csv_data.py

"""
CSV data models for ESG data upload and processing.
"""

from typing import List, Optional, Dict, Any
from pydantic import BaseModel, Field, validator
from datetime import datetime
from enum import Enum


class CSVValidationStatus(str, Enum):
    """CSV validation status."""
    VALID = "valid"
    INVALID = "invalid"
    PARTIAL = "partial"
    PROCESSED = "processed"


class CSVColumnMapping(BaseModel):
    """CSV column mapping configuration."""
    csv_column: str
    esg_field: str
    data_type: str  # numeric, percentage, boolean, text
    required: bool = False
    validation_rules: Optional[Dict[str, Any]] = None


class CSVValidationError(BaseModel):
    """CSV validation error model."""
    row: int
    column: str
    value: Any
    error_type: str
    message: str
    suggested_fix: Optional[str] = None


class CSVProcessingResult(BaseModel):
    """CSV processing result model."""
    status: CSVValidationStatus
    total_rows: int
    valid_rows: int
    invalid_rows: int
    errors: List[CSVValidationError]
    warnings: List[str]
    processed_data: List[Dict[str, Any]]
    llm_suggestions: List[Dict[str, Any]]
    esg_score: Optional[float] = None


class CSVUploadRequest(BaseModel):
    """CSV upload request model."""
    use_llm_for_missing: bool = True
    column_mapping: Optional[Dict[str, str]] = None
    industry: str = "retail"
    company_name: Optional[str] = None


class CSVUploadResponse(BaseModel):
    """CSV upload response model."""
    upload_id: str
    filename: str
    processing_result: CSVProcessingResult
    download_url: Optional[str] = None
    created_at: datetime
    
    class Config:
        json_encoders = {
            datetime: lambda v: v.isoformat()
        }


# Default CSV column mappings for ESG data
DEFAULT_CSV_MAPPINGS = [
    CSVColumnMapping(
        csv_column="energy_consumption",
        esg_field="annual_energy_consumption",
        data_type="numeric",
        required=False,
        validation_rules={"min": 0, "max": 1000000}
    ),
    CSVColumnMapping(
        csv_column="co2_emissions",
        esg_field="co2_emissions",
        data_type="numeric",
        required=False,
        validation_rules={"min": 0, "max": 10000}
    ),
    CSVColumnMapping(
        csv_column="water_usage",
        esg_field="water_usage",
        data_type="numeric",
        required=False,
        validation_rules={"min": 0}
    ),
    CSVColumnMapping(
        csv_column="waste_generated",
        esg_field="waste_generated",
        data_type="numeric",
        required=False,
        validation_rules={"min": 0}
    ),
    CSVColumnMapping(
        csv_column="recycling_rate",
        esg_field="recycling_rate",
        data_type="percentage",
        required=False,
        validation_rules={"min": 0, "max": 100}
    ),
    CSVColumnMapping(
        csv_column="renewable_energy_percentage",
        esg_field="renewable_energy_percentage",
        data_type="percentage",
        required=False,
        validation_rules={"min": 0, "max": 100}
    ),
    CSVColumnMapping(
        csv_column="packaging_recyclability",
        esg_field="packaging_recyclability",
        data_type="percentage",
        required=False,
        validation_rules={"min": 0, "max": 100}
    ),
    CSVColumnMapping(
        csv_column="employee_count",
        esg_field="employee_count",
        data_type="numeric",
        required=False,
        validation_rules={"min": 1, "max": 10000}
    ),
    CSVColumnMapping(
        csv_column="diversity_percentage",
        esg_field="diversity_percentage",
        data_type="percentage",
        required=False,
        validation_rules={"min": 0, "max": 100}
    ),
    CSVColumnMapping(
        csv_column="female_leadership_percentage",
        esg_field="female_leadership_percentage",
        data_type="percentage",
        required=False,
        validation_rules={"min": 0, "max": 100}
    ),
    CSVColumnMapping(
        csv_column="training_hours_per_employee",
        esg_field="training_hours_per_employee",
        data_type="numeric",
        required=False,
        validation_rules={"min": 0, "max": 200}
    ),
    CSVColumnMapping(
        csv_column="employee_satisfaction_score",
        esg_field="employee_satisfaction_score",
        data_type="numeric",
        required=False,
        validation_rules={"min": 0, "max": 10}
    ),
    CSVColumnMapping(
        csv_column="community_investment",
        esg_field="community_investment",
        data_type="numeric",
        required=False,
        validation_rules={"min": 0}
    ),
    CSVColumnMapping(
        csv_column="board_independence",
        esg_field="board_independence",
        data_type="percentage",
        required=False,
        validation_rules={"min": 0, "max": 100}
    ),
    CSVColumnMapping(
        csv_column="ethics_training_completion",
        esg_field="ethics_training_completion",
        data_type="percentage",
        required=False,
        validation_rules={"min": 0, "max": 100}
    ),
    CSVColumnMapping(
        csv_column="data_privacy_compliance",
        esg_field="data_privacy_compliance",
        data_type="boolean",
        required=False
    ),
    CSVColumnMapping(
        csv_column="supplier_code_of_conduct",
        esg_field="supplier_code_of_conduct",
        data_type="boolean",
        required=False
    ),
    CSVColumnMapping(
        csv_column="transparency_reporting",
        esg_field="transparency_reporting",
        data_type="boolean",
        required=False
    )
]


class ESGAnswer(BaseModel):
    """ESG answer model for questionnaire responses."""
    question_id: str
    value: Optional[Any] = None
    is_llm_suggested: bool = False
    source: str = "user_input"
    confidence: Optional[float] = None
    
    class Config:
        json_encoders = {
            datetime: lambda v: v.isoformat()
        }


class ESGCategory(str, Enum):
    """ESG category enumeration."""
    ENVIRONMENTAL = "environmental"
    SOCIAL = "social"
    GOVERNANCE = "governance"


class ESGQuestionType(str, Enum):
    """ESG question type enumeration."""
    NUMERIC = "numeric"
    PERCENTAGE = "percentage"
    BOOLEAN = "boolean"
    TEXT = "text"


class ESGQuestion(BaseModel):
    """ESG question model."""
    id: str
    question: str
    category: ESGCategory
    question_type: ESGQuestionType
    weight: float = 1.0
    industry_default: Optional[float] = None
    description: Optional[str] = None
    unit: Optional[str] = None


# Default ESG questions for retail SMBs
DEFAULT_ESG_QUESTIONS = [
    ESGQuestion(
        id="energy_consumption",
        question="What is your annual energy consumption (kWh)?",
        category=ESGCategory.ENVIRONMENTAL,
        question_type=ESGQuestionType.NUMERIC,
        weight=1.0,
        industry_default=45000,
        description="Total energy consumption across all business operations",
        unit="kWh"
    ),
    ESGQuestion(
        id="co2_emissions",
        question="What are your annual CO2 emissions (metric tons)?",
        category=ESGCategory.ENVIRONMENTAL,
        question_type=ESGQuestionType.NUMERIC,
        weight=1.2,
        industry_default=8.5,
        description="Direct and indirect CO2 emissions",
        unit="metric tons"
    ),
    ESGQuestion(
        id="packaging_recyclability",
        question="What percentage of your packaging is recyclable?",
        category=ESGCategory.ENVIRONMENTAL,
        question_type=ESGQuestionType.PERCENTAGE,
        weight=0.8,
        industry_default=70,
        description="Percentage of packaging materials that can be recycled"
    ),
    ESGQuestion(
        id="diversity_percentage",
        question="What percentage of your workforce represents diverse backgrounds?",
        category=ESGCategory.SOCIAL,
        question_type=ESGQuestionType.PERCENTAGE,
        weight=1.0,
        industry_default=35,
        description="Diversity in terms of gender, ethnicity, and other factors"
    ),
    ESGQuestion(
        id="female_leadership",
        question="What percentage of leadership positions are held by women?",
        category=ESGCategory.SOCIAL,
        question_type=ESGQuestionType.PERCENTAGE,
        weight=0.9,
        industry_default=30,
        description="Female representation in management and leadership roles"
    ),
    ESGQuestion(
        id="employee_satisfaction",
        question="What is your employee satisfaction score (1-10)?",
        category=ESGCategory.SOCIAL,
        question_type=ESGQuestionType.NUMERIC,
        weight=1.1,
        industry_default=7.8,
        description="Employee satisfaction based on surveys or feedback",
        unit="score (1-10)"
    ),
    ESGQuestion(
        id="data_privacy_compliance",
        question="Do you have data privacy compliance measures in place?",
        category=ESGCategory.GOVERNANCE,
        question_type=ESGQuestionType.BOOLEAN,
        weight=1.0,
        description="GDPR, CCPA or other data privacy compliance"
    ),
    ESGQuestion(
        id="ethics_training",
        question="What percentage of employees completed ethics training?",
        category=ESGCategory.GOVERNANCE,
        question_type=ESGQuestionType.PERCENTAGE,
        weight=0.8,
        industry_default=85,
        description="Percentage of staff who completed ethics and compliance training"
    ),
    ESGQuestion(
        id="supplier_code",
        question="Do you have a supplier code of conduct?",
        category=ESGCategory.GOVERNANCE,
        question_type=ESGQuestionType.BOOLEAN,
        weight=0.7,
        description="Written code of conduct for suppliers and partners"
    ),
    ESGQuestion(
        id="transparency_reporting",
        question="Do you publish sustainability/ESG reports?",
        category=ESGCategory.GOVERNANCE,
        question_type=ESGQuestionType.BOOLEAN,
        weight=0.6,
        description="Regular publication of ESG performance reports"
    )
]


class CSVTemplate(BaseModel):
    """CSV template model for download."""
    filename: str
    headers: List[str]
    sample_data: List[Dict[str, Any]]
    description: str
    mappings: List[CSVColumnMapping]