = "days_remaining"
TEXT = "text"
INTEGER = "integer"
FLOAT = "float"
BOOLEAN = "boolean"
class ContractStatus(str, Enum):
"""Contract status values."""
ACTIVE = "active"
INACTIVE = "inactive"
EXPIRED = "expired"
PENDING = "pending"
RENEWED = "renewed"
class QueryOperator(str, Enum):
"""Query filter operators."""
EQUALS = "equals"
NOT_EQUALS = "not_equals"
GREATER_THAN = "greater_than"
GREATER_THAN_OR_EQUAL = "greater_than_or_equal"
LESS_THAN = "less_than"
LESS_THAN_OR_EQUAL = "less_than_or_equal"
BETWEEN = "between"
IN = "in"
NOT_IN = "not_in"
CONTAINS = "contains"
STARTS_WITH = "starts_with"
WITHIN_NEXT_DAYS = "within_next_days"
DATE_RANGE = "date_range"
class QueryIntent(str, Enum):
"""Query intent types."""
FIND_CONTRACTS = "find_contracts"
COUNT_CONTRACTS = "count_contracts"
AGGREGATE_VALUES = "aggregate_values"
COMPARE_CUSTOMERS = "compare_customers"
GROUP_BY = "group_by"
# Schema Models
class SchemaColumn(BaseModel):
"""Represents a column in a database schema."""
name: str = Field(..., description="Column name")
data_type: str = Field(..., description="SQL data type (TEXT, INTEGER, REAL, DATE, etc.)")
semantic_meaning: Optional[str] = Field(None, description="Semantic concept this column represents")
semantic_type: Optional[SemanticType] = Field(None, description="Semantic type of values")
transformations: List[str] = Field(default_factory=list, description="Required transformations")
sample_values: List[Any] = Field(default_factory=list, description="Sample values from this column")
is_primary_key: bool = Field(default=False, description="Whether this is a primary key")
is_foreign_key: bool = Field(default=False, description="Whether this is a foreign key")
foreign_key_table: Optional[str] = Field(None, description="Referenced table if foreign key")
model_config = {"use_enum_values": True}
class SchemaTable(BaseModel):
"""Represents a table in a database schema."""
name: str = Field(..., description="Table name")
columns: List[SchemaColumn] = Field(..., description="List of columns in this table")
relationships: Dict[str, str] = Field(
default_factory=dict,
description="Relationships to other tables (table_name -> join_column)"
)
def get_column(self, name: str) -> Optional[SchemaColumn]:
"""Get a column by name."""
for col in self.columns:
if col.name == name:
return col
return None
class CustomerSchema(BaseModel):
"""Represents the complete schema for a customer database."""
customer_id: str = Field(..., description="Unique customer identifier (e.g., customer_a)")
customer_name: Optional[str] = Field(None, description="Optional customer display name")
tables: List[SchemaTable] = Field(..., description="Tables in this schema")
semantic_notes: Dict[str, str] = Field(
default_factory=dict,
description="Notes about semantic meanings specific to this customer"
)
last_analyzed: Optional[datetime] = Field(None, description="When schema was last analyzed")
def get_table(self, name: str) -> Optional[SchemaTable]:
"""Get a table by name."""
for table in self.tables:
if table.name == name:
return table
return None
# Semantic Concept Models
class ConceptMapping(BaseModel):
"""Maps a concept to a specific customer's schema."""
customer_id: str = Field(..., description="Customer identifier")
table_name: str = Field(..., description="Table containing this concept")
column_name: str = Field(..., description="Column representing this concept")
data_type: str = Field(..., description="SQL data type")
semantic_type: SemanticType = Field(..., description="Semantic interpretation")
transformation: Optional[str] = Field(None, description="SQL transformation needed")
join_requirements: List[str] = Field(
default_factory=list,
description="Additional tables needed for JOIN"
)
model_config = {"use_enum_values": True}
class SemanticConcept(BaseModel):
"""Represents a semantic concept that spans multiple customer schemas."""
concept_id: str = Field(..., description="Unique concept identifier")
concept_name: str = Field(..., description="Human-readable concept name")
description: str = Field(..., description="Description of this concept")
aliases: List[str] = Field(default_factory=list, description="Alternative names for this concept")
customer_mappings: Dict[str, ConceptMapping] = Field(
default_factory=dict,
description="Mappings per customer (customer_id -> mapping)"
)
def get_mapping(self, customer_id: str) -> Optional[ConceptMapping]:
"""Get mapping for a specific customer."""
return self.customer_mappings.get(customer_id)
# Query Models
class QueryFilter(BaseModel):
"""Represents a filter condition in a query."""
concept: str = Field(..., description="Semantic concept to filter on")
operator: QueryOperator = Field(..., description="Filter operator")
value: Any = Field(..., description="Filter value(s)")
semantic_note: Optional[str] = Field(None, description="Note about semantic interpretation")
model_config = {"use_enum_values": True}
class QueryAggregation(BaseModel):
"""Represents an aggregation in a query."""
function: str = Field(..., description="Aggregation function (SUM, COUNT, AVG, MIN, MAX)")
concept: str = Field(..., description="Concept to aggregate")
alias: Optional[str] = Field(None, description="Alias for result column")
class SemanticQueryPlan(BaseModel):
"""Schema-independent query representation."""
intent: QueryIntent = Field(..., description="Query intent")
filters: List[QueryFilter] = Field(default_factory=list, description="Filter conditions")
projections: List[str] = Field(default_factory=list, description="Concepts to return")
aggregations: Optional[List[QueryAggregation]] = Field(None, description="Aggregations to perform")
group_by: Optional[List[str]] = Field(None, description="Concepts to group by")
order_by: Optional[List[tuple[str, str]]] = Field(
None,
description="Ordering (concept, direction) pairs"
)
limit: Optional[int] = Field(None, description="Maximum number of results")
target_customers: Optional[List[str]] = Field(
None,
description="Specific customer databases to query (e.g., ['customer_a', 'customer_b']). None means all customers."
)
model_config = {"use_enum_values": True}
# Result Models
class QueryResult(BaseModel):
"""Result from executing a query against a customer database."""
customer_id: str = Field(..., description="Customer this result is from")
data: List[Dict[str, Any]] = Field(..., description="Query result rows")
sql_executed: str = Field(..., description="SQL that was executed")
execution_time_ms: float = Field(..., description="Execution time in milliseconds")
row_count: int = Field(..., description="Number of rows returned")
error: Optional[str] = Field(None, description="Error message if query failed")
@property
def success(self) -> bool:
"""Whether query executed successfully."""
return self.error is None
class NormalizedValue(BaseModel):
"""Represents a value with both original and normalized forms."""
original_value: Any = Field(..., description="Original value from database")
normalized_value: Any = Field(..., description="Normalized value")
original_type: str = Field(..., description="Original semantic type")
normalized_type: str = Field(..., description="Normalized semantic type")
transformation_applied: Optional[str] = Field(None, description="Transformation that was applied")
class HarmonizedRow(BaseModel):
"""A single row with harmonized/normalized values."""
customer_id: str = Field(..., description="Source customer")
data: Dict[str, Any] = Field(..., description="Harmonized field values")
metadata: Dict[str, Any] = Field(
default_factory=dict,
description="Additional metadata about normalization"
)
class HarmonizedResult(BaseModel):
"""Harmonized results from multiple customers."""
results: List[HarmonizedRow] = Field(..., description="Harmonized result rows")
total_count: int = Field(..., description="Total number of results")
customers_queried: List[str] = Field(..., description="List of customer IDs queried")
customers_succeeded: List[str] = Field(..., description="Customers with successful queries")
customers_failed: List[str] = Field(default_factory=list, description="Customers with failed queries")
errors: Dict[str, str] = Field(
default_factory=dict,
description="Error messages per customer (customer_id -> error)"
)
execution_time_ms: float = Field(..., description="Total execution time")
@property
def success_rate(self) -> float:
"""Percentage of customers that returned results successfully."""
if not self.customers_queried:
return 0.0
return len(self.customers_succeeded) / len(self.customers_queried) * 100
# Feedback and Learning Models
class QueryFeedback(BaseModel):
"""User feedback on a query result."""
query_text: str = Field(..., description="Original query text")
semantic_plan: SemanticQueryPlan = Field(..., description="Semantic query plan used")
feedback_type: str = Field(..., description="Type of feedback (incorrect, missing, good)")
feedback_text: Optional[str] = Field(None, description="User's feedback comment")
correct_result: Optional[Any] = Field(None, description="What the correct result should be")
timestamp: datetime = Field(default_factory=lambda: datetime.now(timezone.utc), description="When feedback was given")
class SchemaChange(BaseModel):
"""Detected change in a customer schema."""
customer_id: str = Field(..., description="Customer with schema change")
change_type: str = Field(..., description="Type of change (added_column, removed_column, type_change)")
table_name: str = Field(..., description="Affected table")
column_name: Optional[str] = Field(None, description="Affected column")
old_value: Optional[Any] = Field(None, description="Previous value")
new_value: Optional[Any] = Field(None, description="New value")
detected_at: datetime = Field(default_factory=lambda: datetime.now(timezone.utc), description="When change was detected")
requires_remapping: bool = Field(default=False, description="Whether concept mappings need update")
|