aiecs 1.1.0__py3-none-any.whl → 1.2.1__py3-none-any.whl

aiecs/domain/community/decision_engine.py

@@ -15,7 +15,7 @@ from .models.community_models import (
     CommunityDecision, CommunityMember, AgentCommunity,
     DecisionStatus, GovernanceType
 )
-from ..core.exceptions.task_exceptions import TaskValidationError
+from .exceptions import CommunityValidationError as TaskValidationError
 
 logger = logging.getLogger(__name__)
 
@@ -334,21 +334,467 @@ class DecisionEngine:
         return await resolver_func(decision_id, community_id)
 
     async def _mediation_resolution(self, decision_id: str, community_id: str) -> Dict[str, Any]:
-        """Mediation-based conflict resolution."""
-        # TODO: Implement mediation logic
-        return {"strategy": "mediation", "status": "pending", "mediator": None}
+        """
+        Mediation-based conflict resolution.
+        
+        Process:
+        1. Select neutral mediator (high reputation, not involved in voting)
+        2. Facilitate structured discussion between opposing sides
+        3. Identify core concerns and interests
+        4. Propose compromise solutions
+        5. Build consensus on mediated outcome
+        """
+        if not self.community_manager:
+            return {"strategy": "mediation", "status": "failed", "reason": "Community manager not available"}
+        
+        decision = self.community_manager.decisions.get(decision_id)
+        community = self.community_manager.communities.get(community_id)
+        
+        if not decision or not community:
+            return {"strategy": "mediation", "status": "failed", "reason": "Decision or community not found"}
+        
+        # Step 1: Select mediator
+        mediator_id = await self._select_mediator(decision, community)
+        if not mediator_id:
+            return {"strategy": "mediation", "status": "failed", "reason": "No suitable mediator found"}
+        
+        # Step 2: Identify opposing sides
+        for_side = decision.votes_for
+        against_side = decision.votes_against
+        
+        # Step 3: Analyze core concerns (simulate by examining vote distribution)
+        concerns = {
+            "support_concerns": self._extract_concerns(for_side, community),
+            "opposition_concerns": self._extract_concerns(against_side, community)
+        }
+        
+        # Step 4: Propose compromise
+        compromise_proposal = {
+            "original_proposal": decision.title,
+            "modifications": [
+                "Address key concerns from opposition",
+                "Maintain core value from supporters",
+                "Add safeguards or conditions",
+                "Phased implementation approach"
+            ],
+            "mediator_recommendation": "Modified proposal with balanced approach"
+        }
+        
+        # Step 5: Set up for re-vote
+        result = {
+            "strategy": "mediation",
+            "status": "mediation_completed",
+            "mediator_id": mediator_id,
+            "mediator": self.community_manager.members.get(mediator_id).agent_id if mediator_id else None,
+            "concerns_identified": concerns,
+            "compromise_proposal": compromise_proposal,
+            "next_steps": "Re-vote on mediated proposal",
+            "recommended_threshold": "simple_majority"
+        }
+        
+        logger.info(f"Mediation completed for decision {decision_id} by mediator {mediator_id}")
+        return result
+    
+    async def _select_mediator(self, decision: CommunityDecision, community: AgentCommunity) -> Optional[str]:
+        """Select a neutral mediator for conflict resolution."""
+        # Find members who didn't vote or abstained, with high reputation
+        candidates = []
+        
+        all_voters = set(decision.votes_for + decision.votes_against)
+        
+        for member_id in community.members:
+            if member_id not in all_voters:
+                member = self.community_manager.members.get(member_id)
+                if member and member.reputation > 0.5:  # High reputation threshold
+                    candidates.append((member_id, member.reputation))
+        
+        # Also consider abstentions with very high reputation
+        for member_id in decision.abstentions:
+            member = self.community_manager.members.get(member_id)
+            if member and member.reputation > 0.7:
+                candidates.append((member_id, member.reputation))
+        
+        if not candidates:
+            return None
+        
+        # Select highest reputation member
+        candidates.sort(key=lambda x: x[1], reverse=True)
+        return candidates[0][0]
+    
+    def _extract_concerns(self, voter_ids: List[str], community: AgentCommunity) -> List[str]:
+        """Extract concerns from voter groups based on their roles and specializations."""
+        concerns = []
+        role_distribution = {}
+        
+        for voter_id in voter_ids:
+            member = self.community_manager.members.get(voter_id) if self.community_manager else None
+            if member:
+                role = member.community_role.value
+                role_distribution[role] = role_distribution.get(role, 0) + 1
+        
+        # Generate concerns based on roles
+        if "leader" in role_distribution:
+            concerns.append("Strategic alignment and governance impact")
+        if "specialist" in role_distribution:
+            concerns.append("Technical feasibility and implementation details")
+        if "contributor" in role_distribution:
+            concerns.append("Practical implications and workload impact")
+        
+        return concerns if concerns else ["General concerns about proposal"]
 
     async def _arbitration_resolution(self, decision_id: str, community_id: str) -> Dict[str, Any]:
-        """Arbitration-based conflict resolution."""
-        # TODO: Implement arbitration logic
-        return {"strategy": "arbitration", "status": "pending", "arbitrator": None}
+        """
+        Arbitration-based conflict resolution.
+        
+        Process:
+        1. Select authoritative arbitrator (leader or senior coordinator)
+        2. Review all arguments and evidence
+        3. Make binding decision
+        4. Provide detailed rationale
+        """
+        if not self.community_manager:
+            return {"strategy": "arbitration", "status": "failed", "reason": "Community manager not available"}
+        
+        decision = self.community_manager.decisions.get(decision_id)
+        community = self.community_manager.communities.get(community_id)
+        
+        if not decision or not community:
+            return {"strategy": "arbitration", "status": "failed", "reason": "Decision or community not found"}
+        
+        # Step 1: Select arbitrator (prefer leader, then coordinator with highest reputation)
+        arbitrator_id = await self._select_arbitrator(community)
+        if not arbitrator_id:
+            return {"strategy": "arbitration", "status": "failed", "reason": "No suitable arbitrator found"}
+        
+        # Step 2: Analyze decision context
+        votes_for = len(decision.votes_for)
+        votes_against = len(decision.votes_against)
+        total_votes = votes_for + votes_against
+        
+        # Step 3: Make arbitration decision (simulate based on vote distribution and priority)
+        # In practice, this would involve the actual arbitrator's judgment
+        support_ratio = votes_for / total_votes if total_votes > 0 else 0
+        
+        # Arbitrator considers: vote distribution, decision priority, community impact
+        arbitration_decision = "approved" if support_ratio >= 0.4 else "rejected"  # Lower threshold with rationale
+        
+        # Step 4: Provide rationale
+        rationale = self._generate_arbitration_rationale(
+            arbitration_decision, support_ratio, decision, community
+        )
+        
+        result = {
+            "strategy": "arbitration",
+            "status": "arbitration_completed",
+            "arbitrator_id": arbitrator_id,
+            "arbitrator": self.community_manager.members.get(arbitrator_id).agent_id if arbitrator_id else None,
+            "binding_decision": arbitration_decision,
+            "rationale": rationale,
+            "votes_for": votes_for,
+            "votes_against": votes_against,
+            "support_ratio": support_ratio,
+            "is_binding": True,
+            "appeal_allowed": False
+        }
+        
+        # Update decision status based on arbitration
+        if arbitration_decision == "approved":
+            decision.status = DecisionStatus.APPROVED
+        else:
+            decision.status = DecisionStatus.REJECTED
+        
+        logger.info(f"Arbitration completed for decision {decision_id}: {arbitration_decision}")
+        return result
+    
+    async def _select_arbitrator(self, community: AgentCommunity) -> Optional[str]:
+        """Select an authoritative arbitrator."""
+        # Prefer leaders first
+        if community.leaders:
+            # Select leader with highest reputation
+            best_leader = None
+            best_reputation = -1
+            
+            for leader_id in community.leaders:
+                member = self.community_manager.members.get(leader_id)
+                if member and member.reputation > best_reputation:
+                    best_reputation = member.reputation
+                    best_leader = leader_id
+            
+            if best_leader:
+                return best_leader
+        
+        # Fall back to coordinators
+        if community.coordinators:
+            best_coordinator = None
+            best_reputation = -1
+            
+            for coordinator_id in community.coordinators:
+                member = self.community_manager.members.get(coordinator_id)
+                if member and member.reputation > best_reputation:
+                    best_reputation = member.reputation
+                    best_coordinator = coordinator_id
+            
+            if best_coordinator:
+                return best_coordinator
+        
+        return None
+    
+    def _generate_arbitration_rationale(
+        self, 
+        decision: str, 
+        support_ratio: float, 
+        proposal: CommunityDecision,
+        community: AgentCommunity
+    ) -> str:
+        """Generate detailed rationale for arbitration decision."""
+        rationale_parts = []
+        
+        rationale_parts.append(f"After careful review of the proposal '{proposal.title}', ")
+        rationale_parts.append(f"with {support_ratio:.1%} support from voting members, ")
+        
+        if decision == "approved":
+            rationale_parts.append("this arbitration approves the proposal. ")
+            rationale_parts.append("The decision aligns with community interests and demonstrates sufficient support. ")
+            if support_ratio < 0.5:
+                rationale_parts.append("While not achieving majority, the strategic importance warrants approval. ")
+        else:
+            rationale_parts.append("this arbitration rejects the proposal. ")
+            rationale_parts.append("The concerns raised outweigh the benefits, and insufficient consensus exists. ")
+        
+        rationale_parts.append(f"Priority level: {proposal.priority}. ")
+        rationale_parts.append("This decision is binding and final.")
+        
+        return "".join(rationale_parts)
 
     async def _compromise_resolution(self, decision_id: str, community_id: str) -> Dict[str, Any]:
-        """Compromise-based conflict resolution."""
-        # TODO: Implement compromise logic
-        return {"strategy": "compromise", "status": "pending", "compromise_proposal": None}
+        """
+        Compromise-based conflict resolution.
+        
+        Process:
+        1. Analyze opposing positions
+        2. Identify negotiable vs non-negotiable elements
+        3. Generate compromise alternatives
+        4. Create hybrid solution
+        5. Test acceptance with stakeholders
+        """
+        if not self.community_manager:
+            return {"strategy": "compromise", "status": "failed", "reason": "Community manager not available"}
+        
+        decision = self.community_manager.decisions.get(decision_id)
+        community = self.community_manager.communities.get(community_id)
+        
+        if not decision or not community:
+            return {"strategy": "compromise", "status": "failed", "reason": "Decision or community not found"}
+        
+        votes_for = len(decision.votes_for)
+        votes_against = len(decision.votes_against)
+        total_votes = votes_for + votes_against
+        
+        if total_votes == 0:
+            return {"strategy": "compromise", "status": "failed", "reason": "No votes to analyze"}
+        
+        # Step 1: Analyze positions
+        support_ratio = votes_for / total_votes
+        opposition_ratio = votes_against / total_votes
+        
+        position_analysis = {
+            "support_strength": support_ratio,
+            "opposition_strength": opposition_ratio,
+            "balance": "balanced" if 0.4 <= support_ratio <= 0.6 else "polarized"
+        }
+        
+        # Step 2: Identify elements
+        elements = {
+            "core_proposal": decision.title,
+            "negotiable_elements": [
+                "Implementation timeline",
+                "Resource allocation",
+                "Scope limitations",
+                "Review checkpoints"
+            ],
+            "non_negotiable_elements": [
+                "Core objectives",
+                "Safety requirements",
+                "Community values alignment"
+            ]
+        }
+        
+        # Step 3 & 4: Generate compromise alternatives
+        compromise_options = []
+        
+        # Option 1: Phased approach
+        compromise_options.append({
+            "option": "phased_implementation",
+            "description": "Implement in phases with review points",
+            "modifications": [
+                "Start with pilot/trial phase",
+                "Review after initial phase",
+                "Full rollout conditional on pilot success"
+            ],
+            "acceptance_probability": 0.75
+        })
+        
+        # Option 2: Conditional approval
+        compromise_options.append({
+            "option": "conditional_approval",
+            "description": "Approve with conditions addressing concerns",
+            "modifications": [
+                "Add oversight committee",
+                "Include opposition representatives",
+                "Establish success metrics and review schedule"
+            ],
+            "acceptance_probability": 0.70
+        })
+        
+        # Option 3: Scaled-down version
+        compromise_options.append({
+            "option": "scaled_down",
+            "description": "Reduced scope addressing primary concerns",
+            "modifications": [
+                "Limit scope to less contentious areas",
+                "Reduce resource commitment",
+                "Extend timeline for gradual adoption"
+            ],
+            "acceptance_probability": 0.65
+        })
+        
+        # Select best compromise based on vote distribution
+        recommended_option = compromise_options[0] if support_ratio > 0.45 else compromise_options[2]
+        
+        result = {
+            "strategy": "compromise",
+            "status": "compromise_proposed",
+            "position_analysis": position_analysis,
+            "elements": elements,
+            "compromise_options": compromise_options,
+            "recommended_option": recommended_option,
+            "next_steps": "Review compromise options and vote on preferred alternative",
+            "requires_revote": True,
+            "expected_consensus": recommended_option["acceptance_probability"]
+        }
+        
+        logger.info(f"Compromise resolution generated for decision {decision_id}")
+        return result
 
     async def _escalation_resolution(self, decision_id: str, community_id: str) -> Dict[str, Any]:
-        """Escalation-based conflict resolution."""
-        # TODO: Implement escalation logic
-        return {"strategy": "escalation", "status": "pending", "escalation_level": 1}
+        """
+        Escalation-based conflict resolution.
+        
+        Process:
+        1. Determine current escalation level
+        2. Escalate to higher authority/broader group
+        3. Apply progressively stronger resolution mechanisms
+        4. Track escalation path and outcomes
+        
+        Escalation Levels:
+        - Level 1: Community-wide discussion and re-vote
+        - Level 2: Coordinator council review
+        - Level 3: Leader decision
+        - Level 4: External arbitration or parent community
+        """
+        if not self.community_manager:
+            return {"strategy": "escalation", "status": "failed", "reason": "Community manager not available"}
+        
+        decision = self.community_manager.decisions.get(decision_id)
+        community = self.community_manager.communities.get(community_id)
+        
+        if not decision or not community:
+            return {"strategy": "escalation", "status": "failed", "reason": "Decision or community not found"}
+        
+        # Determine current escalation level from metadata
+        current_level = decision.metadata.get("escalation_level", 0)
+        next_level = current_level + 1
+        
+        # Define escalation path
+        escalation_path = {
+            1: {
+                "level": 1,
+                "name": "Community Discussion",
+                "authority": "All active members",
+                "process": "Open discussion with extended voting period",
+                "threshold": "supermajority (67%)",
+                "timeline": "7 days"
+            },
+            2: {
+                "level": 2,
+                "name": "Coordinator Council",
+                "authority": "Community coordinators",
+                "process": "Coordinator review and recommendation",
+                "threshold": "coordinator consensus",
+                "timeline": "3 days"
+            },
+            3: {
+                "level": 3,
+                "name": "Leadership Decision",
+                "authority": "Community leaders",
+                "process": "Leadership panel makes binding decision",
+                "threshold": "leader majority or single leader decision",
+                "timeline": "1 day"
+            },
+            4: {
+                "level": 4,
+                "name": "External Review",
+                "authority": "External arbitrator or parent community",
+                "process": "Independent third-party review",
+                "threshold": "external arbitrator decision",
+                "timeline": "As needed"
+            }
+        }
+        
+        if next_level > 4:
+            return {
+                "strategy": "escalation",
+                "status": "max_escalation_reached",
+                "message": "Maximum escalation level reached. Decision must be resolved or abandoned.",
+                "recommendation": "Consider abandoning or significantly revising the proposal"
+            }
+        
+        current_escalation = escalation_path[next_level]
+        
+        # Determine escalation authority
+        authority_members = []
+        if next_level == 1:
+            authority_members = community.members
+        elif next_level == 2:
+            authority_members = community.coordinators
+        elif next_level == 3:
+            authority_members = community.leaders
+        elif next_level == 4:
+            authority_members = []  # External
+        
+        # Update decision metadata
+        decision.metadata["escalation_level"] = next_level
+        decision.metadata["escalation_timestamp"] = datetime.utcnow().isoformat()
+        decision.metadata["escalation_history"] = decision.metadata.get("escalation_history", [])
+        decision.metadata["escalation_history"].append({
+            "from_level": current_level,
+            "to_level": next_level,
+            "timestamp": datetime.utcnow().isoformat(),
+            "reason": "Unresolved conflict"
+        })
+        
+        result = {
+            "strategy": "escalation",
+            "status": "escalated",
+            "previous_level": current_level,
+            "current_level": next_level,
+            "escalation_details": current_escalation,
+            "authority_members": authority_members,
+            "authority_count": len(authority_members),
+            "escalation_history": decision.metadata["escalation_history"],
+            "next_steps": f"Proceed with {current_escalation['name']} process",
+            "required_action": current_escalation["process"],
+            "decision_threshold": current_escalation["threshold"],
+            "timeline": current_escalation["timeline"]
+        }
+        
+        # Reset voting for re-evaluation at new level
+        if next_level < 4:
+            decision.votes_for = []
+            decision.votes_against = []
+            decision.abstentions = []
+            decision.status = DecisionStatus.PROPOSED
+            decision.voting_ends_at = datetime.utcnow() + timedelta(days=int(current_escalation["timeline"].split()[0]))
+        
+        logger.info(f"Decision {decision_id} escalated to level {next_level}: {current_escalation['name']}")
+        return result