Initial commit

email_dlp/models.py

@@ -0,0 +1,52 @@
+"""Pydantic output schema for DLP analysis results."""
+
+from enum import Enum
+from pydantic import BaseModel, Field
+
+
+class RiskLevel(str, Enum):
+    CRITICAL = "CRITICAL"
+    HIGH = "HIGH"
+    MEDIUM = "MEDIUM"
+    LOW = "LOW"
+
+
+class ViolationType(str, Enum):
+    PII = "PII"
+    FINANCIAL_DATA = "FINANCIAL_DATA"
+    SOURCE_CODE = "SOURCE_CODE"
+    REGULATORY_DOCUMENT = "REGULATORY_DOCUMENT"
+    LEGAL_CONTRACT = "LEGAL_CONTRACT"
+    PAYROLL_RECORD = "PAYROLL_RECORD"
+    CUSTOMER_LIST = "CUSTOMER_LIST"
+    INTERNAL_MEMO = "INTERNAL_MEMO"
+    NONE = "NONE"
+
+
+class ActionClass(str, Enum):
+    PASS_ = "PASS"
+    ALERT = "ALERT"
+    BLOCK = "BLOCK"
+
+
+class AttachmentResult(BaseModel):
+    filename: str
+    content_type: str
+    extracted_text_chars: int = 0
+    conversion_status: str = "ok"  # "ok" | "failed" | "skipped"
+
+
+class DLPResult(BaseModel):
+    email_file: str
+    subject: str
+    sender: str
+    recipient: str
+    date: str
+    risk_level: RiskLevel
+    risk_score: int = Field(ge=0, le=100)
+    violation_types: list[ViolationType]
+    action: ActionClass
+    summary: str
+    evidence: list[str]
+    attachments: list[AttachmentResult]
+    processing_errors: list[str] = Field(default_factory=list)