配置全面迁移数据库：PromptConfig、TaskConfig动态调度、敏感词/清洗规则/趋势映射/平台标签/痛点模板全部可编辑

- 新增 PromptConfig 模型 + API，支持提示词在线编辑（16条默认） - 调度器动态读取 TaskConfig.schedule，admin 可调执行时间 - 新增 KeywordDomainMap、SensitiveWord、ContentCleanRule、TrendFieldMapping 表 - DOMAINS、TREND_DOMAIN_MAP、PLATFORM_TAGS、china_pains、RSS关键词、priority_weights 全部迁移到 DB - tasks.html 重构：卡片网格+配置/产出/历史/提示词四个Tab，折叠显示 - 清理冗余代码：DEFAULT_PROMPTS死代码、collector.py unreachable代码、compliance_checker bug - strip_thinking_html 改用 DB 规则优先
2026-05-22 11:18:23 +08:00
parent a8e0a76e07
commit 1855f190f5
31 changed files with 2927 additions and 1127 deletions
@@ -471,24 +471,59 @@ class TaskLog(Base):
    __tablename__ = "task_logs"

    id = Column(Integer, primary_key=True, index=True, autoincrement=True)
+    module_id = Column(String, nullable=False, index=True)  # scheduled_collect / scheduled_generate 等
    task_name = Column(String, nullable=False)
-    topic_id = Column(String, nullable=True)
-    status = Column(String, nullable=False)
+    topic_id = Column(String, nullable=True, index=True)
+    status = Column(String, nullable=False)  # pending / running / success / failed / cancelled
    message = Column(Text, nullable=True)
+    error_trace = Column(Text, nullable=True)
+    triggered_by = Column(String, default="scheduler")  # scheduler / manual / api
+    result_data = Column(JSON, default=dict)  # 产出摘要：{topics_found, articles_created, issues_fixed, ...}
    started_at = Column(DateTime(timezone=True), server_default=func.now())
    finished_at = Column(DateTime(timezone=True), nullable=True)
-    duration = Column(Integer, nullable=True)
+    duration = Column(Integer, nullable=True)  # seconds
+    next_run_time = Column(DateTime(timezone=True), nullable=True)

    def to_dict(self):
        return {
            "id": self.id,
+            "module_id": self.module_id,
            "task_name": self.task_name,
            "topic_id": self.topic_id,
            "status": self.status,
            "message": self.message,
+            "error_trace": self.error_trace,
+            "triggered_by": self.triggered_by,
+            "result_data": self.result_data or {},
            "started_at": self.started_at.isoformat() if self.started_at else None,
            "finished_at": self.finished_at.isoformat() if self.finished_at else None,
            "duration": self.duration,
+            "next_run_time": self.next_run_time.isoformat() if self.next_run_time else None,
+        }
+
+
+class TaskConfig(Base):
+    __tablename__ = "task_configs"
+
+    id = Column(Integer, primary_key=True, index=True, autoincrement=True)
+    module_id = Column(String, unique=True, nullable=False)
+    enabled = Column(Boolean, default=True)
+    params = Column(JSON, default=dict)  # 各任务自定义参数，JSON 格式
+    schedule = Column(String, nullable=True)  # cron 表达式，覆盖默认
+    last_modified_by = Column(String, nullable=True)
+    created_at = Column(DateTime(timezone=True), server_default=func.now())
+    updated_at = Column(DateTime(timezone=True), server_default=func.now(), onupdate=func.now())
+
+    def to_dict(self):
+        return {
+            "id": self.id,
+            "module_id": self.module_id,
+            "enabled": self.enabled,
+            "params": self.params or {},
+            "schedule": self.schedule,
+            "last_modified_by": self.last_modified_by,
+            "created_at": self.created_at.isoformat() if self.created_at else None,
+            "updated_at": self.updated_at.isoformat() if self.updated_at else None,
        }


@@ -527,6 +562,101 @@ class LLMConfig(Base):
        }


+class PromptConfig(Base):
+    __tablename__ = "prompt_configs"
+
+    id = Column(Integer, primary_key=True, index=True, autoincrement=True)
+    key = Column(String, unique=True, nullable=False, index=True)
+    module_id = Column(String, nullable=True, index=True)
+    category = Column(String, default="prompt")  # prompt / rule / template
+    version = Column(String, default="v1")
+    content = Column(Text, nullable=False)
+    variables = Column(JSON, default=[])  # [{name, description, default_value}]
+    description = Column(String, nullable=True)
+    enabled = Column(Boolean, default=True)
+    temperature = Column(Float, nullable=True)
+    max_tokens = Column(Integer, nullable=True)
+    created_by = Column(String, nullable=True)
+    created_at = Column(DateTime(timezone=True), server_default=func.now())
+    updated_at = Column(DateTime(timezone=True), server_default=func.now(), onupdate=func.now())
+
+    def to_dict(self):
+        return {
+            "id": self.id,
+            "key": self.key,
+            "module_id": self.module_id,
+            "category": self.category,
+            "version": self.version,
+            "content": self.content,
+            "variables": self.variables or [],
+            "description": self.description,
+            "enabled": self.enabled,
+            "temperature": self.temperature,
+            "max_tokens": self.max_tokens,
+            "created_by": self.created_by,
+            "created_at": self.created_at.isoformat() if self.created_at else None,
+            "updated_at": self.updated_at.isoformat() if self.updated_at else None,
+        }
+
+
+class KeywordDomainMap(Base):
+    __tablename__ = "keyword_domain_map"
+
+    id = Column(Integer, primary_key=True, index=True, autoincrement=True)
+    pattern = Column(String, nullable=False)
+    domain = Column(String, nullable=False)
+    sort_order = Column(Integer, default=0)
+    is_active = Column(Boolean, default=True)
+    created_at = Column(DateTime(timezone=True), server_default=func.now())
+    updated_at = Column(DateTime(timezone=True), onupdate=func.now())
+
+    def to_dict(self):
+        return {
+            "id": self.id, "pattern": self.pattern, "domain": self.domain,
+            "sort_order": self.sort_order, "is_active": self.is_active,
+            "created_at": self.created_at.isoformat() if self.created_at else None,
+            "updated_at": self.updated_at.isoformat() if self.updated_at else None,
+        }
+
+
+class SensitiveWord(Base):
+    __tablename__ = "sensitive_words"
+
+    id = Column(Integer, primary_key=True, index=True, autoincrement=True)
+    word = Column(String, nullable=False)
+    category = Column(String, default="general")
+    is_active = Column(Boolean, default=True)
+    added_by = Column(String, nullable=True)
+    created_at = Column(DateTime(timezone=True), server_default=func.now())
+
+    def to_dict(self):
+        return {
+            "id": self.id, "word": self.word, "category": self.category,
+            "is_active": self.is_active, "added_by": self.added_by,
+            "created_at": self.created_at.isoformat() if self.created_at else None,
+        }
+
+
+class ContentCleanRule(Base):
+    __tablename__ = "content_clean_rules"
+
+    id = Column(Integer, primary_key=True, index=True, autoincrement=True)
+    rule_type = Column(String, nullable=False)  # thinking / preface / verbosity / html_thinking
+    pattern = Column(Text, nullable=False)
+    description = Column(String, nullable=True)
+    is_active = Column(Boolean, default=True)
+    sort_order = Column(Integer, default=0)
+    created_at = Column(DateTime(timezone=True), server_default=func.now())
+
+    def to_dict(self):
+        return {
+            "id": self.id, "rule_type": self.rule_type, "pattern": self.pattern,
+            "description": self.description, "is_active": self.is_active,
+            "sort_order": self.sort_order,
+            "created_at": self.created_at.isoformat() if self.created_at else None,
+        }
+
+
 class SystemConfig(Base):
    __tablename__ = "system_configs"

@@ -557,6 +687,7 @@ class CollectorCategory(Base):
    name = Column(String, unique=True, nullable=False)
    description = Column(Text, nullable=True)
    search_query = Column(String, nullable=True)
+    pain_template = Column(Text, nullable=True)
    sort_order = Column(Integer, default=0)
    is_active = Column(Boolean, default=True)
    created_at = Column(DateTime(timezone=True), server_default=func.now())
@@ -570,6 +701,7 @@ class CollectorCategory(Base):
            "name": self.name,
            "description": self.description,
            "search_query": self.search_query,
+            "pain_template": self.pain_template,
            "sort_order": self.sort_order,
            "is_active": self.is_active,
            "created_at": self.created_at.isoformat() if self.created_at else None,
@@ -577,6 +709,24 @@ class CollectorCategory(Base):
        }


+class TrendFieldMapping(Base):
+    __tablename__ = "trend_field_mappings"
+
+    id = Column(Integer, primary_key=True, index=True, autoincrement=True)
+    trend_keyword = Column(String, nullable=False)
+    field_name = Column(String, nullable=False)
+    sort_order = Column(Integer, default=0)
+    is_active = Column(Boolean, default=True)
+    created_at = Column(DateTime(timezone=True), server_default=func.now())
+
+    def to_dict(self):
+        return {
+            "id": self.id, "trend_keyword": self.trend_keyword, "field_name": self.field_name,
+            "sort_order": self.sort_order, "is_active": self.is_active,
+            "created_at": self.created_at.isoformat() if self.created_at else None,
+        }
+
+
 class CollectorSource(Base):
    """采集信息源（可在运营管理中动态编辑）"""
    __tablename__ = "collector_sources"