devsquad 3.6.0__py3-none-any.whl

scripts/collaboration/models.py

@@ -0,0 +1,537 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+协作系统数据模型
+
+定义 Coordinator + Scratchpad + Worker 协作模式的所有核心数据结构。
+
+设计决策（门禁条件解决）：
+- 门禁1: Scratchpad 并发写入 → 采用"时间戳排序+版本号"方案，简单可靠
+- 门禁2: Consensus 升级 → 采用"权重投票+否决权+升级到人工"机制
+- 门禁3: 存储选型 → Phase 1 采用内存+JSON文件持久化，无外部依赖
+- 异常场景: 每个组件都有 timeout/retry/cancel 支持
+"""
+
+from enum import Enum
+from dataclasses import dataclass, field
+from typing import Dict, List, Optional, Any, Set
+from datetime import datetime
+import uuid
+
+
+class EntryType(Enum):
+    FINDING = "finding"
+    DECISION = "decision"
+    CONFLICT = "conflict"
+    QUESTION = "question"
+    SUGGESTION = "suggestion"
+    WARNING = "warning"
+
+
+class EntryStatus(Enum):
+    ACTIVE = "active"
+    RESOLVED = "resolved"
+    SUPERSEDED = "superseded"
+    REJECTED = "rejected"
+
+
+class ReferenceType(Enum):
+    SUPPORTS = "supports"
+    CONTRADICTS = "contradicts"
+    EXTENDS = "extends"
+    CLARIFIES = "clarifies"
+
+
+@dataclass
+class Reference:
+    reference_type: ReferenceType
+    target_entry_id: str
+    summary: str = ""
+
+
+@dataclass
+class ScratchpadEntry:
+    entry_id: str = field(default_factory=lambda: f"entry-{uuid.uuid4().hex[:12]}")
+    worker_id: str = ""
+    role_id: str = ""
+    timestamp: datetime = field(default_factory=datetime.now)
+    entry_type: EntryType = EntryType.FINDING
+    content: str = ""
+    confidence: float = 0.5
+    tags: List[str] = field(default_factory=list)
+    references: List[Reference] = field(default_factory=list)
+    status: EntryStatus = EntryStatus.ACTIVE
+    version: int = 1
+
+    def to_dict(self) -> Dict[str, Any]:
+        return {
+            "entry_id": self.entry_id,
+            "worker_id": self.worker_id,
+            "role_id": self.role_id,
+            "timestamp": self.timestamp.isoformat(),
+            "entry_type": self.entry_type.value,
+            "content": self.content,
+            "confidence": self.confidence,
+            "tags": self.tags,
+            "references": [
+                {"type": r.reference_type.value, "target": r.target_entry_id, "summary": r.summary}
+                for r in self.references
+            ],
+            "status": self.status.value,
+            "version": self.version,
+        }
+
+    @classmethod
+    def from_dict(cls, data: Dict[str, Any]) -> "ScratchpadEntry":
+        refs = [
+            Reference(
+                reference_type=ReferenceType(r["type"]),
+                target_entry_id=r["target"],
+                summary=r.get("summary", ""),
+            )
+            for r in data.get("references", [])
+        ]
+        return cls(
+            entry_id=data.get("entry_id", f"entry-{uuid.uuid4().hex[:12]}"),
+            worker_id=data.get("worker_id", ""),
+            role_id=data.get("role_id", ""),
+            timestamp=datetime.fromisoformat(data["timestamp"]) if "timestamp" in data else datetime.now(),
+            entry_type=EntryType(data.get("entry_type", "finding")),
+            content=data.get("content", ""),
+            confidence=data.get("confidence", 0.5),
+            tags=data.get("tags", []),
+            references=refs,
+            status=EntryStatus(data.get("status", "active")),
+            version=data.get("version", 1),
+        )
+
+
+@dataclass
+class TaskNotification:
+    from_worker: str
+    to_workers: List[str]
+    notification_type: str
+    priority: str = "medium"
+    timestamp: datetime = field(default_factory=datetime.now)
+    summary: str = ""
+    details: str = ""
+    references: List[str] = field(default_factory=list)
+    action_required: str = ""
+
+    def to_xml(self) -> str:
+        refs_xml = "".join(f"<ref>{r}</ref>" for r in self.references)
+        to_xml = ",".join(self.to_workers)
+        return (
+            f'<task-notification\n'
+            f'    from-worker="{self.from_worker}"\n'
+            f'    to-workers="{to_xml}"\n'
+            f'    type="{self.notification_type}"\n'
+            f'    priority="{self.priority}"\n'
+            f'    timestamp="{self.timestamp.isoformat()}">\n'
+            f'    <summary>{self.summary}</summary>\n'
+            f'    <details>{self.details}</details>\n'
+            f'    <references>{refs_xml}</references>\n'
+            f'    <action-required>{self.action_required}</action-required>\n'
+            f'</task-notification>'
+        )
+
+
+@dataclass
+class TaskDefinition:
+    task_id: str = field(default_factory=lambda: f"task-{uuid.uuid4().hex[:8]}")
+    description: str = ""
+    role_id: str = ""
+    role_prompt: str = ""
+    stage_id: Optional[str] = None
+    input_data: Dict[str, Any] = field(default_factory=dict)
+    dependencies: List[str] = field(default_factory=list)
+    is_read_only: bool = True
+    timeout_seconds: int = 300
+    retry_count: int = 3
+
+
+@dataclass
+class WorkerResult:
+    worker_id: str
+    task_id: str
+    success: bool
+    output: Any = None
+    error: Optional[str] = None
+    scratchpad_entries_written: int = 0
+    notifications_sent: int = 0
+    duration_seconds: float = 0.0
+
+
+@dataclass
+class Vote:
+    voter_id: str
+    voter_role: str
+    decision: bool
+    reason: str = ""
+    weight: float = 1.0
+    confidence: float = 0.7
+    timestamp: datetime = field(default_factory=datetime.now)
+
+
+@dataclass
+class DecisionProposal:
+    proposal_id: str = field(default_factory=lambda: f"prop-{uuid.uuid4().hex[:8]}")
+    topic: str = ""
+    proposer_id: str = ""
+    proposal_content: str = ""
+    options: List[str] = field(default_factory=list)
+    deadline: Optional[datetime] = None
+    votes: List[Vote] = field(default_factory=list)
+    status: str = "open"
+
+
+class DecisionOutcome(Enum):
+    APPROVED = "approved"
+    REJECTED = "rejected"
+    SPLIT = "split"
+    ESCALATED = "escalated"
+    TIMEOUT = "timeout"
+
+
+@dataclass
+class ConsensusRecord:
+    record_id: str = field(default_factory=lambda: f"consensus-{uuid.uuid4().hex[:8]}")
+    topic: str = ""
+    outcome: DecisionOutcome = DecisionOutcome.APPROVED
+    final_decision: str = ""
+    votes_for: int = 0
+    votes_against: int = 0
+    votes_abstain: int = 0
+    total_weight_for: float = 0.0
+    total_weight_against: float = 0.0
+    participants: List[str] = field(default_factory=list)
+    escalation_reason: Optional[str] = None
+    timestamp: datetime = field(default_factory=datetime.now)
+
+
+@dataclass
+class ExecutionPlan:
+    plan_id: str = field(default_factory=lambda: f"plan-{uuid.uuid4().hex[:8]}")
+    batches: List[Any] = field(default_factory=list)
+    total_tasks: int = 0
+    estimated_parallelism: float = 0.0
+
+
+class BatchMode(Enum):
+    PARALLEL = "parallel"
+    SERIAL = "serial"
+
+
+@dataclass
+class TaskBatch:
+    batch_id: str = field(default_factory=lambda: f"batch-{uuid.uuid4().hex[:8]}")
+    mode: BatchMode = BatchMode.PARALLEL
+    tasks: List[TaskDefinition] = field(default_factory=list)
+    max_concurrency: int = 5
+    dependencies: List[str] = field(default_factory=list)
+    timeout_seconds: int = 600
+
+
+@dataclass
+class ScheduleResult:
+    success: bool = False
+    total_tasks: int = 0
+    completed_tasks: int = 0
+    failed_tasks: int = 0
+    results: List[WorkerResult] = field(default_factory=list)
+    duration_seconds: float = 0.0
+    errors: List[str] = field(default_factory=list)
+
+
+ROLE_WEIGHTS = {
+    "architect": 1.5,
+    "product-manager": 1.2,
+    "security": 1.1,
+    "tester": 1.0,
+    "solo-coder": 1.0,
+    "devops": 1.0,
+    "ui-designer": 0.9,
+}
+
+
+@dataclass
+class RoleDefinition:
+    role_id: str
+    name: str
+    aliases: List[str]
+    prompt: str
+    keywords: List[str]
+    weight: float
+    description: str
+    status: str = "core"
+
+
+ROLE_REGISTRY: Dict[str, RoleDefinition] = {
+    "architect": RoleDefinition(
+        role_id="architect",
+        name="架构师",
+        aliases=["arch"],
+        prompt="你是系统架构师。负责：\n1. 系统架构设计（分层、模块化、接口定义）\n2. 技术选型和评估\n3. 性能架构设计（缓存架构、CDN策略、分库分表方案）\n4. 安全架构设计（认证授权方案、加密策略、安全边界）\n5. 数据架构设计（数据模型、数据仓库架构、ETL架构）\n6. 输出：架构文档、技术方案、模块设计",
+        keywords=["架构", "设计", "选型", "性能", "模块", "接口", "微服务", "数据架构",
+                   "architecture", "design", "microservice", "module", "interface", "performance", "scalability", "system"],
+        weight=1.5,
+        description="System design, tech stack, API design, performance/security/data architecture",
+        status="core",
+    ),
+    "product-manager": RoleDefinition(
+        role_id="product-manager",
+        name="产品经理",
+        aliases=["pm"],
+        prompt="你是产品经理。负责：\n1. 需求分析和PRD编写\n2. 用户故事和验收标准\n3. 竞品分析\n4. 输出：需求文档、用户故事、功能规格",
+        keywords=["需求", "PRD", "用户故事", "竞品", "验收", "体验", "功能",
+                   "requirement", "prd", "user story", "acceptance", "feature", "product", "specification"],
+        weight=1.2,
+        description="Requirements analysis, user stories, acceptance criteria",
+        status="core",
+    ),
+    "tester": RoleDefinition(
+        role_id="tester",
+        name="测试专家",
+        aliases=["test", "qa"],
+        prompt="你是测试专家。负责：\n1. 测试策略和用例设计\n2. 自动化测试方案\n3. 质量评估和缺陷追踪\n4. 输出：测试计划、测试用例、质量报告",
+        keywords=["测试", "质量", "验收", "自动化", "性能测试", "缺陷", "门禁",
+                   "test", "quality", "qa", "automated", "coverage", "bug", "validation"],
+        weight=1.0,
+        description="Test strategy, quality assurance, edge cases",
+        status="core",
+    ),
+    "solo-coder": RoleDefinition(
+        role_id="solo-coder",
+        name="独立开发者",
+        aliases=["coder", "dev"],
+        prompt="你是全栈开发者。负责：\n1. 功能实现和代码编写\n2. 代码审查与质量把关（风格一致性、最佳实践、设计模式合规）\n3. 性能优化实现（算法优化、内存优化、并发优化、SQL调优）\n4. 代码重构和优化\n5. Bug修复\n6. 数据迁移实现\n7. 输出：源代码、测试、技术文档",
+        keywords=["实现", "开发", "代码", "修复", "优化", "重构", "审查", "最佳实践",
+                   "implement", "develop", "code", "fix", "optimize", "refactor", "review", "debug"],
+        weight=1.0,
+        description="Implementation, code review, performance optimization, refactoring",
+        status="core",
+    ),
+    "ui-designer": RoleDefinition(
+        role_id="ui-designer",
+        name="UI设计师",
+        aliases=["ui"],
+        prompt="你是UI/UX设计师。负责：\n1. 界面设计和交互原型\n2. 设计系统和组件规范\n3. 视觉稿和设计交付\n4. 输出：设计稿、原型、设计规范",
+        keywords=["UI", "界面", "前端", "视觉", "交互", "原型", "设计",
+                   "ui", "interface", "frontend", "visual", "interaction", "prototype", "ux", "accessibility"],
+        weight=0.9,
+        description="UX design, interaction logic, accessibility",
+        status="core",
+    ),
+    "devops": RoleDefinition(
+        role_id="devops",
+        name="DevOps工程师",
+        aliases=["infra"],
+        prompt="你是DevOps工程师。负责：\n1. CI/CD流水线设计与实现（GitHub Actions、GitLab CI、Jenkins）\n2. 容器化与编排（Docker、Kubernetes、Docker Compose）\n3. 基础设施即代码（Terraform、Pulumi、CloudFormation）\n4. 监控告警体系搭建（Prometheus、Grafana、ELK、Sentry）\n5. 部署策略设计（蓝绿部署、金丝雀发布、滚动更新）\n6. 环境管理（开发/测试/预生产/生产环境配置与隔离）\n7. 输出：CI/CD配置、Dockerfile、K8s Manifests、监控配置、部署文档",
+        keywords=["CI/CD", "部署", "监控", "运维", "Docker", "Kubernetes", "基础设施", "容器",
+                   "deploy", "monitor", "infrastructure", "container", "pipeline", "devops", "ci/cd", "cloud"],
+        weight=1.0,
+        description="CI/CD pipeline, containerization, monitoring, infrastructure",
+        status="core",
+    ),
+    "security": RoleDefinition(
+        role_id="security",
+        name="安全专家",
+        aliases=["sec"],
+        prompt="你是安全专家。负责：\n1. 威胁建模（STRIDE、DREAD攻击树分析）\n2. 漏洞审计（OWASP Top 10、CWE常见弱点枚举）\n3. 认证与授权安全审查（OAuth2、JWT、RBAC/ABAC）\n4. 数据安全评估（加密方案、密钥管理、数据脱敏）\n5. 依赖安全扫描与供应链安全（Snyk、Dependabot、SBOM）\n6. 合规性检查（GDPR、SOC2、HIPAA、PCI-DSS）\n7. 安全编码规范与最佳实践\n8. 输出：威胁模型、漏洞报告、安全建议、合规评估",
+        keywords=["安全", "漏洞", "审计", "威胁", "加密", "认证", "授权", "OWASP",
+                   "security", "vulnerability", "audit", "threat", "encrypt", "auth", "compliance", "owasp"],
+        weight=1.1,
+        description="Threat modeling, vulnerability audit, compliance, security review",
+        status="core",
+    ),
+}
+
+
+def _build_role_aliases() -> Dict[str, str]:
+    aliases = {}
+    for rid, rdef in ROLE_REGISTRY.items():
+        for alias in rdef.aliases:
+            aliases[alias] = rid
+    return aliases
+
+
+ROLE_ALIASES: Dict[str, str] = _build_role_aliases()
+
+
+def resolve_role_id(role_id: str) -> str:
+    if role_id in ROLE_REGISTRY:
+        return role_id
+    return ROLE_ALIASES.get(role_id, role_id)
+
+
+def get_core_roles() -> Dict[str, RoleDefinition]:
+    return {rid: rdef for rid, rdef in ROLE_REGISTRY.items() if rdef.status == "core"}
+
+
+def get_planned_roles() -> Dict[str, RoleDefinition]:
+    return {rid: rdef for rid, rdef in ROLE_REGISTRY.items() if rdef.status == "planned"}
+
+
+def get_all_role_ids() -> List[str]:
+    return list(ROLE_REGISTRY.keys())
+
+
+def get_cli_role_list() -> List[str]:
+    result = []
+    for rid, rdef in ROLE_REGISTRY.items():
+        result.append(rdef.aliases[0] if rdef.aliases else rid)
+    return result
+
+CONSENSUS_THRESHOLDS = {
+    "simple_majority": 0.51,
+    "super_majority": 0.67,
+    "unanimous": 1.0,
+}
+
+
+# ============================================================
+# V3.6.0 "Anchor & Retrospect" Data Models
+# ============================================================
+
+class GoalItemStatus(Enum):
+    PENDING = "pending"
+    PARTIALLY_COVERED = "partially_covered"
+    FULLY_COVERED = "fully_covered"
+    EXCEEDED = "exceeded"
+
+
+class AnchorTrigger(Enum):
+    STEP_COMPLETE = "step_complete"
+    PHASE_GATE = "phase_gate"
+    DIRECTION_CHANGE = "direction_change"
+    CONFLICT = "conflict"
+    MILESTONE = "milestone"
+
+
+class DriftSeverity(Enum):
+    NONE = "none"
+    LOW = "low"
+    MEDIUM = "medium"
+    HIGH = "high"
+    CRITICAL = "critical"
+
+
+@dataclass
+class GoalItem:
+    item_id: str
+    description: str
+    keywords: List[str] = field(default_factory=list)
+    status: GoalItemStatus = GoalItemStatus.PENDING
+    coverage_score: float = 0.0
+    evidence: List[str] = field(default_factory=list)
+
+
+@dataclass
+class StructuredGoal:
+    goal_id: str = ""
+    original_description: str = ""
+    items: List[GoalItem] = field(default_factory=list)
+    created_at: str = ""
+
+    @property
+    def overall_coverage(self) -> float:
+        if not self.items:
+            return 0.0
+        return sum(i.coverage_score for i in self.items) / len(self.items)
+
+    @property
+    def uncovered_items(self) -> List[GoalItem]:
+        return [i for i in self.items if i.status in (GoalItemStatus.PENDING, GoalItemStatus.PARTIALLY_COVERED)]
+
+
+@dataclass
+class DriftItem:
+    content: str
+    severity: DriftSeverity = DriftSeverity.LOW
+    reason: str = ""
+
+
+@dataclass
+class AnchorResult:
+    aligned: bool = True
+    trigger: AnchorTrigger = AnchorTrigger.STEP_COMPLETE
+    coverage: float = 1.0
+    drift_score: float = 0.0
+    drifts: List[DriftItem] = field(default_factory=list)
+    uncovered_goals: List[str] = field(default_factory=list)
+    recommendation: str = ""
+    checked_at: str = ""
+
+    @property
+    def severity(self) -> DriftSeverity:
+        if self.drift_score < 0.1:
+            return DriftSeverity.NONE
+        elif self.drift_score < 0.2:
+            return DriftSeverity.LOW
+        elif self.drift_score < 0.3:
+            return DriftSeverity.MEDIUM
+        elif self.drift_score < 0.5:
+            return DriftSeverity.HIGH
+        return DriftSeverity.CRITICAL
+
+
+@dataclass
+class DeviationRecord:
+    step_description: str
+    deviation_type: str
+    reason: str
+    impact: str = ""
+    suggestion: str = ""
+
+
+@dataclass
+class RetrospectiveReport:
+    task_goal: str = ""
+    goal_id: str = ""
+    deviations: List[DeviationRecord] = field(default_factory=list)
+    redundant_steps: List[str] = field(default_factory=list)
+    improvements: List[str] = field(default_factory=list)
+    anchor_check_count: int = 0
+    anchor_drift_count: int = 0
+    final_coverage: float = 1.0
+    summary: str = ""
+    created_at: str = ""
+
+    def to_dict(self) -> Dict[str, Any]:
+        return {
+            "task_goal": self.task_goal,
+            "goal_id": self.goal_id,
+            "deviation_count": len(self.deviations),
+            "redundant_step_count": len(self.redundant_steps),
+            "improvement_count": len(self.improvements),
+            "anchor_check_count": self.anchor_check_count,
+            "anchor_drift_count": self.anchor_drift_count,
+            "final_coverage": self.final_coverage,
+            "summary": self.summary,
+            "created_at": self.created_at,
+        }
+
+    def to_markdown(self) -> str:
+        lines = [
+            "# Retrospective Report",
+            "",
+            f"**Task**: {self.task_goal}",
+            f"**Goal ID**: {self.goal_id}",
+            f"**Final Coverage**: {self.final_coverage:.0%}",
+            f"**Anchor Checks**: {self.anchor_check_count} (drifts: {self.anchor_drift_count})",
+            "",
+        ]
+        if self.deviations:
+            lines.append("## Deviations")
+            for d in self.deviations:
+                lines.append(f"- **{d.deviation_type}**: {d.reason}")
+                if d.suggestion:
+                    lines.append(f"  → {d.suggestion}")
+            lines.append("")
+        if self.improvements:
+            lines.append("## Improvements for Next Time")
+            for imp in self.improvements:
+                lines.append(f"- {imp}")
+            lines.append("")
+        lines.append(f"---\n*Generated by RetrospectiveEngine V3.6.0*")
+        return "\n".join(lines)