claw/insightflow
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: auto-fix code issues (cron)

Browse Source 
- 修复隐式 Optional 类型注解 (RUF013)
- 修复不必要的赋值后返回 (RET504)
- 优化列表推导式 (PERF401)
- 修复未使用的参数 (ARG002)
- 清理重复导入
- 优化异常处理
This commit is contained in:
AutoFix Bot
2026-03-03 21:11:47 +08:00
parent d17a58ceae
commit 259f2c90d0
36 changed files with 1651 additions and 863 deletions
Download Patch File Download Diff File Expand all files Collapse all files

127
backend/search_manager.py
View File

@@ -385,7 +385,7 @@ class FullTextSearch:
# 排序和分页
scored_results.sort(key=lambda x: x.score, reverse=True)
return scored_results[offset: offset + limit]
return scored_results[offset : offset + limit]
def _parse_boolean_query(self, query: str) -> dict:
"""
@@ -545,19 +545,24 @@ class FullTextSearch:
return results
def _get_content_by_id(
self, conn: sqlite3.Connection, content_id: str, content_type: str,
self,
conn: sqlite3.Connection,
content_id: str,
content_type: str,
) -> str | None:
"""根据ID获取内容"""
try:
if content_type == "transcript":
row = conn.execute(
"SELECT full_text FROM transcripts WHERE id = ?", (content_id,),
"SELECT full_text FROM transcripts WHERE id = ?",
(content_id,),
).fetchone()
return row["full_text"] if row else None
elif content_type == "entity":
row = conn.execute(
"SELECT name, definition FROM entities WHERE id = ?", (content_id,),
"SELECT name, definition FROM entities WHERE id = ?",
(content_id,),
).fetchone()
if row:
return f"{row['name']} {row['definition'] or ''}"
@@ -583,21 +588,27 @@ class FullTextSearch:
return None
def _get_project_id(
self, conn: sqlite3.Connection, content_id: str, content_type: str,
self,
conn: sqlite3.Connection,
content_id: str,
content_type: str,
) -> str | None:
"""获取内容所属的项目ID"""
try:
if content_type == "transcript":
row = conn.execute(
"SELECT project_id FROM transcripts WHERE id = ?", (content_id,),
"SELECT project_id FROM transcripts WHERE id = ?",
(content_id,),
).fetchone()
elif content_type == "entity":
row = conn.execute(
"SELECT project_id FROM entities WHERE id = ?", (content_id,),
"SELECT project_id FROM entities WHERE id = ?",
(content_id,),
).fetchone()
elif content_type == "relation":
row = conn.execute(
"SELECT project_id FROM entity_relations WHERE id = ?", (content_id,),
"SELECT project_id FROM entity_relations WHERE id = ?",
(content_id,),
).fetchone()
else:
return None
@@ -880,7 +891,11 @@ class SemanticSearch:
return None
def index_embedding(
self, content_id: str, content_type: str, project_id: str, text: str,
self,
content_id: str,
content_type: str,
project_id: str,
text: str,
) -> bool:
"""
为内容生成并保存 embedding
@@ -1029,13 +1044,15 @@ class SemanticSearch:
try:
if content_type == "transcript":
row = conn.execute(
"SELECT full_text FROM transcripts WHERE id = ?", (content_id,),
"SELECT full_text FROM transcripts WHERE id = ?",
(content_id,),
).fetchone()
result = row["full_text"] if row else None
elif content_type == "entity":
row = conn.execute(
"SELECT name, definition FROM entities WHERE id = ?", (content_id,),
"SELECT name, definition FROM entities WHERE id = ?",
(content_id,),
).fetchone()
result = f"{row['name']}: {row['definition']}" if row else None
@@ -1067,7 +1084,10 @@ class SemanticSearch:
return None
def find_similar_content(
self, content_id: str, content_type: str, top_k: int = 5,
self,
content_id: str,
content_type: str,
top_k: int = 5,
) -> list[SemanticSearchResult]:
"""
查找与指定内容相似的内容
@@ -1175,7 +1195,10 @@ class EntityPathDiscovery:
return conn
def find_shortest_path(
self, source_entity_id: str, target_entity_id: str, max_depth: int = 5,
self,
source_entity_id: str,
target_entity_id: str,
max_depth: int = 5,
) -> EntityPath | None:
"""
查找两个实体之间的最短路径（BFS算法）
@@ -1192,7 +1215,8 @@ class EntityPathDiscovery:
# 获取项目ID
row = conn.execute(
"SELECT project_id FROM entities WHERE id = ?", (source_entity_id,),
"SELECT project_id FROM entities WHERE id = ?",
(source_entity_id,),
).fetchone()
if not row:
@@ -1250,7 +1274,11 @@ class EntityPathDiscovery:
return None
def find_all_paths(
self, source_entity_id: str, target_entity_id: str, max_depth: int = 4, max_paths: int = 10,
self,
source_entity_id: str,
target_entity_id: str,
max_depth: int = 4,
max_paths: int = 10,
) -> list[EntityPath]:
"""
查找两个实体之间的所有路径（限制数量和深度）
@@ -1268,7 +1296,8 @@ class EntityPathDiscovery:
# 获取项目ID
row = conn.execute(
"SELECT project_id FROM entities WHERE id = ?", (source_entity_id,),
"SELECT project_id FROM entities WHERE id = ?",
(source_entity_id,),
).fetchone()
if not row:
@@ -1280,7 +1309,11 @@ class EntityPathDiscovery:
paths = []
def dfs(
current_id: str, target_id: str, path: list[str], visited: set[str], depth: int,
current_id: str,
target_id: str,
path: list[str],
visited: set[str],
depth: int,
) -> None:
if depth > max_depth:
return
@@ -1328,7 +1361,8 @@ class EntityPathDiscovery:
nodes = []
for entity_id in entity_ids:
row = conn.execute(
"SELECT id, name, type FROM entities WHERE id = ?", (entity_id,),
"SELECT id, name, type FROM entities WHERE id = ?",
(entity_id,),
).fetchone()
if row:
nodes.append({"id": row["id"], "name": row["name"], "type": row["type"]})
@@ -1398,7 +1432,8 @@ class EntityPathDiscovery:
# 获取项目ID
row = conn.execute(
"SELECT project_id, name FROM entities WHERE id = ?", (entity_id,),
"SELECT project_id, name FROM entities WHERE id = ?",
(entity_id,),
).fetchone()
if not row:
@@ -1445,7 +1480,8 @@ class EntityPathDiscovery:
# 获取邻居信息
neighbor_info = conn.execute(
"SELECT name, type FROM entities WHERE id = ?", (neighbor_id,),
"SELECT name, type FROM entities WHERE id = ?",
(neighbor_id,),
).fetchone()
if neighbor_info:
@@ -1458,7 +1494,10 @@ class EntityPathDiscovery:
"relation_type": neighbor["relation_type"],
"evidence": neighbor["evidence"],
"path": self._get_path_to_entity(
entity_id, neighbor_id, project_id, conn,
entity_id,
neighbor_id,
project_id,
conn,
),
},
)
@@ -1470,7 +1509,11 @@ class EntityPathDiscovery:
return relations
def _get_path_to_entity(
self, source_id: str, target_id: str, project_id: str, conn: sqlite3.Connection,
self,
source_id: str,
target_id: str,
project_id: str,
conn: sqlite3.Connection,
) -> list[str]:
"""获取从源实体到目标实体的路径（简化版）"""
# BFS 找路径
@@ -1565,7 +1608,8 @@ class EntityPathDiscovery:
# 获取所有实体
entities = conn.execute(
"SELECT id, name FROM entities WHERE project_id = ?", (project_id,),
"SELECT id, name FROM entities WHERE project_id = ?",
(project_id,),
).fetchall()
# 计算每个实体作为桥梁的次数
@@ -1706,7 +1750,8 @@ class KnowledgeGapDetection:
# 检查每个实体的属性完整性
entities = conn.execute(
"SELECT id, name FROM entities WHERE project_id = ?", (project_id,),
"SELECT id, name FROM entities WHERE project_id = ?",
(project_id,),
).fetchall()
for entity in entities:
@@ -1714,7 +1759,8 @@ class KnowledgeGapDetection:
# 获取实体已有的属性
existing_attrs = conn.execute(
"SELECT template_id FROM entity_attributes WHERE entity_id = ?", (entity_id,),
"SELECT template_id FROM entity_attributes WHERE entity_id = ?",
(entity_id,),
).fetchall()
existing_template_ids = {a["template_id"] for a in existing_attrs}
@@ -1726,7 +1772,8 @@ class KnowledgeGapDetection:
missing_names = []
for template_id in missing_templates:
template = conn.execute(
"SELECT name FROM attribute_templates WHERE id = ?", (template_id,),
"SELECT name FROM attribute_templates WHERE id = ?",
(template_id,),
).fetchone()
if template:
missing_names.append(template["name"])
@@ -1759,7 +1806,8 @@ class KnowledgeGapDetection:
# 获取所有实体及其关系数量
entities = conn.execute(
"SELECT id, name, type FROM entities WHERE project_id = ?", (project_id,),
"SELECT id, name, type FROM entities WHERE project_id = ?",
(project_id,),
).fetchall()
for entity in entities:
@@ -1900,7 +1948,8 @@ class KnowledgeGapDetection:
# 分析转录文本中频繁提及但未提取为实体的词
transcripts = conn.execute(
"SELECT full_text FROM transcripts WHERE project_id = ?", (project_id,),
"SELECT full_text FROM transcripts WHERE project_id = ?",
(project_id,),
).fetchall()
# 合并所有文本
@@ -1908,7 +1957,8 @@ class KnowledgeGapDetection:
# 获取现有实体名称
existing_entities = conn.execute(
"SELECT name FROM entities WHERE project_id = ?", (project_id,),
"SELECT name FROM entities WHERE project_id = ?",
(project_id,),
).fetchall()
existing_names = {e["name"].lower() for e in existing_entities}
@@ -2146,7 +2196,10 @@ class SearchManager:
for t in transcripts:
if t["full_text"] and self.semantic_search.index_embedding(
t["id"], "transcript", t["project_id"], t["full_text"],
t["id"],
"transcript",
t["project_id"],
t["full_text"],
):
semantic_stats["indexed"] += 1
else:
@@ -2179,12 +2232,14 @@ class SearchManager:
# 全文索引统计
fulltext_count = conn.execute(
f"SELECT COUNT(*) as count FROM search_indexes {where_clause}", params,
f"SELECT COUNT(*) as count FROM search_indexes {where_clause}",
params,
).fetchone()["count"]
# 语义索引统计
semantic_count = conn.execute(
f"SELECT COUNT(*) as count FROM embeddings {where_clause}", params,
f"SELECT COUNT(*) as count FROM embeddings {where_clause}",
params,
).fetchone()["count"]
# 按类型统计
@@ -2225,7 +2280,9 @@ def get_search_manager(db_path: str = "insightflow.db") -> SearchManager:
def fulltext_search(
query: str, project_id: str | None = None, limit: int = 20,
query: str,
project_id: str | None = None,
limit: int = 20,
) -> list[SearchResult]:
"""全文搜索便捷函数"""
manager = get_search_manager()
@@ -2233,7 +2290,9 @@ def fulltext_search(
def semantic_search(
query: str, project_id: str | None = None, top_k: int = 10,
query: str,
project_id: str | None = None,
top_k: int = 10,
) -> list[SemanticSearchResult]:
"""语义搜索便捷函数"""
manager = get_search_manager()