empathy-framework 4.6.2__py3-none-any.whl → 4.6.3__py3-none-any.whl

empathy_os/socratic/session 2.py

@@ -0,0 +1,306 @@
+"""Socratic Session Management
+
+Tracks the state of a Socratic questioning session as it progresses
+from initial goal capture through requirements refinement to generation.
+
+Copyright 2026 Smart-AI-Memory
+Licensed under Fair Source License 0.9
+"""
+
+from __future__ import annotations
+
+import uuid
+from dataclasses import dataclass, field
+from datetime import datetime
+from enum import Enum
+from typing import Any
+
+
+class SessionState(Enum):
+    """State machine for Socratic sessions."""
+
+    # Initial state - waiting for goal
+    AWAITING_GOAL = "awaiting_goal"
+
+    # Goal received, analyzing for questions
+    ANALYZING_GOAL = "analyzing_goal"
+
+    # Questions generated, waiting for answers
+    AWAITING_ANSWERS = "awaiting_answers"
+
+    # Answers received, determining if more questions needed
+    PROCESSING_ANSWERS = "processing_answers"
+
+    # Requirements sufficient, ready to generate
+    READY_TO_GENERATE = "ready_to_generate"
+
+    # Generation in progress
+    GENERATING = "generating"
+
+    # Workflow generated successfully
+    COMPLETED = "completed"
+
+    # User cancelled or session expired
+    CANCELLED = "cancelled"
+
+
+@dataclass
+class GoalAnalysis:
+    """Analysis of the user's stated goal.
+
+    Captures what we understand and what needs clarification.
+    """
+
+    # Original goal statement
+    raw_goal: str
+
+    # Extracted intent (what they want to achieve)
+    intent: str
+
+    # Domain detected (e.g., "code_review", "testing", "documentation")
+    domain: str
+
+    # Confidence in our understanding (0-1)
+    confidence: float
+
+    # Identified ambiguities that need clarification
+    ambiguities: list[str] = field(default_factory=list)
+
+    # Assumptions we're making (need validation)
+    assumptions: list[str] = field(default_factory=list)
+
+    # Constraints mentioned or implied
+    constraints: list[str] = field(default_factory=list)
+
+    # Keywords extracted for matching
+    keywords: list[str] = field(default_factory=list)
+
+    def needs_clarification(self) -> bool:
+        """Check if goal needs more clarification."""
+        return self.confidence < 0.8 or len(self.ambiguities) > 0
+
+
+@dataclass
+class RequirementSet:
+    """Accumulated requirements from Socratic questioning."""
+
+    # Core requirements (must have)
+    must_have: list[str] = field(default_factory=list)
+
+    # Nice to have requirements
+    should_have: list[str] = field(default_factory=list)
+
+    # Explicitly excluded requirements
+    must_not_have: list[str] = field(default_factory=list)
+
+    # Technical constraints (languages, frameworks, etc.)
+    technical_constraints: dict[str, Any] = field(default_factory=dict)
+
+    # Quality attributes (performance, security, etc.)
+    quality_attributes: dict[str, float] = field(default_factory=dict)
+
+    # Domain-specific requirements
+    domain_specific: dict[str, Any] = field(default_factory=dict)
+
+    # User preferences
+    preferences: dict[str, Any] = field(default_factory=dict)
+
+    def completeness_score(self) -> float:
+        """Calculate how complete the requirements are (0-1)."""
+        scores = []
+
+        # Must have at least one core requirement
+        if self.must_have:
+            scores.append(1.0)
+        else:
+            scores.append(0.0)
+
+        # Technical constraints help
+        if self.technical_constraints:
+            scores.append(min(len(self.technical_constraints) / 3, 1.0))
+        else:
+            scores.append(0.3)  # Can work without, but less optimal
+
+        # Quality attributes help prioritization
+        if self.quality_attributes:
+            scores.append(min(len(self.quality_attributes) / 2, 1.0))
+        else:
+            scores.append(0.5)  # Can use defaults
+
+        return sum(scores) / len(scores) if scores else 0.0
+
+
+@dataclass
+class SocraticSession:
+    """A Socratic questioning session.
+
+    Tracks the full state of a session from goal to generated workflow.
+    """
+
+    # Unique session identifier
+    session_id: str = field(default_factory=lambda: str(uuid.uuid4()))
+
+    # Current state in the state machine
+    state: SessionState = SessionState.AWAITING_GOAL
+
+    # When the session started
+    created_at: datetime = field(default_factory=datetime.now)
+
+    # Last activity timestamp
+    updated_at: datetime = field(default_factory=datetime.now)
+
+    # The user's original goal statement
+    goal: str = ""
+
+    # Analysis of the goal
+    goal_analysis: GoalAnalysis | None = None
+
+    # Accumulated requirements
+    requirements: RequirementSet = field(default_factory=RequirementSet)
+
+    # History of question rounds
+    question_rounds: list[dict[str, Any]] = field(default_factory=list)
+
+    # Current round number
+    current_round: int = 0
+
+    # Maximum rounds before forcing generation
+    max_rounds: int = 5
+
+    # Generated blueprint (when ready)
+    blueprint: Any = None  # WorkflowBlueprint, imported lazily
+
+    # Generated workflow (when complete)
+    workflow: Any = None  # GeneratedWorkflow, imported lazily
+
+    # Error message if something went wrong
+    error: str | None = None
+
+    # Metadata for extensibility
+    metadata: dict[str, Any] = field(default_factory=dict)
+
+    def touch(self) -> None:
+        """Update the last activity timestamp."""
+        self.updated_at = datetime.now()
+
+    def is_active(self) -> bool:
+        """Check if session is still active."""
+        return self.state not in (SessionState.COMPLETED, SessionState.CANCELLED)
+
+    def can_generate(self) -> bool:
+        """Check if we have enough information to generate."""
+        if self.goal_analysis is None:
+            return False
+
+        # Either high confidence or we've done enough rounds
+        if self.goal_analysis.confidence >= 0.8:
+            return True
+
+        if self.current_round >= self.max_rounds:
+            return True
+
+        if self.requirements.completeness_score() >= 0.7:
+            return True
+
+        return False
+
+    def add_question_round(
+        self,
+        questions: list[dict[str, Any]],
+        answers: dict[str, Any],
+    ) -> None:
+        """Record a round of questions and answers."""
+        self.question_rounds.append({
+            "round": self.current_round,
+            "questions": questions,
+            "answers": answers,
+            "timestamp": datetime.now().isoformat(),
+        })
+        self.current_round += 1
+        self.touch()
+
+    def get_context_summary(self) -> dict[str, Any]:
+        """Get a summary of accumulated context for display."""
+        return {
+            "session_id": self.session_id,
+            "state": self.state.value,
+            "goal": self.goal,
+            "rounds_completed": self.current_round,
+            "requirements_count": len(self.requirements.must_have),
+            "confidence": self.goal_analysis.confidence if self.goal_analysis else 0.0,
+            "ready_to_generate": self.can_generate(),
+        }
+
+    def to_dict(self) -> dict[str, Any]:
+        """Serialize session to dictionary."""
+        return {
+            "session_id": self.session_id,
+            "state": self.state.value,
+            "created_at": self.created_at.isoformat(),
+            "updated_at": self.updated_at.isoformat(),
+            "goal": self.goal,
+            "goal_analysis": {
+                "intent": self.goal_analysis.intent,
+                "domain": self.goal_analysis.domain,
+                "confidence": self.goal_analysis.confidence,
+                "ambiguities": self.goal_analysis.ambiguities,
+                "assumptions": self.goal_analysis.assumptions,
+            } if self.goal_analysis else None,
+            "requirements": {
+                "must_have": self.requirements.must_have,
+                "should_have": self.requirements.should_have,
+                "must_not_have": self.requirements.must_not_have,
+                "technical_constraints": self.requirements.technical_constraints,
+                "quality_attributes": self.requirements.quality_attributes,
+            },
+            "question_rounds": self.question_rounds,
+            "current_round": self.current_round,
+            "error": self.error,
+            "metadata": self.metadata,
+        }
+
+    @classmethod
+    def from_dict(cls, data: dict[str, Any]) -> SocraticSession:
+        """Deserialize session from dictionary."""
+        session = cls(
+            session_id=data.get("session_id", str(uuid.uuid4())),
+            state=SessionState(data.get("state", "awaiting_goal")),
+            goal=data.get("goal", ""),
+            current_round=data.get("current_round", 0),
+            error=data.get("error"),
+            metadata=data.get("metadata", {}),
+        )
+
+        # Parse timestamps
+        if "created_at" in data:
+            session.created_at = datetime.fromisoformat(data["created_at"])
+        if "updated_at" in data:
+            session.updated_at = datetime.fromisoformat(data["updated_at"])
+
+        # Parse goal analysis
+        if data.get("goal_analysis"):
+            ga = data["goal_analysis"]
+            session.goal_analysis = GoalAnalysis(
+                raw_goal=session.goal,
+                intent=ga.get("intent", ""),
+                domain=ga.get("domain", "general"),
+                confidence=ga.get("confidence", 0.0),
+                ambiguities=ga.get("ambiguities", []),
+                assumptions=ga.get("assumptions", []),
+            )
+
+        # Parse requirements
+        if data.get("requirements"):
+            req = data["requirements"]
+            session.requirements = RequirementSet(
+                must_have=req.get("must_have", []),
+                should_have=req.get("should_have", []),
+                must_not_have=req.get("must_not_have", []),
+                technical_constraints=req.get("technical_constraints", {}),
+                quality_attributes=req.get("quality_attributes", {}),
+            )
+
+        # Parse question rounds
+        session.question_rounds = data.get("question_rounds", [])
+
+        return session