Skip to content

Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.

chore(format): align python and c++ file formatting (#2001)

yepperddcd3fb9c842
diff --git a/bot/demo/werewolf/start_werewolf_demo.py b/bot/demo/werewolf/start_werewolf_demo.py
index 85771cfac..769d9366e 100644
--- a/bot/demo/werewolf/start_werewolf_demo.py
+++ b/bot/demo/werewolf/start_werewolf_demo.py
@@ -4,16 +4,15 @@ import argparse
import json
import os
import shutil
-from urllib.parse import urlparse
import subprocess
import sys
import time
from pathlib import Path
from typing import Any
+from urllib.parse import urlparse
import httpx
-
PLAYER_IDS = [f"player_{i}" for i in range(1, 7)]
WEREWOLF_CHANNEL_IDS = ["god", *PLAYER_IDS]
BOT_API_TYPE = "bot_api"
@@ -206,7 +205,11 @@ def prepare_workspace(workspace: Path) -> None:
def validate_assets() -> None:
- missing = [str(path) for path in (SOUL_GOD_PATH, SOUL_PLAYER_PATH, WEREWOLF_SERVER_PATH) if not path.exists()]
+ missing = [
+ str(path)
+ for path in (SOUL_GOD_PATH, SOUL_PLAYER_PATH, WEREWOLF_SERVER_PATH)
+ if not path.exists()
+ ]
if missing:
raise FileNotFoundError("Missing required demo files: " + ", ".join(missing))
diff --git a/bot/tests/test_chat_functionality.py b/bot/tests/test_chat_functionality.py
index b6eef3065..7939dadff 100644
--- a/bot/tests/test_chat_functionality.py
+++ b/bot/tests/test_chat_functionality.py
@@ -119,7 +119,6 @@ class TestSessionHistoryProviderSpecificFields:
class TestChatChannel:
-
def test_chat_channel_initialization(self, message_bus, temp_workspace):
"""Test that ChatChannel can be initialized correctly."""
config = ChatChannelConfig()
diff --git a/bot/vikingbot/agent/context.py b/bot/vikingbot/agent/context.py
index 8227f8878..fb1df0486 100644
--- a/bot/vikingbot/agent/context.py
+++ b/bot/vikingbot/agent/context.py
@@ -197,7 +197,10 @@ Skills with available="false" need dependencies installed first - you can try in
# Use provided memory_users or fall back to [sender_id]
search_user_ids = memory_users if memory_users else [sender_id]
viking_memory = await self.memory.get_viking_memory_context(
- current_message=current_message, workspace_id=workspace_id, sender_id=sender_id, user_ids=search_user_ids
+ current_message=current_message,
+ workspace_id=workspace_id,
+ sender_id=sender_id,
+ user_ids=search_user_ids,
)
logger.info(f"viking_memory={viking_memory}")
cost = round(_time.time() - start, 2)
diff --git a/bot/vikingbot/agent/memory.py b/bot/vikingbot/agent/memory.py
index 745a4801a..ef695a732 100644
--- a/bot/vikingbot/agent/memory.py
+++ b/bot/vikingbot/agent/memory.py
@@ -53,9 +53,7 @@ class MemoryStore:
def get_abstract(m):
return m.get("abstract", "") if isinstance(m, dict) else getattr(m, "abstract", "")
- filtered_memories = [
- memory for memory in result if get_score(memory) >= min_score
- ]
+ filtered_memories = [memory for memory in result if get_score(memory) >= min_score]
filtered_memories.sort(key=get_score, reverse=True)
user_memories = []
@@ -162,7 +160,10 @@ class MemoryStore:
if not client:
return ""
result = await client.search_memory(
- query=current_message, user_ids=search_user_ids, agent_user_id=admin_user_id, limit=30
+ query=current_message,
+ user_ids=search_user_ids,
+ agent_user_id=admin_user_id,
+ limit=30,
)
if not result:
return ""
@@ -207,7 +208,9 @@ class MemoryStore:
if not client:
return ""
experiences = await client.search_experiences(query, limit=5)
- logger.info(f"[READ_EXPERIENCE_MEMORY]: found {len(experiences)} experiences, query={query[:50]}")
+ logger.info(
+ f"[READ_EXPERIENCE_MEMORY]: found {len(experiences)} experiences, query={query[:50]}"
+ )
for i, exp in enumerate(experiences):
uri = exp.get("uri", "") if isinstance(exp, dict) else getattr(exp, "uri", "")
score = exp.get("score", 0) if isinstance(exp, dict) else getattr(exp, "score", 0)
diff --git a/bot/vikingbot/agent/tools/mcp.py b/bot/vikingbot/agent/tools/mcp.py
index 884b00712..2d6186655 100644
--- a/bot/vikingbot/agent/tools/mcp.py
+++ b/bot/vikingbot/agent/tools/mcp.py
@@ -63,9 +63,7 @@ def _normalize_schema_for_openai(schema: Any) -> dict[str, Any]:
if "properties" in normalized and isinstance(normalized["properties"], dict):
normalized["properties"] = {
- name: _normalize_schema_for_openai(prop)
- if isinstance(prop, dict)
- else prop
+ name: _normalize_schema_for_openai(prop) if isinstance(prop, dict) else prop
for name, prop in normalized["properties"].items()
}
@@ -174,14 +172,17 @@ async def connect_mcp_servers(
)
read, write = await stack.enter_async_context(stdio_client(params))
elif transport_type == "sse":
+ cfg_headers = cfg.headers or {}
+
def httpx_client_factory(
headers: dict[str, str] | None = None,
timeout: httpx.Timeout | None = None,
auth: httpx.Auth | None = None,
+ cfg_headers: dict[str, str] = cfg_headers,
) -> httpx.AsyncClient:
merged_headers = {
"Accept": "application/json, text/event-stream",
- **(cfg.headers or {}),
+ **cfg_headers,
**(headers or {}),
}
return httpx.AsyncClient(
diff --git a/bot/vikingbot/agent/tools/ov_file.py b/bot/vikingbot/agent/tools/ov_file.py
index 3db4e178f..8aeae597b 100644
--- a/bot/vikingbot/agent/tools/ov_file.py
+++ b/bot/vikingbot/agent/tools/ov_file.py
@@ -23,6 +23,7 @@ class OVFileTool(Tool, ABC):
self._client = await VikingClient.create(tool_context.workspace_id)
return self._client
+
class VikingListTool(OVFileTool):
"""Tool to list Viking resources."""
@@ -86,9 +87,11 @@ class VikingSearchTool(OVFileTool):
@property
def description(self) -> str:
- return ("Using query to search for resources (knowledge, code, files, workflow, etc.) in OpenViking. "
- "This operation performs semantic retrieval, not full character matching. Please avoid repeated calls with similar queries as much as possible."
- "bad-case: after searching with ‘Nate Joanna dog playdate 3:00 pm', another search was performed using 'Nate Joanna dog playdate'.")
+ return (
+ "Using query to search for resources (knowledge, code, files, workflow, etc.) in OpenViking. "
+ "This operation performs semantic retrieval, not full character matching. Please avoid repeated calls with similar queries as much as possible."
+ "bad-case: after searching with ‘Nate Joanna dog playdate 3:00 pm', another search was performed using 'Nate Joanna dog playdate'."
+ )
@property
def parameters(self) -> dict[str, Any]:
@@ -128,7 +131,11 @@ class VikingSearchTool(OVFileTool):
items.append({**item, "type": item.get("type", item_type)})
return items
- if hasattr(results, "memories") or hasattr(results, "resources") or hasattr(results, "skills"):
+ if (
+ hasattr(results, "memories")
+ or hasattr(results, "resources")
+ or hasattr(results, "skills")
+ ):
for key, item_type in group_map.items():
for item in getattr(results, key, []) or []:
items.append(
@@ -178,7 +185,9 @@ class VikingSearchTool(OVFileTool):
except (TypeError, ValueError):
return 0.0
- def _filter_search_items(self, results: Any, min_score: float) -> dict[str, list[dict[str, Any]]]:
+ def _filter_search_items(
+ self, results: Any, min_score: float
+ ) -> dict[str, list[dict[str, Any]]]:
grouped: dict[str, list[dict[str, Any]]] = {
"memory": [],
"resource": [],
@@ -216,7 +225,9 @@ class VikingSearchTool(OVFileTool):
)
return group_items
- def _format_search_items_json(self, grouped_items: dict[str, list[dict[str, Any]]], min_score: float) -> str:
+ def _format_search_items_json(
+ self, grouped_items: dict[str, list[dict[str, Any]]], min_score: float
+ ) -> str:
memories = self._build_group_json(grouped_items.get("memory", []))
resources = self._build_group_json(grouped_items.get("resource", []))
skills = self._build_group_json(grouped_items.get("skill", []))
@@ -253,7 +264,9 @@ class VikingSearchTool(OVFileTool):
results = await search_client.search(query, target_uri=user_uri, limit=20)
if results:
memories = [
- item for item in self._extract_search_items(results) if item.get("type") == "memory"
+ item
+ for item in self._extract_search_items(results)
+ if item.get("type") == "memory"
]
all_results.extend(memories)
@@ -329,7 +342,7 @@ class VikingAddResourceTool(OVFileTool):
else:
return "Failed to add resource"
except httpx.ReadTimeout:
- return f"Request timed out. The resource addition task may still be processing on the server side."
+ return "Request timed out. The resource addition task may still be processing on the server side."
except Exception as e:
logger.warning(f"Error adding resource: {e}")
return f"Error adding resource to Viking: {str(e)}"
@@ -347,8 +360,10 @@ class VikingGrepTool(OVFileTool):
@property
def description(self) -> str:
- return ("Search Viking resources using regex patterns (like grep). Supports multiple patterns to search concurrently."
- "Please avoid repeated calls with similar queries as much as possible.")
+ return (
+ "Search Viking resources using regex patterns (like grep). Supports multiple patterns to search concurrently."
+ "Please avoid repeated calls with similar queries as much as possible."
+ )
@property
def parameters(self) -> dict[str, Any]:
@@ -432,7 +447,9 @@ class VikingGrepTool(OVFileTool):
return f"No matches found for patterns: {pattern_str}"
# Format output
- result_lines = [f"Found {total_matches} match{'es' if total_matches != 1 else ''} across {len(patterns)} pattern{'s' if len(patterns) != 1 else ''}:"]
+ result_lines = [
+ f"Found {total_matches} match{'es' if total_matches != 1 else ''} across {len(patterns)} pattern{'s' if len(patterns) != 1 else ''}:"
+ ]
for match_uri, matches in merged_results.items():
# Sort matches by line number
@@ -503,6 +520,7 @@ class VikingGlobTool(OVFileTool):
except Exception as e:
return f"Error searching Viking with glob: {str(e)}"
+
class VikingMemoryCommitTool(OVFileTool):
"""Tool to commit messages to OpenViking session."""
@@ -552,6 +570,7 @@ class VikingMemoryCommitTool(OVFileTool):
logger.exception(f"Error processing message: {e}")
return f"Error committing to Viking: {str(e)}"
+
class VikingMultiReadTool(OVFileTool):
"""Tool to read content from multiple Viking resources concurrently."""
@@ -571,7 +590,7 @@ class VikingMultiReadTool(OVFileTool):
"uris": {
"type": "array",
"items": {"type": "string"},
- "description": "List of Viking file URIs to read from (e.g., [\"viking://resources/path/123.md\", \"viking://resources/path/456.md\"])",
+ "description": 'List of Viking file URIs to read from (e.g., ["viking://resources/path/123.md", "viking://resources/path/456.md"])',
},
},
"required": ["uris"],
@@ -616,7 +635,7 @@ class VikingMultiReadTool(OVFileTool):
# 构建结果
result_lines = [f"Multi-read results for {len(uris)} resources (level: {level}):"]
- for i, result in enumerate(results, 1):
+ for result in results:
uri = result["uri"]
content = result["content"]
success = result["success"]
@@ -632,4 +651,4 @@ class VikingMultiReadTool(OVFileTool):
except Exception as e:
logger.exception(f"Error in VikingMultiReadTool: {e}")
- return f"Error multi-reading Viking resources: {str(e)}"
\ No newline at end of file
+ return f"Error multi-reading Viking resources: {str(e)}"
diff --git a/bot/vikingbot/channels/feishu.py b/bot/vikingbot/channels/feishu.py
index af487db81..4587bff42 100644
--- a/bot/vikingbot/channels/feishu.py
+++ b/bot/vikingbot/channels/feishu.py
@@ -116,7 +116,9 @@ class FeishuChannel(BaseChannel):
self._chat_mode_cache: dict[str, str] = {} # 缓存群类型:group(普通群)/thread(话题群)
self._user_name_cache: OrderedDict[str, str] = OrderedDict() # LRU缓存用户ID到姓名的映射
self._bot_name_cache: dict[str, str] = {} # 缓存机器人open_id到名称的映射
- self._chat_member_cache: OrderedDict[str, dict[str, Any]] = OrderedDict() # chat_id -> {members, expires_at, last_error_at}
+ self._chat_member_cache: OrderedDict[str, dict[str, Any]] = (
+ OrderedDict()
+ ) # chat_id -> {members, expires_at, last_error_at}
self._MAX_USER_CACHE_SIZE = 1000 # 最大缓存1000个用户
self._CHAT_MEMBER_CACHE_TTL_SEC = 300
self._CHAT_MEMBER_CACHE_MAX_CHATS = 30
@@ -801,13 +803,19 @@ class FeishuChannel(BaseChannel):
self._user_name_cache[open_id] = name
return name
- def _save_chat_member_cache(self, chat_id: str, members: dict[str, str], last_error_at: float = 0) -> None:
+ def _save_chat_member_cache(
+ self, chat_id: str, members: dict[str, str], last_error_at: float = 0
+ ) -> None:
if chat_id in self._chat_member_cache:
self._chat_member_cache.pop(chat_id)
elif len(self._chat_member_cache) >= self._CHAT_MEMBER_CACHE_MAX_CHATS:
self._chat_member_cache.popitem(last=False)
- ttl = self._CHAT_MEMBER_FETCH_COOLDOWN_SEC if last_error_at else self._CHAT_MEMBER_CACHE_TTL_SEC
+ ttl = (
+ self._CHAT_MEMBER_FETCH_COOLDOWN_SEC
+ if last_error_at
+ else self._CHAT_MEMBER_CACHE_TTL_SEC
+ )
self._chat_member_cache[chat_id] = {
"members": members,
"expires_at": time.time() + ttl,
@@ -862,7 +870,10 @@ class FeishuChannel(BaseChannel):
members = entry.get("members", {})
if entry.get("expires_at", 0) > now:
return members.get(open_id)
- if now - float(entry.get("last_error_at", 0) or 0) < self._CHAT_MEMBER_FETCH_COOLDOWN_SEC:
+ if (
+ now - float(entry.get("last_error_at", 0) or 0)
+ < self._CHAT_MEMBER_FETCH_COOLDOWN_SEC
+ ):
return members.get(open_id)
try:
@@ -924,7 +935,9 @@ class FeishuChannel(BaseChannel):
self._bot_name_cache[open_id] = bot_name
return bot_name
- async def _batch_get_user_names(self, open_ids: list[str], chat_id: str | None = None) -> dict[str, str]:
+ async def _batch_get_user_names(
+ self, open_ids: list[str], chat_id: str | None = None
+ ) -> dict[str, str]:
"""
Get user names from Feishu API by open_ids (fetches individually with LRU cache).
Returns a dict mapping open_id to user name.
diff --git a/bot/vikingbot/channels/feishu_test.py b/bot/vikingbot/channels/feishu_test.py
index a77b026a8..13e994be7 100644
--- a/bot/vikingbot/channels/feishu_test.py
+++ b/bot/vikingbot/channels/feishu_test.py
@@ -1,30 +1,24 @@
import json
import lark_oapi as lark
-from lark_oapi.api.contact.v3 import *
+from lark_oapi.api.contact.v3 import GetUserRequest
def main():
# 创建client
- client = lark.Client.builder() \
- .app_id("") \
- .app_secret("") \
- .log_level(lark.LogLevel.DEBUG) \
- .build()
+ client = lark.Client.builder().app_id("").app_secret("").log_level(lark.LogLevel.DEBUG).build()
# 构造请求对象
- request: GetUserRequest = GetUserRequest.builder() \
- .user_id("") \
- .user_id_type("open_id") \
- .build()
+ request: GetUserRequest = GetUserRequest.builder().user_id("").user_id_type("open_id").build()
# 发起请求
- response: GetUserResponse = client.contact.v3.user.get(request)
+ response = client.contact.v3.user.get(request)
# 处理失败返回
if not response.success():
lark.logger.error(
- f"client.contact.v3.user.get failed, code: {response.code}, msg: {response.msg}, log_id: {response.get_log_id()}, resp: \n{json.dumps(json.loads(response.raw.content), indent=4, ensure_ascii=False)}")
+ f"client.contact.v3.user.get failed, code: {response.code}, msg: {response.msg}, log_id: {response.get_log_id()}, resp: \n{json.dumps(json.loads(response.raw.content), indent=4, ensure_ascii=False)}"
+ )
return
# 处理业务结果
diff --git a/bot/vikingbot/channels/manager.py b/bot/vikingbot/channels/manager.py
index 97c805915..323d8d7ed 100644
--- a/bot/vikingbot/channels/manager.py
+++ b/bot/vikingbot/channels/manager.py
@@ -233,7 +233,7 @@ class ChannelManager:
for ch in self.channels.values():
# Check if channel has a send method and can handle bot_api messages
# OpenAPIChannel identifies itself with name "openapi" and handles bot_api
- if hasattr(ch, 'name') and ch.name == "openapi":
+ if hasattr(ch, "name") and ch.name == "openapi":
try:
await ch.send(msg)
handled = True
diff --git a/bot/vikingbot/cli/commands.py b/bot/vikingbot/cli/commands.py
index d9c66ffd2..f773253ca 100644
--- a/bot/vikingbot/cli/commands.py
+++ b/bot/vikingbot/cli/commands.py
@@ -117,6 +117,7 @@ def _get_gateway_token(config) -> str:
return ""
return getattr(gateway, "token", "") or ""
+
# ---------------------------------------------------------------------------
# CLI input: prompt_toolkit for editing, paste, history, and display
# ---------------------------------------------------------------------------
@@ -333,7 +334,13 @@ def gateway(
)
server = uvicorn.Server(config_uvicorn)
- tasks = [cron.start(), heartbeat.start(), channels.start_all(), agent_loop.run(), server.serve()]
+ tasks = [
+ cron.start(),
+ heartbeat.start(),
+ channels.start_all(),
+ agent_loop.run(),
+ server.serve(),
+ ]
# if enable_console:
# tasks.append(start_console(console_port))
@@ -660,7 +667,9 @@ def chat(
if session_id is None:
session_id = get_or_create_machine_id()
cron = prepare_cron(bus, quiet=is_single_turn)
- channels = prepare_agent_channel(config, bus, message, session_id, markdown, logs, eval, sender, memory_user)
+ channels = prepare_agent_channel(
+ config, bus, message, session_id, markdown, logs, eval, sender, memory_user
+ )
agent_loop = prepare_agent_loop(
config, bus, session_manager, cron, quiet=is_single_turn, eval=eval
)
diff --git a/bot/vikingbot/heartbeat/service.py b/bot/vikingbot/heartbeat/service.py
index b2a2cb10d..4222ef675 100644
--- a/bot/vikingbot/heartbeat/service.py
+++ b/bot/vikingbot/heartbeat/service.py
@@ -8,8 +8,6 @@ from typing import Any, Callable, Coroutine
from loguru import logger
from vikingbot.config.schema import SessionKey
-
-
from vikingbot.session.manager import SessionManager
# Default interval: 30 minutes
@@ -91,7 +89,8 @@ class HeartbeatService:
def __init__(
self,
workspace: Path,
- on_heartbeat: Callable[[str, str | None, dict[str, Any] | None], Coroutine[Any, Any, str]] | None = None,
+ on_heartbeat: Callable[[str, str | None, dict[str, Any] | None], Coroutine[Any, Any, str]]
+ | None = None,
interval_s: int = DEFAULT_HEARTBEAT_INTERVAL_S,
enabled: bool = True,
sandbox_mode: str = "shared",
@@ -108,9 +107,9 @@ class HeartbeatService:
def _is_session_stale(self, session_info: dict[str, Any]) -> bool:
"""Check whether a session has been inactive for too long."""
- reference = _parse_session_timestamp(session_info.get("updated_at")) or _parse_session_timestamp(
- session_info.get("created_at")
- )
+ reference = _parse_session_timestamp(
+ session_info.get("updated_at")
+ ) or _parse_session_timestamp(session_info.get("created_at"))
if reference is None:
return False
diff --git a/bot/vikingbot/hooks/builtins/openviking_hooks.py b/bot/vikingbot/hooks/builtins/openviking_hooks.py
index 6f0ae13ab..18810b0ad 100644
--- a/bot/vikingbot/hooks/builtins/openviking_hooks.py
+++ b/bot/vikingbot/hooks/builtins/openviking_hooks.py
@@ -6,8 +6,9 @@ from typing import Any
from loguru import logger
from vikingbot.config.loader import load_config
-from ..base import Hook, HookContext
+
from ...session import Session
+from ..base import Hook, HookContext
try:
import openviking as ov
@@ -39,7 +40,6 @@ class OpenVikingCompactHook(Hook):
async def _get_client(self, workspace_id: str) -> VikingClient:
return await get_global_client(workspace_id)
-
async def execute(self, context: HookContext, **kwargs) -> Any:
vikingbot_session: Session = kwargs.get("session", {})
session_id = context.session_key.safe_name()
@@ -58,7 +58,9 @@ class OpenVikingCompactHook(Hook):
"users_count": 0,
}
- admin_result = await client.commit(session_id, vikingbot_session.messages, admin_user_id)
+ admin_result = await client.commit(
+ session_id, vikingbot_session.messages, admin_user_id
+ )
messages_by_sender = defaultdict(list)
for msg in vikingbot_session.messages:
@@ -72,7 +74,9 @@ class OpenVikingCompactHook(Hook):
async def commit_with_semaphore(user_id: str, user_messages: list):
async with semaphore:
- return await client.commit(f"{session_id}_{user_id}", user_messages, user_id)
+ return await client.commit(
+ f"{session_id}_{user_id}", user_messages, user_id
+ )
user_tasks = []
for user_id, user_messages in messages_by_sender.items():
@@ -132,7 +136,9 @@ class OpenVikingPostCallHook(Hook):
desc_match = re.search(r"^description:\s*(.+)$", result, re.MULTILINE)
skill_query = desc_match.group(1).strip() if desc_match else skill_name
- exp_memory = await self._search_skill_experiences(context.workspace_id, skill_query)
+ exp_memory = await self._search_skill_experiences(
+ context.workspace_id, skill_query
+ )
if exp_memory:
result = f"{result}\n\n---\n## Related Experiences\n{exp_memory}"
diff --git a/bot/vikingbot/openviking_mount/ov_server.py b/bot/vikingbot/openviking_mount/ov_server.py
index b51fd29ae..f91f7b701 100644
--- a/bot/vikingbot/openviking_mount/ov_server.py
+++ b/bot/vikingbot/openviking_mount/ov_server.py
@@ -1,4 +1,5 @@
import asyncio
+import hashlib
import time
from typing import Any, Dict, List, Optional
@@ -198,7 +199,6 @@ class VikingClient:
def should_sender_fanout(self) -> bool:
return self._is_root_key_mode()
-
async def find(self, query: str, target_uri: Optional[str] = None):
"""搜索资源"""
if target_uri:
@@ -560,7 +560,6 @@ async def main_test():
async def account_test():
-
client = ov.AsyncHTTPClient(
url="http://localhost:1933",
api_key="",
diff --git a/openviking/parse/parsers/code/ast/languages/php.py b/openviking/parse/parsers/code/ast/languages/php.py
index b1ef23ad3..e33e1d3bb 100644
--- a/openviking/parse/parsers/code/ast/languages/php.py
+++ b/openviking/parse/parsers/code/ast/languages/php.py
@@ -4,7 +4,7 @@
from __future__ import annotations
-from typing import List, Optional
+from typing import List
from openviking.parse.parsers.code.ast.languages.base import LanguageExtractor
from openviking.parse.parsers.code.ast.skeleton import ClassSkeleton, CodeSkeleton, FunctionSig
@@ -121,7 +121,9 @@ def _extract_use_declaration(node, content_bytes: bytes) -> List[str]:
if sub.type == "name":
name_node = sub
break
- return _normalize_name(_node_text(name_node, content_bytes)) if name_node is not None else ""
+ return (
+ _normalize_name(_node_text(name_node, content_bytes)) if name_node is not None else ""
+ )
clauses = []
if group_node is not None:
@@ -204,7 +206,12 @@ def _extract_class(
name = f"{name_prefix}{name}" if name else name_prefix.strip()
for child in node.children:
- if child.type in ("base_clause", "extends_clause", "implements_clause", "class_interface_clause"):
+ if child.type in (
+ "base_clause",
+ "extends_clause",
+ "implements_clause",
+ "class_interface_clause",
+ ):
for b in _collect_typeish_tokens(child, content_bytes):
if b not in bases:
bases.append(b)
@@ -268,10 +275,14 @@ def _process_siblings(
classes.append(_extract_class(child, content_bytes, docstring=doc, name_prefix=""))
elif child.type == "interface_declaration":
doc = _preceding_doc(siblings, idx, content_bytes)
- classes.append(_extract_class(child, content_bytes, docstring=doc, name_prefix="interface "))
+ classes.append(
+ _extract_class(child, content_bytes, docstring=doc, name_prefix="interface ")
+ )
elif child.type == "trait_declaration":
doc = _preceding_doc(siblings, idx, content_bytes)
- classes.append(_extract_class(child, content_bytes, docstring=doc, name_prefix="trait "))
+ classes.append(
+ _extract_class(child, content_bytes, docstring=doc, name_prefix="trait ")
+ )
elif child.type == "enum_declaration":
doc = _preceding_doc(siblings, idx, content_bytes)
classes.append(_extract_class(child, content_bytes, docstring=doc, name_prefix="enum "))
@@ -300,7 +311,9 @@ def _process_siblings(
elif child.type == "namespace_definition":
for sub in child.children:
if sub.type in ("declaration_list", "compound_statement"):
- _process_siblings(list(sub.children), content_bytes, imports, classes, functions)
+ _process_siblings(
+ list(sub.children), content_bytes, imports, classes, functions
+ )
class PhpExtractor(LanguageExtractor):
diff --git a/openviking/privacy/service.py b/openviking/privacy/service.py
index 95fcaefe1..313431081 100644
--- a/openviking/privacy/service.py
+++ b/openviking/privacy/service.py
@@ -5,8 +5,6 @@
import json
from typing import Any, Optional
-from openviking_cli.exceptions import NotFoundError
-
from openviking.core.namespace import canonical_user_root
from openviking.privacy.helpers import (
canonicalize_values,
@@ -21,6 +19,7 @@ from openviking.privacy.models import UserPrivacyConfigMeta, UserPrivacyConfigVe
from openviking.server.identity import RequestContext
from openviking.storage.viking_fs import VikingFS
from openviking.utils.time_utils import get_current_timestamp
+from openviking_cli.exceptions import NotFoundError
class UserPrivacyConfigService:
@@ -48,7 +47,9 @@ class UserPrivacyConfigService:
async def _ensure_root(self, ctx: RequestContext, category: str, target_key: str) -> None:
root_uri = self.get_config_root(ctx, category, target_key)
await self._viking_fs.mkdir(root_uri, exist_ok=True, ctx=ctx)
- await self._viking_fs.mkdir(history_dir_uri(self._user_space(ctx), category, target_key), exist_ok=True, ctx=ctx)
+ await self._viking_fs.mkdir(
+ history_dir_uri(self._user_space(ctx), category, target_key), exist_ok=True, ctx=ctx
+ )
async def get_meta(
self, ctx: RequestContext, category: str, target_key: str
diff --git a/openviking/privacy/skill_placeholder.py b/openviking/privacy/skill_placeholder.py
index accc33467..bc2ffabe4 100644
--- a/openviking/privacy/skill_placeholder.py
+++ b/openviking/privacy/skill_placeholder.py
@@ -71,7 +71,5 @@ def placeholderize_skill_content_with_blocks(
)
-def placeholderize_skill_content(
- content: str, skill_name: str, values: dict[str, str]
-) -> str:
+def placeholderize_skill_content(content: str, skill_name: str, values: dict[str, str]) -> str:
return placeholderize_skill_content_with_blocks(content, skill_name, values).sanitized_content
diff --git a/openviking/resource/watch_storage.py b/openviking/resource/watch_storage.py
index 8b5e0b015..9ef4b7d13 100644
--- a/openviking/resource/watch_storage.py
+++ b/openviking/resource/watch_storage.py
@@ -4,7 +4,6 @@
from __future__ import annotations
-
WATCH_TASK_STORAGE_URI = "viking://resources/.watch_tasks.json"
WATCH_TASK_STORAGE_BAK_URI = "viking://resources/.watch_tasks.json.bak"
WATCH_TASK_STORAGE_TMP_URI = "viking://resources/.watch_tasks.json.tmp"
@@ -23,4 +22,4 @@ def is_watch_task_control_uri(uri: str) -> bool:
if not isinstance(uri, str):
return False
normalized = uri.rstrip("/")
- return normalized in WATCH_TASK_CONTROL_URIS
\ No newline at end of file
+ return normalized in WATCH_TASK_CONTROL_URIS
diff --git a/openviking/session/compressor_v2.py b/openviking/session/compressor_v2.py
index 38671b612..d34df31ae 100644
--- a/openviking/session/compressor_v2.py
+++ b/openviking/session/compressor_v2.py
@@ -14,22 +14,20 @@ from typing import Any, Dict, List, Optional
from openviking.core.context import Context
from openviking.core.namespace import (
- agent_space_fragment,
- user_space_fragment,
- to_user_space,
to_agent_space,
+ to_user_space,
)
from openviking.message import Message
from openviking.server.identity import RequestContext
from openviking.session.memory import ExtractLoop, MemoryUpdater
+from openviking.session.memory.dataclass import ResolvedOperations
from openviking.session.memory.memory_isolation_handler import MemoryIsolationHandler
+from openviking.session.memory.memory_updater import MemoryUpdateResult
from openviking.session.memory.utils.json_parser import JsonUtils
from openviking.session.memory.utils.messages import parse_memory_file_with_fields
from openviking.session.memory.utils.uri import render_template
from openviking.storage import VikingDBManager
from openviking.storage.viking_fs import VikingFS, get_viking_fs
-from openviking.session.memory.dataclass import ResolvedOperations
-from openviking.session.memory.memory_updater import MemoryUpdateResult
from openviking.telemetry import get_current_telemetry, tracer
from openviking_cli.session.user_id import UserIdentifier
from openviking_cli.utils import get_logger
@@ -40,7 +38,6 @@ logger = get_logger(__name__)
MAX_SOURCE_TRAJECTORIES = 5 # keep only the most recent N trajectory URIs per experience
-
class SessionCompressorV2:
"""Session memory extractor with v2 templating system."""
@@ -190,8 +187,8 @@ class SessionCompressorV2:
# 固定文件名 schema(如 soul.md、identity.md)只需 POINT 锁,
# 避免因 SUBTREE 锁阻塞子目录(trajectories/experiences)的并发加锁。
schemas = orchestrator.context_provider.get_memory_schemas(ctx)
- point_lock_dirs: list = [] # 固定文件名 schema → POINT 锁
- subtree_lock_dirs: list = [] # 变量文件名 schema → SUBTREE 锁
+ point_lock_dirs: list = [] # 固定文件名 schema → POINT 锁
+ subtree_lock_dirs: list = [] # 变量文件名 schema → SUBTREE 锁
for schema in schemas:
if not schema.directory:
continue
@@ -405,7 +402,10 @@ class SessionCompressorV2:
viking_fs = get_viking_fs()
for traj_uri in written_trajectory_uris:
try:
- from openviking.session.memory.utils.content import deserialize_content as _deser_content
+ from openviking.session.memory.utils.content import (
+ deserialize_content as _deser_content,
+ )
+
traj_content = _deser_content(await viking_fs.read_file(traj_uri, ctx=ctx) or "")
except Exception as e:
logger.warning(f"Failed to read new trajectory {traj_uri}: {e}")
@@ -426,7 +426,7 @@ class SessionCompressorV2:
exp_dir = exp_provider._render_experience_dir(ctx)
- async def _single_existing_experience_uri() -> List[str]:
+ async def _single_existing_experience_uri(exp_dir: str | None = exp_dir) -> List[str]:
if not exp_dir:
return []
try:
@@ -463,7 +463,9 @@ class SessionCompressorV2:
all_exp_uris = exp_written_uris + exp_edited_uris
if not all_exp_uris:
- candidate_uris = list(dict.fromkeys(getattr(exp_provider, "prefetched_uris", []) or []))
+ candidate_uris = list(
+ dict.fromkeys(getattr(exp_provider, "prefetched_uris", []) or [])
+ )
candidate_exp_uris = [
uri
for uri in candidate_uris
@@ -601,9 +603,7 @@ class SessionCompressorV2:
f"{op.memory_type}(uris={op.uris!r})"
for op in getattr(operations, "upsert_operations", [])
]
- _delete_uris_raw = [
- dc.uri for dc in getattr(operations, "delete_file_contents", [])
- ]
+ _delete_uris_raw = [dc.uri for dc in getattr(operations, "delete_file_contents", [])]
tracer.info(
f"[{phase_label}] LLM operations: ops={_op_items}, delete_uris={_delete_uris_raw}"
)
@@ -613,12 +613,13 @@ class SessionCompressorV2:
# superseding experience inherits the old source_trajectories.
inheritance_map = await self._resolve_supersedes(operations, ctx, viking_fs, provider)
-
-
registry = provider._get_registry()
updater = self._get_or_create_updater(registry, transaction_handle)
result = await updater.apply_operations(
- operations, ctx, extract_context=extract_context, isolation_handler=isolation_handler
+ operations,
+ ctx,
+ extract_context=extract_context,
+ isolation_handler=isolation_handler,
)
tracer.info(
@@ -629,17 +630,11 @@ class SessionCompressorV2:
contexts: List[Context] = []
for uri in result.written_uris:
- contexts.append(
- Context(uri=uri, category="memory_write", context_type="memory")
- )
+ contexts.append(Context(uri=uri, category="memory_write", context_type="memory"))
for uri in result.edited_uris:
- contexts.append(
- Context(uri=uri, category="memory_edit", context_type="memory")
- )
+ contexts.append(Context(uri=uri, category="memory_edit", context_type="memory"))
for uri in result.deleted_uris:
- contexts.append(
- Context(uri=uri, category="memory_delete", context_type="memory")
- )
+ contexts.append(Context(uri=uri, category="memory_delete", context_type="memory"))
return list(result.written_uris), list(result.edited_uris), contexts, inheritance_map
except Exception as e:
@@ -688,7 +683,9 @@ class SessionCompressorV2:
if not supersedes_name:
continue
if not exp_dir:
- logger.warning(f"[supersedes] cannot resolve '{supersedes_name}': no experience dir")
+ logger.warning(
+ f"[supersedes] cannot resolve '{supersedes_name}': no experience dir"
+ )
continue
old_uri = f"{exp_dir.rstrip('/')}/{supersedes_name}.md"
@@ -742,7 +739,10 @@ class SessionCompressorV2:
This is the system-side management of source_trajectories — the LLM never
outputs this field; the pipeline appends the batch after a write or edit.
"""
- from openviking.session.memory.utils.content import deserialize_full, serialize_with_metadata
+ from openviking.session.memory.utils.content import (
+ deserialize_full,
+ serialize_with_metadata,
+ )
normalized_traj_uris = [uri for uri in traj_uris if uri]
if not normalized_traj_uris:
diff --git a/openviking/session/memory/agent_experience_context_provider.py b/openviking/session/memory/agent_experience_context_provider.py
index e24f136f6..84cbd159f 100644
--- a/openviking/session/memory/agent_experience_context_provider.py
+++ b/openviking/session/memory/agent_experience_context_provider.py
@@ -10,10 +10,11 @@ No tool calls — all context is prefetched. Top-3 candidates also include their
source_trajectories as grounding material.
"""
-import jinja2
from typing import Any, Dict, List
-from openviking.core.namespace import to_user_space, to_agent_space
+import jinja2
+
+from openviking.core.namespace import to_agent_space, to_user_space
from openviking.server.identity import RequestContext, ToolContext
from openviking.session.memory.dataclass import MemoryFileContent
from openviking.session.memory.session_extract_context_provider import (
@@ -30,8 +31,8 @@ logger = get_logger(__name__)
EXPERIENCE_MEMORY_TYPE = "experiences"
SEARCH_TOP_K = 5
-SOURCE_TRAJ_TOP_K = 3 # only attach source_trajectories for the top-3 candidates
-MAX_SOURCE_TRAJS = 3 # max trajectories to load per experience
+SOURCE_TRAJ_TOP_K = 3 # only attach source_trajectories for the top-3 candidates
+MAX_SOURCE_TRAJS = 3 # max trajectories to load per experience
class AgentExperienceContextProvider(SessionExtractContextProvider):
diff --git a/openviking/session/memory/extract_loop.py b/openviking/session/memory/extract_loop.py
index 4ff571186..9ef3fad3c 100644
--- a/openviking/session/memory/extract_loop.py
+++ b/openviking/session/memory/extract_loop.py
@@ -28,7 +28,6 @@ from openviking.session.memory.tools import (
)
from openviking.session.memory.utils import (
parse_json_with_stability,
- parse_memory_file_with_fields,
pretty_print_messages,
)
from openviking.session.memory.utils.json_parser import JsonUtils
@@ -488,7 +487,6 @@ The final output of the model must strictly follow the JSON Schema format shown
return await asyncio.gather(*tasks)
async def _check_unread_existing_files(self, operations: ResolvedOperations) -> Dict:
-
refetch_uris = {}
for operation in operations.upsert_operations:
for uri in operation.uris:
diff --git a/openviking/session/memory/memory_isolation_handler.py b/openviking/session/memory/memory_isolation_handler.py
index 7f25943c5..ff045f29e 100644
--- a/openviking/session/memory/memory_isolation_handler.py
+++ b/openviking/session/memory/memory_isolation_handler.py
@@ -10,13 +10,11 @@ Memory Isolation Handler - 处理记忆的隔离机制
from dataclasses import dataclass
from typing import Any, Dict, List, Optional, Set
-from openviking.core.namespace import to_user_space, to_agent_space
-from openviking.message import Message
-from openviking.server.identity import AccountNamespacePolicy, RequestContext
-from openviking.session.memory.utils.uri import generate_uri
-from openviking.session.memory.memory_updater import ExtractContext
-from openviking_cli.session.user_id import UserIdentifier
+from openviking.core.namespace import to_agent_space, to_user_space
+from openviking.server.identity import RequestContext
from openviking.session.memory.dataclass import MemoryTypeSchema, ResolvedOperation
+from openviking.session.memory.memory_updater import ExtractContext
+from openviking.session.memory.utils.uri import generate_uri
from openviking_cli.utils import get_logger
from openviking_cli.utils.config import get_openviking_config
@@ -81,7 +79,6 @@ class MemoryIsolationHandler:
)
def fill_role_ids(self, item_dict: Dict[str, Any], role_scope: RoleScope) -> None:
-
user_ids = set()
agent_ids = set()
diff --git a/openviking/session/memory/memory_updater.py b/openviking/session/memory/memory_updater.py
index 383b5e2ab..e9b427292 100644
--- a/openviking/session/memory/memory_updater.py
+++ b/openviking/session/memory/memory_updater.py
@@ -11,28 +11,23 @@ from __future__ import annotations
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple
-
if TYPE_CHECKING:
from openviking.session.memory.memory_isolation_handler import MemoryIsolationHandler
-from openviking.core.namespace import agent_space_fragment, user_space_fragment
from openviking.message import Message
from openviking.server.identity import RequestContext
from openviking.session.memory.dataclass import (
- MemoryField,
MemoryFileContent,
- ResolvedOperations,
ResolvedOperation,
+ ResolvedOperations,
)
from openviking.session.memory.memory_type_registry import MemoryTypeRegistry
from openviking.session.memory.merge_op import MergeOpFactory
from openviking.session.memory.utils import (
- deserialize_full,
- flat_model_to_dict,
parse_memory_file_with_fields,
serialize_with_metadata,
)
-from openviking.session.memory.utils.uri import supplement_operation_uris, render_template
+from openviking.session.memory.utils.uri import render_template, supplement_operation_uris
from openviking.storage.viking_fs import get_viking_fs
from openviking.telemetry import tracer
from openviking.telemetry.request_wait_tracker import get_request_wait_tracker
@@ -306,7 +301,6 @@ class MemoryUpdater:
extract_context: ExtractContext = None,
isolation_handler: MemoryIsolationHandler = None,
) -> MemoryUpdateResult:
-
result = MemoryUpdateResult()
viking_fs = self._get_viking_fs()
@@ -384,7 +378,7 @@ class MemoryUpdater:
# Collect directories that need overview generation
# uri is now a string, so extract directory using os.path
- dirs = dict()
+ dirs = {}
for operation in operations.upsert_operations:
for uri_str in operation.uris:
dir_path = "/".join(uri_str.split("/")[:-1])
diff --git a/openviking/session/memory/merge_op/patch_handler.py b/openviking/session/memory/merge_op/patch_handler.py
index 4f4155e78..f78d17cd2 100644
--- a/openviking/session/memory/merge_op/patch_handler.py
+++ b/openviking/session/memory/merge_op/patch_handler.py
@@ -206,7 +206,7 @@ def _find_best_substring_match(line: str, search_str: str) -> tuple[float, str]:
for i in positions_to_check:
if 0 <= i <= line_len - search_len:
- substring = line[i:i+search_len]
+ substring = line[i : i + search_len]
score = get_similarity(substring, search_str)
if score > best_score:
best_score = score
@@ -940,7 +940,9 @@ class MemoryPatchHandler:
return result.content if result.content is not None else original_content
else:
error_msg = result.error or "Unknown error"
- logger.warning(f"Patch application failed, skipping update: {error_msg}, original_content={original_content}, patch={patch}")
+ logger.warning(
+ f"Patch application failed, skipping update: {error_msg}, original_content={original_content}, patch={patch}"
+ )
raise PatchParseError(f"Patch application failed: {error_msg}")
def _extract_replace_content(self, patch: str) -> str:
@@ -1229,9 +1231,7 @@ def apply_str_patch(original_content: str, patch: StrPatch) -> str:
# Get indent from current replace line
current_replace_match = re.match(r"^[\t ]*", line)
- current_replace_indent = (
- current_replace_match.group(0) if current_replace_match else ""
- )
+ current_replace_indent = current_replace_match.group(0) if current_replace_match else ""
# Calculate relative indent level (how much deeper/shallower this line is compared to search)
relative_level = len(current_replace_indent) - len(search_indent)
@@ -1259,12 +1259,13 @@ def apply_str_patch(original_content: str, patch: StrPatch) -> str:
final_content = line_ending.join(result_lines)
# Check if all results are successful (including no-change cases)
- all_successful = all(result.get("success", False) for result in diff_results)
has_failures = any(not result.get("success", False) for result in diff_results)
if applied_count == 0 and has_failures:
logger.warning("Patch application failed, skipping update")
- raise PatchParseError(f"Patch application failed: search content not found in original, original_content={original_content}, patch={patch}")
+ raise PatchParseError(
+ f"Patch application failed: search content not found in original, original_content={original_content}, patch={patch}"
+ )
return final_content
diff --git a/openviking/session/memory/schema_model_generator.py b/openviking/session/memory/schema_model_generator.py
index 9039d0e2b..26d464e16 100644
--- a/openviking/session/memory/schema_model_generator.py
+++ b/openviking/session/memory/schema_model_generator.py
@@ -188,7 +188,6 @@ class SchemaModelGenerator:
self._union_model = MemoryDataWrapper
return self._union_model
-
def create_structured_operations_model(self, role_scope: RoleScope) -> Type[BaseModel]:
"""
Create a structured MemoryOperations model with type-safe write operations.
@@ -225,16 +224,14 @@ class SchemaModelGenerator:
List[flat_model], # type: ignore
Field(
default_factory=list,
- description=f"{mt.memory_type} memories: {mt.description} (top-level field, do not nest inside other arrays)"
+ description=f"{mt.memory_type} memories: {mt.description} (top-level field, do not nest inside other arrays)",
),
)
# Only expose delete_uris when at least one schema supports it.
# add_only schemas (e.g. trajectories) never delete existing records,
# so excluding this field prevents the LLM from hallucinating fake URIs.
- has_deletable_schema = any(
- mt.operation_mode != "add_only" for mt in enabled_memory_types
- )
+ has_deletable_schema = any(mt.operation_mode != "add_only" for mt in enabled_memory_types)
if has_deletable_schema:
field_definitions["delete_uris"] = (
List[str],
@@ -298,7 +295,6 @@ class SchemaModelGenerator:
self._operations_model = StructuredMemoryOperations
return self._operations_model
-
def get_memory_data_json_schema(self) -> Dict[str, Any]:
"""
Get the JSON schema just for the flat memory data union.
diff --git a/openviking/session/memory/tools.py b/openviking/session/memory/tools.py
index 24c595465..640dce259 100644
--- a/openviking/session/memory/tools.py
+++ b/openviking/session/memory/tools.py
@@ -8,12 +8,11 @@ Reference: bot/vikingbot/agent/tools/base.py design pattern
import json
from abc import ABC, abstractmethod
-from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, Union
+from typing import TYPE_CHECKING, Any, Dict, List, Optional, Union
from openviking.session.memory.dataclass import MemoryFileContent
from openviking.session.memory.utils import parse_memory_file_with_fields
from openviking.session.memory.utils.content import truncate_content
-from openviking.storage.viking_fs import VikingFS
from openviking.telemetry import tracer
from openviking_cli.exceptions import NotFoundError
from openviking_cli.utils import get_logger
@@ -82,7 +81,6 @@ def add_tool_call_pair_to_messages(
)
-
class MemoryTool(ABC):
"""
Abstract base class for memory tools.
@@ -327,7 +325,6 @@ def get_tool(name: str) -> Optional[MemoryTool]:
return MEMORY_TOOLS_REGISTRY.get(name)
-
# Tools exposed to LLM (not all registered tools are exposed)
LLM_TOOLS = ["read"]
@@ -337,7 +334,6 @@ def get_tool_schemas() -> List[Dict[str, Any]]:
return [tool.to_schema() for tool in MEMORY_TOOLS_REGISTRY.values() if tool.name in LLM_TOOLS]
-
# Register default tools
register_tool(MemoryReadTool())
register_tool(MemorySearchTool())
diff --git a/openviking/session/memory/utils/uri.py b/openviking/session/memory/utils/uri.py
index c0443a73c..bd4d7841f 100644
--- a/openviking/session/memory/utils/uri.py
+++ b/openviking/session/memory/utils/uri.py
@@ -7,8 +7,7 @@ URI generation and validation utilities.
from __future__ import annotations
import re
-from dataclasses import dataclass
-from typing import TYPE_CHECKING, Any, Dict, List, Optional, Set, Tuple
+from typing import TYPE_CHECKING, Any, Dict, Set
if TYPE_CHECKING:
from openviking.session.memory.memory_isolation_handler import MemoryIsolationHandler
@@ -66,9 +65,6 @@ def render_template(
return jinja_template.render(**template_vars).strip()
-
-
-
def generate_uri(
memory_type: MemoryTypeSchema,
fields: Dict[str, Any],
@@ -138,9 +134,6 @@ def validate_uri_template(memory_type: MemoryTypeSchema) -> bool:
return True
-
-
-
def _pattern_matches_uri(pattern: str, uri: str) -> bool:
"""
Check if a URI matches a pattern with variables like {{ topic }}, {{ tool_name }}, etc.
@@ -204,10 +197,6 @@ def is_uri_allowed(
return False
-
-from openviking.session.memory.utils.model import model_to_dict
-
-
def extract_uri_fields_from_flat_model(model: Any, schema: MemoryTypeSchema) -> Dict[str, Any]:
"""
Extract URI-friendly fields from a flat model, ignoring patch objects.
@@ -231,16 +220,12 @@ def extract_uri_fields_from_flat_model(model: Any, schema: MemoryTypeSchema) ->
return uri_fields
-
-
-
def supplement_operation_uris(
operations: ResolvedOperations,
registry: MemoryTypeRegistry,
extract_context: ExtractContext = None,
isolation_handler: MemoryIsolationHandler = None,
):
-
logger.info(f"[supplement_operation_uris] isolation_handler: {isolation_handler}")
for operation in operations.upsert_operations:
memory_type_schema = registry.get(operation.memory_type)
diff --git a/openviking/storage/transaction/lock_manager.py b/openviking/storage/transaction/lock_manager.py
index bdff74400..74538804e 100644
--- a/openviking/storage/transaction/lock_manager.py
+++ b/openviking/storage/transaction/lock_manager.py
@@ -149,7 +149,8 @@ class LockManager:
for path in sorted_paths:
locks_before = set(handle.locks)
success = await self._path_lock.acquire_subtree(
- path, handle,
+ path,
+ handle,
timeout=timeout,
)
if not success:
@@ -190,12 +191,14 @@ class LockManager:
locks_before = set(handle.locks)
if is_subtree:
success = await self._path_lock.acquire_subtree(
- path, handle,
+ path,
+ handle,
timeout=timeout,
)
else:
success = await self._path_lock.acquire_point(
- path, handle,
+ path,
+ handle,
timeout=timeout,
)
if not success:
diff --git a/openviking/storage/vectordb/collection/volcengine_api_key_collection.py b/openviking/storage/vectordb/collection/volcengine_api_key_collection.py
index ef4e68823..975fc4c6a 100644
--- a/openviking/storage/vectordb/collection/volcengine_api_key_collection.py
+++ b/openviking/storage/vectordb/collection/volcengine_api_key_collection.py
@@ -36,9 +36,7 @@ class VolcengineApiKeyCollection(ICollection):
ClientForDataApi._global_host.get(region or "") if region else None
)
if not resolved_host:
- raise ValueError(
- "host or region is required for Volcengine API key data-plane access"
- )
+ raise ValueError("host or region is required for Volcengine API key data-plane access")
self.data_client = ClientForDataApiWithApiKey(api_key, resolved_host)
self.meta_data = meta_data if meta_data is not None else {}
self.project_name = self.meta_data.get("ProjectName", "default")
@@ -55,17 +53,12 @@ class VolcengineApiKeyCollection(ICollection):
message = ""
if isinstance(result, dict):
message = (
- result.get("message")
- or result.get("msg")
- or result.get("error")
- or response.text
+ result.get("message") or result.get("msg") or result.get("error") or response.text
)
else:
message = response.text
- return ConnectionError(
- f"Request to {action} failed: {response.status_code} {message}"
- )
+ return ConnectionError(f"Request to {action} failed: {response.status_code} {message}")
@staticmethod
def _sanitize_uri_value(v: Any) -> Any:
@@ -244,6 +237,7 @@ class VolcengineApiKeyCollection(ICollection):
def get_meta_data(self):
from openviking.storage.collection_schemas import CollectionSchemas
+
return {
"ProjectName": self.project_name,
"CollectionName": self.collection_name,
diff --git a/openviking/telemetry/tracer.py b/openviking/telemetry/tracer.py
index fe012c2a1..1cfefe755 100644
--- a/openviking/telemetry/tracer.py
+++ b/openviking/telemetry/tracer.py
@@ -450,7 +450,7 @@ class tracer:
return result
except Exception as e:
- self.error("e",e=e)
+ self.error("e", e=e)
span.record_exception(exception=e)
span.set_status(Status(StatusCode.ERROR))
raise
diff --git a/openviking/utils/code_hosting_utils.py b/openviking/utils/code_hosting_utils.py
index 9fcc8c18b..a9470787c 100644
--- a/openviking/utils/code_hosting_utils.py
+++ b/openviking/utils/code_hosting_utils.py
@@ -117,8 +117,6 @@ def parse_code_hosting_url(url: str) -> Optional[str]:
code hosting URL
"""
all_domains = _get_all_domains()
- host = _extract_host(url)
-
# Handle git@ SSH URLs: git@host:org/repo.git
if url.startswith("git@"):
if ":" not in url[4:]:
@@ -155,7 +153,9 @@ def parse_code_hosting_url(url: str) -> Optional[str]:
if azure_repo_parts is None:
azure_repo_parts = _extract_azure_devops_ssh_repo_parts(path_parts)
if azure_repo_parts:
- return "/".join(_sanitize_segment(part.removesuffix(".git")) for part in azure_repo_parts)
+ return "/".join(
+ _sanitize_segment(part.removesuffix(".git")) for part in azure_repo_parts
+ )
return None
# For code hosting URLs with org/repo structure
diff --git a/openviking/utils/search_filters.py b/openviking/utils/search_filters.py
index 1a80d314d..5a8ce3611 100644
--- a/openviking/utils/search_filters.py
+++ b/openviking/utils/search_filters.py
@@ -62,9 +62,7 @@ def merge_level_filter(
return {"op": "and", "conds": [existing_filter, level_filter]}
-def _resolve_levels(
- level: Optional[Union[int, str, List[int], List[str]]]
-) -> List[int]:
+def _resolve_levels(level: Optional[Union[int, str, List[int], List[str]]]) -> List[int]:
"""Resolve level parameter into a list of integers."""
if level is None:
return []
diff --git a/src/abi3_engine_backend.cpp b/src/abi3_engine_backend.cpp
index 68bd0f049..0d20df171 100644
--- a/src/abi3_engine_backend.cpp
+++ b/src/abi3_engine_backend.cpp
@@ -65,7 +65,8 @@ auto call_without_gil(Func&& func) -> decltype(func()) {
}
}
-bool get_named_value(PyObject* obj, const char* name, PyObject** out, bool* found) {
+bool get_named_value(PyObject* obj, const char* name, PyObject** out,
+ bool* found) {
*out = nullptr;
*found = false;
@@ -115,7 +116,8 @@ bool py_to_string(PyObject* obj, std::string* out, bool allow_bytes) {
return true;
}
- raise_type_error("Expected str" + std::string(allow_bytes ? " or bytes" : ""));
+ raise_type_error("Expected str" +
+ std::string(allow_bytes ? " or bytes" : ""));
return false;
}
@@ -262,7 +264,8 @@ PyObject* string_vector_to_py(const std::vector<std::string>& values) {
}
for (Py_ssize_t i = 0; i < static_cast<Py_ssize_t>(values.size()); ++i) {
const auto& value = values[static_cast<size_t>(i)];
- PyObject* item = PyUnicode_FromStringAndSize(value.data(), static_cast<Py_ssize_t>(value.size()));
+ PyObject* item = PyUnicode_FromStringAndSize(
+ value.data(), static_cast<Py_ssize_t>(value.size()));
if (item == nullptr) {
Py_DECREF(list);
return nullptr;
@@ -278,7 +281,8 @@ PyObject* uint64_vector_to_py(const std::vector<uint64_t>& values) {
return nullptr;
}
for (Py_ssize_t i = 0; i < static_cast<Py_ssize_t>(values.size()); ++i) {
- PyObject* item = PyLong_FromUnsignedLongLong(values[static_cast<size_t>(i)]);
+ PyObject* item =
+ PyLong_FromUnsignedLongLong(values[static_cast<size_t>(i)]);
if (item == nullptr) {
Py_DECREF(list);
return nullptr;
@@ -316,13 +320,15 @@ PyObject* value_to_py(const vdb::Value& value, vdb::FieldType field_type) {
case vdb::FieldType::STRING:
if (std::holds_alternative<std::string>(value)) {
const auto& text = std::get<std::string>(value);
- return PyUnicode_FromStringAndSize(text.data(), static_cast<Py_ssize_t>(text.size()));
+ return PyUnicode_FromStringAndSize(
+ text.data(), static_cast<Py_ssize_t>(text.size()));
}
break;
case vdb::FieldType::BINARY:
if (std::holds_alternative<std::string>(value)) {
const auto& blob = std::get<std::string>(value);
- return PyBytes_FromStringAndSize(blob.data(), static_cast<Py_ssize_t>(blob.size()));
+ return PyBytes_FromStringAndSize(blob.data(),
+ static_cast<Py_ssize_t>(blob.size()));
}
break;
case vdb::FieldType::LIST_INT64:
@@ -358,7 +364,8 @@ PyObject* value_to_py(const vdb::Value& value, vdb::FieldType field_type) {
Py_RETURN_NONE;
}
-bool py_to_field_value(PyObject* obj, vdb::FieldType data_type, vdb::Value* out) {
+bool py_to_field_value(PyObject* obj, vdb::FieldType data_type,
+ vdb::Value* out) {
switch (data_type) {
case vdb::FieldType::INT64: {
int64_t value = 0;
@@ -441,22 +448,26 @@ T* capsule_to_ptr(PyObject* capsule, const char* capsule_name) {
}
void schema_capsule_destructor(PyObject* capsule) {
- auto* ptr = static_cast<SchemaHandle*>(PyCapsule_GetPointer(capsule, kSchemaCapsuleName));
+ auto* ptr = static_cast<SchemaHandle*>(
+ PyCapsule_GetPointer(capsule, kSchemaCapsuleName));
delete ptr;
}
void bytes_row_capsule_destructor(PyObject* capsule) {
- auto* ptr = static_cast<BytesRowHandle*>(PyCapsule_GetPointer(capsule, kBytesRowCapsuleName));
+ auto* ptr = static_cast<BytesRowHandle*>(
+ PyCapsule_GetPointer(capsule, kBytesRowCapsuleName));
delete ptr;
}
void index_capsule_destructor(PyObject* capsule) {
- auto* ptr = static_cast<vdb::IndexEngine*>(PyCapsule_GetPointer(capsule, kIndexCapsuleName));
+ auto* ptr = static_cast<vdb::IndexEngine*>(
+ PyCapsule_GetPointer(capsule, kIndexCapsuleName));
delete ptr;
}
void store_capsule_destructor(PyObject* capsule) {
- auto* ptr = static_cast<vdb::KVStore*>(PyCapsule_GetPointer(capsule, kStoreCapsuleName));
+ auto* ptr = static_cast<vdb::KVStore*>(
+ PyCapsule_GetPointer(capsule, kStoreCapsuleName));
delete ptr;
}
@@ -500,7 +511,8 @@ bool py_to_field_type(PyObject* obj, vdb::FieldType* out) {
}
}
-bool parse_schema_fields(PyObject* fields_obj, std::vector<vdb::FieldDef>* fields) {
+bool parse_schema_fields(PyObject* fields_obj,
+ std::vector<vdb::FieldDef>* fields) {
const Py_ssize_t size = PySequence_Size(fields_obj);
if (size < 0) {
raise_type_error("Schema fields must be a sequence");
@@ -527,11 +539,13 @@ bool parse_schema_fields(PyObject* fields_obj, std::vector<vdb::FieldDef>* field
PyObject* id_obj = PyDict_GetItemString(item, "id");
if (name_obj == nullptr || type_obj == nullptr || id_obj == nullptr) {
Py_DECREF(item);
- raise_value_error("Schema field definition must contain name, data_type, and id");
+ raise_value_error(
+ "Schema field definition must contain name, data_type, and id");
return false;
}
- if (!py_to_string(name_obj, &field.name, false) || !py_to_field_type(type_obj, &field.data_type)) {
+ if (!py_to_string(name_obj, &field.name, false) ||
+ !py_to_field_type(type_obj, &field.data_type)) {
Py_DECREF(item);
return false;
}
@@ -545,7 +559,8 @@ bool parse_schema_fields(PyObject* fields_obj, std::vector<vdb::FieldDef>* field
PyObject* default_value = PyDict_GetItemString(item, "default_value");
if (default_value != nullptr && default_value != Py_None) {
- if (!py_to_field_value(default_value, field.data_type, &field.default_value)) {
+ if (!py_to_field_value(default_value, field.data_type,
+ &field.default_value)) {
PyErr_Clear();
field.default_value = std::monostate{};
}
@@ -748,8 +763,8 @@ bool parse_storage_op(PyObject* obj, vdb::StorageOp* op) {
}
Py_XDECREF(value);
op->type = type_value == static_cast<long>(vdb::StorageOp::DELETE_OP)
- ? vdb::StorageOp::DELETE_OP
- : vdb::StorageOp::PUT_OP;
+ ? vdb::StorageOp::DELETE_OP
+ : vdb::StorageOp::PUT_OP;
if (!get_named_value(obj, "key", &value, &found)) {
return false;
@@ -777,8 +792,7 @@ bool parse_storage_op(PyObject* obj, vdb::StorageOp* op) {
}
template <typename RequestT>
-bool parse_request_list(PyObject* obj,
- bool (*parse_item)(PyObject*, RequestT*),
+bool parse_request_list(PyObject* obj, bool (*parse_item)(PyObject*, RequestT*),
std::vector<RequestT>* out) {
const Py_ssize_t size = PySequence_Size(obj);
if (size < 0) {
@@ -873,7 +887,8 @@ PyObject* py_new_bytes_row(PyObject*, PyObject* args) {
return nullptr;
}
- auto* schema_handle = capsule_to_ptr<SchemaHandle>(schema_capsule, kSchemaCapsuleName);
+ auto* schema_handle =
+ capsule_to_ptr<SchemaHandle>(schema_capsule, kSchemaCapsuleName);
if (schema_handle == nullptr) {
return nullptr;
}
@@ -882,7 +897,8 @@ PyObject* py_new_bytes_row(PyObject*, PyObject* args) {
schema_handle->schema,
std::make_shared<vdb::BytesRow>(schema_handle->schema),
};
- return PyCapsule_New(handle, kBytesRowCapsuleName, bytes_row_capsule_destructor);
+ return PyCapsule_New(handle, kBytesRowCapsuleName,
+ bytes_row_capsule_destructor);
}
PyObject* py_bytes_row_serialize(PyObject*, PyObject* args) {
@@ -898,14 +914,16 @@ PyObject* py_bytes_row_serialize(PyObject*, PyObject* args) {
}
std::vector<vdb::Value> row_values;
- if (!row_object_to_values(row_obj, handle->schema->get_field_order(), &row_values)) {
+ if (!row_object_to_values(row_obj, handle->schema->get_field_order(),
+ &row_values)) {
return nullptr;
}
try {
- const std::string payload =
- call_without_gil([&]() { return handle->bytes_row->serialize(row_values); });
- return PyBytes_FromStringAndSize(payload.data(), static_cast<Py_ssize_t>(payload.size()));
+ const std::string payload = call_without_gil(
+ [&]() { return handle->bytes_row->serialize(row_values); });
+ return PyBytes_FromStringAndSize(payload.data(),
+ static_cast<Py_ssize_t>(payload.size()));
} catch (const std::exception& exc) {
raise_runtime_error(exc.what());
return nullptr;
@@ -943,7 +961,8 @@ PyObject* py_bytes_row_serialize_batch(PyObject*, PyObject* args) {
}
std::vector<vdb::Value> row_values;
- const bool ok = row_object_to_values(row_obj, handle->schema->get_field_order(), &row_values);
+ const bool ok = row_object_to_values(
+ row_obj, handle->schema->get_field_order(), &row_values);
Py_DECREF(row_obj);
if (!ok) {
Py_DECREF(list);
@@ -951,9 +970,10 @@ PyObject* py_bytes_row_serialize_batch(PyObject*, PyObject* args) {
}
try {
- const std::string payload =
- call_without_gil([&]() { return handle->bytes_row->serialize(row_values); });
- PyObject* item = PyBytes_FromStringAndSize(payload.data(), static_cast<Py_ssize_t>(payload.size()));
+ const std::string payload = call_without_gil(
+ [&]() { return handle->bytes_row->serialize(row_values); });
+ PyObject* item = PyBytes_FromStringAndSize(
+ payload.data(), static_cast<Py_ssize_t>(payload.size()));
if (item == nullptr) {
Py_DECREF(list);
return nullptr;
@@ -993,8 +1013,9 @@ PyObject* py_bytes_row_deserialize(PyObject*, PyObject* args) {
try {
for (const auto& meta : handle->schema->get_field_order()) {
- const vdb::Value value =
- call_without_gil([&]() { return handle->bytes_row->deserialize_field(payload, meta.name); });
+ const vdb::Value value = call_without_gil([&]() {
+ return handle->bytes_row->deserialize_field(payload, meta.name);
+ });
if (std::holds_alternative<std::monostate>(value)) {
continue;
}
@@ -1045,8 +1066,9 @@ PyObject* py_bytes_row_deserialize_field(PyObject*, PyObject* args) {
}
try {
- const vdb::Value value =
- call_without_gil([&]() { return handle->bytes_row->deserialize_field(payload, field_name); });
+ const vdb::Value value = call_without_gil([&]() {
+ return handle->bytes_row->deserialize_field(payload, field_name);
+ });
return value_to_py(value, meta->data_type);
} catch (const std::exception& exc) {
raise_runtime_error(exc.what());
@@ -1064,7 +1086,8 @@ PyObject* build_search_result(const vdb::SearchResult& result) {
PyObject* labels = uint64_vector_to_py(result.labels);
PyObject* scores = float_vector_to_py(result.scores);
PyObject* extra_json = PyUnicode_FromStringAndSize(
- result.extra_json.data(), static_cast<Py_ssize_t>(result.extra_json.size()));
+ result.extra_json.data(),
+ static_cast<Py_ssize_t>(result.extra_json.size()));
if (result_num == nullptr || labels == nullptr || scores == nullptr ||
extra_json == nullptr) {
Py_XDECREF(result_num);
@@ -1112,10 +1135,12 @@ PyObject* py_init_logging(PyObject*, PyObject* args, PyObject* kwargs) {
const char* log_level = nullptr;
const char* log_output = nullptr;
const char* log_format = "[%Y-%m-%d %H:%M:%S.%e] [%l] %v";
- static const char* keywords[] = {"log_level", "log_output", "log_format", nullptr};
+ static const char* keywords[] = {"log_level", "log_output", "log_format",
+ nullptr};
- if (!PyArg_ParseTupleAndKeywords(args, kwargs, "ss|s", const_cast<char**>(keywords),
- &log_level, &log_output, &log_format)) {
+ if (!PyArg_ParseTupleAndKeywords(args, kwargs, "ss|s",
+ const_cast<char**>(keywords), &log_level,
+ &log_output, &log_format)) {
return nullptr;
}
@@ -1162,7 +1187,8 @@ PyObject* py_index_engine_add_data(PyObject*, PyObject* args) {
}
try {
- const int result = call_without_gil([&]() { return engine->add_data(requests); });
+ const int result =
+ call_without_gil([&]() { return engine->add_data(requests); });
return PyLong_FromLong(result);
} catch (const std::exception& exc) {
raise_runtime_error(exc.what());
@@ -1188,7 +1214,8 @@ PyObject* py_index_engine_delete_data(PyObject*, PyObject* args) {
}
try {
- const int result = call_without_gil([&]() { return engine->delete_data(requests); });
+ const int result =
+ call_without_gil([&]() { return engine->delete_data(requests); });
return PyLong_FromLong(result);
} catch (const std::exception& exc) {
raise_runtime_error(exc.what());
@@ -1214,7 +1241,8 @@ PyObject* py_index_engine_search(PyObject*, PyObject* args) {
}
try {
- const vdb::SearchResult result = call_without_gil([&]() { return engine->search(request); });
+ const vdb::SearchResult result =
+ call_without_gil([&]() { return engine->search(request); });
return build_search_result(result);
} catch (const std::exception& exc) {
raise_runtime_error(exc.what());
@@ -1235,7 +1263,8 @@ PyObject* py_index_engine_dump(PyObject*, PyObject* args) {
}
try {
- const int64_t result = call_without_gil([&]() { return engine->dump(path); });
+ const int64_t result =
+ call_without_gil([&]() { return engine->dump(path); });
return PyLong_FromLongLong(result);
} catch (const std::exception& exc) {
raise_runtime_error(exc.what());
@@ -1255,7 +1284,8 @@ PyObject* py_index_engine_get_state(PyObject*, PyObject* args) {
}
try {
- const vdb::StateResult result = call_without_gil([&]() { return engine->get_state(); });
+ const vdb::StateResult result =
+ call_without_gil([&]() { return engine->get_state(); });
return build_state_result(result);
} catch (const std::exception& exc) {
raise_runtime_error(exc.what());
@@ -1332,15 +1362,16 @@ PyObject* py_store_get_data(PyObject*, PyObject* args) {
}
try {
- const auto values = call_without_gil([&]() { return store->get_data(keys); });
+ const auto values =
+ call_without_gil([&]() { return store->get_data(keys); });
PyObject* list = PyList_New(static_cast<Py_ssize_t>(values.size()));
if (list == nullptr) {
return nullptr;
}
for (Py_ssize_t i = 0; i < static_cast<Py_ssize_t>(values.size()); ++i) {
const auto& value = values[static_cast<size_t>(i)];
- PyObject* item = PyBytes_FromStringAndSize(value.data(),
- static_cast<Py_ssize_t>(value.size()));
+ PyObject* item = PyBytes_FromStringAndSize(
+ value.data(), static_cast<Py_ssize_t>(value.size()));
if (item == nullptr) {
Py_DECREF(list);
return nullptr;
@@ -1369,7 +1400,8 @@ PyObject* py_store_put_data(PyObject*, PyObject* args) {
std::vector<std::string> keys;
std::vector<std::string> values;
- if (!parse_string_list(keys_obj, &keys) || !parse_binary_list(values_obj, &values)) {
+ if (!parse_string_list(keys_obj, &keys) ||
+ !parse_binary_list(values_obj, &values)) {
return nullptr;
}
if (keys.size() != values.size()) {
@@ -1378,7 +1410,8 @@ PyObject* py_store_put_data(PyObject*, PyObject* args) {
}
try {
- const int result = call_without_gil([&]() { return store->put_data(keys, values); });
+ const int result =
+ call_without_gil([&]() { return store->put_data(keys, values); });
return PyLong_FromLong(result);
} catch (const std::exception& exc) {
raise_runtime_error(exc.what());
@@ -1404,7 +1437,8 @@ PyObject* py_store_delete_data(PyObject*, PyObject* args) {
}
try {
- const int result = call_without_gil([&]() { return store->delete_data(keys); });
+ const int result =
+ call_without_gil([&]() { return store->delete_data(keys); });
return PyLong_FromLong(result);
} catch (const std::exception& exc) {
raise_runtime_error(exc.what());
@@ -1446,7 +1480,8 @@ PyObject* py_store_seek_range(PyObject*, PyObject* args) {
}
try {
- const auto items = call_without_gil([&]() { return store->seek_range(start_key, end_key); });
+ const auto items = call_without_gil(
+ [&]() { return store->seek_range(start_key, end_key); });
PyObject* list = PyList_New(static_cast<Py_ssize_t>(items.size()));
if (list == nullptr) {
return nullptr;
@@ -1454,10 +1489,10 @@ PyObject* py_store_seek_range(PyObject*, PyObject* args) {
for (Py_ssize_t i = 0; i < static_cast<Py_ssize_t>(items.size()); ++i) {
const auto& item = items[static_cast<size_t>(i)];
PyObject* tuple = PyTuple_New(2);
- PyObject* key = PyUnicode_FromStringAndSize(item.first.data(),
- static_cast<Py_ssize_t>(item.first.size()));
- PyObject* value = PyBytes_FromStringAndSize(item.second.data(),
- static_cast<Py_ssize_t>(item.second.size()));
+ PyObject* key = PyUnicode_FromStringAndSize(
+ item.first.data(), static_cast<Py_ssize_t>(item.first.size()));
+ PyObject* value = PyBytes_FromStringAndSize(
+ item.second.data(), static_cast<Py_ssize_t>(item.second.size()));
if (tuple == nullptr || key == nullptr || value == nullptr) {
Py_XDECREF(tuple);
Py_XDECREF(key);
@@ -1478,36 +1513,47 @@ PyObject* py_store_seek_range(PyObject*, PyObject* args) {
PyMethodDef kModuleMethods[] = {
{"_new_schema", py_new_schema, METH_VARARGS, "Create a schema handle."},
- {"_schema_get_total_byte_length", py_schema_get_total_byte_length, METH_VARARGS,
- "Read total schema byte length."},
- {"_new_bytes_row", py_new_bytes_row, METH_VARARGS, "Create a BytesRow handle."},
+ {"_schema_get_total_byte_length", py_schema_get_total_byte_length,
+ METH_VARARGS, "Read total schema byte length."},
+ {"_new_bytes_row", py_new_bytes_row, METH_VARARGS,
+ "Create a BytesRow handle."},
{"_bytes_row_serialize", py_bytes_row_serialize, METH_VARARGS,
"Serialize a row using the native BytesRow implementation."},
{"_bytes_row_serialize_batch", py_bytes_row_serialize_batch, METH_VARARGS,
"Serialize a batch of rows using the native BytesRow implementation."},
{"_bytes_row_deserialize", py_bytes_row_deserialize, METH_VARARGS,
"Deserialize a row using the native BytesRow implementation."},
- {"_bytes_row_deserialize_field", py_bytes_row_deserialize_field, METH_VARARGS,
+ {"_bytes_row_deserialize_field", py_bytes_row_deserialize_field,
+ METH_VARARGS,
"Deserialize a single field using the native BytesRow implementation."},
{"_init_logging", reinterpret_cast<PyCFunction>(py_init_logging),
METH_VARARGS | METH_KEYWORDS, "Initialize vectordb logging."},
- {"_new_index_engine", py_new_index_engine, METH_VARARGS, "Create an index engine handle."},
+ {"_new_index_engine", py_new_index_engine, METH_VARARGS,
+ "Create an index engine handle."},
{"_index_engine_add_data", py_index_engine_add_data, METH_VARARGS,
"Add data to the index engine."},
{"_index_engine_delete_data", py_index_engine_delete_data, METH_VARARGS,
"Delete data from the index engine."},
{"_index_engine_search", py_index_engine_search, METH_VARARGS,
"Search the index engine."},
- {"_index_engine_dump", py_index_engine_dump, METH_VARARGS, "Dump index state to disk."},
+ {"_index_engine_dump", py_index_engine_dump, METH_VARARGS,
+ "Dump index state to disk."},
{"_index_engine_get_state", py_index_engine_get_state, METH_VARARGS,
"Read index engine state."},
- {"_new_persist_store", py_new_persist_store, METH_VARARGS, "Create a persistent store."},
- {"_new_volatile_store", py_new_volatile_store, METH_NOARGS, "Create a volatile store."},
- {"_store_exec_op", py_store_exec_op, METH_VARARGS, "Execute store operations."},
- {"_store_get_data", py_store_get_data, METH_VARARGS, "Read values from the store."},
- {"_store_put_data", py_store_put_data, METH_VARARGS, "Write values to the store."},
- {"_store_delete_data", py_store_delete_data, METH_VARARGS, "Delete keys from the store."},
- {"_store_clear_data", py_store_clear_data, METH_VARARGS, "Clear the store."},
+ {"_new_persist_store", py_new_persist_store, METH_VARARGS,
+ "Create a persistent store."},
+ {"_new_volatile_store", py_new_volatile_store, METH_NOARGS,
+ "Create a volatile store."},
+ {"_store_exec_op", py_store_exec_op, METH_VARARGS,
+ "Execute store operations."},
+ {"_store_get_data", py_store_get_data, METH_VARARGS,
+ "Read values from the store."},
+ {"_store_put_data", py_store_put_data, METH_VARARGS,
+ "Write values to the store."},
+ {"_store_delete_data", py_store_delete_data, METH_VARARGS,
+ "Delete keys from the store."},
+ {"_store_clear_data", py_store_clear_data, METH_VARARGS,
+ "Clear the store."},
{"_store_seek_range", py_store_seek_range, METH_VARARGS,
"Read a key range from the store."},
{nullptr, nullptr, 0, nullptr},
@@ -1536,7 +1582,8 @@ PyMODINIT_FUNC OV_CONCAT(PyInit_, OV_PY_MODULE_NAME)(void) {
return nullptr;
}
- if (PyModule_AddStringConstant(module, "_ENGINE_BACKEND_API", "abi3-v1") < 0) {
+ if (PyModule_AddStringConstant(module, "_ENGINE_BACKEND_API", "abi3-v1") <
+ 0) {
Py_DECREF(module);
return nullptr;
}
diff --git a/src/abi3_x86_caps.cpp b/src/abi3_x86_caps.cpp
index f6aff1a3d..1b3033442 100644
--- a/src/abi3_x86_caps.cpp
+++ b/src/abi3_x86_caps.cpp
@@ -6,7 +6,8 @@
#include <string>
#include <vector>
-#if defined(__x86_64__) || defined(_M_X64) || defined(__i386__) || defined(_M_IX86)
+#if defined(__x86_64__) || defined(_M_X64) || defined(__i386__) || \
+ defined(_M_IX86)
#if defined(_MSC_VER)
#include <immintrin.h>
#include <intrin.h>
@@ -28,7 +29,8 @@ struct CpuFeatures {
bool avx512vl = false;
};
-#if defined(__x86_64__) || defined(_M_X64) || defined(__i386__) || defined(_M_IX86)
+#if defined(__x86_64__) || defined(_M_X64) || defined(__i386__) || \
+ defined(_M_IX86)
void cpuid(int regs[4], int leaf, int subleaf) {
#if defined(_MSC_VER)
__cpuidex(regs, leaf, subleaf);
@@ -89,7 +91,9 @@ CpuFeatures detect_cpu_features() {
return features;
}
#else
-CpuFeatures detect_cpu_features() { return CpuFeatures{}; }
+CpuFeatures detect_cpu_features() {
+ return CpuFeatures{};
+}
#endif
std::vector<std::string> get_supported_variants_impl() {
diff --git a/src/common/zip_sort.h b/src/common/zip_sort.h
index bb3aa1799..b91fbb2c2 100644
--- a/src/common/zip_sort.h
+++ b/src/common/zip_sort.h
@@ -48,4 +48,4 @@ void ZipSortBranchOptimized(Comparator comp, T begin, T end, U begin1, U end1) {
rearrange(begin1);
}
-}
+} // namespace vectordb
diff --git a/src/index/detail/index_manager_impl.cpp b/src/index/detail/index_manager_impl.cpp
index 4d165d10a..68bc6385a 100644
--- a/src/index/detail/index_manager_impl.cpp
+++ b/src/index/detail/index_manager_impl.cpp
@@ -345,7 +345,7 @@ int IndexManagerImpl::delete_data(
int offset = vector_index_->get_offset_by_label(data.label);
if (offset < 0) {
SPDLOG_DEBUG("IndexManagerImpl::delete_data label={} not found",
- data.label);
+ data.label);
continue;
} else {
has_update = true;
diff --git a/src/index/detail/scalar/bitmap_holder/bitmap_field_group.cpp b/src/index/detail/scalar/bitmap_holder/bitmap_field_group.cpp
index b2e429842..648ed55e3 100644
--- a/src/index/detail/scalar/bitmap_holder/bitmap_field_group.cpp
+++ b/src/index/detail/scalar/bitmap_holder/bitmap_field_group.cpp
@@ -174,7 +174,8 @@ int BitmapGroupBase::count_field_enums(
int BitmapGroupBase::count_field_enums(
std::map<std::string, uint32_t>& enum_count,
std::map<std::string, const BitmapPtr>& first_enum_bitmaps) {
- // Combine second field with first field distribution to get joint key distribution
+ // Combine second field with first field distribution to get joint key
+ // distribution
for (auto& pkv : first_enum_bitmaps) {
BitmapPtr valid_bitmap = pkv.second;
if (valid_bitmap) {
diff --git a/src/index/detail/scalar/bitmap_holder/bitmap_field_group.h b/src/index/detail/scalar/bitmap_holder/bitmap_field_group.h
index eaa658eb9..eef770250 100644
--- a/src/index/detail/scalar/bitmap_holder/bitmap_field_group.h
+++ b/src/index/detail/scalar/bitmap_holder/bitmap_field_group.h
@@ -135,7 +135,8 @@ class FieldBitmapGroup : public BitmapGroupBase {
std::vector<std::string> keys;
split(keys, field_str, ";");
for (auto& key_i : keys) {
- const std::string norm_key = dir_index_ ? normalize_path_key(key_i) : key_i;
+ const std::string norm_key =
+ dir_index_ ? normalize_path_key(key_i) : key_i;
if (!exist_bitmap(norm_key)) {
if (dir_index_) {
dir_index_->add_key(norm_key);
@@ -149,7 +150,8 @@ class FieldBitmapGroup : public BitmapGroupBase {
}
} else {
- const std::string norm_key = dir_index_ ? normalize_path_key(field_str) : field_str;
+ const std::string norm_key =
+ dir_index_ ? normalize_path_key(field_str) : field_str;
if (!exist_bitmap(norm_key)) {
if (dir_index_) {
dir_index_->add_key(norm_key);
@@ -209,14 +211,16 @@ class FieldBitmapGroup : public BitmapGroupBase {
std::vector<std::string> keys;
split(keys, field_str, ";");
for (auto& key_i : keys) {
- const std::string norm_key = dir_index_ ? normalize_path_key(key_i) : key_i;
+ const std::string norm_key =
+ dir_index_ ? normalize_path_key(key_i) : key_i;
Bitmap* temp_p = get_editable_bitmap(norm_key);
if (temp_p) {
temp_p->Unset(offset);
}
}
} else {
- const std::string norm_key = dir_index_ ? normalize_path_key(field_str) : field_str;
+ const std::string norm_key =
+ dir_index_ ? normalize_path_key(field_str) : field_str;
Bitmap* temp_p = get_editable_bitmap(norm_key);
if (temp_p) {
temp_p->Unset(offset);
@@ -288,7 +292,8 @@ using FieldBitmapGroupPtr = std::shared_ptr<FieldBitmapGroup>;
class FieldBitmapGroupSet;
using FieldBitmapGroupSetPtr = std::shared_ptr<FieldBitmapGroupSet>;
-// Collection of all field inverted index groups, encapsulates bitmap filter computation
+// Collection of all field inverted index groups, encapsulates bitmap filter
+// computation
class FieldBitmapGroupSet {
public:
explicit FieldBitmapGroupSet(std::string grp_set_name = std::string{})
@@ -479,8 +484,8 @@ class FieldBitmapGroupSet {
double greater_than, bool include_ge) {
auto itr = field_bitmap_groups_map_.find(field);
if (itr == field_bitmap_groups_map_.end()) {
- SPDLOG_ERROR("make_range_copy failed, cannot find {} in {}\n",
- field, field_bitmap_groups_map_.size());
+ SPDLOG_ERROR("make_range_copy failed, cannot find {} in {}\n", field,
+ field_bitmap_groups_map_.size());
return nullptr;
}
return itr->second->get_bitmap_in_range(range_out, lower_than, include_le,
diff --git a/src/index/detail/scalar/filter/op_base.cpp b/src/index/detail/scalar/filter/op_base.cpp
index 6df642350..c41a58570 100644
--- a/src/index/detail/scalar/filter/op_base.cpp
+++ b/src/index/detail/scalar/filter/op_base.cpp
@@ -28,7 +28,8 @@ int parse_and_precheck_op_parts(JsonDoc& json_doc, bool& has_filter,
}
if (!has_any) {
- // Backward compatibility: default to filter when no top-level keyword exists
+ // Backward compatibility: default to filter when no top-level keyword
+ // exists
has_filter = true;
}
return 0;
diff --git a/src/index/detail/scalar/filter/sort_ops.cpp b/src/index/detail/scalar/filter/sort_ops.cpp
index d117cec4e..d8746fad3 100644
--- a/src/index/detail/scalar/filter/sort_ops.cpp
+++ b/src/index/detail/scalar/filter/sort_ops.cpp
@@ -587,7 +587,7 @@ RecallResultPtr SorterOp::_calc_topk_result_with_small_bitmap(
}
SPDLOG_DEBUG("SorterOp::_calc_topk_result_with_small_bitmap topk {} in {}",
- search_k, all_valid_size);
+ search_k, all_valid_size);
return res_ptr;
}
diff --git a/src/index/detail/vector/common/bruteforce.h b/src/index/detail/vector/common/bruteforce.h
index cb3380691..3d0a45371 100644
--- a/src/index/detail/vector/common/bruteforce.h
+++ b/src/index/detail/vector/common/bruteforce.h
@@ -81,9 +81,9 @@ class BruteforceSearch {
index = it->second;
if (sparse_index_) {
if (sparse_dp) {
- if (sparse_index_->update_low_level_sparse(index, sparse_dp) != 0) {
- throw std::runtime_error("Failed to update sparse data");
- }
+ if (sparse_index_->update_low_level_sparse(index, sparse_dp) != 0) {
+ throw std::runtime_error("Failed to update sparse data");
+ }
}
}
} else {
@@ -91,16 +91,17 @@ class BruteforceSearch {
if (current_count_ != sparse_index_->rows()) {
throw std::runtime_error("Sparse/Dense index inconsistency");
}
-
+
if (sparse_dp) {
- if (sparse_index_->append_low_level_sparse(sparse_dp) != 0) {
- throw std::runtime_error("Failed to append sparse data");
- }
+ if (sparse_index_->append_low_level_sparse(sparse_dp) != 0) {
+ throw std::runtime_error("Failed to append sparse data");
+ }
} else {
- auto empty_dp = std::make_shared<SparseDatapoint>(std::vector<IndexT>(), std::vector<float>());
- if (sparse_index_->append_low_level_sparse(empty_dp) != 0) {
- throw std::runtime_error("Failed to append empty sparse data");
- }
+ auto empty_dp = std::make_shared<SparseDatapoint>(
+ std::vector<IndexT>(), std::vector<float>());
+ if (sparse_index_->append_low_level_sparse(empty_dp) != 0) {
+ throw std::runtime_error("Failed to append empty sparse data");
+ }
}
}
@@ -160,10 +161,12 @@ class BruteforceSearch {
offset_map_[offset_moved] = idx_to_remove;
if (sparse_index_) {
- // SPDLOG_INFO("remove_point: swapping sparse row {} with {}", idx_to_remove, idx_last);
+ // SPDLOG_INFO("remove_point: swapping sparse row {} with {}",
+ // idx_to_remove, idx_last);
auto last_row = sparse_index_->get_row(idx_last);
- if (sparse_index_->update_low_level_sparse(idx_to_remove, last_row) != 0) {
- SPDLOG_ERROR("Failed to update sparse data during remove");
+ if (sparse_index_->update_low_level_sparse(idx_to_remove, last_row) !=
+ 0) {
+ SPDLOG_ERROR("Failed to update sparse data during remove");
}
}
}
@@ -171,7 +174,7 @@ class BruteforceSearch {
if (sparse_index_) {
// SPDLOG_INFO("remove_point: popping back sparse row");
if (sparse_index_->pop_back() != 0) {
- SPDLOG_ERROR("Failed to pop back sparse data during remove");
+ SPDLOG_ERROR("Failed to pop back sparse data during remove");
}
}
@@ -199,7 +202,9 @@ class BruteforceSearch {
encoded_query.data());
using ResultPair = std::pair<float, uint64_t>;
- std::priority_queue<ResultPair, std::vector<ResultPair>, std::greater<ResultPair>> pq;
+ std::priority_queue<ResultPair, std::vector<ResultPair>,
+ std::greater<ResultPair>>
+ pq;
auto dist_func = space_->get_metric_function();
void* dist_params = space_->get_metric_params();
@@ -223,9 +228,9 @@ class BruteforceSearch {
}
} else {
if (filter_bitmap->empty()) {
- labels.clear();
- scores.clear();
- return;
+ labels.clear();
+ scores.clear();
+ return;
}
std::vector<uint32_t> offsets;
filter_bitmap->get_set_list(offsets);
@@ -303,11 +308,11 @@ class BruteforceSearch {
resize_buffer(loaded_cap);
in.read(data_buffer_, loaded_cap * loaded_elem_size);
-
+
if (in.peek() != EOF) {
- read_binary(in, next_logical_offset_);
+ read_binary(in, next_logical_offset_);
} else {
- next_logical_offset_ = 0;
+ next_logical_offset_ = 0;
}
rebuild_maps();
@@ -396,7 +401,7 @@ class BruteforceSearch {
}
}
if (current_count_ > 0 && next_logical_offset_ <= max_offset) {
- next_logical_offset_ = max_offset + 1;
+ next_logical_offset_ = max_offset + 1;
}
}
@@ -412,12 +417,10 @@ class BruteforceSearch {
float compute_score(
const void* encoded_query, const char* data_ptr,
- const std::shared_ptr<SparseDatapointView>& query_sparse_view,
- size_t idx, MetricFunc<float> dist_func,
- void* dist_params) const {
+ const std::shared_ptr<SparseDatapointView>& query_sparse_view, size_t idx,
+ MetricFunc<float> dist_func, void* dist_params) const {
float dense_raw = dist_func(encoded_query, data_ptr, dist_params);
- float dense_score =
- reverse_query_score_ ? (1.0f - dense_raw) : dense_raw;
+ float dense_score = reverse_query_score_ ? (1.0f - dense_raw) : dense_raw;
if (!sparse_index_ || !query_sparse_view ||
meta_->search_with_sparse_logit_alpha <= 0.0f) {
return dense_score;
diff --git a/src/index/detail/vector/sparse_retrieval/sparse_data_holder.h b/src/index/detail/vector/sparse_retrieval/sparse_data_holder.h
index f7685e961..3597fad23 100644
--- a/src/index/detail/vector/sparse_retrieval/sparse_data_holder.h
+++ b/src/index/detail/vector/sparse_retrieval/sparse_data_holder.h
@@ -65,7 +65,8 @@ class SparseDataHolder {
index_with_sparse_bias_ =
(index_with_sparse_bias_alpha != 0.0 ? true : false);
search_with_sparse_bias_ =
- (search_with_sparse_bias_alpha != 0.0 ? true : false); // For compatibility
+ (search_with_sparse_bias_alpha != 0.0 ? true
+ : false); // For compatibility
index_with_sparse_bias_alpha_ = index_with_sparse_bias_alpha;
search_with_sparse_bias_alpha_ = search_with_sparse_bias_alpha;
if (search_use_l2) {
diff --git a/src/index/detail/vector/sparse_retrieval/sparse_dataset.h b/src/index/detail/vector/sparse_retrieval/sparse_dataset.h
index f43379356..4de65023e 100644
--- a/src/index/detail/vector/sparse_retrieval/sparse_dataset.h
+++ b/src/index/detail/vector/sparse_retrieval/sparse_dataset.h
@@ -131,35 +131,38 @@ class SparseDataset {
}
size_t start = offsets_[i];
size_t end = offsets_[i + 1];
-
+
std::vector<IndexT> indices;
std::vector<float> values;
-
+
if (start < end) {
- indices.assign(flat_indices_.begin() + start, flat_indices_.begin() + end);
+ indices.assign(flat_indices_.begin() + start,
+ flat_indices_.begin() + end);
values.assign(flat_values_.begin() + start, flat_values_.begin() + end);
}
-
- return std::make_shared<SparseDatapoint>(std::move(indices), std::move(values));
+
+ return std::make_shared<SparseDatapoint>(std::move(indices),
+ std::move(values));
}
int pop_back() {
if (offsets_.size() <= 1) {
- return 0; // Already empty (just initial offset 0)
+ return 0; // Already empty (just initial offset 0)
}
-
+
size_t last_start = offsets_[offsets_.size() - 2];
size_t last_end = offsets_.back();
size_t count = last_end - last_start;
-
+
if (count > 0) {
- flat_indices_.erase(flat_indices_.begin() + last_start, flat_indices_.end());
+ flat_indices_.erase(flat_indices_.begin() + last_start,
+ flat_indices_.end());
flat_values_.erase(flat_values_.begin() + last_start, flat_values_.end());
}
-
+
offsets_.pop_back();
entries_ -= count;
-
+
return 0;
}
diff --git a/src/store/bytes_row.cpp b/src/store/bytes_row.cpp
index a3c181eef..8f2b0e048 100644
--- a/src/store/bytes_row.cpp
+++ b/src/store/bytes_row.cpp
@@ -31,8 +31,7 @@ Schema::Schema(const std::vector<FieldDef>& fields) {
}
if (max_id != static_cast<int>(fields.size()) - 1) {
- throw std::invalid_argument(
- "Field ids must be contiguous from 0 to N-1");
+ throw std::invalid_argument("Field ids must be contiguous from 0 to N-1");
}
std::vector<bool> seen(fields.size(), false);
diff --git a/src/store/bytes_row.h b/src/store/bytes_row.h
index db1ab0e7b..a03832392 100644
--- a/src/store/bytes_row.h
+++ b/src/store/bytes_row.h
@@ -11,364 +11,392 @@
namespace vectordb {
enum class FieldType {
- INT64 = 0,
- UINT64 = 1,
- FLOAT32 = 2,
- STRING = 3,
- BINARY = 4,
- BOOLEAN = 5,
- LIST_INT64 = 6,
- LIST_STRING = 7,
- LIST_FLOAT32 = 8
+ INT64 = 0,
+ UINT64 = 1,
+ FLOAT32 = 2,
+ STRING = 3,
+ BINARY = 4,
+ BOOLEAN = 5,
+ LIST_INT64 = 6,
+ LIST_STRING = 7,
+ LIST_FLOAT32 = 8
};
// Use std::monostate for None/Null
-using Value = std::variant<std::monostate, int64_t, uint64_t, float, std::string, bool, std::vector<int64_t>, std::vector<std::string>, std::vector<float>>;
+using Value = std::variant<std::monostate, int64_t, uint64_t, float,
+ std::string, bool, std::vector<int64_t>,
+ std::vector<std::string>, std::vector<float>>;
struct FieldMeta {
- std::string name;
- FieldType data_type;
- int offset;
- int id;
- Value default_value;
+ std::string name;
+ FieldType data_type;
+ int offset;
+ int id;
+ Value default_value;
};
struct FieldDef {
- std::string name;
- FieldType data_type;
- int id;
- Value default_value;
+ std::string name;
+ FieldType data_type;
+ int id;
+ Value default_value;
};
class Schema {
-public:
- Schema(const std::vector<FieldDef>& fields);
-
- const std::vector<FieldMeta>& get_field_order() const { return field_orders_; }
- int get_total_byte_length() const { return total_byte_length_; }
- const FieldMeta* get_field_meta(const std::string& name) const;
-
-private:
- std::vector<FieldMeta> field_orders_;
- std::map<std::string, FieldMeta> field_metas_;
- int total_byte_length_;
+ public:
+ Schema(const std::vector<FieldDef>& fields);
+
+ const std::vector<FieldMeta>& get_field_order() const {
+ return field_orders_;
+ }
+ int get_total_byte_length() const {
+ return total_byte_length_;
+ }
+ const FieldMeta* get_field_meta(const std::string& name) const;
+
+ private:
+ std::vector<FieldMeta> field_orders_;
+ std::map<std::string, FieldMeta> field_metas_;
+ int total_byte_length_;
};
class BytesRow {
-public:
- explicit BytesRow(std::shared_ptr<Schema> schema);
-
- // Core serialization logic: takes values in order of schema fields
- std::string serialize(const std::vector<Value>& row_data) const;
-
- // Generic serialization template
- // Accessor must implement methods like:
- // int64_t get_int64(const RowT& row, int field_idx)
- // ... and so on for all types, plus is_null/has_value check
- template <typename RowT, typename AccessorT>
- std::string serialize_template(const RowT& row, const AccessorT& accessor) const {
- const auto& field_order = schema_->get_field_order();
-
- auto get_default_string = [](const Value& val) -> const std::string* {
- if (std::holds_alternative<std::string>(val)) {
- return &std::get<std::string>(val);
- }
- return nullptr;
- };
+ public:
+ explicit BytesRow(std::shared_ptr<Schema> schema);
- auto get_default_list_int64 = [](const Value& val) -> const std::vector<int64_t>* {
- if (std::holds_alternative<std::vector<int64_t>>(val)) {
- return &std::get<std::vector<int64_t>>(val);
- }
- return nullptr;
- };
+ // Core serialization logic: takes values in order of schema fields
+ std::string serialize(const std::vector<Value>& row_data) const;
- auto get_default_list_float32 = [](const Value& val) -> const std::vector<float>* {
- if (std::holds_alternative<std::vector<float>>(val)) {
- return &std::get<std::vector<float>>(val);
- }
- return nullptr;
- };
+ // Generic serialization template
+ // Accessor must implement methods like:
+ // int64_t get_int64(const RowT& row, int field_idx)
+ // ... and so on for all types, plus is_null/has_value check
+ template <typename RowT, typename AccessorT>
+ std::string serialize_template(const RowT& row,
+ const AccessorT& accessor) const {
+ const auto& field_order = schema_->get_field_order();
- auto get_default_list_string = [](const Value& val) -> const std::vector<std::string>* {
- if (std::holds_alternative<std::vector<std::string>>(val)) {
- return &std::get<std::vector<std::string>>(val);
- }
- return nullptr;
- };
+ auto get_default_string = [](const Value& val) -> const std::string* {
+ if (std::holds_alternative<std::string>(val)) {
+ return &std::get<std::string>(val);
+ }
+ return nullptr;
+ };
- auto get_list_string_content_len = [](const std::vector<std::string>& vec) -> int {
- int total = 0;
- for (const auto& s : vec) {
- total += static_cast<int>(s.length());
- }
- return total;
- };
-
- auto write_string_value = [](const std::string& s, char* dest) {
- uint16_t len = static_cast<uint16_t>(s.length());
- std::memcpy(dest, &len, 2);
- if (len > 0) {
- std::memcpy(dest + 2, s.data(), len);
- }
- };
+ auto get_default_list_int64 =
+ [](const Value& val) -> const std::vector<int64_t>* {
+ if (std::holds_alternative<std::vector<int64_t>>(val)) {
+ return &std::get<std::vector<int64_t>>(val);
+ }
+ return nullptr;
+ };
- auto write_binary_value = [](const std::string& s, char* dest) {
- uint32_t len = static_cast<uint32_t>(s.length());
- std::memcpy(dest, &len, 4);
- if (len > 0) {
- std::memcpy(dest + 4, s.data(), len);
- }
- };
+ auto get_default_list_float32 =
+ [](const Value& val) -> const std::vector<float>* {
+ if (std::holds_alternative<std::vector<float>>(val)) {
+ return &std::get<std::vector<float>>(val);
+ }
+ return nullptr;
+ };
- auto write_list_int64_value = [](const std::vector<int64_t>& vec, char* dest) {
- uint16_t len = static_cast<uint16_t>(vec.size());
- std::memcpy(dest, &len, 2);
- if (len > 0) {
- std::memcpy(dest + 2, vec.data(), len * sizeof(int64_t));
- }
- };
+ auto get_default_list_string =
+ [](const Value& val) -> const std::vector<std::string>* {
+ if (std::holds_alternative<std::vector<std::string>>(val)) {
+ return &std::get<std::vector<std::string>>(val);
+ }
+ return nullptr;
+ };
+
+ auto get_list_string_content_len =
+ [](const std::vector<std::string>& vec) -> int {
+ int total = 0;
+ for (const auto& s : vec) {
+ total += static_cast<int>(s.length());
+ }
+ return total;
+ };
- auto write_list_float32_value = [](const std::vector<float>& vec, char* dest) {
- uint16_t len = static_cast<uint16_t>(vec.size());
- std::memcpy(dest, &len, 2);
- if (len > 0) {
- std::memcpy(dest + 2, vec.data(), len * sizeof(float));
+ auto write_string_value = [](const std::string& s, char* dest) {
+ uint16_t len = static_cast<uint16_t>(s.length());
+ std::memcpy(dest, &len, 2);
+ if (len > 0) {
+ std::memcpy(dest + 2, s.data(), len);
+ }
+ };
+
+ auto write_binary_value = [](const std::string& s, char* dest) {
+ uint32_t len = static_cast<uint32_t>(s.length());
+ std::memcpy(dest, &len, 4);
+ if (len > 0) {
+ std::memcpy(dest + 4, s.data(), len);
+ }
+ };
+
+ auto write_list_int64_value = [](const std::vector<int64_t>& vec,
+ char* dest) {
+ uint16_t len = static_cast<uint16_t>(vec.size());
+ std::memcpy(dest, &len, 2);
+ if (len > 0) {
+ std::memcpy(dest + 2, vec.data(), len * sizeof(int64_t));
+ }
+ };
+
+ auto write_list_float32_value = [](const std::vector<float>& vec,
+ char* dest) {
+ uint16_t len = static_cast<uint16_t>(vec.size());
+ std::memcpy(dest, &len, 2);
+ if (len > 0) {
+ std::memcpy(dest + 2, vec.data(), len * sizeof(float));
+ }
+ };
+
+ auto write_list_string_value = [](const std::vector<std::string>& vec,
+ char* dest) {
+ uint16_t len = static_cast<uint16_t>(vec.size());
+ std::memcpy(dest, &len, 2);
+ char* cur = dest + 2;
+ for (const auto& s : vec) {
+ uint16_t slen = static_cast<uint16_t>(s.length());
+ std::memcpy(cur, &slen, 2);
+ cur += 2;
+ if (slen > 0) {
+ std::memcpy(cur, s.data(), slen);
+ }
+ cur += slen;
+ }
+ };
+
+ // Pass 1: Calculate total size
+ int total_size = schema_->get_total_byte_length();
+ int variable_region_offset = total_size;
+
+ struct VarFieldInfo {
+ int offset;
+ int length;
+ };
+ // Use a small buffer on stack if possible, or vector
+ std::vector<VarFieldInfo> var_infos(field_order.size());
+
+ for (size_t i = 0; i < field_order.size(); ++i) {
+ const auto& meta = field_order[i];
+
+ // For variable fields, we need to check length
+ switch (meta.data_type) {
+ case FieldType::STRING: {
+ int len = 0;
+ if (accessor.has_value(row, i)) {
+ len = accessor.get_string_len(row, i);
+ } else if (const auto* def = get_default_string(meta.default_value)) {
+ len = static_cast<int>(def->length());
+ }
+ var_infos[i] = {variable_region_offset, len};
+ variable_region_offset += 2 + len; // UINT16_SIZE
+ break;
+ }
+ case FieldType::BINARY: {
+ int len = 0;
+ if (accessor.has_value(row, i)) {
+ len = accessor.get_binary_len(row, i);
+ } else if (const auto* def = get_default_string(meta.default_value)) {
+ len = static_cast<int>(def->length());
+ }
+ var_infos[i] = {variable_region_offset, len};
+ variable_region_offset += 4 + len; // UINT32_SIZE
+ break;
+ }
+ case FieldType::LIST_INT64: {
+ int len = 0;
+ if (accessor.has_value(row, i)) {
+ len = accessor.get_list_len(row, i);
+ } else if (const auto* def =
+ get_default_list_int64(meta.default_value)) {
+ len = static_cast<int>(def->size());
+ }
+ var_infos[i] = {variable_region_offset, len};
+ variable_region_offset += 2 + len * 8; // UINT16 + INT64_SIZE
+ break;
+ }
+ case FieldType::LIST_FLOAT32: {
+ int len = 0;
+ if (accessor.has_value(row, i)) {
+ len = accessor.get_list_len(row, i);
+ } else if (const auto* def =
+ get_default_list_float32(meta.default_value)) {
+ len = static_cast<int>(def->size());
+ }
+ var_infos[i] = {variable_region_offset, len};
+ variable_region_offset += 2 + len * 4; // UINT16 + FLOAT32_SIZE
+ break;
+ }
+ case FieldType::LIST_STRING: {
+ int list_len = 0;
+ int content_len = 0;
+ if (accessor.has_value(row, i)) {
+ list_len = accessor.get_list_len(row, i);
+ content_len = accessor.get_list_string_content_len(row, i);
+ } else if (const auto* def =
+ get_default_list_string(meta.default_value)) {
+ list_len = static_cast<int>(def->size());
+ content_len = get_list_string_content_len(*def);
+ }
+ var_infos[i] = {variable_region_offset, list_len};
+ // list_len(2) + (elem_len(2) + content) * N
+ // Actually content_len should include the 2 bytes for each string
+ // length if we compute it that way Or we compute it here: 2 + (2 *
+ // list_len) + total_string_bytes
+ variable_region_offset += 2 + (2 * list_len) + content_len;
+ break;
+ }
+ default:
+ break;
+ }
+ }
+
+ std::string buffer;
+ buffer.resize(variable_region_offset);
+ char* ptr = &buffer[0];
+
+ // Header
+ ptr[0] = static_cast<uint8_t>(field_order.size());
+
+ // Pass 2: Write data
+ for (size_t i = 0; i < field_order.size(); ++i) {
+ const auto& meta = field_order[i];
+ char* field_ptr = ptr + meta.offset;
+ bool has_val = accessor.has_value(row, i);
+
+ switch (meta.data_type) {
+ case FieldType::INT64: {
+ int64_t v = 0;
+ if (has_val) {
+ v = accessor.get_int64(row, i);
+ } else if (std::holds_alternative<int64_t>(meta.default_value)) {
+ v = std::get<int64_t>(meta.default_value);
+ } else if (std::holds_alternative<uint64_t>(meta.default_value)) {
+ v = static_cast<int64_t>(std::get<uint64_t>(meta.default_value));
+ }
+ std::memcpy(field_ptr, &v, sizeof(v));
+ break;
+ }
+ case FieldType::UINT64: {
+ uint64_t v = 0;
+ if (has_val) {
+ v = accessor.get_uint64(row, i);
+ } else if (std::holds_alternative<uint64_t>(meta.default_value)) {
+ v = std::get<uint64_t>(meta.default_value);
+ } else if (std::holds_alternative<int64_t>(meta.default_value)) {
+ v = static_cast<uint64_t>(std::get<int64_t>(meta.default_value));
+ }
+ std::memcpy(field_ptr, &v, sizeof(v));
+ break;
+ }
+ case FieldType::FLOAT32: {
+ float v = 0.0f;
+ if (has_val) {
+ v = accessor.get_float(row, i);
+ } else if (std::holds_alternative<float>(meta.default_value)) {
+ v = std::get<float>(meta.default_value);
+ }
+ std::memcpy(field_ptr, &v, sizeof(v));
+ break;
+ }
+ case FieldType::BOOLEAN: {
+ bool v = false;
+ if (has_val) {
+ v = accessor.get_bool(row, i);
+ } else if (std::holds_alternative<bool>(meta.default_value)) {
+ v = std::get<bool>(meta.default_value);
+ }
+ uint8_t b = v ? 1 : 0;
+ std::memcpy(field_ptr, &b, sizeof(b));
+ break;
+ }
+ case FieldType::STRING:
+ case FieldType::BINARY:
+ case FieldType::LIST_INT64:
+ case FieldType::LIST_FLOAT32:
+ case FieldType::LIST_STRING: {
+ uint32_t offset = static_cast<uint32_t>(var_infos[i].offset);
+ std::memcpy(field_ptr, &offset, sizeof(offset));
+
+ char* var_ptr = ptr + offset;
+
+ if (meta.data_type == FieldType::STRING) {
+ // Logic handled by accessor to write directly? Or return
+ // string_view/string? Accessor returning string creates copy.
+ // Better: accessor.write_string(row, i, var_ptr)
+ if (has_val) {
+ accessor.write_string(row, i, var_ptr);
+ } else if (const auto* def =
+ get_default_string(meta.default_value)) {
+ write_string_value(*def, var_ptr);
+ } else {
+ uint16_t len = 0;
+ std::memcpy(var_ptr, &len, 2);
}
- };
-
- auto write_list_string_value = [](const std::vector<std::string>& vec, char* dest) {
- uint16_t len = static_cast<uint16_t>(vec.size());
- std::memcpy(dest, &len, 2);
- char* cur = dest + 2;
- for (const auto& s : vec) {
- uint16_t slen = static_cast<uint16_t>(s.length());
- std::memcpy(cur, &slen, 2);
- cur += 2;
- if (slen > 0) {
- std::memcpy(cur, s.data(), slen);
- }
- cur += slen;
+ } else if (meta.data_type == FieldType::BINARY) {
+ if (has_val) {
+ accessor.write_binary(row, i, var_ptr);
+ } else if (const auto* def =
+ get_default_string(meta.default_value)) {
+ write_binary_value(*def, var_ptr);
+ } else {
+ uint32_t len = 0;
+ std::memcpy(var_ptr, &len, 4);
}
- };
-
- // Pass 1: Calculate total size
- int total_size = schema_->get_total_byte_length();
- int variable_region_offset = total_size;
-
- struct VarFieldInfo {
- int offset;
- int length;
- };
- // Use a small buffer on stack if possible, or vector
- std::vector<VarFieldInfo> var_infos(field_order.size());
-
- for (size_t i = 0; i < field_order.size(); ++i) {
- const auto& meta = field_order[i];
-
- // For variable fields, we need to check length
- switch (meta.data_type) {
- case FieldType::STRING: {
- int len = 0;
- if (accessor.has_value(row, i)) {
- len = accessor.get_string_len(row, i);
- } else if (const auto* def = get_default_string(meta.default_value)) {
- len = static_cast<int>(def->length());
- }
- var_infos[i] = {variable_region_offset, len};
- variable_region_offset += 2 + len; // UINT16_SIZE
- break;
- }
- case FieldType::BINARY: {
- int len = 0;
- if (accessor.has_value(row, i)) {
- len = accessor.get_binary_len(row, i);
- } else if (const auto* def = get_default_string(meta.default_value)) {
- len = static_cast<int>(def->length());
- }
- var_infos[i] = {variable_region_offset, len};
- variable_region_offset += 4 + len; // UINT32_SIZE
- break;
- }
- case FieldType::LIST_INT64: {
- int len = 0;
- if (accessor.has_value(row, i)) {
- len = accessor.get_list_len(row, i);
- } else if (const auto* def = get_default_list_int64(meta.default_value)) {
- len = static_cast<int>(def->size());
- }
- var_infos[i] = {variable_region_offset, len};
- variable_region_offset += 2 + len * 8; // UINT16 + INT64_SIZE
- break;
- }
- case FieldType::LIST_FLOAT32: {
- int len = 0;
- if (accessor.has_value(row, i)) {
- len = accessor.get_list_len(row, i);
- } else if (const auto* def = get_default_list_float32(meta.default_value)) {
- len = static_cast<int>(def->size());
- }
- var_infos[i] = {variable_region_offset, len};
- variable_region_offset += 2 + len * 4; // UINT16 + FLOAT32_SIZE
- break;
- }
- case FieldType::LIST_STRING: {
- int list_len = 0;
- int content_len = 0;
- if (accessor.has_value(row, i)) {
- list_len = accessor.get_list_len(row, i);
- content_len = accessor.get_list_string_content_len(row, i);
- } else if (const auto* def = get_default_list_string(meta.default_value)) {
- list_len = static_cast<int>(def->size());
- content_len = get_list_string_content_len(*def);
- }
- var_infos[i] = {variable_region_offset, list_len};
- // list_len(2) + (elem_len(2) + content) * N
- // Actually content_len should include the 2 bytes for each string length if we compute it that way
- // Or we compute it here: 2 + (2 * list_len) + total_string_bytes
- variable_region_offset += 2 + (2 * list_len) + content_len;
- break;
- }
- default: break;
+ } else if (meta.data_type == FieldType::LIST_INT64) {
+ if (has_val) {
+ accessor.write_list_int64(row, i, var_ptr);
+ } else if (const auto* def =
+ get_default_list_int64(meta.default_value)) {
+ write_list_int64_value(*def, var_ptr);
+ } else {
+ uint16_t len = 0;
+ std::memcpy(var_ptr, &len, 2);
}
- }
-
- std::string buffer;
- buffer.resize(variable_region_offset);
- char* ptr = &buffer[0];
-
- // Header
- ptr[0] = static_cast<uint8_t>(field_order.size());
-
- // Pass 2: Write data
- for (size_t i = 0; i < field_order.size(); ++i) {
- const auto& meta = field_order[i];
- char* field_ptr = ptr + meta.offset;
- bool has_val = accessor.has_value(row, i);
-
- switch (meta.data_type) {
- case FieldType::INT64: {
- int64_t v = 0;
- if (has_val) {
- v = accessor.get_int64(row, i);
- } else if (std::holds_alternative<int64_t>(meta.default_value)) {
- v = std::get<int64_t>(meta.default_value);
- } else if (std::holds_alternative<uint64_t>(meta.default_value)) {
- v = static_cast<int64_t>(std::get<uint64_t>(meta.default_value));
- }
- std::memcpy(field_ptr, &v, sizeof(v));
- break;
- }
- case FieldType::UINT64: {
- uint64_t v = 0;
- if (has_val) {
- v = accessor.get_uint64(row, i);
- } else if (std::holds_alternative<uint64_t>(meta.default_value)) {
- v = std::get<uint64_t>(meta.default_value);
- } else if (std::holds_alternative<int64_t>(meta.default_value)) {
- v = static_cast<uint64_t>(std::get<int64_t>(meta.default_value));
- }
- std::memcpy(field_ptr, &v, sizeof(v));
- break;
- }
- case FieldType::FLOAT32: {
- float v = 0.0f;
- if (has_val) {
- v = accessor.get_float(row, i);
- } else if (std::holds_alternative<float>(meta.default_value)) {
- v = std::get<float>(meta.default_value);
- }
- std::memcpy(field_ptr, &v, sizeof(v));
- break;
- }
- case FieldType::BOOLEAN: {
- bool v = false;
- if (has_val) {
- v = accessor.get_bool(row, i);
- } else if (std::holds_alternative<bool>(meta.default_value)) {
- v = std::get<bool>(meta.default_value);
- }
- uint8_t b = v ? 1 : 0;
- std::memcpy(field_ptr, &b, sizeof(b));
- break;
- }
- case FieldType::STRING:
- case FieldType::BINARY:
- case FieldType::LIST_INT64:
- case FieldType::LIST_FLOAT32:
- case FieldType::LIST_STRING: {
- uint32_t offset = static_cast<uint32_t>(var_infos[i].offset);
- std::memcpy(field_ptr, &offset, sizeof(offset));
-
- char* var_ptr = ptr + offset;
-
- if (meta.data_type == FieldType::STRING) {
- // Logic handled by accessor to write directly? Or return string_view/string?
- // Accessor returning string creates copy.
- // Better: accessor.write_string(row, i, var_ptr)
- if (has_val) {
- accessor.write_string(row, i, var_ptr);
- } else if (const auto* def = get_default_string(meta.default_value)) {
- write_string_value(*def, var_ptr);
- } else {
- uint16_t len = 0;
- std::memcpy(var_ptr, &len, 2);
- }
- } else if (meta.data_type == FieldType::BINARY) {
- if (has_val) {
- accessor.write_binary(row, i, var_ptr);
- } else if (const auto* def = get_default_string(meta.default_value)) {
- write_binary_value(*def, var_ptr);
- } else {
- uint32_t len = 0;
- std::memcpy(var_ptr, &len, 4);
- }
- } else if (meta.data_type == FieldType::LIST_INT64) {
- if (has_val) {
- accessor.write_list_int64(row, i, var_ptr);
- } else if (const auto* def = get_default_list_int64(meta.default_value)) {
- write_list_int64_value(*def, var_ptr);
- } else {
- uint16_t len = 0;
- std::memcpy(var_ptr, &len, 2);
- }
- } else if (meta.data_type == FieldType::LIST_FLOAT32) {
- if (has_val) {
- accessor.write_list_float32(row, i, var_ptr);
- } else if (const auto* def = get_default_list_float32(meta.default_value)) {
- write_list_float32_value(*def, var_ptr);
- } else {
- uint16_t len = 0;
- std::memcpy(var_ptr, &len, 2);
- }
- } else if (meta.data_type == FieldType::LIST_STRING) {
- if (has_val) {
- accessor.write_list_string(row, i, var_ptr);
- } else if (const auto* def = get_default_list_string(meta.default_value)) {
- write_list_string_value(*def, var_ptr);
- } else {
- uint16_t len = 0;
- std::memcpy(var_ptr, &len, 2);
- }
- }
- break;
- }
+ } else if (meta.data_type == FieldType::LIST_FLOAT32) {
+ if (has_val) {
+ accessor.write_list_float32(row, i, var_ptr);
+ } else if (const auto* def =
+ get_default_list_float32(meta.default_value)) {
+ write_list_float32_value(*def, var_ptr);
+ } else {
+ uint16_t len = 0;
+ std::memcpy(var_ptr, &len, 2);
+ }
+ } else if (meta.data_type == FieldType::LIST_STRING) {
+ if (has_val) {
+ accessor.write_list_string(row, i, var_ptr);
+ } else if (const auto* def =
+ get_default_list_string(meta.default_value)) {
+ write_list_string_value(*def, var_ptr);
+ } else {
+ uint16_t len = 0;
+ std::memcpy(var_ptr, &len, 2);
}
+ }
+ break;
}
- return buffer;
+ }
}
+ return buffer;
+ }
+
+ // Deserialize to a map
+ std::map<std::string, Value> deserialize(
+ const std::string& serialized_data) const;
+
+ // Deserialize a single field
+ Value deserialize_field(const std::string& serialized_data,
+ const std::string& field_name) const;
+
+ // Get schema
+ const Schema& get_schema() const {
+ return *schema_;
+ }
- // Deserialize to a map
- std::map<std::string, Value> deserialize(const std::string& serialized_data) const;
-
- // Deserialize a single field
- Value deserialize_field(const std::string& serialized_data, const std::string& field_name) const;
-
- // Get schema
- const Schema& get_schema() const { return *schema_; }
-
-private:
- std::shared_ptr<Schema> schema_;
+ private:
+ std::shared_ptr<Schema> schema_;
};
-} // namespace vectordb
+} // namespace vectordb
diff --git a/tests/agfs/conftest.py b/tests/agfs/conftest.py
index 20dbe1f35..0b41a614e 100644
--- a/tests/agfs/conftest.py
+++ b/tests/agfs/conftest.py
@@ -6,6 +6,7 @@ from pathlib import Path
import pytest
+
@pytest.fixture(scope="session")
def agfs_test_root():
"""Root directory for AGFS tests."""
diff --git a/tests/api_test/filesystem/test_fs_read_write.py b/tests/api_test/filesystem/test_fs_read_write.py
index 07f59bd2e..4c1f29853 100644
--- a/tests/api_test/filesystem/test_fs_read_write.py
+++ b/tests/api_test/filesystem/test_fs_read_write.py
@@ -1,7 +1,6 @@
import json
import pytest
-import requests
class TestFsReadWrite:
@@ -10,16 +9,17 @@ class TestFsReadWrite:
test_file_uri = "viking://resources/test_fs_read_write_test.txt"
test_content = "This is a test file created for fs_read test."
-
try:
write_response = api_client.fs_write(test_file_uri, test_content, wait=True)
print(f"\nCreated test file: {test_file_uri}")
print(f"Write response status: {write_response.status_code}")
write_data = write_response.json()
print(f"Write response: {json.dumps(write_data, indent=2, ensure_ascii=False)}")
-
+
if write_data.get("status") != "ok":
- pytest.skip(f"fs_write failed on this environment: {write_data.get('error')}. This may be due to AGFS service not being available.")
+ pytest.skip(
+ f"fs_write failed on this environment: {write_data.get('error')}. This may be due to AGFS service not being available."
+ )
response = api_client.fs_read(test_file_uri)
print(f"\nFS read API status code: {response.status_code}")
@@ -32,11 +32,15 @@ class TestFsReadWrite:
print("=" * 80 + "\n")
if data.get("status") != "ok":
- pytest.skip(f"fs_read failed on this environment: {data.get('error')}. This may be due to AGFS service not being available.")
+ pytest.skip(
+ f"fs_read failed on this environment: {data.get('error')}. This may be due to AGFS service not being available."
+ )
assert data.get("error") is None, f"Expected error to be null, got {data.get('error')}"
assert "result" in data, "'result' field should exist"
- assert data["result"] == test_content, f"Expected content '{test_content}', got {data.get('result')}"
+ assert data["result"] == test_content, (
+ f"Expected content '{test_content}', got {data.get('result')}"
+ )
except Exception as e:
print(f"Error: {e}")
diff --git a/tests/api_test/filesystem/test_get_overview.py b/tests/api_test/filesystem/test_get_overview.py
index b6c2af056..3c683ba99 100644
--- a/tests/api_test/filesystem/test_get_overview.py
+++ b/tests/api_test/filesystem/test_get_overview.py
@@ -20,7 +20,9 @@ class TestGetOverview:
print("=" * 80)
print(json.dumps(data, indent=2, ensure_ascii=False))
print("=" * 80 + "\n")
- pytest.skip("Overview file not found on this server. This may be due to AGFS service not being available or no .overview.md file exists.")
+ pytest.skip(
+ "Overview file not found on this server. This may be due to AGFS service not being available or no .overview.md file exists."
+ )
if response.status_code >= 500:
data = response.json() if response.text else {}
@@ -29,10 +31,14 @@ class TestGetOverview:
print("=" * 80)
print(json.dumps(data, indent=2, ensure_ascii=False))
print("=" * 80 + "\n")
- pytest.skip(f"Server error on this environment: {data.get('error', 'Unknown error')}. This may be due to AGFS service not being available.")
+ pytest.skip(
+ f"Server error on this environment: {data.get('error', 'Unknown error')}. This may be due to AGFS service not being available."
+ )
if response.status_code != 200:
- pytest.skip(f"Unexpected status code {response.status_code}. This may be due to environment configuration.")
+ pytest.skip(
+ f"Unexpected status code {response.status_code}. This may be due to environment configuration."
+ )
data = response.json()
print("\n" + "=" * 80)
diff --git a/tests/api_test/scenarios/conftest.py b/tests/api_test/scenarios/conftest.py
index f199f0cfd..22a506990 100644
--- a/tests/api_test/scenarios/conftest.py
+++ b/tests/api_test/scenarios/conftest.py
@@ -5,28 +5,30 @@ import uuid
def create_test_file(content=None, suffix=".txt"):
if content is None:
- content = f"测试文件内容 - {uuid.uuid4()}\n这是一个用于API测试的临时文件。\n包含一些测试数据。"
-
+ content = (
+ f"测试文件内容 - {uuid.uuid4()}\n这是一个用于API测试的临时文件。\n包含一些测试数据。"
+ )
+
temp_dir = tempfile.mkdtemp()
test_file_path = os.path.join(temp_dir, f"test_file_{str(uuid.uuid4())[:8]}{suffix}")
-
+
with open(test_file_path, "w", encoding="utf-8") as f:
f.write(content)
-
+
return test_file_path, temp_dir
def create_test_directory():
temp_dir = tempfile.mkdtemp()
-
+
for i in range(3):
file_path = os.path.join(temp_dir, f"file_{i}.txt")
with open(file_path, "w", encoding="utf-8") as f:
f.write(f"测试文件 {i} 的内容\n一些测试数据 {uuid.uuid4()}")
-
+
subdir = os.path.join(temp_dir, "subdir")
os.makedirs(subdir)
with open(os.path.join(subdir, "nested_file.txt"), "w", encoding="utf-8") as f:
f.write("嵌套文件的内容")
-
+
return temp_dir
diff --git a/tests/api_test/scenarios/resources_retrieval/test_build_error_handling.py b/tests/api_test/scenarios/resources_retrieval/test_build_error_handling.py
index 457353d05..6c36ca798 100644
--- a/tests/api_test/scenarios/resources_retrieval/test_build_error_handling.py
+++ b/tests/api_test/scenarios/resources_retrieval/test_build_error_handling.py
@@ -140,14 +140,10 @@ class TestBuildErrorHandling:
try:
response = api_client.add_resource(path=zip_path, wait=True)
- assert response.status_code == 500, (
- f"损坏ZIP应返回 500, 实际: {response.status_code}"
- )
+ assert response.status_code == 500, f"损坏ZIP应返回 500, 实际: {response.status_code}"
data = response.json()
- assert data.get("status") == "error", (
- f"损坏ZIP应返回 error, 实际: {data.get('status')}"
- )
+ assert data.get("status") == "error", f"损坏ZIP应返回 error, 实际: {data.get('status')}"
print("✓ TC-E16 损坏的ZIP文件处理通过")
finally:
diff --git a/tests/api_test/scenarios/resources_retrieval/test_delete_sync.py b/tests/api_test/scenarios/resources_retrieval/test_delete_sync.py
index 8d25c46ac..2320648cc 100644
--- a/tests/api_test/scenarios/resources_retrieval/test_delete_sync.py
+++ b/tests/api_test/scenarios/resources_retrieval/test_delete_sync.py
@@ -1,94 +1,94 @@
-import pytest
-import json
-import uuid
-import time
import os
import shutil
+import time
+import uuid
+
from conftest import create_test_file
class TestDeleteSync:
"""TC-R04 资源删除索引同步
-
+
根据API文档:
- 删除资源使用 DELETE /api/v1/fs?uri={uri}&recursive={bool}
- 删除后应该同步更新向量索引
"""
-
+
def test_resource_deletion_index_sync(self, api_client):
"""资源删除索引同步:添加资源 -> 等待索引 -> 删除资源 -> 验证删除"""
random_id = str(uuid.uuid4())[:8]
unique_keyword = f"delete_test_{random_id}"
-
+
# 1. 创建临时测试文件
test_file_path, temp_dir = create_test_file(
content=f"测试文件 {random_id}\n这是一个用于删除同步测试的文件。\n包含唯一关键词:{unique_keyword}、test、删除、同步。"
)
-
+
try:
# 2. 添加该文件到资源
response = api_client.add_resource(path=test_file_path, wait=True)
assert response.status_code == 200
-
+
add_data = response.json()
- assert add_data.get('status') == 'ok'
-
- add_result = add_data.get('result', {})
- resource_uri = add_result.get('root_uri')
+ assert add_data.get("status") == "ok"
+
+ add_result = add_data.get("result", {})
+ resource_uri = add_result.get("root_uri")
assert resource_uri is not None, "Add resource should return root_uri"
print(f"资源添加成功,URI: {resource_uri}")
-
+
# 3. 等待处理完成
response = api_client.wait_processed()
assert response.status_code == 200
-
+
# 额外等待索引同步
time.sleep(3)
-
+
# 4. 验证能搜索到资源
response = api_client.find(unique_keyword)
assert response.status_code == 200
data = response.json()
- assert data.get('status') == 'ok'
- assert 'result' in data
-
- search_result = data['result']
-
+ assert data.get("status") == "ok"
+ assert "result" in data
+
+ search_result = data["result"]
+
# 验证搜索结果不为空
total_results = 0
- for field in ['resources', 'memories', 'matches']:
+ for field in ["resources", "memories", "matches"]:
if field in search_result:
items = search_result[field]
total_results += len(items)
-
+
# 搜索应该返回结果
- assert total_results > 0, \
+ assert total_results > 0, (
f"Search should return results before deletion, keyword: {unique_keyword}"
+ )
print(f"删除前搜索结果数: {total_results}")
-
+
# 5. 删除资源
response = api_client.fs_rm(resource_uri, recursive=True)
assert response.status_code == 200
delete_data = response.json()
- assert delete_data.get('status') == 'ok'
+ assert delete_data.get("status") == "ok"
print(f"资源已删除: {resource_uri}")
-
+
# 6. 等待索引同步
time.sleep(3)
response = api_client.wait_processed()
assert response.status_code == 200
-
+
# 7. 验证删除后资源不存在于文件系统
response = api_client.fs_stat(resource_uri)
# 资源应该不存在
if response.status_code != 200:
- print(f"删除后资源不存在于文件系统 ✓")
+ print("删除后资源不存在于文件系统 ✓")
else:
stat_data = response.json()
- if stat_data.get('status') == 'error':
- print(f"删除后资源不存在于文件系统 ✓")
-
- print(f"✓ 资源删除索引同步测试通过")
+ if stat_data.get("status") == "error":
+ print("删除后资源不存在于文件系统 ✓")
+
+ print("✓ 资源删除索引同步测试通过")
finally:
# 清理临时文件
if os.path.exists(temp_dir):
diff --git a/tests/api_test/scenarios/resources_retrieval/test_grep_validation.py b/tests/api_test/scenarios/resources_retrieval/test_grep_validation.py
index 4819861d8..60fe46cd5 100644
--- a/tests/api_test/scenarios/resources_retrieval/test_grep_validation.py
+++ b/tests/api_test/scenarios/resources_retrieval/test_grep_validation.py
@@ -1,76 +1,76 @@
-import pytest
import json
-import uuid
import os
import shutil
+import uuid
+
from conftest import create_test_file
class TestGrepValidation:
"""TC-R03 正则检索验证 (Grep)
-
+
根据API文档:grep用于文本搜索,支持正则表达式匹配。
API: GET /api/v1/search/grep?uri={uri}&pattern={pattern}
"""
-
+
def test_grep_pattern_match(self, api_client):
"""正则检索验证:添加资源 -> grep搜索 -> 验证匹配结果"""
random_id = str(uuid.uuid4())[:8]
unique_pattern = f"GrepTest{random_id}"
-
+
# 1. 创建临时测试文件,包含特定模式
test_file_path, temp_dir = create_test_file(
content=f"测试文件 {random_id}\n这是一个用于grep测试的文件。\n{unique_pattern} pattern matching.\n包含Test关键词。\nAnother {unique_pattern} occurrence."
)
-
+
try:
# 2. 添加该文件到资源
response = api_client.add_resource(path=test_file_path, wait=True)
assert response.status_code == 200
-
+
add_data = response.json()
- assert add_data.get('status') == 'ok'
-
+ assert add_data.get("status") == "ok"
+
# 获取导入后的 URI
- add_result = add_data.get('result', {})
- imported_uri = add_result.get('root_uri')
+ add_result = add_data.get("result", {})
+ imported_uri = add_result.get("root_uri")
assert imported_uri is not None, "Add resource should return root_uri"
print(f"资源添加成功,URI: {imported_uri}")
-
+
# 3. 等待处理完成
response = api_client.wait_processed()
assert response.status_code == 200
-
+
# 4. 执行grep搜索
response = api_client.grep(imported_uri, unique_pattern)
assert response.status_code == 200
-
+
grep_data = response.json()
- assert grep_data.get('status') == 'ok'
- assert 'result' in grep_data
-
- grep_result = grep_data['result']
-
+ assert grep_data.get("status") == "ok"
+ assert "result" in grep_data
+
+ grep_result = grep_data["result"]
+
# 5. 业务逻辑验证:grep结果应该包含匹配
# 根据API文档,grep返回匹配的文本行
found_match = False
-
- if 'matches' in grep_result:
- matches = grep_result['matches']
+
+ if "matches" in grep_result:
+ matches = grep_result["matches"]
assert isinstance(matches, list), "Matches should be a list"
-
+
for match in matches:
if isinstance(match, dict):
- if 'text' in match and unique_pattern in match['text']:
+ if "text" in match and unique_pattern in match["text"]:
found_match = True
print(f"找到匹配: {match.get('text', '')[:100]}")
elif isinstance(match, str) and unique_pattern in match:
found_match = True
print(f"找到匹配: {match[:100]}")
-
+
# 如果没有matches字段,检查其他可能的字段
if not found_match:
- for field in ['results', 'lines', 'content']:
+ for field in ["results", "lines", "content"]:
if field in grep_result:
content = grep_result[field]
if isinstance(content, list):
@@ -80,15 +80,15 @@ class TestGrepValidation:
break
elif unique_pattern in str(content):
found_match = True
-
+
# 验证grep找到了匹配(如果API支持grep功能)
# 注意:grep可能需要特定配置才能工作
print(f"Grep结果: {json.dumps(grep_result, ensure_ascii=False)[:500]}")
-
+
# 6. 验证搜索结果结构正确
assert grep_result is not None, "Grep result should not be None"
-
- print(f"✓ Grep验证测试通过")
+
+ print("✓ Grep验证测试通过")
finally:
# 清理临时文件
if os.path.exists(temp_dir):
diff --git a/tests/api_test/scenarios/resources_retrieval/test_intent_extended_search.py b/tests/api_test/scenarios/resources_retrieval/test_intent_extended_search.py
index e65c6d1ff..0d5d01be3 100644
--- a/tests/api_test/scenarios/resources_retrieval/test_intent_extended_search.py
+++ b/tests/api_test/scenarios/resources_retrieval/test_intent_extended_search.py
@@ -1,98 +1,104 @@
-import pytest
-import json
-import uuid
import os
import shutil
+import uuid
+
from conftest import create_test_file
class TestIntentExtendedSearch:
"""TC-R06 意图扩展搜索 (Search)
-
+
根据API文档:
- search() 带会话上下文和意图分析
- 参数 session_id 用于上下文感知搜索
- 与 find() 的区别:search 支持意图分析、会话上下文、查询扩展
"""
-
+
def test_intent_extended_search(self, api_client):
"""意图扩展搜索:create_session -> add_message -> search(with session_id)"""
random_id = str(uuid.uuid4())[:8]
unique_keyword = f"intent_search_{random_id}"
-
+
# 1. 创建临时测试文件
test_file_path, temp_dir = create_test_file(
content=f"测试文件 {random_id}\n这是一个用于意图扩展搜索测试的文件。\n包含唯一关键词:{unique_keyword}、test、搜索、意图。\nOAuth认证相关内容。"
)
-
+
try:
# 2. 添加该文件到资源
response = api_client.add_resource(path=test_file_path, wait=True)
assert response.status_code == 200
-
+
add_data = response.json()
- assert add_data.get('status') == 'ok'
- print(f"资源添加成功")
-
+ assert add_data.get("status") == "ok"
+ print("资源添加成功")
+
# 3. 等待处理完成
response = api_client.wait_processed()
assert response.status_code == 200
-
+
# 4. 创建会话
response = api_client.create_session()
assert response.status_code == 200
create_data = response.json()
- assert create_data.get('status') == 'ok'
-
- session_id = create_data['result']['session_id']
+ assert create_data.get("status") == "ok"
+
+ session_id = create_data["result"]["session_id"]
assert session_id is not None
print(f"会话创建成功: {session_id}")
-
+
# 5. 添加对话上下文(模拟用户讨论OAuth)
- response = api_client.add_message(session_id, "user",
- f"我正在实现OAuth认证功能,需要查看相关文档。{random_id}")
+ response = api_client.add_message(
+ session_id, "user", f"我正在实现OAuth认证功能,需要查看相关文档。{random_id}"
+ )
assert response.status_code == 200
msg_data = response.json()
- assert msg_data.get('status') == 'ok'
- print(f"消息添加成功")
-
+ assert msg_data.get("status") == "ok"
+ print("消息添加成功")
+
# 6. 执行搜索(带会话上下文)
# 根据API文档,search支持session_id参数
search_query = "认证"
response = api_client.search(search_query)
assert response.status_code == 200
-
+
search_data = response.json()
- assert search_data.get('status') == 'ok'
- assert 'result' in search_data
-
- search_result = search_data['result']
-
+ assert search_data.get("status") == "ok"
+ assert "result" in search_data
+
+ search_result = search_data["result"]
+
# 7. 验证搜索结果结构
- assert 'memories' in search_result or 'resources' in search_result or 'results' in search_result
-
+ assert (
+ "memories" in search_result
+ or "resources" in search_result
+ or "results" in search_result
+ )
+
total_results = 0
- for field in ['memories', 'resources', 'results']:
+ for field in ["memories", "resources", "results"]:
if field in search_result:
items = search_result[field]
assert isinstance(items, list), f"{field} should be a list"
total_results += len(items)
-
+
print(f"搜索结果数量: {total_results}")
-
+
# 8. 业务逻辑验证:搜索应该返回相关结果
- assert total_results > 0, \
+ assert total_results > 0, (
"Search should return at least one result when resources exist"
-
+ )
+
# 9. 验证搜索结果的相关性分数(如果返回)
- for field in ['resources', 'memories']:
+ for field in ["resources", "memories"]:
if field in search_result:
for item in search_result[field]:
- if 'score' in item:
- assert 0 <= item['score'] <= 1, \
+ if "score" in item:
+ assert 0 <= item["score"] <= 1, (
f"Score should be between 0 and 1, got {item['score']}"
-
- print(f"✓ 意图扩展搜索测试通过")
+ )
+
+ print("✓ 意图扩展搜索测试通过")
finally:
# 清理临时文件
if os.path.exists(temp_dir):
diff --git a/tests/api_test/scenarios/resources_retrieval/test_pack_consistency.py b/tests/api_test/scenarios/resources_retrieval/test_pack_consistency.py
index bd4f29b99..7484aa765 100644
--- a/tests/api_test/scenarios/resources_retrieval/test_pack_consistency.py
+++ b/tests/api_test/scenarios/resources_retrieval/test_pack_consistency.py
@@ -1,90 +1,88 @@
-import pytest
-import json
-import uuid
-import time
import os
import shutil
+import time
+import uuid
+
from conftest import create_test_file
class TestPackConsistency:
"""TC-R05 批量导入导出一致性
-
+
根据API文档:
- export_ovpack: POST /api/v1/pack/export
- import_ovpack: POST /api/v1/pack/import
- 用于资源的批量导出和导入
"""
-
+
def test_pack_export_import_consistency(self, api_client):
"""批量导入导出一致性:添加资源 -> 验证资源存在 -> 验证搜索正常"""
random_id = str(uuid.uuid4())[:8]
unique_keyword = f"pack_test_{random_id}"
-
+
# 1. 创建临时测试文件
test_file_path, temp_dir = create_test_file(
content=f"测试文件 {random_id}\n这是一个用于pack测试的文件。\n包含唯一关键词:{unique_keyword}、test、pack、导出。"
)
-
+
try:
# 2. 添加该文件到资源(确保有资源可导出)
response = api_client.add_resource(path=test_file_path, wait=True)
assert response.status_code == 200
-
+
add_data = response.json()
- assert add_data.get('status') == 'ok'
-
- add_result = add_data.get('result', {})
- resource_uri = add_result.get('root_uri')
+ assert add_data.get("status") == "ok"
+
+ add_result = add_data.get("result", {})
+ resource_uri = add_result.get("root_uri")
assert resource_uri is not None, "Add resource should return root_uri"
print(f"资源添加成功,URI: {resource_uri}")
-
+
# 3. 等待处理完成
response = api_client.wait_processed()
assert response.status_code == 200
-
+
# 额外等待索引同步
time.sleep(3)
-
+
# 4. 验证资源存在于文件系统
response = api_client.fs_ls("viking://resources/")
assert response.status_code == 200
ls_data = response.json()
- assert ls_data.get('status') == 'ok'
-
- ls_result = ls_data.get('result', [])
+ assert ls_data.get("status") == "ok"
+
+ ls_result = ls_data.get("result", [])
assert isinstance(ls_result, list), "fs_ls result should be a list"
print(f"资源目录列表: {len(ls_result)} 个条目")
-
+
# 5. 验证搜索能找到资源
response = api_client.find(unique_keyword)
assert response.status_code == 200
-
+
search_data = response.json()
- assert search_data.get('status') == 'ok'
- assert 'result' in search_data
-
- search_result = search_data['result']
-
+ assert search_data.get("status") == "ok"
+ assert "result" in search_data
+
+ search_result = search_data["result"]
+
# 验证搜索结果不为空
total_results = 0
- for field in ['resources', 'memories', 'matches']:
+ for field in ["resources", "memories", "matches"]:
if field in search_result:
items = search_result[field]
total_results += len(items)
-
- assert total_results > 0, \
- f"Search should return results for keyword: {unique_keyword}"
+
+ assert total_results > 0, f"Search should return results for keyword: {unique_keyword}"
print(f"搜索结果数: {total_results}")
-
+
# 6. 验证资源状态
response = api_client.fs_stat(resource_uri)
if response.status_code == 200:
stat_data = response.json()
- if stat_data.get('status') == 'ok':
- print(f"资源状态验证成功 ✓")
-
- print(f"✓ Pack一致性测试通过,资源已正确索引")
+ if stat_data.get("status") == "ok":
+ print("资源状态验证成功 ✓")
+
+ print("✓ Pack一致性测试通过,资源已正确索引")
finally:
# 清理临时文件
if os.path.exists(temp_dir):
diff --git a/tests/api_test/scenarios/resources_retrieval/test_relation_link.py b/tests/api_test/scenarios/resources_retrieval/test_relation_link.py
index ca14dc8a6..8913bf7d5 100644
--- a/tests/api_test/scenarios/resources_retrieval/test_relation_link.py
+++ b/tests/api_test/scenarios/resources_retrieval/test_relation_link.py
@@ -1,24 +1,23 @@
-import pytest
-import json
-import uuid
import os
import shutil
+import uuid
+
from conftest import create_test_file
class TestRelationLink:
"""TC-R07 关系链接验证
-
+
根据API文档:
- link(): POST /api/v1/relations/link
- 用于建立资源之间的关系
- 参数:from_uri, to_uris, reason
"""
-
+
def test_relation_link(self, api_client):
"""关系链接验证:添加资源A和B -> link(A, B) -> 验证关系建立"""
random_id = str(uuid.uuid4())[:8]
-
+
# 1. 创建两个临时测试文件
test_file_a, temp_dir_a = create_test_file(
content=f"资源A {random_id}\n这是资源A的内容。\n包含关键词:testA、资源、链接。"
@@ -26,73 +25,73 @@ class TestRelationLink:
test_file_b, temp_dir_b = create_test_file(
content=f"资源B {random_id}\n这是资源B的内容。\n包含关键词:testB、资源、链接。\n与资源A相关。"
)
-
+
try:
# 2. 添加资源A
response = api_client.add_resource(path=test_file_a, wait=True)
assert response.status_code == 200
add_data_a = response.json()
- assert add_data_a.get('status') == 'ok'
- uri_a = add_data_a.get('result', {}).get('root_uri')
+ assert add_data_a.get("status") == "ok"
+ uri_a = add_data_a.get("result", {}).get("root_uri")
assert uri_a is not None
print(f"资源A添加成功: {uri_a}")
-
+
# 3. 添加资源B
response = api_client.add_resource(path=test_file_b, wait=True)
assert response.status_code == 200
add_data_b = response.json()
- assert add_data_b.get('status') == 'ok'
- uri_b = add_data_b.get('result', {}).get('root_uri')
+ assert add_data_b.get("status") == "ok"
+ uri_b = add_data_b.get("result", {}).get("root_uri")
assert uri_b is not None
print(f"资源B添加成功: {uri_b}")
-
+
# 4. 等待处理完成
response = api_client.wait_processed()
assert response.status_code == 200
-
+
# 5. 建立关系链接 A -> B
response = api_client.link(
- from_uri=uri_a,
- to_uris=[uri_b],
- reason=f"测试关系链接 {random_id}"
+ from_uri=uri_a, to_uris=[uri_b], reason=f"测试关系链接 {random_id}"
)
-
+
# 验证link API调用成功
if response.status_code == 200:
link_data = response.json()
- assert link_data.get('status') == 'ok'
+ assert link_data.get("status") == "ok"
print(f"关系链接建立成功: {uri_a} -> {uri_b}")
else:
print(f"关系链接API返回: {response.status_code}")
-
+
# 6. 验证搜索功能正常
response = api_client.search("testA")
assert response.status_code == 200
-
+
search_data = response.json()
- assert search_data.get('status') == 'ok'
- assert 'result' in search_data
-
- search_result = search_data['result']
-
+ assert search_data.get("status") == "ok"
+ assert "result" in search_data
+
+ search_result = search_data["result"]
+
# 7. 验证搜索结果结构
- assert 'memories' in search_result or 'resources' in search_result or 'results' in search_result
-
+ assert (
+ "memories" in search_result
+ or "resources" in search_result
+ or "results" in search_result
+ )
+
# 8. 验证资源A能被搜索到
- found_a = False
- for field in ['resources', 'memories', 'results']:
+ for field in ["resources", "memories", "results"]:
if field in search_result:
items = search_result[field]
for item in items:
- if 'uri' in item and uri_a in item['uri']:
- found_a = True
+ if "uri" in item and uri_a in item["uri"]:
# 验证关系是否被正确返回
- if 'relations' in item:
- relations = item['relations']
+ if "relations" in item:
+ relations = item["relations"]
print(f"资源A的关系: {relations}")
break
-
- print(f"✓ 关系链接测试通过")
+
+ print("✓ 关系链接测试通过")
finally:
# 清理临时文件
if os.path.exists(temp_dir_a):
diff --git a/tests/api_test/scenarios/resources_retrieval/test_resource_swap.py b/tests/api_test/scenarios/resources_retrieval/test_resource_swap.py
index 96bcb04c5..137e63ebf 100644
--- a/tests/api_test/scenarios/resources_retrieval/test_resource_swap.py
+++ b/tests/api_test/scenarios/resources_retrieval/test_resource_swap.py
@@ -1,73 +1,71 @@
-import pytest
-import json
-import uuid
-import time
import os
import shutil
+import time
+import uuid
+
from conftest import create_test_file
class TestResourceSwap:
"""TC-R02 资源增量更新
-
+
根据API文档:当你为同一个资源 URI 反复调用 add_resource() 时,
系统会走"增量更新"而不是每次全量重建。
触发条件:请求里显式指定 target,且该 target 在知识库中已存在。
"""
-
+
def test_resource_incremental_update(self, api_client):
"""资源增量更新:添加资源 -> 等待索引 -> 验证能搜索到"""
random_id = str(uuid.uuid4())[:8]
unique_keyword = f"incremental_{random_id}"
-
+
# 1. 创建临时测试文件
test_file_path, temp_dir = create_test_file(
content=f"测试文件 {random_id}\n这是一个用于资源增量更新测试的文件。\n包含关键词:{unique_keyword}、test、更新、资源。"
)
-
+
try:
# 2. 添加该文件到资源
response = api_client.add_resource(path=test_file_path, wait=True)
assert response.status_code == 200
-
+
add_data = response.json()
- assert add_data.get('status') == 'ok'
-
+ assert add_data.get("status") == "ok"
+
# 验证返回结果包含root_uri
- add_result = add_data.get('result', {})
- assert 'root_uri' in add_result, "Add resource should return root_uri"
- root_uri = add_result['root_uri']
+ add_result = add_data.get("result", {})
+ assert "root_uri" in add_result, "Add resource should return root_uri"
+ root_uri = add_result["root_uri"]
print(f"资源添加成功,root_uri: {root_uri}")
-
+
# 3. 等待处理完成
response = api_client.wait_processed()
assert response.status_code == 200
-
+
# 额外等待索引同步
time.sleep(3)
-
+
# 4. 验证find能搜索到该资源
response = api_client.find(unique_keyword)
assert response.status_code == 200
-
+
search_data = response.json()
- assert search_data.get('status') == 'ok'
- assert 'result' in search_data
-
- search_result = search_data['result']
-
+ assert search_data.get("status") == "ok"
+ assert "result" in search_data
+
+ search_result = search_data["result"]
+
# 5. 验证搜索结果结构正确
total_results = 0
- for field in ['resources', 'memories', 'matches']:
+ for field in ["resources", "memories", "matches"]:
if field in search_result:
items = search_result[field]
assert isinstance(items, list), f"{field} should be a list"
total_results += len(items)
-
+
# 6. 业务逻辑验证:搜索应该返回结果
- assert total_results > 0, \
- f"Search should return results for keyword: {unique_keyword}"
-
+ assert total_results > 0, f"Search should return results for keyword: {unique_keyword}"
+
print(f"✓ 资源增量更新测试通过,搜索结果数: {total_results}")
finally:
# 清理临时文件
diff --git a/tests/api_test/scenarios/resources_retrieval/test_semantic_retrieval.py b/tests/api_test/scenarios/resources_retrieval/test_semantic_retrieval.py
index c3f8fcc4b..0dff473bc 100644
--- a/tests/api_test/scenarios/resources_retrieval/test_semantic_retrieval.py
+++ b/tests/api_test/scenarios/resources_retrieval/test_semantic_retrieval.py
@@ -1,129 +1,133 @@
-import pytest
-import json
-import uuid
import os
+import uuid
+
from conftest import create_test_file
class TestSemanticRetrieval:
"""TC-R01 语义检索全链路验证"""
-
+
def test_semantic_retrieval_end_to_end(self, api_client):
"""语义检索全链路验证:添加资源 -> 等待处理 -> 搜索验证"""
random_id = str(uuid.uuid4())[:8]
unique_keyword = f"unique_keyword_{random_id}"
-
+
# 1. 创建临时测试文件,包含唯一关键词
test_file_path, temp_dir = create_test_file(
content=f"测试文件 {random_id}\n这是一个用于语义检索测试的文件。\n包含唯一关键词:{unique_keyword}、test、测试、检索。"
)
-
+
try:
# 2. 添加该文件到资源
response = api_client.add_resource(path=test_file_path, wait=True)
assert response.status_code == 200
-
+
add_data = response.json()
- assert add_data.get('status') == 'ok'
-
+ assert add_data.get("status") == "ok"
+
# 验证添加资源的返回结果
- add_result = add_data.get('result', {})
- assert 'root_uri' in add_result or 'resource_id' in add_result, \
+ add_result = add_data.get("result", {})
+ assert "root_uri" in add_result or "resource_id" in add_result, (
"Add resource should return root_uri or resource_id"
-
+ )
+
# 保存添加的资源URI,用于后续验证
- added_resource_uri = add_result.get('root_uri') or add_result.get('resource_id')
-
+ added_resource_uri = add_result.get("root_uri") or add_result.get("resource_id")
+
# 3. 等待处理完成
response = api_client.wait_processed()
assert response.status_code == 200
-
+
# 4. 执行语义搜索,使用唯一关键词
search_query = unique_keyword
response = api_client.find(search_query)
assert response.status_code == 200
-
+
search_data = response.json()
- assert search_data.get('status') == 'ok'
- assert 'result' in search_data
-
- search_result = search_data['result']
-
+ assert search_data.get("status") == "ok"
+ assert "result" in search_data
+
+ search_result = search_data["result"]
+
# 5. 验证搜索结果结构正确
found_added_resource = False
total_results = 0
-
- for field in ['resources', 'memories', 'matches']:
+
+ for field in ["resources", "memories", "matches"]:
if field in search_result:
items = search_result[field]
assert isinstance(items, list), f"{field} should be a list"
total_results += len(items)
-
+
# 如果有结果,验证每个结果的结构
for item in items:
- assert 'score' in item or 'uri' in item, \
+ assert "score" in item or "uri" in item, (
"Each search result should have score or uri"
-
+ )
+
# 验证是否找到添加的资源
- if 'uri' in item and added_resource_uri:
- if added_resource_uri in item['uri']:
+ if "uri" in item and added_resource_uri:
+ if added_resource_uri in item["uri"]:
found_added_resource = True
-
+
# 记录搜索结果数量
print(f"Total search results: {total_results}")
-
+
# 6. 验证业务逻辑:搜索结果应该包含刚添加的资源
# 这是一个重要的业务逻辑验证,应该保持失败状态以发现问题
if added_resource_uri:
- assert found_added_resource, \
- f"Search result should contain the added resource: {added_resource_uri}. " \
+ assert found_added_resource, (
+ f"Search result should contain the added resource: {added_resource_uri}. "
f"This indicates that the resource was not correctly indexed or the search algorithm has issues."
-
+ )
+
# 7. 验证搜索结果的相关性(如果返回了score)
- for field in ['resources', 'memories', 'matches']:
+ for field in ["resources", "memories", "matches"]:
if field in search_result:
items = search_result[field]
for item in items:
- if 'score' in item:
+ if "score" in item:
# 验证score是合理的范围(0-1)
- assert 0 <= item['score'] <= 1, \
+ assert 0 <= item["score"] <= 1, (
f"Score should be between 0 and 1, got {item['score']}"
-
+ )
+
# 8. 业务逻辑验证:验证资源是否被正确索引
# 使用更通用的关键词进行搜索
response = api_client.find("test")
assert response.status_code == 200
general_search_data = response.json()
- assert general_search_data.get('status') == 'ok'
-
+ assert general_search_data.get("status") == "ok"
+
# 记录通用搜索的结果数量
- general_search_result = general_search_data.get('result', {})
+ general_search_result = general_search_data.get("result", {})
general_total = 0
- for field in ['resources', 'memories', 'matches']:
+ for field in ["resources", "memories", "matches"]:
if field in general_search_result:
general_total += len(general_search_result[field])
-
+
print(f"General search results: {general_total}")
-
+
# 9. 业务逻辑验证:验证资源列表
response = api_client.fs_ls("viking://")
assert response.status_code == 200
ls_data = response.json()
- assert ls_data.get('status') == 'ok'
-
- ls_result = ls_data.get('result', [])
+ assert ls_data.get("status") == "ok"
+
+ ls_result = ls_data.get("result", [])
print(f"Total resources in root: {len(ls_result)}")
-
+
# 10. 业务逻辑验证:验证系统状态
response = api_client.is_healthy()
assert response.status_code == 200
health_data = response.json()
- assert health_data.get('status') == 'ok'
+ assert health_data.get("status") == "ok"
print("✓ System is healthy")
-
+
print("✓ Semantic retrieval test passed")
finally:
# 清理临时文件
import shutil
+
if os.path.exists(temp_dir):
shutil.rmtree(temp_dir)
diff --git a/tests/api_test/scenarios/resources_retrieval/test_watch_update.py b/tests/api_test/scenarios/resources_retrieval/test_watch_update.py
index 94d2cd654..3c309b563 100644
--- a/tests/api_test/scenarios/resources_retrieval/test_watch_update.py
+++ b/tests/api_test/scenarios/resources_retrieval/test_watch_update.py
@@ -1,86 +1,83 @@
-import pytest
-import json
-import uuid
-import time
import os
import shutil
+import uuid
+
from conftest import create_test_file
class TestWatchUpdate:
"""TC-R08 定时监听更新 (Watch)
-
+
根据API文档:
- add_resource() 支持 watch_interval 参数
- watch_interval: 定时更新间隔(分钟)。>0 开启/更新定时任务;<=0 关闭定时任务
- 仅在指定 target 时生效
"""
-
+
def test_watch_update(self, api_client):
"""定时监听更新:add_resource -> 验证资源索引 -> 验证搜索"""
random_id = str(uuid.uuid4())[:8]
unique_keyword = f"watch_test_{random_id}"
-
+
# 1. 创建临时测试文件
test_file_path, temp_dir = create_test_file(
content=f"测试文件 {random_id}\n这是一个用于监听更新测试的文件。\n包含唯一关键词:{unique_keyword}、test、监听、更新。"
)
-
+
try:
# 2. 添加该文件到资源
response = api_client.add_resource(path=test_file_path, wait=True)
assert response.status_code == 200
-
+
add_data = response.json()
- assert add_data.get('status') == 'ok'
-
- add_result = add_data.get('result', {})
- resource_uri = add_result.get('root_uri')
+ assert add_data.get("status") == "ok"
+
+ add_result = add_data.get("result", {})
+ resource_uri = add_result.get("root_uri")
assert resource_uri is not None, "Add resource should return root_uri"
print(f"资源添加成功: {resource_uri}")
-
+
# 3. 等待处理完成
response = api_client.wait_processed()
assert response.status_code == 200
-
+
# 4. 验证资源已被正确索引
response = api_client.fs_stat(resource_uri)
if response.status_code == 200:
stat_data = response.json()
- if stat_data.get('status') == 'ok':
- print(f"资源状态验证成功")
-
+ if stat_data.get("status") == "ok":
+ print("资源状态验证成功")
+
# 5. 执行搜索验证
response = api_client.find(unique_keyword)
assert response.status_code == 200
-
+
search_data = response.json()
- assert search_data.get('status') == 'ok'
- assert 'result' in search_data
-
- search_result = search_data['result']
-
+ assert search_data.get("status") == "ok"
+ assert "result" in search_data
+
+ search_result = search_data["result"]
+
# 6. 验证搜索结果包含刚添加的资源
found_resource = False
- for field in ['resources', 'memories', 'matches']:
+ for field in ["resources", "memories", "matches"]:
if field in search_result:
items = search_result[field]
assert isinstance(items, list), f"{field} should be a list"
for item in items:
- if 'uri' in item and resource_uri in item['uri']:
+ if "uri" in item and resource_uri in item["uri"]:
found_resource = True
break
-
- assert found_resource, \
- f"Search should find the added resource: {resource_uri}"
-
+
+ assert found_resource, f"Search should find the added resource: {resource_uri}"
+
# 7. 验证系统状态
response = api_client.is_healthy()
assert response.status_code == 200
health_data = response.json()
- assert health_data.get('status') == 'ok'
-
- print(f"✓ 定时监听更新测试通过,资源已被正确索引")
+ assert health_data.get("status") == "ok"
+
+ print("✓ 定时监听更新测试通过,资源已被正确索引")
finally:
# 清理临时文件
if os.path.exists(temp_dir):
diff --git a/tests/api_test/scenarios/sessions/test_long_context_recall.py b/tests/api_test/scenarios/sessions/test_long_context_recall.py
index dd55ef65b..dd31368bc 100644
--- a/tests/api_test/scenarios/sessions/test_long_context_recall.py
+++ b/tests/api_test/scenarios/sessions/test_long_context_recall.py
@@ -1,41 +1,40 @@
-import pytest
-import json
-import uuid
import os
import shutil
+import uuid
+
from conftest import create_test_file
class TestLongContextRecall:
"""TC-S04 长程上下文召回 (Recall)
-
+
根据API文档:
- 会话commit是异步操作,需要等待任务完成
- get_session_context() 可以获取会话上下文
- search() 可以搜索会话记忆
"""
-
+
def test_long_context_recall(self, api_client):
"""长程上下文召回:add_msg(Turn 1..5) -> commit -> 等待完成 -> search"""
random_id = str(uuid.uuid4())[:8]
specific_content = f"特定记忆点内容_{random_id}"
-
+
# 1. 创建临时测试文件
test_file_path, temp_dir = create_test_file(
content=f"测试文件 {random_id}\n这是一个用于长程上下文召回测试的文件。\n包含关键词:test、上下文、召回。\n特定记忆点:{specific_content}"
)
-
+
try:
# 2. 创建会话
response = api_client.create_session()
assert response.status_code == 200
create_data = response.json()
- assert create_data.get('status') == 'ok'
-
- session_id = create_data['result']['session_id']
+ assert create_data.get("status") == "ok"
+
+ session_id = create_data["result"]["session_id"]
assert session_id is not None
print(f"会话创建成功: {session_id}")
-
+
# 3. 添加5轮对话,第3轮包含特定记忆点
added_messages = 0
for i in range(1, 6):
@@ -43,73 +42,78 @@ class TestLongContextRecall:
message = f"对话第3轮,包含特定记忆点:{specific_content}"
else:
message = f"对话第{i}轮,普通内容 {random_id}"
-
+
response = api_client.add_message(session_id, "user", message)
assert response.status_code == 200
msg_data = response.json()
- assert msg_data.get('status') == 'ok'
+ assert msg_data.get("status") == "ok"
added_messages += 1
-
+
print(f"添加了 {added_messages} 条消息")
-
+
# 4. 验证消息数量
response = api_client.get_session(session_id)
assert response.status_code == 200
session_info = response.json()
- assert session_info.get('status') == 'ok'
- message_count = session_info['result'].get('message_count', 0)
- assert message_count == added_messages, \
+ assert session_info.get("status") == "ok"
+ message_count = session_info["result"].get("message_count", 0)
+ assert message_count == added_messages, (
f"Message count should be {added_messages}, got {message_count}"
+ )
print(f"消息数量验证通过: {message_count}")
-
+
# 5. 提交会话
response = api_client.session_commit(session_id)
assert response.status_code == 200
commit_data = response.json()
- assert commit_data.get('status') == 'ok'
-
- commit_result = commit_data['result']
- assert commit_result.get('archived') == True, "Messages should be archived"
- print(f"会话提交成功,archived=True")
-
+ assert commit_data.get("status") == "ok"
+
+ commit_result = commit_data["result"]
+ assert commit_result.get("archived"), "Messages should be archived"
+ print("会话提交成功,archived=True")
+
# 6. 等待异步任务完成
- task_id = commit_result.get('task_id')
+ task_id = commit_result.get("task_id")
if task_id:
task_result = api_client.wait_for_task(task_id, timeout=60.0)
- task_status = task_result.get('status')
- assert task_status == 'completed', \
+ task_status = task_result.get("status")
+ assert task_status == "completed", (
f"Task should complete successfully, got status: {task_status}"
+ )
print(f"异步任务完成: {task_status}")
-
+
# 7. 验证commit_count更新
response = api_client.get_session(session_id)
assert response.status_code == 200
session_info = response.json()
- commit_count = session_info['result'].get('commit_count', 0)
- assert commit_count >= 1, \
- f"Commit count should be at least 1, got {commit_count}"
+ commit_count = session_info["result"].get("commit_count", 0)
+ assert commit_count >= 1, f"Commit count should be at least 1, got {commit_count}"
print(f"commit_count验证通过: {commit_count}")
-
+
# 8. 添加临时文件到资源
response = api_client.add_resource(path=test_file_path, wait=True)
assert response.status_code == 200
-
+
response = api_client.wait_processed()
assert response.status_code == 200
-
+
# 9. 执行搜索,验证能否找到特定记忆点
search_query = specific_content
response = api_client.search(search_query)
assert response.status_code == 200
-
+
search_data = response.json()
- assert search_data.get('status') == 'ok'
- assert 'result' in search_data
-
- search_result = search_data['result']
- assert 'memories' in search_result or 'resources' in search_result or 'results' in search_result
-
- print(f"✓ 长程上下文召回测试通过")
+ assert search_data.get("status") == "ok"
+ assert "result" in search_data
+
+ search_result = search_data["result"]
+ assert (
+ "memories" in search_result
+ or "resources" in search_result
+ or "results" in search_result
+ )
+
+ print("✓ 长程上下文召回测试通过")
finally:
# 清理临时文件
if os.path.exists(temp_dir):
diff --git a/tests/api_test/scenarios/sessions/test_session_delete_cleanup.py b/tests/api_test/scenarios/sessions/test_session_delete_cleanup.py
index 4834e79bc..1496760b6 100644
--- a/tests/api_test/scenarios/sessions/test_session_delete_cleanup.py
+++ b/tests/api_test/scenarios/sessions/test_session_delete_cleanup.py
@@ -1,92 +1,92 @@
-import pytest
-import json
import uuid
class TestSessionDeleteCleanup:
"""TC-S05 会话删除与清理
-
+
根据API文档:
- DELETE /api/v1/sessions/{session_id} 删除会话
- 删除后再次获取会话应返回 NOT_FOUND 错误
"""
-
+
def test_session_delete_cleanup(self, api_client):
"""会话删除与清理:创建会话 -> 验证存在 -> 删除 -> 验证不存在"""
random_id = str(uuid.uuid4())[:8]
-
+
# 1. 创建会话
response = api_client.create_session()
assert response.status_code == 200
create_data = response.json()
- assert create_data.get('status') == 'ok'
-
- session_id = create_data['result']['session_id']
+ assert create_data.get("status") == "ok"
+
+ session_id = create_data["result"]["session_id"]
assert session_id is not None
print(f"会话创建成功: {session_id}")
-
+
# 2. 验证会话存在
response = api_client.get_session(session_id)
assert response.status_code == 200
session_data = response.json()
- assert session_data.get('status') == 'ok'
-
- session_result = session_data['result']
- assert 'session_id' in session_result
- assert session_result['session_id'] == session_id
- print(f"会话验证存在 ✓")
-
+ assert session_data.get("status") == "ok"
+
+ session_result = session_data["result"]
+ assert "session_id" in session_result
+ assert session_result["session_id"] == session_id
+ print("会话验证存在 ✓")
+
# 3. 添加消息(验证删除后消息也被清理)
response = api_client.add_message(session_id, "user", f"测试消息 {random_id}")
assert response.status_code == 200
msg_data = response.json()
- assert msg_data.get('status') == 'ok'
- print(f"消息添加成功")
-
+ assert msg_data.get("status") == "ok"
+ print("消息添加成功")
+
# 4. 再次验证会话存在
response = api_client.get_session(session_id)
assert response.status_code == 200
session_data = response.json()
- message_count = session_data['result'].get('message_count', 0)
+ message_count = session_data["result"].get("message_count", 0)
assert message_count >= 1, "Message count should be at least 1"
print(f"消息数量: {message_count}")
-
+
# 5. 删除会话
response = api_client.delete_session(session_id)
assert response.status_code == 200
delete_data = response.json()
- assert delete_data.get('status') == 'ok'
- print(f"会话删除成功")
-
+ assert delete_data.get("status") == "ok"
+ print("会话删除成功")
+
# 6. 验证删除后无法获取会话
response = api_client.get_session(session_id)
-
+
# 根据API文档,删除后应返回 NOT_FOUND 错误
if response.status_code == 200:
data = response.json()
# 如果返回200但状态是error,也视为正确
- if data.get('status') == 'error':
- error_info = data.get('error', {})
- assert error_info.get('code') == 'NOT_FOUND', \
+ if data.get("status") == "error":
+ error_info = data.get("error", {})
+ assert error_info.get("code") == "NOT_FOUND", (
f"Error code should be NOT_FOUND, got {error_info.get('code')}"
- print(f"删除后获取会话返回 NOT_FOUND 错误 ✓")
+ )
+ print("删除后获取会话返回 NOT_FOUND 错误 ✓")
else:
# 如果没有返回错误,可能是API行为不同
- print(f"⚠️ 警告:删除后仍能获取会话,API行为可能不符合预期")
+ print("⚠️ 警告:删除后仍能获取会话,API行为可能不符合预期")
else:
# 非200状态码也是预期的
- assert response.status_code in [404, 410], \
+ assert response.status_code in [404, 410], (
f"Expected 404 or 410 after deletion, got {response.status_code}"
+ )
print(f"删除后获取会话返回 {response.status_code} ✓")
-
+
# 7. 验证删除后无法添加消息
response = api_client.add_message(session_id, "user", "Another message")
# 应该返回错误
if response.status_code != 200:
- print(f"删除后无法添加消息 ✓")
+ print("删除后无法添加消息 ✓")
else:
data = response.json()
- if data.get('status') == 'error':
- print(f"删除后添加消息返回错误 ✓")
-
- print(f"✓ 会话删除与清理测试通过")
+ if data.get("status") == "error":
+ print("删除后添加消息返回错误 ✓")
+
+ print("✓ 会话删除与清理测试通过")
diff --git a/tests/api_test/scenarios/stability_error/test_account_isolation.py b/tests/api_test/scenarios/stability_error/test_account_isolation.py
index 4744b34da..a2e612f20 100644
--- a/tests/api_test/scenarios/stability_error/test_account_isolation.py
+++ b/tests/api_test/scenarios/stability_error/test_account_isolation.py
@@ -1,182 +1,184 @@
-import pytest
-import json
-import uuid
-import time
import os
import shutil
-from config import Config
+import time
+import uuid
+
from conftest import create_test_file
class TestAccountIsolation:
"""TC-ER03 账户隔离完整性验证
-
+
测试场景:验证资源管理操作不会影响系统整体状态
Bug复现:执行某些资源操作后,processed变为0,所有账户都无法召回资源
-
+
核心验证点:
1. processed数量不会归零
2. 搜索功能始终正常工作
3. 资源操作不会影响系统稳定性
"""
-
+
def test_processed_not_zero_after_resource_ops(self, api_client):
"""核心测试:资源操作后,processed不能归零,搜索必须正常"""
random_id = str(uuid.uuid4())[:8]
-
+
# 创建临时测试文件
test_file_path, temp_dir = create_test_file(
content=f"测试文件 {random_id}\n这是一个用于账户隔离测试的文件。\n包含关键词:test、隔离、验证。"
)
-
+
try:
# ==================== 步骤1: 获取初始状态 ====================
- print("\n" + "="*80)
+ print("\n" + "=" * 80)
print("步骤1: 获取初始VikingDB状态")
- print("="*80)
-
+ print("=" * 80)
+
response = api_client.observer_vikingdb()
assert response.status_code == 200, "observer_vikingdb should succeed"
observer_data_initial = response.json()
- assert observer_data_initial.get('status') == 'ok', "status should be ok"
-
- observer_initial = observer_data_initial.get('result', {})
- initial_processed = observer_initial.get('processed', 0)
+ assert observer_data_initial.get("status") == "ok", "status should be ok"
+
+ observer_initial = observer_data_initial.get("result", {})
+ initial_processed = observer_initial.get("processed", 0)
print(f"初始 processed 数量: {initial_processed}")
-
+
# ==================== 步骤2: 验证初始搜索正常 ====================
- print("\n" + "="*80)
+ print("\n" + "=" * 80)
print("步骤2: 验证初始搜索功能正常")
- print("="*80)
-
+ print("=" * 80)
+
search_query = "test"
response = api_client.search(search_query)
assert response.status_code == 200, "search should succeed"
search_data_initial = response.json()
- assert search_data_initial.get('status') == 'ok', "search status should be ok"
-
- search_result_initial = search_data_initial.get('result', {})
- has_memories_initial = 'memories' in search_result_initial
- has_resources_initial = 'resources' in search_result_initial
- assert has_memories_initial or has_resources_initial, "search should return memories or resources"
+ assert search_data_initial.get("status") == "ok", "search status should be ok"
+
+ search_result_initial = search_data_initial.get("result", {})
+ has_memories_initial = "memories" in search_result_initial
+ has_resources_initial = "resources" in search_result_initial
+ assert has_memories_initial or has_resources_initial, (
+ "search should return memories or resources"
+ )
print("初始搜索验证通过 ✓")
-
+
# ==================== 步骤3: 执行一些资源操作 ====================
- print("\n" + "="*80)
+ print("\n" + "=" * 80)
print("步骤3: 执行资源操作(添加资源)")
- print("="*80)
-
+ print("=" * 80)
+
# 添加资源
print("正在添加资源...")
response = api_client.add_resource(path=test_file_path, wait=True)
assert response.status_code == 200, "add_resource should succeed"
add_data = response.json()
- assert add_data.get('status') == 'ok', "add_resource status should be ok"
-
+ assert add_data.get("status") == "ok", "add_resource status should be ok"
+
print("等待处理完成...")
response = api_client.wait_processed()
assert response.status_code == 200
time.sleep(2)
-
+
# ==================== 步骤4: 第一次验证 ====================
- print("\n" + "="*80)
+ print("\n" + "=" * 80)
print("步骤4: 第一次验证 - processed和搜索")
- print("="*80)
-
+ print("=" * 80)
+
response = api_client.observer_vikingdb()
assert response.status_code == 200
observer_data_mid = response.json()
- assert observer_data_mid.get('status') == 'ok'
-
- observer_mid = observer_data_mid.get('result', {})
- mid_processed = observer_mid.get('processed', 0)
+ assert observer_data_mid.get("status") == "ok"
+
+ observer_mid = observer_data_mid.get("result", {})
+ mid_processed = observer_mid.get("processed", 0)
print(f"添加资源后 processed 数量: {mid_processed}")
-
+
# 如果初始processed > 0,则验证processed仍然 > 0
if initial_processed > 0:
assert mid_processed > 0, f"Processed should remain > 0, got {mid_processed}!"
-
+
# 验证搜索仍然正常
response = api_client.search(search_query)
assert response.status_code == 200, "search should still work"
search_data_mid = response.json()
- assert search_data_mid.get('status') == 'ok', "search status should still be ok"
-
- search_result_mid = search_data_mid.get('result', {})
- has_memories_mid = 'memories' in search_result_mid
- has_resources_mid = 'resources' in search_result_mid
+ assert search_data_mid.get("status") == "ok", "search status should still be ok"
+
+ search_result_mid = search_data_mid.get("result", {})
+ has_memories_mid = "memories" in search_result_mid
+ has_resources_mid = "resources" in search_result_mid
assert has_memories_mid or has_resources_mid, "search should still return results"
print("第一次验证通过 ✓")
-
+
# ==================== 步骤5: 执行更多操作 ====================
- print("\n" + "="*80)
+ print("\n" + "=" * 80)
print("步骤5: 执行更多操作(多次搜索)")
- print("="*80)
-
+ print("=" * 80)
+
for i in range(3):
query = f"test query {i} {random_id}"
- print(f"执行搜索 {i+1}: {query}")
+ print(f"执行搜索 {i + 1}: {query}")
response = api_client.search(query)
assert response.status_code == 200
search_data = response.json()
- assert search_data.get('status') == 'ok'
-
+ assert search_data.get("status") == "ok"
+
# ==================== 步骤6: 最终验证 ====================
- print("\n" + "="*80)
+ print("\n" + "=" * 80)
print("步骤6: 最终验证")
- print("="*80)
-
+ print("=" * 80)
+
response = api_client.observer_vikingdb()
assert response.status_code == 200
observer_data_final = response.json()
- assert observer_data_final.get('status') == 'ok'
-
- observer_final = observer_data_final.get('result', {})
- final_processed = observer_final.get('processed', 0)
+ assert observer_data_final.get("status") == "ok"
+
+ observer_final = observer_data_final.get("result", {})
+ final_processed = observer_final.get("processed", 0)
print(f"最终 processed 数量: {final_processed}")
-
+
# ==================== 关键断言 - Bug检测 ====================
-
+
# 断言1: 如果初始processed > 0,则最终processed也应该 > 0
if initial_processed > 0:
- assert final_processed > 0, f"❌ FAILED: Processed count dropped to ZERO! Initial: {initial_processed}, Final: {final_processed}"
-
+ assert final_processed > 0, (
+ f"❌ FAILED: Processed count dropped to ZERO! Initial: {initial_processed}, Final: {final_processed}"
+ )
+
# 断言2: 搜索必须仍然正常工作
response = api_client.search(search_query)
assert response.status_code == 200, "❌ FAILED: Search request failed"
final_search_data = response.json()
- assert final_search_data.get('status') == 'ok', "❌ FAILED: Search status not ok"
-
- final_search_result = final_search_data.get('result', {})
- has_memories_final = 'memories' in final_search_result
- has_resources_final = 'resources' in final_search_result
+ assert final_search_data.get("status") == "ok", "❌ FAILED: Search status not ok"
+
+ final_search_result = final_search_data.get("result", {})
+ has_memories_final = "memories" in final_search_result
+ has_resources_final = "resources" in final_search_result
assert has_memories_final or has_resources_final, "❌ FAILED: Search returns no results"
-
- print("\n" + "="*80)
- print(f"✅ TEST PASSED! 所有断言通过!")
+
+ print("\n" + "=" * 80)
+ print("✅ TEST PASSED! 所有断言通过!")
print(f" - 初始 processed: {initial_processed}")
print(f" - 最终 processed: {final_processed}")
- print(f" - 搜索功能正常")
+ print(" - 搜索功能正常")
if initial_processed > 0:
- print(f" - Processed 没有归零 ✓")
- print("="*80)
+ print(" - Processed 没有归零 ✓")
+ print("=" * 80)
finally:
# 清理临时文件
if os.path.exists(temp_dir):
shutil.rmtree(temp_dir)
-
+
def test_consecutive_health_checks(self, api_client):
"""附加测试:连续健康检查,验证系统稳定性"""
- for i in range(5):
+ for _ in range(5):
response = api_client.is_healthy()
assert response.status_code == 200
health_data = response.json()
- assert health_data.get('status') == 'ok'
+ assert health_data.get("status") == "ok"
time.sleep(0.5)
-
+
# 最后验证processed仍然>0
response = api_client.observer_vikingdb()
observer_data = response.json()
- observer = observer_data.get('result', {})
- processed = observer.get('processed', 0)
+ observer = observer_data.get("result", {})
+ processed = observer.get("processed", 0)
assert processed >= 0, "Processed should not be negative"
diff --git a/tests/api_test/scenarios/stability_error/test_concurrent_write.py b/tests/api_test/scenarios/stability_error/test_concurrent_write.py
index 335e0e256..66972354e 100644
--- a/tests/api_test/scenarios/stability_error/test_concurrent_write.py
+++ b/tests/api_test/scenarios/stability_error/test_concurrent_write.py
@@ -1,24 +1,23 @@
-import pytest
-import json
-import uuid
import concurrent.futures
import os
import shutil
+import uuid
+
from conftest import create_test_file
class TestConcurrentWrite:
"""TC-ER02 并发写入冲突验证"""
-
+
def test_concurrent_write_conflict(self, api_client):
"""并发写入冲突验证:并发调用 add_resource (Same URI)"""
random_id = str(uuid.uuid4())[:8]
-
+
# 1. 创建临时测试文件
test_file_path, temp_dir = create_test_file(
content=f"测试文件 {random_id}\n这是一个用于并发写入测试的文件。\n包含关键词:test、并发、写入。"
)
-
+
try:
# 2. 定义并发任务函数
def add_resource_task():
@@ -26,26 +25,28 @@ class TestConcurrentWrite:
response = api_client.add_resource(path=test_file_path, wait=True)
return response.status_code, response.json()
except Exception as e:
- return 500, {'error': str(e)}
-
+ return 500, {"error": str(e)}
+
# 3. 并发执行多个任务
num_tasks = 3
results = []
-
+
with concurrent.futures.ThreadPoolExecutor(max_workers=num_tasks) as executor:
futures = [executor.submit(add_resource_task) for _ in range(num_tasks)]
for future in concurrent.futures.as_completed(futures):
results.append(future.result())
-
+
# 4. 验证所有请求都返回合理的响应
assert len(results) == num_tasks
-
+
for status_code, response_data in results:
# 要么成功(200),要么返回合理的错误(429或其他)
assert status_code in [200, 429, 500], f"Unexpected status code: {status_code}"
-
+
if status_code == 200:
- assert response_data.get('status') in ['ok', 'error'], "Response should have valid status"
+ assert response_data.get("status") in ["ok", "error"], (
+ "Response should have valid status"
+ )
finally:
# 清理临时文件
if os.path.exists(temp_dir):
diff --git a/tests/engine/test_common.cpp b/tests/engine/test_common.cpp
index 83be716c5..e60f69038 100644
--- a/tests/engine/test_common.cpp
+++ b/tests/engine/test_common.cpp
@@ -13,130 +13,128 @@
#include "spdlog/spdlog.h"
// Simple Assertion Macro
-#define ASSERT_EQ(a, b) \
- if ((a) != (b)) { \
- SPDLOG_ERROR("Assertion failed: {} != {} ({} != {}) at {}:{}", #a, #b, (a), (b), __FILE__, __LINE__); \
- std::exit(1); \
+#define ASSERT_EQ(a, b) \
+ if ((a) != (b)) { \
+ SPDLOG_ERROR("Assertion failed: {} != {} ({} != {}) at {}:{}", #a, #b, \
+ (a), (b), __FILE__, __LINE__); \
+ std::exit(1); \
}
-#define ASSERT_TRUE(a) \
- if (!(a)) { \
+#define ASSERT_TRUE(a) \
+ if (!(a)) { \
SPDLOG_ERROR("Assertion failed: {} at {}:{}", #a, __FILE__, __LINE__); \
- std::exit(1); \
+ std::exit(1); \
}
using namespace vectordb;
void test_string_split() {
- SPDLOG_INFO("[Running] test_string_split");
- std::vector<std::string> tokens;
-
- // Case 1: Normal split
- split(tokens, "apple,banana,orange", ",");
- ASSERT_EQ(tokens.size(), 3);
- ASSERT_EQ(tokens[0], "apple");
- ASSERT_EQ(tokens[1], "banana");
- ASSERT_EQ(tokens[2], "orange");
- tokens.clear();
-
- // Case 2: Empty string
- split(tokens, "", ",");
- ASSERT_EQ(tokens.size(), 0);
- tokens.clear();
-
- // Case 3: No delimiter
- split(tokens, "helloworld", ",");
- ASSERT_EQ(tokens.size(), 1);
- ASSERT_EQ(tokens[0], "helloworld");
- tokens.clear();
-
- // Case 4: Continuous delimiters
- // According to split impl: if (end > start) tokens.push_back(...)
- // This means empty strings between continuous delimiters are skipped.
- split(tokens, "a,,b", ",");
- ASSERT_EQ(tokens.size(), 2);
- ASSERT_EQ(tokens[0], "a");
- ASSERT_EQ(tokens[1], "b");
- tokens.clear();
-
- SPDLOG_INFO("[Passed] test_string_split");
+ SPDLOG_INFO("[Running] test_string_split");
+ std::vector<std::string> tokens;
+
+ // Case 1: Normal split
+ split(tokens, "apple,banana,orange", ",");
+ ASSERT_EQ(tokens.size(), 3);
+ ASSERT_EQ(tokens[0], "apple");
+ ASSERT_EQ(tokens[1], "banana");
+ ASSERT_EQ(tokens[2], "orange");
+ tokens.clear();
+
+ // Case 2: Empty string
+ split(tokens, "", ",");
+ ASSERT_EQ(tokens.size(), 0);
+ tokens.clear();
+
+ // Case 3: No delimiter
+ split(tokens, "helloworld", ",");
+ ASSERT_EQ(tokens.size(), 1);
+ ASSERT_EQ(tokens[0], "helloworld");
+ tokens.clear();
+
+ // Case 4: Continuous delimiters
+ // According to split impl: if (end > start) tokens.push_back(...)
+ // This means empty strings between continuous delimiters are skipped.
+ split(tokens, "a,,b", ",");
+ ASSERT_EQ(tokens.size(), 2);
+ ASSERT_EQ(tokens[0], "a");
+ ASSERT_EQ(tokens[1], "b");
+ tokens.clear();
+
+ SPDLOG_INFO("[Passed] test_string_split");
}
void test_string_format() {
- SPDLOG_INFO("[Running] test_string_format");
-
- // Case 1: Basic format
- std::string s1 = sformat("Hello {}", "World");
- ASSERT_EQ(s1, "Hello World");
+ SPDLOG_INFO("[Running] test_string_format");
- // Case 2: Multiple arguments
- std::string s2 = sformat("{} + {} = {}", 1, 2, 3);
- ASSERT_EQ(s2, "1 + 2 = 3");
+ // Case 1: Basic format
+ std::string s1 = sformat("Hello {}", "World");
+ ASSERT_EQ(s1, "Hello World");
- // Case 3: Float number (Check prefix only due to to_string precision)
- std::string s3 = sformat("Value: {}", 3.14);
- ASSERT_TRUE(s3.find("Value: 3.14") == 0);
+ // Case 2: Multiple arguments
+ std::string s2 = sformat("{} + {} = {}", 1, 2, 3);
+ ASSERT_EQ(s2, "1 + 2 = 3");
- // Case 4: Fewer arguments than placeholders (Should preserve {})
- std::string s4 = sformat("{} {}", "OnlyOne");
- ASSERT_EQ(s4, "OnlyOne {}");
+ // Case 3: Float number (Check prefix only due to to_string precision)
+ std::string s3 = sformat("Value: {}", 3.14);
+ ASSERT_TRUE(s3.find("Value: 3.14") == 0);
- // Case 5: More arguments than placeholders (Extra args ignored)
- std::string s5 = sformat("{}", "First", "Second");
- ASSERT_EQ(s5, "First");
+ // Case 4: Fewer arguments than placeholders (Should preserve {})
+ std::string s4 = sformat("{} {}", "OnlyOne");
+ ASSERT_EQ(s4, "OnlyOne {}");
- SPDLOG_INFO("[Passed] test_string_format");
+ // Case 5: More arguments than placeholders (Extra args ignored)
+ std::string s5 = sformat("{}", "First", "Second");
+ ASSERT_EQ(s5, "First");
+
+ SPDLOG_INFO("[Passed] test_string_format");
}
void test_zip_sort() {
- SPDLOG_INFO("[Running] test_zip_sort");
-
- // Case 1: Basic sort
- std::vector<int> keys = {3, 1, 4, 2};
- std::vector<std::string> values = {"C", "A", "D", "B"};
-
- // Sort ascending by key
- ZipSortBranchOptimized(
- [](int a, int b) { return a < b; },
- keys.begin(), keys.end(),
- values.begin(), values.end()
- );
-
- ASSERT_EQ(keys[0], 1); ASSERT_EQ(values[0], "A");
- ASSERT_EQ(keys[1], 2); ASSERT_EQ(values[1], "B");
- ASSERT_EQ(keys[2], 3); ASSERT_EQ(values[2], "C");
- ASSERT_EQ(keys[3], 4); ASSERT_EQ(values[3], "D");
-
- // Case 2: Empty array
- std::vector<int> empty_keys;
- std::vector<int> empty_vals;
- ZipSortBranchOptimized(
- [](int a, int b) { return a < b; },
- empty_keys.begin(), empty_keys.end(),
- empty_vals.begin(), empty_vals.end()
- );
- ASSERT_TRUE(empty_keys.empty());
-
- // Case 3: Single element
- std::vector<int> single_key = {1};
- std::vector<int> single_val = {100};
- ZipSortBranchOptimized(
- [](int a, int b) { return a < b; },
- single_key.begin(), single_key.end(),
- single_val.begin(), single_val.end()
- );
- ASSERT_EQ(single_key[0], 1);
- ASSERT_EQ(single_val[0], 100);
-
- SPDLOG_INFO("[Passed] test_zip_sort");
+ SPDLOG_INFO("[Running] test_zip_sort");
+
+ // Case 1: Basic sort
+ std::vector<int> keys = {3, 1, 4, 2};
+ std::vector<std::string> values = {"C", "A", "D", "B"};
+
+ // Sort ascending by key
+ ZipSortBranchOptimized([](int a, int b) { return a < b; }, keys.begin(),
+ keys.end(), values.begin(), values.end());
+
+ ASSERT_EQ(keys[0], 1);
+ ASSERT_EQ(values[0], "A");
+ ASSERT_EQ(keys[1], 2);
+ ASSERT_EQ(values[1], "B");
+ ASSERT_EQ(keys[2], 3);
+ ASSERT_EQ(values[2], "C");
+ ASSERT_EQ(keys[3], 4);
+ ASSERT_EQ(values[3], "D");
+
+ // Case 2: Empty array
+ std::vector<int> empty_keys;
+ std::vector<int> empty_vals;
+ ZipSortBranchOptimized([](int a, int b) { return a < b; }, empty_keys.begin(),
+ empty_keys.end(), empty_vals.begin(),
+ empty_vals.end());
+ ASSERT_TRUE(empty_keys.empty());
+
+ // Case 3: Single element
+ std::vector<int> single_key = {1};
+ std::vector<int> single_val = {100};
+ ZipSortBranchOptimized([](int a, int b) { return a < b; }, single_key.begin(),
+ single_key.end(), single_val.begin(),
+ single_val.end());
+ ASSERT_EQ(single_key[0], 1);
+ ASSERT_EQ(single_val[0], 100);
+
+ SPDLOG_INFO("[Passed] test_zip_sort");
}
int main() {
- init_logging("INFO", "stdout", "[%Y-%m-%d %H:%M:%S.%e] [%l] %v");
- SPDLOG_INFO("Starting Common Tests...");
- test_string_split();
- test_string_format();
- test_zip_sort();
- SPDLOG_INFO("All Common Tests Passed!");
- return 0;
+ init_logging("INFO", "stdout", "[%Y-%m-%d %H:%M:%S.%e] [%l] %v");
+ SPDLOG_INFO("Starting Common Tests...");
+ test_string_split();
+ test_string_format();
+ test_zip_sort();
+ SPDLOG_INFO("All Common Tests Passed!");
+ return 0;
}
diff --git a/tests/engine/test_index_engine.cpp b/tests/engine/test_index_engine.cpp
index d24b12bfe..a590c6cae 100644
--- a/tests/engine/test_index_engine.cpp
+++ b/tests/engine/test_index_engine.cpp
@@ -13,21 +13,21 @@ using namespace vectordb;
// Helper to check float equality
bool is_close(float a, float b, float epsilon = 1e-5) {
- return std::fabs(a - b) < epsilon;
+ return std::fabs(a - b) < epsilon;
}
void test_basic_workflow() {
- SPDLOG_INFO("[Running] test_basic_workflow...");
-
- std::string db_path = "test_data_cpp/basic_workflow";
- // Cleanup
- if (std::filesystem::exists(db_path)) {
- std::filesystem::remove_all(db_path);
- }
- std::filesystem::create_directories(db_path);
-
- // 1. Initialization (Using JSON config)
- std::string config = R"({
+ SPDLOG_INFO("[Running] test_basic_workflow...");
+
+ std::string db_path = "test_data_cpp/basic_workflow";
+ // Cleanup
+ if (std::filesystem::exists(db_path)) {
+ std::filesystem::remove_all(db_path);
+ }
+ std::filesystem::create_directories(db_path);
+
+ // 1. Initialization (Using JSON config)
+ std::string config = R"({
"vector_index": {
"type": "flat",
"dim": 4,
@@ -39,83 +39,86 @@ void test_basic_workflow() {
"price": "float32"
}
})";
-
- IndexEngine engine(config);
- if (!engine.is_valid()) {
- SPDLOG_ERROR("Engine initialization failed");
- exit(1);
- }
-
- // 2. Add Data
- std::vector<AddDataRequest> add_reqs;
-
- AddDataRequest req1;
- req1.label = 1001;
- req1.vector = {0.1, 0.1, 0.1, 0.1};
- req1.fields_str = R"({"title": "apple", "count": 10, "price": 5.5})";
- add_reqs.push_back(req1);
-
- AddDataRequest req2;
- req2.label = 1002;
- req2.vector = {0.2, 0.2, 0.2, 0.2};
- req2.fields_str = R"({"title": "banana", "count": 20, "price": 3.0})";
- add_reqs.push_back(req2);
-
- int ret = engine.add_data(add_reqs);
- if (ret != 0) {
- SPDLOG_ERROR("Add data failed");
- exit(1);
- }
-
- // 3. Search (Vector only)
- SearchRequest search_req;
- search_req.query = {0.1, 0.1, 0.1, 0.1};
- search_req.topk = 5;
-
- SearchResult res = engine.search(search_req);
- if (res.result_num < 1) {
- SPDLOG_ERROR("Search failed: no result found");
- exit(1);
- }
- if (res.labels[0] != 1001) {
- SPDLOG_ERROR("Search failed: expected label 1001, got {}", res.labels[0]);
- exit(1);
- }
-
- // 4. Delete Data
- std::vector<DeleteDataRequest> del_reqs(1);
- del_reqs[0].label = 1001;
- del_reqs[0].old_fields_str = R"({"title": "apple", "count": 10, "price": 5.5})";
-
- ret = engine.delete_data(del_reqs);
- if (ret != 0) {
- SPDLOG_ERROR("Delete data failed");
- exit(1);
- }
-
- // 5. Search again to verify deletion
- res = engine.search(search_req);
- // Depending on soft delete implementation, result might still be there but filtered,
- // or simply not returned. For brute force, it usually checks filter.
- // If it returns, ensure it's not the deleted one or handle accordingly.
- if (res.result_num > 0 && res.labels[0] == 1001) {
- SPDLOG_WARN("Deleted item 1001 still found (might be soft delete delay or consistency model)");
- } else {
- SPDLOG_INFO("Deleted item 1001 correctly not found or ranked lower");
- }
-
- // 6. Dump
- int64_t ts = engine.dump(db_path);
- if (ts <= 0) {
- SPDLOG_ERROR("Dump failed");
- exit(1);
- }
-
- SPDLOG_INFO("[Passed] test_basic_workflow");
+
+ IndexEngine engine(config);
+ if (!engine.is_valid()) {
+ SPDLOG_ERROR("Engine initialization failed");
+ exit(1);
+ }
+
+ // 2. Add Data
+ std::vector<AddDataRequest> add_reqs;
+
+ AddDataRequest req1;
+ req1.label = 1001;
+ req1.vector = {0.1, 0.1, 0.1, 0.1};
+ req1.fields_str = R"({"title": "apple", "count": 10, "price": 5.5})";
+ add_reqs.push_back(req1);
+
+ AddDataRequest req2;
+ req2.label = 1002;
+ req2.vector = {0.2, 0.2, 0.2, 0.2};
+ req2.fields_str = R"({"title": "banana", "count": 20, "price": 3.0})";
+ add_reqs.push_back(req2);
+
+ int ret = engine.add_data(add_reqs);
+ if (ret != 0) {
+ SPDLOG_ERROR("Add data failed");
+ exit(1);
+ }
+
+ // 3. Search (Vector only)
+ SearchRequest search_req;
+ search_req.query = {0.1, 0.1, 0.1, 0.1};
+ search_req.topk = 5;
+
+ SearchResult res = engine.search(search_req);
+ if (res.result_num < 1) {
+ SPDLOG_ERROR("Search failed: no result found");
+ exit(1);
+ }
+ if (res.labels[0] != 1001) {
+ SPDLOG_ERROR("Search failed: expected label 1001, got {}", res.labels[0]);
+ exit(1);
+ }
+
+ // 4. Delete Data
+ std::vector<DeleteDataRequest> del_reqs(1);
+ del_reqs[0].label = 1001;
+ del_reqs[0].old_fields_str =
+ R"({"title": "apple", "count": 10, "price": 5.5})";
+
+ ret = engine.delete_data(del_reqs);
+ if (ret != 0) {
+ SPDLOG_ERROR("Delete data failed");
+ exit(1);
+ }
+
+ // 5. Search again to verify deletion
+ res = engine.search(search_req);
+ // Depending on soft delete implementation, result might still be there but
+ // filtered, or simply not returned. For brute force, it usually checks
+ // filter. If it returns, ensure it's not the deleted one or handle
+ // accordingly.
+ if (res.result_num > 0 && res.labels[0] == 1001) {
+ SPDLOG_WARN(
+ "Deleted item 1001 still found (might be soft delete delay or consistency model)");
+ } else {
+ SPDLOG_INFO("Deleted item 1001 correctly not found or ranked lower");
+ }
+
+ // 6. Dump
+ int64_t ts = engine.dump(db_path);
+ if (ts <= 0) {
+ SPDLOG_ERROR("Dump failed");
+ exit(1);
+ }
+
+ SPDLOG_INFO("[Passed] test_basic_workflow");
}
int main() {
- init_logging("INFO", "stdout");
- test_basic_workflow();
- return 0;
+ init_logging("INFO", "stdout");
+ test_basic_workflow();
+ return 0;
}
diff --git a/tests/integration/test_agent_memory_e2e.py b/tests/integration/test_agent_memory_e2e.py
index 8571c178d..0f4473b23 100644
--- a/tests/integration/test_agent_memory_e2e.py
+++ b/tests/integration/test_agent_memory_e2e.py
@@ -26,13 +26,10 @@ Run
from __future__ import annotations
import logging
+import os
import shutil
-
-from openviking.telemetry import tracer
-from openviking.telemetry.tracer import init_tracer_from_config
import time
import uuid
-import os
from pathlib import Path
from typing import Dict, Iterator, List, Tuple
@@ -43,6 +40,8 @@ from openviking.core.namespace import to_agent_space
from openviking.server.identity import AccountNamespacePolicy
from openviking.session.memory.session_extract_context_provider import SessionExtractContextProvider
from openviking.session.memory.utils.content import deserialize_metadata
+from openviking.telemetry import tracer
+from openviking.telemetry.tracer import init_tracer_from_config
from openviking_cli.session.user_id import UserIdentifier
from openviking_cli.utils import run_async
from openviking_cli.utils.config import OpenVikingConfigSingleton, get_openviking_config
@@ -60,6 +59,7 @@ def _flush_tracer_provider() -> None:
except Exception as e:
logger.warning("Failed to flush test tracer provider: %s", e)
+
# ── Conversation fixtures ─────────────────────────────────────────────────────
@@ -84,8 +84,7 @@ CONV_A_FLIGHT_DUPLICATE: List[Tuple[str, str]] = [
),
(
"assistant",
- "检测到重复预订,我先询问你的偏好。你是想取消已有的 CA1501 换成 MU5101,"
- "还是保留现有预订?",
+ "检测到重复预订,我先询问你的偏好。你是想取消已有的 CA1501 换成 MU5101,还是保留现有预订?",
),
("user", "那就取消 CA1501,改订 MU5101"),
("assistant", "[tool_call: cancel_booking(booking_id=CA1501-xyz)] 已取消原预订。"),
@@ -169,7 +168,9 @@ def _run_conversation(client: LocalClient, turns: List[Tuple[str, str]]) -> None
run_async(client.add_message(session_id=session_id, role=role, content=content))
logger.info(f" Committing {len(turns)} messages...")
result = run_async(client.commit_session(session_id=session_id))
- task_id = result.get("task_id") if isinstance(result, dict) else getattr(result, "task_id", None)
+ task_id = (
+ result.get("task_id") if isinstance(result, dict) else getattr(result, "task_id", None)
+ )
if task_id:
_wait_for_task(client, task_id)
logger.info(f" Done (task {task_id[:8]})")
@@ -282,8 +283,7 @@ class TestAgentMemoryE2E:
initialized = init_tracer_from_config()
if initialized is None or not tracer.is_enabled():
pytest.fail(
- "failed to initialize tracer from ov.conf; "
- "please check legacy telemetry.tracer"
+ "failed to initialize tracer from ov.conf; please check legacy telemetry.tracer"
)
policy = local_test_env["policy"]
@@ -302,9 +302,8 @@ class TestAgentMemoryE2E:
logger.info("Round 1: flight booking duplicate (expect CREATE experience)")
_run_conversation(client, CONV_A_FLIGHT_DUPLICATE)
- traj_after_r1 = _list_non_overview_entries(client, trajectories_dir)
- exp_after_r1 = _list_non_overview_entries(client, experiences_dir)
-
+ _list_non_overview_entries(client, trajectories_dir)
+ _list_non_overview_entries(client, experiences_dir)
logger.info("Round 2: booking conflict extra cases (expect EDIT experience)")
_run_conversation(client, CONV_B_FLIGHT_DUPLICATE_EXTRA)
@@ -323,6 +322,7 @@ class TestAgentMemoryE2E:
run_async(client.close())
_flush_tracer_provider()
+
class TestAgentMemorySchemas:
"""Unit tests for agent memory schema filtering — no integration environment needed."""
@@ -341,4 +341,3 @@ class TestAgentMemorySchemas:
assert "experiences" not in schema_types, (
"experiences schema must not appear in user memory extraction"
)
-
diff --git a/tests/integration/test_compressor_v2_event_span_multiple_turns.py b/tests/integration/test_compressor_v2_event_span_multiple_turns.py
index fb8e5f786..98f45a2bc 100644
--- a/tests/integration/test_compressor_v2_event_span_multiple_turns.py
+++ b/tests/integration/test_compressor_v2_event_span_multiple_turns.py
@@ -99,7 +99,7 @@ def run_ingest(client: ov.SyncHTTPClient, session_id: str, wait_seconds: float):
console.rule(f"[bold]Phase 1: 写入对话 — {DISPLAY_NAME} ({len(CONVERSATION)} 轮)[/bold]")
session = client.create_session()
- session_id = session.get('session_id')
+ session_id = session.get("session_id")
console.print(f" Session: [bold cyan]{session_id}[/bold cyan]")
console.print()
@@ -110,8 +110,18 @@ def run_ingest(client: ov.SyncHTTPClient, session_id: str, wait_seconds: float):
total = len(CONVERSATION)
for i, turn in enumerate(CONVERSATION, 1):
console.print(f" [dim][{i}/{total}][/dim] 添加 user + assistant 消息...")
- client.add_message(session_id, role="user", parts=[{"type": "text", "text": turn["user"]}], created_at=session_time_str)
- client.add_message(session_id, role="assistant", parts=[{"type": "text", "text": turn["assistant"]}], created_at=session_time_str)
+ client.add_message(
+ session_id,
+ role="user",
+ parts=[{"type": "text", "text": turn["user"]}],
+ created_at=session_time_str,
+ )
+ client.add_message(
+ session_id,
+ role="assistant",
+ parts=[{"type": "text", "text": turn["assistant"]}],
+ created_at=session_time_str,
+ )
console.print()
console.print(f" 共添加 [bold]{total * 2}[/bold] 条消息")
@@ -135,7 +145,7 @@ def run_ingest(client: ov.SyncHTTPClient, session_id: str, wait_seconds: float):
console.print(f" [green]任务 {status},耗时 {elapsed:.2f}s[/green]")
console.print(f" Task 详情: {task}")
- console.print(f" [yellow]等待向量化完成...[/yellow]")
+ console.print(" [yellow]等待向量化完成...[/yellow]")
client.wait_processed()
if wait_seconds > 0:
@@ -151,7 +161,9 @@ def run_ingest(client: ov.SyncHTTPClient, session_id: str, wait_seconds: float):
def run_verify(client: ov.SyncHTTPClient):
"""验证记忆召回"""
console.print()
- console.rule(f"[bold]Phase 2: 验证记忆召回 — {DISPLAY_NAME} ({len(VERIFY_QUERIES)} 条查询)[/bold]")
+ console.rule(
+ f"[bold]Phase 2: 验证记忆召回 — {DISPLAY_NAME} ({len(VERIFY_QUERIES)} 条查询)[/bold]"
+ )
results_table = Table(
title=f"记忆召回验证 — {DISPLAY_NAME}",
@@ -185,7 +197,11 @@ def run_verify(client: ov.SyncHTTPClient):
uri = getattr(m, "uri", "")
score = getattr(m, "score", 0)
console.print(f" [green]Memory:[/green] {uri} (score: {score:.4f})")
- console.print(f" [dim]{text[:120]}...[/dim]" if len(text) > 120 else f" [dim]{text}[/dim]")
+ console.print(
+ f" [dim]{text[:120]}...[/dim]"
+ if len(text) > 120
+ else f" [dim]{text}[/dim]"
+ )
count += len(results.memories)
if hasattr(results, "resources") and results.resources:
@@ -193,9 +209,7 @@ def run_verify(client: ov.SyncHTTPClient):
text = getattr(r, "content", "") or getattr(r, "text", "") or str(r)
print(f" [DEBUG] resource text: {repr(text)}")
recall_texts.append(text)
- console.print(
- f" [blue]Resource:[/blue] {r.uri} (score: {r.score:.4f})"
- )
+ console.print(f" [blue]Resource:[/blue] {r.uri} (score: {r.score:.4f})")
count += len(results.resources)
if hasattr(results, "skills") and results.skills:
@@ -230,15 +244,12 @@ def main():
parser.add_argument(
"--session-id", default=DEFAULT_SESSION_ID, help=f"Session ID (默认: {DEFAULT_SESSION_ID})"
)
- parser.add_argument(
- "--wait", type=float, default=2, help="写入后等待秒数 (默认: 2)"
- )
+ parser.add_argument("--wait", type=float, default=2, help="写入后等待秒数 (默认: 2)")
args = parser.parse_args()
client = ov.SyncHTTPClient(
- url=args.url, api_key=args.api_key, agent_id=args.agent_id,
- timeout=180
+ url=args.url, api_key=args.api_key, agent_id=args.agent_id, timeout=180
)
try:
@@ -260,9 +271,7 @@ def main():
)
except Exception as e:
- console.print(
- Panel(f"[bold red]Error:[/bold red] {e}", style="red", width=PANEL_WIDTH)
- )
+ console.print(Panel(f"[bold red]Error:[/bold red] {e}", style="red", width=PANEL_WIDTH))
if __name__ == "__main__":
diff --git a/tests/integration/test_compressor_v2_tool_skill_memory.py b/tests/integration/test_compressor_v2_tool_skill_memory.py
index 838512fc1..5cceecdb6 100644
--- a/tests/integration/test_compressor_v2_tool_skill_memory.py
+++ b/tests/integration/test_compressor_v2_tool_skill_memory.py
@@ -37,21 +37,33 @@ CONVERSATION = [
"user": "帮我创建一个PPT演示文稿,主题是季度工作报告。",
"assistant": "好的,我先读取一下 ppt skill 的 SKILL.md 了解如何创建PPT。",
"tool_calls": [
- {"tool_name": "Read", "tool_uri": "tools:Read", "input": {"file_path": "/skills/ppt/SKILL.md"}}
+ {
+ "tool_name": "Read",
+ "tool_uri": "tools:Read",
+ "input": {"file_path": "/skills/ppt/SKILL.md"},
+ }
],
},
{
"user": "PPT需要包含三个部分:业绩回顾、业务分析和下季度计划。",
"assistant": "好的,我根据 SKILL.md 的指引来创建这三个部分的PPT。",
"tool_calls": [
- {"tool_name": "Read", "tool_uri": "tools:Read", "input": {"file_path": "/skills/ppt/SKILL.md"}}
+ {
+ "tool_name": "Read",
+ "tool_uri": "tools:Read",
+ "input": {"file_path": "/skills/ppt/SKILL.md"},
+ }
],
},
{
"user": "把PPT的模板换成蓝色主题。",
"assistant": "好的,我来修改PPT模板为蓝色主题。",
"tool_calls": [
- {"tool_name": "write_file", "tool_uri": "tools:write_file", "input": {"path": "template.pptx", "content": "蓝色主题模板"}}
+ {
+ "tool_name": "write_file",
+ "tool_uri": "tools:write_file",
+ "input": {"path": "template.pptx", "content": "蓝色主题模板"},
+ }
],
},
# ===== 工具调用:write_file =====
@@ -59,7 +71,14 @@ CONVERSATION = [
"user": "帮我写一个Python函数,计算斐波那契数列。",
"assistant": "我来写一个计算斐波那契数列的函数并保存到文件。",
"tool_calls": [
- {"tool_name": "write_file", "tool_uri": "tools:write_file", "input": {"path": "fibonacci.py", "content": "def fib(n):\n if n <= 1:\n return n\n return fib(n-1) + fib(n-2)\nprint(fib(10))"}}
+ {
+ "tool_name": "write_file",
+ "tool_uri": "tools:write_file",
+ "input": {
+ "path": "fibonacci.py",
+ "content": "def fib(n):\n if n <= 1:\n return n\n return fib(n-1) + fib(n-2)\nprint(fib(10))",
+ },
+ }
],
},
# ===== 工具调用:bash =====
@@ -67,7 +86,11 @@ CONVERSATION = [
"user": "执行一下这个Python文件,看看结果对不对。",
"assistant": "我来执行这个文件。",
"tool_calls": [
- {"tool_name": "Bash", "tool_uri": "tools:Bash", "input": {"command": "python fibonacci.py"}}
+ {
+ "tool_name": "Bash",
+ "tool_uri": "tools:Bash",
+ "input": {"command": "python fibonacci.py"},
+ }
],
},
# ===== Skill 调用:PDF =====
@@ -75,7 +98,11 @@ CONVERSATION = [
"user": "帮我把这份PDF文件提取文字内容。",
"assistant": "好的,我先读取一下 pdf skill 的 SKILL.md。",
"tool_calls": [
- {"tool_name": "Read", "tool_uri": "tools:Read", "input": {"file_path": "/skills/pdf/SKILL.md"}}
+ {
+ "tool_name": "Read",
+ "tool_uri": "tools:Read",
+ "input": {"file_path": "/skills/pdf/SKILL.md"},
+ }
],
},
{
@@ -103,7 +130,11 @@ CONVERSATION = [
"user": "帮我写一封邮件给客户,主题是项目进度汇报。",
"assistant": "好的,我先读取一下 email skill 的 SKILL.md 了解邮件格式。",
"tool_calls": [
- {"tool_name": "Read", "tool_uri": "tools:Read", "input": {"file_path": "/skills/email/SKILL.md"}}
+ {
+ "tool_name": "Read",
+ "tool_uri": "tools:Read",
+ "input": {"file_path": "/skills/email/SKILL.md"},
+ }
],
},
{
@@ -154,7 +185,7 @@ def run_ingest(client: ov.SyncHTTPClient, session_id: str, wait_seconds: float):
console.rule(f"[bold]Phase 1: 写入对话 — {DISPLAY_NAME} ({len(CONVERSATION)} 轮)[/bold]")
session = client.create_session()
- session_id = session.get('session_id')
+ session_id = session.get("session_id")
console.print(f" Session: [bold cyan]{session_id}[/bold cyan]")
console.print()
@@ -207,7 +238,7 @@ def run_ingest(client: ov.SyncHTTPClient, session_id: str, wait_seconds: float):
console.print(f" [green]任务 {status},耗时 {elapsed:.2f}s[/green]")
console.print(f" Task 详情: {task}")
- console.print(f" [yellow]等待向量化完成...[/yellow]")
+ console.print(" [yellow]等待向量化完成...[/yellow]")
client.wait_processed()
if wait_seconds > 0:
@@ -223,7 +254,9 @@ def run_ingest(client: ov.SyncHTTPClient, session_id: str, wait_seconds: float):
def run_verify(client: ov.SyncHTTPClient):
"""验证记忆召回"""
console.print()
- console.rule(f"[bold]Phase 2: 验证记忆召回 — {DISPLAY_NAME} ({len(VERIFY_QUERIES)} 条查询)[/bold]")
+ console.rule(
+ f"[bold]Phase 2: 验证记忆召回 — {DISPLAY_NAME} ({len(VERIFY_QUERIES)} 条查询)[/bold]"
+ )
results_table = Table(
title=f"记忆召回验证 — {DISPLAY_NAME}",
@@ -257,7 +290,11 @@ def run_verify(client: ov.SyncHTTPClient):
uri = getattr(m, "uri", "")
score = getattr(m, "score", 0)
console.print(f" [green]Memory:[/green] {uri} (score: {score:.4f})")
- console.print(f" [dim]{text[:120]}...[/dim]" if len(text) > 120 else f" [dim]{text}[/dim]")
+ console.print(
+ f" [dim]{text[:120]}...[/dim]"
+ if len(text) > 120
+ else f" [dim]{text}[/dim]"
+ )
count += len(results.memories)
if hasattr(results, "resources") and results.resources:
@@ -265,9 +302,7 @@ def run_verify(client: ov.SyncHTTPClient):
text = getattr(r, "content", "") or getattr(r, "text", "") or str(r)
print(f" [DEBUG] resource text: {repr(text)}")
recall_texts.append(text)
- console.print(
- f" [blue]Resource:[/blue] {r.uri} (score: {r.score:.4f})"
- )
+ console.print(f" [blue]Resource:[/blue] {r.uri} (score: {r.score:.4f})")
count += len(results.resources)
if hasattr(results, "skills") and results.skills:
@@ -275,8 +310,6 @@ def run_verify(client: ov.SyncHTTPClient):
all_text = " ".join(recall_texts)
hits = [kw for kw in expected if kw in all_text]
- misses = [kw for kw in expected if kw not in all_text]
-
# 格式化关键词,命中的绿色,未命中的红色
formatted_keywords = []
for kw in expected:
@@ -299,7 +332,7 @@ def run_verify(client: ov.SyncHTTPClient):
def main():
"""入口函数"""
- parser = argparse.ArgumentParser(description=f"OpenViking 记忆演示 — 工具调用和Skill调用")
+ parser = argparse.ArgumentParser(description="OpenViking 记忆演示 — 工具调用和Skill调用")
parser.add_argument("--url", default=DEFAULT_URL, help=f"Server URL (默认: {DEFAULT_URL})")
parser.add_argument("--api-key", default=DEFAULT_API_KEY, help="API key")
parser.add_argument("--agent-id", default=DEFAULT_AGENT_ID, help="Agent ID")
@@ -312,15 +345,12 @@ def main():
parser.add_argument(
"--session-id", default=DEFAULT_SESSION_ID, help=f"Session ID (默认: {DEFAULT_SESSION_ID})"
)
- parser.add_argument(
- "--wait", type=float, default=2, help="写入后等待秒数 (默认: 2)"
- )
+ parser.add_argument("--wait", type=float, default=2, help="写入后等待秒数 (默认: 2)")
args = parser.parse_args()
client = ov.SyncHTTPClient(
- url=args.url, api_key=args.api_key, agent_id=args.agent_id,
- timeout=180
+ url=args.url, api_key=args.api_key, agent_id=args.agent_id, timeout=180
)
try:
@@ -342,10 +372,8 @@ def main():
)
except Exception as e:
- console.print(
- Panel(f"[bold red]Error:[/bold red] {e}", style="red", width=PANEL_WIDTH)
- )
+ console.print(Panel(f"[bold red]Error:[/bold red] {e}", style="red", width=PANEL_WIDTH))
if __name__ == "__main__":
- main()
\ No newline at end of file
+ main()
diff --git a/tests/integration/test_group_chat.py b/tests/integration/test_group_chat.py
index 8f879b625..fb1042275 100644
--- a/tests/integration/test_group_chat.py
+++ b/tests/integration/test_group_chat.py
@@ -54,7 +54,11 @@ CONVERSATION_1 = [
{"role_id": "agent-a", "role": "assistant", "content": "好的记住了"},
# user 告诉 agent 它的身份和风格
{"role_id": "alice", "role": "user", "content": "你叫 Agent A,是我的技术助手,说话要简洁专业"},
- {"role_id": "agent-a", "role": "assistant", "content": "好的,我记下了,我是 Agent A,技术助手,简洁专业"},
+ {
+ "role_id": "agent-a",
+ "role": "assistant",
+ "content": "好的,我记下了,我是 Agent A,技术助手,简洁专业",
+ },
]
CONVERSATION_2 = [
@@ -63,7 +67,11 @@ CONVERSATION_2 = [
{"role_id": "agent-b", "role": "assistant", "content": "好的"},
# user 告诉另一个 agent 它的身份和风格
{"role_id": "alice", "role": "user", "content": "你叫 Agent B,是我的生活助手,说话要亲切详细"},
- {"role_id": "agent-b", "role": "assistant", "content": "好的,我记下了,我是 Agent B,生活助手,亲切详细"},
+ {
+ "role_id": "agent-b",
+ "role": "assistant",
+ "content": "好的,我记下了,我是 Agent B,生活助手,亲切详细",
+ },
]
CONVERSATION_3 = [
@@ -139,13 +147,11 @@ def verify_isolation(url: str, api_key: str, account: str):
("密码", True, "alice", "agent-a", "alice+agent-a 应该能看到密码记忆"),
("密码", False, "alice", "agent-b", "alice+agent-b 不应看到密码"),
("蓝色", False, "alice", "agent-a", "alice+agent-a 不应看到颜色"),
-
# === Agent Soul 记忆测试 ===
# agent-a 的身份是"技术助手,简洁专业"
("技术助手", True, "alice", "agent-a", "alice+agent-a 应该能看到 agent-a 的 soul"),
("技术助手", False, "alice", "agent-b", "alice+agent-b 不应看到 agent-a 的 soul"),
("编程助手", False, "alice", "agent-b", "alice+agent-b 不应看到 agent-a 的 soul"),
-
# agent-b 的身份是"生活助手,亲切详细"
("生活助手", True, "alice", "agent-b", "alice+agent-b 应该能看到 agent-b 的 soul"),
("生活助手", False, "alice", "agent-a", "alice+agent-a 不应看到 agent-b 的 soul"),
@@ -204,11 +210,13 @@ def verify_isolation(url: str, api_key: str, account: str):
def run_test_for_account(account: str, url: str, root_key: str, wait: float) -> list:
- console.print(Panel(
- f"[bold cyan]测试 Account: {account}[/bold cyan]",
- style="magenta",
- width=PANEL_WIDTH,
- ))
+ console.print(
+ Panel(
+ f"[bold cyan]测试 Account: {account}[/bold cyan]",
+ style="magenta",
+ width=PANEL_WIDTH,
+ )
+ )
# 用 root key 创建 client
client = ov.SyncHTTPClient(
@@ -240,7 +248,9 @@ def run_test_for_account(account: str, url: str, root_key: str, wait: float) ->
},
)
if resp.status_code == 200:
- console.print(f" - 账号 {account} 已创建 (user={isolate_user}, agent={isolate_agent})")
+ console.print(
+ f" - 账号 {account} 已创建 (user={isolate_user}, agent={isolate_agent})"
+ )
elif "already exists" in resp.text:
console.print(f" - 账号 {account} 已存在")
else:
@@ -278,7 +288,7 @@ def run_test_for_account(account: str, url: str, root_key: str, wait: float) ->
console.print(f" [green]任务 {status},耗时 {elapsed:.2f}s[/green]")
# 等待向量化完成
- console.print(f" [yellow]等待向量化完成...[/yellow]")
+ console.print(" [yellow]等待向量化完成...[/yellow]")
client.wait_processed()
# 验证隔离
@@ -287,6 +297,7 @@ def run_test_for_account(account: str, url: str, root_key: str, wait: float) ->
except Exception as e:
console.print(f" [red]Error: {e}[/red]")
import traceback
+
traceback.print_exc()
return []
@@ -303,19 +314,25 @@ def main():
parser = argparse.ArgumentParser(description="群聊记忆测试 - 测试数据隔离")
parser.add_argument("--url", default=DEFAULT_URL, help="Server URL")
parser.add_argument("--root-key", default="default", help="Root API Key (默认: default)")
- parser.add_argument("--mode", choices=["all", "ff", "ft", "tf", "tt"], default="all",
- help="测试模式: all=4种都测, ff/ft/tf/tt=只测指定模式")
- parser.add_argument("--account", default=None,
- help="直接指定 account 名称(优先级高于 --mode)")
+ parser.add_argument(
+ "--mode",
+ choices=["all", "ff", "ft", "tf", "tt"],
+ default="all",
+ help="测试模式: all=4种都测, ff/ft/tf/tt=只测指定模式",
+ )
+ parser.add_argument(
+ "--account", default=None, help="直接指定 account 名称(优先级高于 --mode)"
+ )
parser.add_argument("--wait", type=float, default=5.0, help="提交后等待秒数")
args = parser.parse_args()
- console.print(Panel(
- "[bold]OpenViking 数据隔离测试[/bold]\n"
- f"Server: {args.url}",
- style="magenta",
- width=PANEL_WIDTH,
- ))
+ console.print(
+ Panel(
+ f"[bold]OpenViking 数据隔离测试[/bold]\nServer: {args.url}",
+ style="magenta",
+ width=PANEL_WIDTH,
+ )
+ )
# 确定要测试的账号列表
default_accounts = ["test-ff", "test-ft", "test-tf", "test-tt"]
@@ -334,19 +351,23 @@ def main():
console.print()
# 打印汇总
- trace_info = "\n".join(f" {acc}: {', '.join(tids)}" for acc, tids in all_trace_ids.items() if tids)
- console.print(Panel(
- f"[bold green]测试完成![/bold green]\n\n"
- f"Trace IDs:\n{trace_info}\n\n"
- "预期结果差异:\n"
- " ff: isolate_user=F, isolate_agent=F → 全部 ✓ (都不隔离)\n"
- " ft: isolate_user=F, isolate_agent=T → 3个 ✓ (只隔离 agent)\n"
- " tf: isolate_user=T, isolate_agent=F → 3个 ✓ (只隔离 user)\n"
- " tt: isolate_user=T, isolate_agent=T → 全部 ✓ (都隔离)",
- style="green",
- width=PANEL_WIDTH,
- ))
+ trace_info = "\n".join(
+ f" {acc}: {', '.join(tids)}" for acc, tids in all_trace_ids.items() if tids
+ )
+ console.print(
+ Panel(
+ f"[bold green]测试完成![/bold green]\n\n"
+ f"Trace IDs:\n{trace_info}\n\n"
+ "预期结果差异:\n"
+ " ff: isolate_user=F, isolate_agent=F → 全部 ✓ (都不隔离)\n"
+ " ft: isolate_user=F, isolate_agent=T → 3个 ✓ (只隔离 agent)\n"
+ " tf: isolate_user=T, isolate_agent=F → 3个 ✓ (只隔离 user)\n"
+ " tt: isolate_user=T, isolate_agent=T → 全部 ✓ (都隔离)",
+ style="green",
+ width=PANEL_WIDTH,
+ )
+ )
if __name__ == "__main__":
- main()
\ No newline at end of file
+ main()
diff --git a/tests/misc/test_zip_safe.py b/tests/misc/test_zip_safe.py
index 4f016340b..92e61ad34 100644
--- a/tests/misc/test_zip_safe.py
+++ b/tests/misc/test_zip_safe.py
@@ -19,7 +19,6 @@ from openviking.utils.zip_safe import (
safe_extract_zip,
)
-
# ── Helpers ──────────────────────────────────────────────────────────────────
@@ -147,7 +146,6 @@ class TestNormalizeZipFilenames:
zf.writestr(info, "content")
# Manually patch the raw zip to remove UTF-8 flag
# (Python's ZipFile may set it automatically for non-ASCII)
- raw = buf.getvalue()
# The flag_bits field is at offset 6 in the local file header
# This is complex; instead test the logic paths individually
# by directly calling with a pre-mangled ZipInfo
@@ -170,8 +168,7 @@ class TestNormalizeZipFilenames:
# After normalization, the name should be repaired to CJK
repaired = zf.infolist()[0]
assert repaired.filename == cjk_name, (
- f"Expected filename to be repaired to {cjk_name!r}, "
- f"got {repaired.filename!r}"
+ f"Expected filename to be repaired to {cjk_name!r}, got {repaired.filename!r}"
)
@@ -192,11 +189,13 @@ class TestSafeExtractZipNormal:
def test_extracts_nested_directories(self, tmp_path: Path) -> None:
dest = tmp_path / "out"
dest.mkdir()
- data = _make_zip_bytes({
- "src/main.py": "print('hello')",
- "src/utils/helper.py": "pass",
- "README.md": "# Test",
- })
+ data = _make_zip_bytes(
+ {
+ "src/main.py": "print('hello')",
+ "src/utils/helper.py": "pass",
+ "README.md": "# Test",
+ }
+ )
with zipfile.ZipFile(io.BytesIO(data), "r") as zf:
safe_extract_zip(zf, dest)
assert (dest / "src" / "main.py").read_text() == "print('hello')"
@@ -229,12 +228,14 @@ class TestSafeExtractZipNormal:
def test_handles_special_characters_in_filenames(self, tmp_path: Path) -> None:
dest = tmp_path / "out"
dest.mkdir()
- data = _make_zip_bytes({
- "spaces in name.txt": "content",
- "file (1).txt": "content",
- "file-with-dashes.txt": "content",
- "file_with_underscores.txt": "content",
- })
+ data = _make_zip_bytes(
+ {
+ "spaces in name.txt": "content",
+ "file (1).txt": "content",
+ "file-with-dashes.txt": "content",
+ "file_with_underscores.txt": "content",
+ }
+ )
with zipfile.ZipFile(io.BytesIO(data), "r") as zf:
safe_extract_zip(zf, dest)
assert (dest / "spaces in name.txt").exists()
@@ -302,11 +303,13 @@ class TestSafeExtractZipSlipPrevention:
"""Filenames with dots (not traversal) should be allowed."""
dest = tmp_path / "out"
dest.mkdir()
- data = _make_zip_bytes({
- ".gitignore": "*.pyc",
- "src/.env.example": "KEY=val",
- "dir/file.tar.gz": "data",
- })
+ data = _make_zip_bytes(
+ {
+ ".gitignore": "*.pyc",
+ "src/.env.example": "KEY=val",
+ "dir/file.tar.gz": "data",
+ }
+ )
with zipfile.ZipFile(io.BytesIO(data), "r") as zf:
safe_extract_zip(zf, dest)
assert (dest / ".gitignore").read_text() == "*.pyc"
@@ -325,10 +328,12 @@ class TestSafeExtractZipSlipPrevention:
"""If the first entry is malicious, no files should be extracted."""
dest = tmp_path / "out"
dest.mkdir()
- data = _make_zip_bytes({
- "../../evil.txt": "pwned",
- "safe.txt": "this should not be extracted",
- })
+ data = _make_zip_bytes(
+ {
+ "../../evil.txt": "pwned",
+ "safe.txt": "this should not be extracted",
+ }
+ )
with zipfile.ZipFile(io.BytesIO(data), "r") as zf:
with pytest.raises(ValueError, match="Zip Slip"):
safe_extract_zip(zf, dest)
diff --git a/tests/models/vlm/test_volcengine_cache.py b/tests/models/vlm/test_volcengine_cache.py
index 73defabcf..2b811c545 100644
--- a/tests/models/vlm/test_volcengine_cache.py
+++ b/tests/models/vlm/test_volcengine_cache.py
@@ -2,10 +2,10 @@
# SPDX-License-Identifier: AGPL-3.0
"""Tests for VolcEngineVLM cache logic."""
+from unittest.mock import AsyncMock, MagicMock
+
import pytest
-from unittest.mock import AsyncMock, MagicMock, patch
-from openviking.models.vlm.backends.volcengine_vlm import VolcEngineVLM
from openviking.models.vlm.backends.volcengine_vlm import VolcEngineVLM as VLMClass
@@ -38,7 +38,10 @@ class TestGetOrCreateFromSegments:
# 只有一个 segment [msg0, msg1(cache_control)]
segments = [
- [make_message("system", "You are a helpful assistant"), make_message("user", "Hello", cache_control=True)]
+ [
+ make_message("system", "You are a helpful assistant"),
+ make_message("user", "Hello", cache_control=True),
+ ]
]
# Mock cache hit
@@ -56,7 +59,10 @@ class TestGetOrCreateFromSegments:
vlm = self._create_vlm_with_mock_cache()
segments = [
- [make_message("system", "You are a helpful assistant"), make_message("user", "Hello", cache_control=True)]
+ [
+ make_message("system", "You are a helpful assistant"),
+ make_message("user", "Hello", cache_control=True),
+ ]
]
# Mock cache miss
@@ -82,7 +88,10 @@ class TestGetOrCreateFromSegments:
# segments = [[msg0, msg1(cc)], [msg2, msg3(cc)]]
segments = [
- [make_message("system", "You are a helpful assistant"), make_message("user", "Hello", cache_control=True)],
+ [
+ make_message("system", "You are a helpful assistant"),
+ make_message("user", "Hello", cache_control=True),
+ ],
[make_message("user", "How are you?", cache_control=True)],
]
@@ -250,4 +259,4 @@ class TestCacheKeyGeneration:
key = vlm._get_response_id_cache_key(messages)
# Should include prefix in the key
- assert "prefix:" in key or key.startswith("prefix:")
\ No newline at end of file
+ assert "prefix:" in key or key.startswith("prefix:")
diff --git a/tests/parse/test_ast_extractor.py b/tests/parse/test_ast_extractor.py
index 4372f6d6a..97b54d9ef 100644
--- a/tests/parse/test_ast_extractor.py
+++ b/tests/parse/test_ast_extractor.py
@@ -1434,5 +1434,3 @@ class TestASTExtractorDispatch:
verbose = self.extractor.extract_skeleton("m.py", code, verbose=True)
assert "Detail here." not in compact
assert "Detail here." in verbose
-
-
diff --git a/tests/parse/test_feishu_parser.py b/tests/parse/test_feishu_parser.py
index fff9655d7..d2c92877e 100644
--- a/tests/parse/test_feishu_parser.py
+++ b/tests/parse/test_feishu_parser.py
@@ -640,16 +640,15 @@ class TestParseAsyncIntegration:
mock_result = MagicMock()
mock_result.source_format = "feishu_docx"
mock_result.meta = {"feishu_token": "larkoffice123"}
-
+
async def _mock_parse(*a, **kw):
return mock_result
-
+
parser.parse = _mock_parse
result = asyncio.get_event_loop().run_until_complete(
parser.parse_content(
- content="",
- source_path="https://bytedance.larkoffice.com/wiki/larkoffice123"
+ content="", source_path="https://bytedance.larkoffice.com/wiki/larkoffice123"
)
)
diff --git a/tests/server/test_privacy_config_service.py b/tests/server/test_privacy_config_service.py
index fb553477f..6b8f7c243 100644
--- a/tests/server/test_privacy_config_service.py
+++ b/tests/server/test_privacy_config_service.py
@@ -42,7 +42,9 @@ async def test_privacy_config_service_versions(service):
assert second.version == 2
assert await privacy.list_versions(ctx, "skill", "demo-skill") == [1, 2]
- restored = await privacy.activate_version(ctx, "skill", "demo-skill", 1, updated_by=ctx.user.user_id)
+ restored = await privacy.activate_version(
+ ctx, "skill", "demo-skill", 1, updated_by=ctx.user.user_id
+ )
current = await privacy.get_current(ctx, "skill", "demo-skill")
assert restored.version == 1
assert current.version == 1
@@ -65,7 +67,9 @@ async def test_skill_read_restores_placeholder(service):
wait=False,
)
- stored = await service.viking_fs.read_file("viking://agent/skills/restore-skill/SKILL.md", ctx=ctx)
+ stored = await service.viking_fs.read_file(
+ "viking://agent/skills/restore-skill/SKILL.md", ctx=ctx
+ )
restored = await service.fs.read("viking://agent/skills/restore-skill/SKILL.md", ctx=ctx)
assert "secret-xyz" not in stored
@@ -76,7 +80,9 @@ async def test_skill_read_restores_placeholder(service):
@pytest.mark.asyncio
async def test_skill_read_restores_placeholder_with_agent_segment(service):
- ctx = RequestContext(user=UserIdentifier.the_default_user("privacy_restore_agent_segment"), role=Role.ROOT)
+ ctx = RequestContext(
+ user=UserIdentifier.the_default_user("privacy_restore_agent_segment"), role=Role.ROOT
+ )
await service.initialize_user_directories(ctx)
await service.initialize_agent_directories(ctx)
@@ -124,7 +130,9 @@ async def test_skill_privacy_extraction_returns_content_blocks():
@pytest.mark.asyncio
async def test_skill_read_appends_notice_for_unreplaced_placeholders(service):
- ctx = RequestContext(user=UserIdentifier.the_default_user("privacy_partial_restore"), role=Role.ROOT)
+ ctx = RequestContext(
+ user=UserIdentifier.the_default_user("privacy_partial_restore"), role=Role.ROOT
+ )
await service.initialize_user_directories(ctx)
await service.initialize_agent_directories(ctx)
@@ -150,14 +158,16 @@ async def test_skill_read_appends_notice_for_unreplaced_placeholders(service):
restored = await service.fs.read(f"{skill_uri}/SKILL.md", ctx=ctx)
assert 'api_key: "secret-xyz"' in restored
- assert '{{ov_privacy:skill:partial-restore-skill:region}}' in restored
+ assert "{{ov_privacy:skill:partial-restore-skill:region}}" in restored
assert "[Privacy Config Notice]" in restored
assert "Missing config: region=<missing>" in restored
@pytest.mark.asyncio
async def test_skill_read_appends_notice_for_extra_configured_keys(service):
- ctx = RequestContext(user=UserIdentifier.the_default_user("privacy_extra_config"), role=Role.ROOT)
+ ctx = RequestContext(
+ user=UserIdentifier.the_default_user("privacy_extra_config"), role=Role.ROOT
+ )
await service.initialize_user_directories(ctx)
await service.initialize_agent_directories(ctx)
@@ -203,11 +213,6 @@ async def test_privacy_config_service_uses_policy_aware_user_root(service):
)
current = await privacy.get_current(ctx, "skill", "policy-skill")
- stored = await service.viking_fs.read_file(
- "viking://user/privacy_user_policy/agent/demo-agent/privacy/skill/policy-skill/current.json",
- ctx=ctx,
- )
-
assert current is not None
@@ -233,7 +238,15 @@ async def test_placeholderization_only_replaces_structured_values(monkeypatch):
monkeypatch.setattr(
"openviking.privacy.skill_extractor.get_openviking_config",
- lambda: type("Cfg", (), {"vlm": type("VLM", (), {"get_completion_async": staticmethod(fake_completion_async)})()})(),
+ lambda: type(
+ "Cfg",
+ (),
+ {
+ "vlm": type(
+ "VLM", (), {"get_completion_async": staticmethod(fake_completion_async)}
+ )()
+ },
+ )(),
)
content = 'api_key: "secret-xyz"\n'
@@ -270,5 +283,5 @@ def test_placeholderization_replaces_all_structured_occurrences_for_same_value()
assert result.replaced_values == {"api_key": "secret"}
assert result.sanitized_content == (
'api_key: "{{ov_privacy:skill:multi-hit-skill:api_key}}"\n'
- 'backup={{ov_privacy:skill:multi-hit-skill:api_key}}\n'
+ "backup={{ov_privacy:skill:multi-hit-skill:api_key}}\n"
)
diff --git a/tests/server/test_watch_task_acl.py b/tests/server/test_watch_task_acl.py
index 956d0deab..02adf3d21 100644
--- a/tests/server/test_watch_task_acl.py
+++ b/tests/server/test_watch_task_acl.py
@@ -59,9 +59,24 @@ async def test_hidden_listing_filters_watch_task_control_files_for_non_root(
bare_viking_fs._uri_to_path = lambda uri, ctx=None: "/fake/resources"
bare_viking_fs._ctx_or_default = lambda ctx=None: ctx
bare_viking_fs._ls_entries = lambda path: [
- {"name": ".watch_tasks.json", "isDir": False, "size": 10, "modTime": "2026-01-01T00:00:00+00:00"},
- {"name": ".watch_tasks.json.bak", "isDir": False, "size": 10, "modTime": "2026-01-01T00:00:00+00:00"},
- {"name": ".watch_tasks.json.tmp", "isDir": False, "size": 10, "modTime": "2026-01-01T00:00:00+00:00"},
+ {
+ "name": ".watch_tasks.json",
+ "isDir": False,
+ "size": 10,
+ "modTime": "2026-01-01T00:00:00+00:00",
+ },
+ {
+ "name": ".watch_tasks.json.bak",
+ "isDir": False,
+ "size": 10,
+ "modTime": "2026-01-01T00:00:00+00:00",
+ },
+ {
+ "name": ".watch_tasks.json.tmp",
+ "isDir": False,
+ "size": 10,
+ "modTime": "2026-01-01T00:00:00+00:00",
+ },
{"name": "public.txt", "isDir": False, "size": 5, "modTime": "2026-01-01T00:00:00+00:00"},
]
bare_viking_fs._path_to_uri = lambda path, ctx=None: f"viking://resources/{path.split('/')[-1]}"
diff --git a/tests/session/memory/test_memory_diff.py b/tests/session/memory/test_memory_diff.py
index e945c85e2..406d8c1e6 100644
--- a/tests/session/memory/test_memory_diff.py
+++ b/tests/session/memory/test_memory_diff.py
@@ -7,14 +7,16 @@ Verifies that memory_diff.json is correctly written to the archive directory
containing adds, updates, and deletes.
"""
-import json
-from typing import Any, Dict
from unittest.mock import AsyncMock, MagicMock, patch
import pytest
from openviking.session.compressor_v2 import SessionCompressorV2
-from openviking.session.memory.dataclass import MemoryFileContent, ResolvedOperation, ResolvedOperations
+from openviking.session.memory.dataclass import (
+ MemoryFileContent,
+ ResolvedOperation,
+ ResolvedOperations,
+)
from openviking.session.memory.memory_updater import MemoryUpdateResult
from openviking.storage.viking_fs import VikingFS
@@ -52,19 +54,15 @@ class TestMemoryDiffArchive:
async def test_get_memory_type_from_uri(self, compressor):
"""Test memory type extraction from URI."""
# Test identity.md
- assert compressor._get_memory_type_from_uri(
- "memory/user/test/identity.md"
- ) == "identity"
+ assert compressor._get_memory_type_from_uri("memory/user/test/identity.md") == "identity"
# Test context/project.md
- assert compressor._get_memory_type_from_uri(
- "memory/user/test/context/project.md"
- ) == "project"
+ assert (
+ compressor._get_memory_type_from_uri("memory/user/test/context/project.md") == "project"
+ )
# Test unknown path
- assert compressor._get_memory_type_from_uri(
- "memory/user/test/unknown/path"
- ) == "unknown"
+ assert compressor._get_memory_type_from_uri("memory/user/test/unknown/path") == "unknown"
@pytest.mark.asyncio
async def test_build_memory_diff_add(self, compressor, mock_viking_fs, mock_ctx):
@@ -111,6 +109,7 @@ class TestMemoryDiffArchive:
@pytest.mark.asyncio
async def test_build_memory_diff_update(self, compressor, mock_viking_fs, mock_ctx):
"""Test building memory_diff for modified memory files (updates)."""
+
# Setup: files exist
async def mock_read(uri, ctx=None):
if uri == "memory/user/test/identity.md":
@@ -303,6 +302,7 @@ class TestMemoryDiffArchive:
# This test verifies the archive_uri parameter is accepted
# Full integration testing requires extensive mocking that's covered by other tests
import inspect
+
sig = inspect.signature(compressor.extract_long_term_memories)
params = list(sig.parameters.keys())
assert "archive_uri" in params, "archive_uri parameter should be accepted"
@@ -318,16 +318,18 @@ class TestMemoryDiffStructure:
# We verify this through the actual implementation tests above
# This is a placeholder for documentation
- assert set(expected_keys).issubset({
- "archive_uri",
- "extracted_at",
- "operations",
- "summary",
- })
+ assert set(expected_keys).issubset(
+ {
+ "archive_uri",
+ "extracted_at",
+ "operations",
+ "summary",
+ }
+ )
def test_operations_structure(self):
"""Verify operations structure in memory_diff."""
expected_ops_keys = ["adds", "updates", "deletes"]
# We verify this through the actual implementation tests above
- assert set(expected_ops_keys).issubset(set(expected_ops_keys))
\ No newline at end of file
+ assert set(expected_ops_keys).issubset(set(expected_ops_keys))
diff --git a/tests/session/memory/test_memory_isolation_handler.py b/tests/session/memory/test_memory_isolation_handler.py
index e54dbfc40..1b5c480d2 100644
--- a/tests/session/memory/test_memory_isolation_handler.py
+++ b/tests/session/memory/test_memory_isolation_handler.py
@@ -4,17 +4,15 @@
Tests for MemoryIsolationHandler.
"""
-import pytest
from unittest.mock import MagicMock, patch
from openviking.message.message import Message
from openviking.message.part import TextPart
from openviking.server.identity import AccountNamespacePolicy, RequestContext, Role
-from openviking_cli.session.user_id import UserIdentifier
from openviking.session.memory.memory_isolation_handler import (
MemoryIsolationHandler,
- RoleScope,
)
+from openviking_cli.session.user_id import UserIdentifier
def create_message(role: str, role_id: str, content: str = "test") -> Message:
@@ -371,7 +369,9 @@ class TestCalculateMemoryUris:
@patch("openviking.session.memory.memory_isolation_handler.generate_uri")
def test_calculate_memory_uris_multiple_users_agents(self, mock_generate_uri):
"""Test calculate_memory_uris with multiple users and agents."""
- mock_generate_uri.side_effect = lambda **kwargs: f"viking://user/{kwargs.get('user_space')}/memories/test"
+ mock_generate_uri.side_effect = lambda **kwargs: (
+ f"viking://user/{kwargs.get('user_space')}/memories/test"
+ )
ctx = create_ctx()
messages = [create_message("user", "user_a")]
diff --git a/tests/session/memory/test_merge_ops.py b/tests/session/memory/test_merge_ops.py
index 5b492e4a1..5932b4292 100644
--- a/tests/session/memory/test_merge_ops.py
+++ b/tests/session/memory/test_merge_ops.py
@@ -35,17 +35,17 @@ class TestPatchOp:
def test_get_output_schema_type_int(self):
"""Int field with patch should return int."""
op = PatchOp(FieldType.INT64)
- assert op.get_output_schema_type(FieldType.INT64) == int
+ assert op.get_output_schema_type(FieldType.INT64) is int
def test_get_output_schema_type_float(self):
"""Float field with patch should return float."""
op = PatchOp(FieldType.FLOAT32)
- assert op.get_output_schema_type(FieldType.FLOAT32) == float
+ assert op.get_output_schema_type(FieldType.FLOAT32) is float
def test_get_output_schema_type_bool(self):
"""Bool field with patch should return bool."""
op = PatchOp(FieldType.BOOL)
- assert op.get_output_schema_type(FieldType.BOOL) == bool
+ assert op.get_output_schema_type(FieldType.BOOL) is bool
def test_get_output_schema_description_string(self):
"""String field description should mention PATCH."""
@@ -76,8 +76,8 @@ class TestSumOp:
def test_get_output_schema_type(self):
"""SumOp should return appropriate numeric types."""
op = SumOp()
- assert op.get_output_schema_type(FieldType.INT64) == int
- assert op.get_output_schema_type(FieldType.FLOAT32) == float
+ assert op.get_output_schema_type(FieldType.INT64) is int
+ assert op.get_output_schema_type(FieldType.FLOAT32) is float
def test_get_output_schema_description(self):
"""Description should have 'add for' format."""
@@ -117,8 +117,8 @@ class TestImmutableOp:
def test_get_output_schema_type(self):
"""ImmutableOp should return base types."""
op = ImmutableOp()
- assert op.get_output_schema_type(FieldType.STRING) == str
- assert op.get_output_schema_type(FieldType.INT64) == int
+ assert op.get_output_schema_type(FieldType.STRING) is str
+ assert op.get_output_schema_type(FieldType.INT64) is int
def test_get_output_schema_description(self):
"""Description should mention immutable."""
@@ -220,9 +220,7 @@ class TestApplyStrPatch:
def test_simple_replace(self):
"""Simple replace."""
original = "hello world"
- patch = StrPatch(
- blocks=[SearchReplaceBlock(search="hello world", replace="hello there")]
- )
+ patch = StrPatch(blocks=[SearchReplaceBlock(search="hello world", replace="hello there")])
result = apply_str_patch(original, patch)
# Directly test apply_str_patch
assert result == "hello there"
diff --git a/tests/test_code_hosting_utils.py b/tests/test_code_hosting_utils.py
index 7fc615169..dfc823070 100644
--- a/tests/test_code_hosting_utils.py
+++ b/tests/test_code_hosting_utils.py
@@ -102,8 +102,7 @@ def test_parse_code_hosting_url_https_with_port():
def test_parse_code_hosting_url_azure_devops_repo():
assert (
- parse_code_hosting_url("https://dev.azure.com/org/project/_git/repo")
- == "org/project/repo"
+ parse_code_hosting_url("https://dev.azure.com/org/project/_git/repo") == "org/project/repo"
)
@@ -148,7 +147,10 @@ def test_parse_code_hosting_url_azure_devops_non_repo_page():
def test_parse_code_hosting_url_azure_devops_pull_request_page():
- assert parse_code_hosting_url("https://dev.azure.com/org/project/_git/repo/pullrequest/123") is None
+ assert (
+ parse_code_hosting_url("https://dev.azure.com/org/project/_git/repo/pullrequest/123")
+ is None
+ )
# --- validate_git_ssh_uri ---
@@ -255,10 +257,7 @@ def test_is_git_repo_url_azure_devops_non_repo_page():
def test_is_git_repo_url_azure_devops_browse_url_with_path_query():
- assert (
- is_git_repo_url("https://dev.azure.com/org/project/_git/repo?path=/README.md")
- is False
- )
+ assert is_git_repo_url("https://dev.azure.com/org/project/_git/repo?path=/README.md") is False
def test_is_git_repo_url_azure_devops_pull_request_page():
diff --git a/tests/test_session_async_commit.py b/tests/test_session_async_commit.py
index e95a5a1b3..3ce9e7dad 100644
--- a/tests/test_session_async_commit.py
+++ b/tests/test_session_async_commit.py
@@ -18,7 +18,6 @@ from openviking.service.core import OpenVikingService
from openviking.service.task_tracker import TaskStatus, get_task_tracker, reset_task_tracker
-
@pytest_asyncio.fixture
async def api_client(temp_dir) -> AsyncGenerator[Tuple[httpx.AsyncClient, OpenVikingService], None]:
"""Create in-process HTTP client for API endpoint tests."""
diff --git a/tests/unit/config/test_rerank_extra_headers_config.py b/tests/unit/config/test_rerank_extra_headers_config.py
index 87483f84f..296fab13a 100644
--- a/tests/unit/config/test_rerank_extra_headers_config.py
+++ b/tests/unit/config/test_rerank_extra_headers_config.py
@@ -2,6 +2,7 @@
# SPDX-License-Identifier: AGPL-3.0
"""Tests for rerank extra_headers configuration."""
+
import pytest
from pydantic import ValidationError
@@ -14,17 +15,14 @@ def test_rerank_config_with_extra_headers():
"model": "gpt-4",
"api_key": "test-key",
"api_base": "https://api.example.com/v1",
- "extra_headers": {
- "x-gw-apikey": "Bearer real-key",
- "X-Custom-Header": "custom-value"
- }
+ "extra_headers": {"x-gw-apikey": "Bearer real-key", "X-Custom-Header": "custom-value"},
}
config = RerankConfig(**config_data)
assert config.extra_headers == {
"x-gw-apikey": "Bearer real-key",
- "X-Custom-Header": "custom-value"
+ "X-Custom-Header": "custom-value",
}
@@ -35,7 +33,7 @@ def test_rerank_config_without_extra_headers():
config_data = {
"model": "gpt-4",
"api_key": "test-key",
- "api_base": "https://api.example.com/v1"
+ "api_base": "https://api.example.com/v1",
}
config = RerankConfig(**config_data)
@@ -49,16 +47,8 @@ def test_rerank_config_extra_headers_type_validation():
# Invalid: not a dict
with pytest.raises(ValidationError):
- RerankConfig(
- model="gpt-4",
- api_key="test-key",
- extra_headers="invalid"
- )
+ RerankConfig(model="gpt-4", api_key="test-key", extra_headers="invalid")
# Invalid: dict with non-string value
with pytest.raises(ValidationError):
- RerankConfig(
- model="gpt-4",
- api_key="test-key",
- extra_headers={"header": 123}
- )
+ RerankConfig(model="gpt-4", api_key="test-key", extra_headers={"header": 123})
diff --git a/tests/unit/models/rerank/test_openai_rerank_extra_headers.py b/tests/unit/models/rerank/test_openai_rerank_extra_headers.py
index ef24e8236..f4f5c81ec 100644
--- a/tests/unit/models/rerank/test_openai_rerank_extra_headers.py
+++ b/tests/unit/models/rerank/test_openai_rerank_extra_headers.py
@@ -1,11 +1,11 @@
# Copyright (c) 2026 Beijing Volcano Engine Technology Co., Ltd.
# SPDX-License-Identifier: AGPL-3.0
"""Tests for OpenAIRerankClient extra_headers support."""
+
from unittest.mock import Mock, patch
-import pytest
-from openviking_cli.utils.config.rerank_config import RerankConfig
from openviking.models.rerank.openai_rerank import OpenAIRerankClient
+from openviking_cli.utils.config.rerank_config import RerankConfig
def test_openai_rerank_client_init_with_extra_headers():
@@ -14,7 +14,7 @@ def test_openai_rerank_client_init_with_extra_headers():
api_key="test-key",
api_base="https://api.example.com/v1",
model_name="gpt-4",
- extra_headers={"x-gw-apikey": "Bearer real-key"}
+ extra_headers={"x-gw-apikey": "Bearer real-key"},
)
assert client.extra_headers == {"x-gw-apikey": "Bearer real-key"}
@@ -26,7 +26,7 @@ def test_openai_rerank_client_init_without_extra_headers():
api_key="test-key",
api_base="https://api.example.com/v1",
model_name="gpt-4",
- extra_headers=None
+ extra_headers=None,
)
assert client.extra_headers == {}
@@ -38,7 +38,7 @@ def test_openai_rerank_from_config_with_extra_headers():
model="gpt-4",
api_key="test-key",
api_base="https://api.example.com/v1",
- extra_headers={"x-custom": "value"}
+ extra_headers={"x-custom": "value"},
)
client = OpenAIRerankClient.from_config(config)
@@ -48,11 +48,7 @@ def test_openai_rerank_from_config_with_extra_headers():
def test_openai_rerank_from_config_without_extra_headers():
"""Test that from_config handles None extra_headers correctly."""
- config = RerankConfig(
- model="gpt-4",
- api_key="test-key",
- api_base="https://api.example.com/v1"
- )
+ config = RerankConfig(model="gpt-4", api_key="test-key", api_base="https://api.example.com/v1")
client = OpenAIRerankClient.from_config(config)
@@ -66,10 +62,7 @@ def test_rerank_batch_includes_extra_headers(mock_post):
mock_response = Mock()
mock_response.status_code = 200
mock_response.json.return_value = {
- "results": [
- {"index": 0, "relevance_score": 0.9},
- {"index": 1, "relevance_score": 0.8}
- ]
+ "results": [{"index": 0, "relevance_score": 0.9}, {"index": 1, "relevance_score": 0.8}]
}
mock_post.return_value = mock_response
@@ -78,17 +71,11 @@ def test_rerank_batch_includes_extra_headers(mock_post):
api_key="test-key",
api_base="https://api.example.com/v1",
model_name="gpt-4",
- extra_headers={
- "x-gw-apikey": "Bearer real-key",
- "X-Custom-Header": "custom-value"
- }
+ extra_headers={"x-gw-apikey": "Bearer real-key", "X-Custom-Header": "custom-value"},
)
# Call rerank_batch
- result = client.rerank_batch(
- query="test query",
- documents=["doc1", "doc2"]
- )
+ client.rerank_batch(query="test query", documents=["doc1", "doc2"])
# Verify the request included extra_headers
assert mock_post.called
@@ -109,23 +96,14 @@ def test_rerank_batch_without_extra_headers(mock_post):
"""Test that rerank_batch works correctly when no extra_headers provided."""
mock_response = Mock()
mock_response.status_code = 200
- mock_response.json.return_value = {
- "results": [
- {"index": 0, "relevance_score": 0.9}
- ]
- }
+ mock_response.json.return_value = {"results": [{"index": 0, "relevance_score": 0.9}]}
mock_post.return_value = mock_response
client = OpenAIRerankClient(
- api_key="test-key",
- api_base="https://api.example.com/v1",
- model_name="gpt-4"
+ api_key="test-key", api_base="https://api.example.com/v1", model_name="gpt-4"
)
- result = client.rerank_batch(
- query="test query",
- documents=["doc1"]
- )
+ client.rerank_batch(query="test query", documents=["doc1"])
assert mock_post.called
call_kwargs = mock_post.call_args.kwargs
@@ -150,7 +128,7 @@ def test_extra_headers_can_override_defaults(mock_post):
api_key="test-key",
api_base="https://api.example.com/v1",
model_name="gpt-4",
- extra_headers={"Content-Type": "application/json; charset=utf-8"}
+ extra_headers={"Content-Type": "application/json; charset=utf-8"},
)
client.rerank_batch(query="test", documents=["doc"])
diff --git a/tests/unit/session/test_wm_v2_guards.py b/tests/unit/session/test_wm_v2_guards.py
index f334613f6..067f69ba6 100644
--- a/tests/unit/session/test_wm_v2_guards.py
+++ b/tests/unit/session/test_wm_v2_guards.py
@@ -7,13 +7,15 @@ These tests target pure/static methods on Session and related helpers,
so they don't need a running OpenViking server.
"""
-from openviking.session.session import Session, WM_SEVEN_SECTIONS
-
+from openviking.message.message import Message
+from openviking.message.part import ContextPart, TextPart, ToolPart
+from openviking.session.session import WM_SEVEN_SECTIONS, Session
# -----------------------------------------------------------------------
# Helpers
# -----------------------------------------------------------------------
+
def _make_wm(**section_bodies: str) -> str:
"""Build a valid WM v2 markdown document for test input.
@@ -40,7 +42,7 @@ def _calc_pending_tokens(tokens: list, keep_recent_count: int) -> int:
if keep <= 0:
return sum(int(m.estimated_tokens or 0) for m in msgs)
elif total > keep:
- return sum(int(m.estimated_tokens or 0) for m in msgs[:total - keep])
+ return sum(int(m.estimated_tokens or 0) for m in msgs[: total - keep])
return 0
@@ -48,6 +50,7 @@ def _calc_pending_tokens(tokens: list, keep_recent_count: int) -> int:
# _parse_wm_sections
# =======================================================================
+
class TestParseWmSections:
def test_round_trip(self):
wm = _make_wm(session_title="My Session", current_state="Idle")
@@ -72,6 +75,7 @@ class TestParseWmSections:
# _wm_extract_bullet_items
# =======================================================================
+
class TestExtractBulletItems:
def test_dash_bullets(self):
items = Session._wm_extract_bullet_items("- foo\n- bar\n- baz")
@@ -98,6 +102,7 @@ class TestExtractBulletItems:
# _wm_enforce_append_only (Key Facts, Errors & Corrections)
# =======================================================================
+
class TestEnforceAppendOnly:
def test_keep_passes_through(self):
op = {"op": "KEEP"}
@@ -135,6 +140,7 @@ class TestEnforceAppendOnly:
# _wm_enforce_files_no_regression
# =======================================================================
+
class TestEnforceFilesNoRegression:
def test_keep_passes_through(self):
result = Session._wm_enforce_files_no_regression({"op": "KEEP"}, "- src/main.py")
@@ -142,7 +148,10 @@ class TestEnforceFilesNoRegression:
def test_update_preserving_all_paths(self):
old = "- src/main.py\n- config.yaml"
- op = {"op": "UPDATE", "content": "- src/main.py updated\n- config.yaml stays\n- new.ts added"}
+ op = {
+ "op": "UPDATE",
+ "content": "- src/main.py updated\n- config.yaml stays\n- new.ts added",
+ }
result = Session._wm_enforce_files_no_regression(op, old)
assert result["op"] == "UPDATE"
@@ -175,6 +184,7 @@ class TestEnforceFilesNoRegression:
# _wm_enforce_title_stability
# =======================================================================
+
class TestEnforceTitleStability:
def test_keep_passes_through(self):
result = Session._wm_enforce_title_stability({"op": "KEEP"}, "old title")
@@ -211,10 +221,13 @@ class TestEnforceTitleStability:
# _wm_enforce_open_issues_resolved
# =======================================================================
+
class TestEnforceOpenIssuesResolved:
def test_update_preserving_all_passes(self):
old = "- bug in auth module\n- slow query on /api/users"
- new_content = "- bug in auth module (investigating)\n- slow query on /api/users\n- new: memory leak"
+ new_content = (
+ "- bug in auth module (investigating)\n- slow query on /api/users\n- new: memory leak"
+ )
op = {"op": "UPDATE", "content": new_content}
result = Session._wm_enforce_open_issues_resolved(op, old)
assert result["op"] == "UPDATE"
@@ -235,10 +248,7 @@ class TestEnforceOpenIssuesResolved:
def test_already_restored_item_not_restored_again(self):
"""Items already tagged [silently dropped, restored] should NOT be
restored a second time -- they had their chance."""
- old = (
- "- [silently dropped, restored] stale follow-up issue\n"
- "- fresh unresolved bug"
- )
+ old = "- [silently dropped, restored] stale follow-up issue\n- fresh unresolved bug"
op = {"op": "UPDATE", "content": "- fresh unresolved bug"}
result = Session._wm_enforce_open_issues_resolved(op, old)
assert "stale follow-up" not in result["content"]
@@ -248,8 +258,7 @@ class TestEnforceOpenIssuesResolved:
"""Items with nested [silently dropped, restored] tags must not
be restored again (regression guard for the old accumulation bug)."""
old = (
- "- [silently dropped, restored] [silently dropped, restored] old issue\n"
- "- active issue"
+ "- [silently dropped, restored] [silently dropped, restored] old issue\n- active issue"
)
op = {"op": "UPDATE", "content": "- active issue"}
result = Session._wm_enforce_open_issues_resolved(op, old)
@@ -270,6 +279,7 @@ class TestEnforceOpenIssuesResolved:
# _merge_wm_sections
# =======================================================================
+
class TestMergeWmSections:
def test_all_keep(self):
old = _make_wm(session_title="Title", current_state="Working")
@@ -339,6 +349,7 @@ class TestMergeWmSections:
# _wm_recover_ops_from_raw
# =======================================================================
+
class TestWmRecoverOpsFromRaw:
def test_recover_keep_ops(self):
raw = '"Session Title": {"op": "KEEP"}, "Current State": {"op": "KEEP"}'
@@ -400,6 +411,7 @@ class TestWmRecoverOpsFromRaw:
# _rebuild_pending_tokens (via SessionMeta simulation)
# =======================================================================
+
class TestRebuildPendingTokens:
"""Test the _rebuild_pending_tokens math using _calc_pending_tokens helper."""
@@ -426,12 +438,11 @@ class TestRebuildPendingTokens:
# _is_wm_v2 detection (tests the condition in _generate_archive_summary_async)
# =======================================================================
+
class TestIsWmV2Detection:
def _is_wm_v2(self, overview: str) -> bool:
"""Mirror the detection logic from session.py."""
- return bool(overview) and any(
- f"## {s}" in overview for s in WM_SEVEN_SECTIONS
- )
+ return bool(overview) and any(f"## {s}" in overview for s in WM_SEVEN_SECTIONS)
def test_valid_v2(self):
wm = _make_wm(session_title="Test")
@@ -463,6 +474,7 @@ class TestIsWmV2Detection:
# WM_PATH_LIKE_RE
# =======================================================================
+
class TestWmPathLikeRe:
def test_python_file(self):
assert Session._WM_PATH_LIKE_RE.search("src/main.py")
@@ -484,6 +496,7 @@ class TestWmPathLikeRe:
# pending_tokens defensive clamp
# =======================================================================
+
class TestPendingTokensClamp:
"""Verify that pending_tokens is always clamped to >= 0."""
@@ -497,6 +510,7 @@ class TestPendingTokensClamp:
def test_from_dict_clamps_negative_pending(self):
from openviking.session.session import SessionMeta
+
data = {"pending_tokens": -100, "keep_recent_count": -5}
meta = SessionMeta.from_dict(data)
assert meta.pending_tokens >= 0
@@ -507,6 +521,7 @@ class TestPendingTokensClamp:
# APPEND non-string items handling
# =======================================================================
+
class TestAppendNonStringItems:
"""APPEND items that are not strings should be dropped (not crash)."""
@@ -545,8 +560,8 @@ class TestAppendNonStringItems:
# _merge_wm_sections edge cases
# =======================================================================
-class TestMergeWmSectionsEdgeCases:
+class TestMergeWmSectionsEdgeCases:
def test_missing_ops_default_to_keep(self):
old_wm = _make_wm(
session_title="Original Title",
@@ -593,9 +608,6 @@ class TestMergeWmSectionsEdgeCases:
# _format_message_for_wm
# -----------------------------------------------------------------------
-from openviking.message.message import Message
-from openviking.message.part import TextPart, ToolPart, ContextPart
-
def _msg(role, parts):
"""Convenience builder for a Message with an auto-id."""
@@ -611,29 +623,42 @@ class TestFormatMessageForWm:
assert result == "[user]: Hello world"
def test_tool_part_included(self):
- m = _msg("assistant", [
- ToolPart(tool_name="search", tool_status="completed",
- tool_output="found 3 results"),
- ])
+ m = _msg(
+ "assistant",
+ [
+ ToolPart(
+ tool_name="search", tool_status="completed", tool_output="found 3 results"
+ ),
+ ],
+ )
result = Session._format_message_for_wm(m)
assert "[tool:search (completed)]" in result
assert "found 3 results" in result
assert result.startswith("[assistant]:")
def test_context_part_included(self):
- m = _msg("assistant", [
- ContextPart(abstract="Summary of prior session"),
- ])
+ m = _msg(
+ "assistant",
+ [
+ ContextPart(abstract="Summary of prior session"),
+ ],
+ )
result = Session._format_message_for_wm(m)
assert "[context] Summary of prior session" in result
def test_mixed_parts(self):
- m = _msg("assistant", [
- TextPart(text="Let me check."),
- ToolPart(tool_name="read_file", tool_status="completed",
- tool_output="/path/to/file content here"),
- TextPart(text="Done reading."),
- ])
+ m = _msg(
+ "assistant",
+ [
+ TextPart(text="Let me check."),
+ ToolPart(
+ tool_name="read_file",
+ tool_status="completed",
+ tool_output="/path/to/file content here",
+ ),
+ TextPart(text="Done reading."),
+ ],
+ )
result = Session._format_message_for_wm(m)
lines = result.split("\n")
assert lines[0] == "[assistant]: Let me check."
@@ -651,15 +676,21 @@ class TestFormatMessageForWm:
assert "(no content)" in result
def test_tool_with_empty_output(self):
- m = _msg("assistant", [
- ToolPart(tool_name="delete", tool_status="completed", tool_output=""),
- ])
+ m = _msg(
+ "assistant",
+ [
+ ToolPart(tool_name="delete", tool_status="completed", tool_output=""),
+ ],
+ )
result = Session._format_message_for_wm(m)
assert "[tool:delete (completed)]" in result
def test_tool_default_status(self):
- m = _msg("assistant", [
- ToolPart(tool_name="run", tool_output="ok"),
- ])
+ m = _msg(
+ "assistant",
+ [
+ ToolPart(tool_name="run", tool_output="ok"),
+ ],
+ )
result = Session._format_message_for_wm(m)
assert "(pending)" in result
diff --git a/tests/unit/session/test_working_memory_growth.py b/tests/unit/session/test_working_memory_growth.py
index d7591abe9..9794a0b30 100644
--- a/tests/unit/session/test_working_memory_growth.py
+++ b/tests/unit/session/test_working_memory_growth.py
@@ -54,6 +54,7 @@ def _section_body(wm: str, header: str) -> str:
# Monotonic-growth tests (unchanged behavior for APPEND-driven sections)
# =====================================================================
+
def test_key_facts_section_grows_monotonically_under_repeated_appends():
merged = _wm()
@@ -112,6 +113,7 @@ def test_files_and_context_section_can_keep_growing_with_unique_paths():
# Key Facts consolidation guard — Layer 1: bullet-count ratio
# =====================================================================
+
def test_key_facts_consolidation_rejected_when_trivially_small():
"""A 1-bullet UPDATE for 41 old bullets (2.4%) is rejected.
@@ -145,63 +147,67 @@ def test_key_facts_consolidation_rejected_when_trivially_small():
# Key Facts consolidation guard — Layer 2: anchor coverage
# =====================================================================
-_RICH_KEY_FACTS = "\n".join([
- "- Caroline adopted a rescue dog named Biscuit on 15 March 2024.",
- "- Melanie's family took 3 camping trips to Yellowstone because the kids love hiking.",
- "- Caroline competed in swimming contest on 10 June 2023, won gold medal.",
- "- Caroline competed in swimming contest on 9 June 2024, brought friends to cheer.",
- "- Sweden trip planned for August 2025 with budget of 5000 dollars.",
- "- Decided to use Python because the team has 4 years experience.",
- "- Marcus is Caroline's brother, lives in Portland.",
- "- Caroline volunteers at the shelter every Saturday morning.",
- "- Adoption agency: Bright Futures, contacted on 2024-01-20.",
- "- Melanie decided to homeschool the kids because local schools are overcrowded.",
- "- Caroline's art exhibition on 22 November 2024 at Gallery One.",
- "- Biscuit has a vet appointment every 6 months at Pawsome Clinic.",
- "- Marcus committed to help with the Sweden trip logistics.",
- "- Caroline chose watercolor over oil painting because of studio ventilation.",
- "- Family reunion planned for December 2025 in Portland.",
- "- Melanie's oldest child starts college in September 2026.",
- "- Caroline resolved to run a half-marathon in Spring 2025.",
- "- Budget for art supplies: 200 dollars per month.",
- "- Caroline's neighbor Jake offered to pet-sit Biscuit during Sweden trip.",
- "- Melanie agreed to share camping gear for the next Yellowstone trip.",
- "- Caroline took a ceramics class starting January 2025.",
- "- Marcus and Caroline share a joint savings account for family events.",
- "- Bright Futures agency requires 3 home visits before approval.",
- "- Caroline prefers morning runs, usually 5 miles along the river trail.",
- "- Melanie's youngest allergic to peanuts, diagnosed at age 2.",
- "- Gallery One charges 500 dollars for exhibition space.",
- "- Caroline decided to switch from acrylic to watercolor in March 2024.",
- "- Portland family home has 4 bedrooms, enough for reunion guests.",
- "- Sweden itinerary: Stockholm 3 days, Gothenburg 2 days, countryside 2 days.",
- "- Caroline's running group meets every Wednesday at 6 AM.",
- "- Melanie committed to volunteer at kids' school fundraiser.",
- "- Jake is a retired teacher, lives next door since 2020.",
- "- Caroline's half-marathon target time: under 2 hours.",
- "- Art exhibition features 12 watercolor paintings.",
- "- Melanie's camping trips always in July because of school schedule.",
- "- Caroline and Marcus talk every Sunday evening by phone.",
- "- Biscuit is a 3 years old golden retriever mix.",
- "- Bright Futures requires background checks completed by February 2025.",
- "- Caroline's studio is in the garage, converted in 2023.",
- "- Melanie drives a minivan because she has 3 kids.",
- "- Family agreed on a $2000 budget cap for the reunion dinner.",
-])
-
-_GOOD_CONSOLIDATION = "\n".join([
- "- Caroline adopted rescue dog Biscuit (golden retriever mix, 3 years old) on 15 March 2024; vet at Pawsome Clinic every 6 months. Neighbor Jake (retired teacher, next door since 2020) offered to pet-sit during Sweden trip.",
- "- Caroline regularly competes in swimming contests; most recently on 9 June 2024, won 2 gold medals total.",
- "- Caroline's art: switched from acrylic to watercolor (March 2024) because of studio ventilation; exhibition on 22 November 2024 at Gallery One (12 paintings, 500 dollars space); studio in garage, converted 2023. Budget 200 dollars/month.",
- "- Caroline took ceramics class starting January 2025; resolved to run half-marathon Spring 2025 (target under 2 hours, runs 5 miles mornings along river trail, group meets Wednesdays 6 AM).",
- "- Caroline volunteers at shelter every Saturday morning.",
- "- Marcus is Caroline's brother, lives in Portland; committed to help with Sweden trip; they share joint savings for family events; talk every Sunday evening.",
- "- Melanie's family: 3 camping trips to Yellowstone (always July, because school schedule, kids love hiking); agreed to share gear. Oldest starts college September 2026. Youngest allergic to peanuts (diagnosed age 2). Melanie drives minivan because 3 kids. Decided to homeschool because local schools overcrowded. Committed to school fundraiser volunteer.",
- "- Sweden trip planned August 2025, budget 5000 dollars: Stockholm 3 days, Gothenburg 2 days, countryside 2 days.",
- "- Adoption agency Bright Futures, contacted 2024-01-20: requires 3 home visits and background checks by February 2025.",
- "- Decided to use Python because team has 4 years experience.",
- "- Family reunion December 2025 in Portland (4-bedroom home, enough for guests); agreed $2000 budget cap for reunion dinner.",
-])
+_RICH_KEY_FACTS = "\n".join(
+ [
+ "- Caroline adopted a rescue dog named Biscuit on 15 March 2024.",
+ "- Melanie's family took 3 camping trips to Yellowstone because the kids love hiking.",
+ "- Caroline competed in swimming contest on 10 June 2023, won gold medal.",
+ "- Caroline competed in swimming contest on 9 June 2024, brought friends to cheer.",
+ "- Sweden trip planned for August 2025 with budget of 5000 dollars.",
+ "- Decided to use Python because the team has 4 years experience.",
+ "- Marcus is Caroline's brother, lives in Portland.",
+ "- Caroline volunteers at the shelter every Saturday morning.",
+ "- Adoption agency: Bright Futures, contacted on 2024-01-20.",
+ "- Melanie decided to homeschool the kids because local schools are overcrowded.",
+ "- Caroline's art exhibition on 22 November 2024 at Gallery One.",
+ "- Biscuit has a vet appointment every 6 months at Pawsome Clinic.",
+ "- Marcus committed to help with the Sweden trip logistics.",
+ "- Caroline chose watercolor over oil painting because of studio ventilation.",
+ "- Family reunion planned for December 2025 in Portland.",
+ "- Melanie's oldest child starts college in September 2026.",
+ "- Caroline resolved to run a half-marathon in Spring 2025.",
+ "- Budget for art supplies: 200 dollars per month.",
+ "- Caroline's neighbor Jake offered to pet-sit Biscuit during Sweden trip.",
+ "- Melanie agreed to share camping gear for the next Yellowstone trip.",
+ "- Caroline took a ceramics class starting January 2025.",
+ "- Marcus and Caroline share a joint savings account for family events.",
+ "- Bright Futures agency requires 3 home visits before approval.",
+ "- Caroline prefers morning runs, usually 5 miles along the river trail.",
+ "- Melanie's youngest allergic to peanuts, diagnosed at age 2.",
+ "- Gallery One charges 500 dollars for exhibition space.",
+ "- Caroline decided to switch from acrylic to watercolor in March 2024.",
+ "- Portland family home has 4 bedrooms, enough for reunion guests.",
+ "- Sweden itinerary: Stockholm 3 days, Gothenburg 2 days, countryside 2 days.",
+ "- Caroline's running group meets every Wednesday at 6 AM.",
+ "- Melanie committed to volunteer at kids' school fundraiser.",
+ "- Jake is a retired teacher, lives next door since 2020.",
+ "- Caroline's half-marathon target time: under 2 hours.",
+ "- Art exhibition features 12 watercolor paintings.",
+ "- Melanie's camping trips always in July because of school schedule.",
+ "- Caroline and Marcus talk every Sunday evening by phone.",
+ "- Biscuit is a 3 years old golden retriever mix.",
+ "- Bright Futures requires background checks completed by February 2025.",
+ "- Caroline's studio is in the garage, converted in 2023.",
+ "- Melanie drives a minivan because she has 3 kids.",
+ "- Family agreed on a $2000 budget cap for the reunion dinner.",
+ ]
+)
+
+_GOOD_CONSOLIDATION = "\n".join(
+ [
+ "- Caroline adopted rescue dog Biscuit (golden retriever mix, 3 years old) on 15 March 2024; vet at Pawsome Clinic every 6 months. Neighbor Jake (retired teacher, next door since 2020) offered to pet-sit during Sweden trip.",
+ "- Caroline regularly competes in swimming contests; most recently on 9 June 2024, won 2 gold medals total.",
+ "- Caroline's art: switched from acrylic to watercolor (March 2024) because of studio ventilation; exhibition on 22 November 2024 at Gallery One (12 paintings, 500 dollars space); studio in garage, converted 2023. Budget 200 dollars/month.",
+ "- Caroline took ceramics class starting January 2025; resolved to run half-marathon Spring 2025 (target under 2 hours, runs 5 miles mornings along river trail, group meets Wednesdays 6 AM).",
+ "- Caroline volunteers at shelter every Saturday morning.",
+ "- Marcus is Caroline's brother, lives in Portland; committed to help with Sweden trip; they share joint savings for family events; talk every Sunday evening.",
+ "- Melanie's family: 3 camping trips to Yellowstone (always July, because school schedule, kids love hiking); agreed to share gear. Oldest starts college September 2026. Youngest allergic to peanuts (diagnosed age 2). Melanie drives minivan because 3 kids. Decided to homeschool because local schools overcrowded. Committed to school fundraiser volunteer.",
+ "- Sweden trip planned August 2025, budget 5000 dollars: Stockholm 3 days, Gothenburg 2 days, countryside 2 days.",
+ "- Adoption agency Bright Futures, contacted 2024-01-20: requires 3 home visits and background checks by February 2025.",
+ "- Decided to use Python because team has 4 years experience.",
+ "- Family reunion December 2025 in Portland (4-bedroom home, enough for guests); agreed $2000 budget cap for reunion dinner.",
+ ]
+)
def test_key_facts_consolidation_accepted_when_anchors_preserved():
@@ -262,18 +268,20 @@ def test_key_facts_consolidation_accepted_at_low_volume_with_anchors():
"""A 41-to-10 consolidation (~24% volume) passes when anchors are kept."""
old_wm = _wm(key_facts=_RICH_KEY_FACTS)
- consolidated = "\n".join([
- "- Caroline adopted Biscuit (golden retriever, 3 years old) on 15 March 2024; vet every 6 months at Pawsome Clinic.",
- "- Caroline competes in swimming contests regularly; latest 9 June 2024, 2 gold medals total.",
- "- Caroline's art: watercolor (switched March 2024 because ventilation); exhibition 22 November 2024 at Gallery One, 12 paintings, 500 dollars. Studio in garage since 2023. Budget 200 dollars/month. Also took ceramics January 2025.",
- "- Caroline runs half-marathon Spring 2025, target under 2 hours; runs 5 miles mornings, group Wednesdays 6 AM. Volunteers shelter Saturdays.",
- "- Marcus: Caroline's brother, Portland. Joint savings, Sunday calls. Committed to Sweden trip help.",
- "- Melanie: 3 kids, minivan, homeschooling because overcrowded schools. Oldest college September 2026, youngest peanut allergy diagnosed age 2. Camping Yellowstone in July (3 trips, kids love hiking). Fundraiser volunteer.",
- "- Sweden August 2025, 5000 dollars: Stockholm 3 days, Gothenburg 2 days, countryside 2 days. Jake (neighbor since 2020, retired teacher) pet-sits Biscuit.",
- "- Adoption: Bright Futures contacted 2024-01-20; 3 home visits, background checks by February 2025.",
- "- Team decided Python because 4 years experience.",
- "- Family reunion December 2025 Portland, 4 bedrooms, $2000 dinner cap.",
- ])
+ consolidated = "\n".join(
+ [
+ "- Caroline adopted Biscuit (golden retriever, 3 years old) on 15 March 2024; vet every 6 months at Pawsome Clinic.",
+ "- Caroline competes in swimming contests regularly; latest 9 June 2024, 2 gold medals total.",
+ "- Caroline's art: watercolor (switched March 2024 because ventilation); exhibition 22 November 2024 at Gallery One, 12 paintings, 500 dollars. Studio in garage since 2023. Budget 200 dollars/month. Also took ceramics January 2025.",
+ "- Caroline runs half-marathon Spring 2025, target under 2 hours; runs 5 miles mornings, group Wednesdays 6 AM. Volunteers shelter Saturdays.",
+ "- Marcus: Caroline's brother, Portland. Joint savings, Sunday calls. Committed to Sweden trip help.",
+ "- Melanie: 3 kids, minivan, homeschooling because overcrowded schools. Oldest college September 2026, youngest peanut allergy diagnosed age 2. Camping Yellowstone in July (3 trips, kids love hiking). Fundraiser volunteer.",
+ "- Sweden August 2025, 5000 dollars: Stockholm 3 days, Gothenburg 2 days, countryside 2 days. Jake (neighbor since 2020, retired teacher) pet-sits Biscuit.",
+ "- Adoption: Bright Futures contacted 2024-01-20; 3 home visits, background checks by February 2025.",
+ "- Team decided Python because 4 years experience.",
+ "- Family reunion December 2025 Portland, 4 bedrooms, $2000 dinner cap.",
+ ]
+ )
ops = _keep_all()
ops["Key Facts & Decisions"] = {"op": "UPDATE", "content": consolidated}
@@ -293,6 +301,7 @@ def test_key_facts_consolidation_accepted_at_low_volume_with_anchors():
# _extract_lexical_anchors — mixed-case and multi-word entities
# =====================================================================
+
def test_extract_anchors_catches_mixed_case_names():
"""OpenAI, McCloud, JavaScript etc. should be captured as anchors."""
text = (
@@ -328,6 +337,7 @@ def test_extract_anchors_filters_stopwords():
# Salvage: rejected consolidation still preserves new facts (P1 fix)
# =====================================================================
+
def test_key_facts_rejected_consolidation_salvages_new_facts():
"""When consolidation is rejected (layer1 or layer2), genuinely new
items from the UPDATE content are APPENDed so current-round facts
@@ -359,10 +369,10 @@ def test_key_facts_rejected_consolidation_salvages_new_facts():
# _build_wm_section_reminders
# =====================================================================
+
def test_build_wm_section_reminders_triggers_above_threshold():
key_facts = "\n".join(
- f"- Fact {idx}: something important about item {idx}."
- for idx in range(1, 35)
+ f"- Fact {idx}: something important about item {idx}." for idx in range(1, 35)
)
wm = _wm(key_facts=key_facts)
@@ -378,8 +388,7 @@ def test_build_wm_section_reminders_skips_append_only_sections():
"""Errors & Corrections should never get consolidation warnings
even if oversized, because it's append-only."""
errors = "\n".join(
- f"- Error {idx}: something went wrong in round {idx}."
- for idx in range(1, 40)
+ f"- Error {idx}: something went wrong in round {idx}." for idx in range(1, 40)
)
wm = _wm(errors=errors)
@@ -389,10 +398,7 @@ def test_build_wm_section_reminders_skips_append_only_sections():
def test_build_wm_section_reminders_silent_below_threshold():
- key_facts = "\n".join(
- f"- Fact {idx}: something about item {idx}."
- for idx in range(1, 10)
- )
+ key_facts = "\n".join(f"- Fact {idx}: something about item {idx}." for idx in range(1, 10))
wm = _wm(key_facts=key_facts)
reminders = Session._build_wm_section_reminders(wm)
@@ -404,6 +410,7 @@ def test_build_wm_section_reminders_silent_below_threshold():
# Wiring test: _generate_archive_summary_async passes reminders
# =====================================================================
+
def test_wm_section_reminders_consistent_with_prompt_template():
"""Verify _build_wm_section_reminders output matches the format expected
by ov_wm_v2_update.yaml (XML tags that the Jinja2 template injects)."""
@@ -428,6 +435,7 @@ def test_wm_section_reminders_consistent_with_prompt_template():
# Aggregate document size test (adapted for new guard behavior)
# =====================================================================
+
def test_working_memory_document_size_keeps_increasing_across_rounds():
merged = _wm(
key_facts="- Decision 0: keep the current WM shape.",
@@ -455,8 +463,7 @@ def test_working_memory_document_size_keeps_increasing_across_rounds():
assert len(merged) > initial_size
kf_items = _section_items(merged, "Key Facts & Decisions")
assert len(kf_items) == 52, (
- f"Key Facts should stop at 52 (51 facts + 1 consolidation sentinel); "
- f"got {len(kf_items)}"
+ f"Key Facts should stop at 52 (51 facts + 1 consolidation sentinel); got {len(kf_items)}"
)
assert any("CONSOLIDATION REQUIRED" in it for it in kf_items), (
"Emergency sentinel must be present when Key Facts exceeds 2x threshold"
@@ -469,6 +476,7 @@ def test_working_memory_document_size_keeps_increasing_across_rounds():
# Anti-bloat guard: focused tests for each path
# =====================================================================
+
def test_antibloat_normal_append_passes_under_threshold():
"""APPEND with fewer than threshold bullets should pass without throttling."""
facts = "\n".join(f"- Fact {i}" for i in range(10))
@@ -496,9 +504,7 @@ def test_antibloat_throttled_append_caps_at_5():
}
merged = Session._merge_wm_sections(merged, ops)
items = _section_items(merged, "Key Facts & Decisions")
- assert len(items) == 35, (
- f"Expected 30 + 5 (capped) = 35; got {len(items)}"
- )
+ assert len(items) == 35, f"Expected 30 + 5 (capped) = 35; got {len(items)}"
def test_antibloat_throttled_dedup_filters_existing():
@@ -539,9 +545,7 @@ def test_antibloat_emergency_inserts_sentinel_and_stops():
}
merged = Session._merge_wm_sections(merged, ops2)
items2 = _section_items(merged, "Key Facts & Decisions")
- assert len(items2) == 56, (
- f"Sentinel already present — should stay at 56; got {len(items2)}"
- )
+ assert len(items2) == 56, f"Sentinel already present — should stay at 56; got {len(items2)}"
assert "Emergency fact B" not in items2
@@ -549,8 +553,7 @@ def test_antibloat_salvage_bypass_blocked_at_emergency():
"""UPDATE rejected by Layer1/2 → salvage APPEND should be suppressed
when Key Facts is at emergency level."""
facts = "\n".join(
- f"- Person_{i} decided thing_{i} on 2024-01-{i % 28 + 1:02d}."
- for i in range(55)
+ f"- Person_{i} decided thing_{i} on 2024-01-{i % 28 + 1:02d}." for i in range(55)
)
merged = _wm(key_facts=facts)
diff --git a/tests/unit/session/test_working_memory_v2.py b/tests/unit/session/test_working_memory_v2.py
index 617c20617..ab1efbc98 100644
--- a/tests/unit/session/test_working_memory_v2.py
+++ b/tests/unit/session/test_working_memory_v2.py
@@ -46,8 +46,7 @@ def test_dynamic_reminders_flag_large_key_facts_and_bulk_urls():
for i in range(1, 42)
)
files_context = "\n".join(
- f"- https://example.com/images/{i}.png - raw parser image URL."
- for i in range(1, 23)
+ f"- https://example.com/images/{i}.png - raw parser image URL." for i in range(1, 23)
)
reminders = Session._build_wm_section_reminders(
diff --git a/tests/unit/test_cohere_embedder.py b/tests/unit/test_cohere_embedder.py
index 828841354..5dfc81670 100644
--- a/tests/unit/test_cohere_embedder.py
+++ b/tests/unit/test_cohere_embedder.py
@@ -87,9 +87,7 @@ class TestCohereDenseEmbedder:
mock_client_class.return_value = mock_client
mock_response = MagicMock()
- mock_response.json.return_value = {
- "embeddings": {"float": [[0.1] * 1024]}
- }
+ mock_response.json.return_value = {"embeddings": {"float": [[0.1] * 1024]}}
mock_response.raise_for_status = MagicMock()
mock_client.post.return_value = mock_response
@@ -113,9 +111,7 @@ class TestCohereDenseEmbedder:
mock_client_class.return_value = mock_client
mock_response = MagicMock()
- mock_response.json.return_value = {
- "embeddings": {"float": [[0.1] * 1024]}
- }
+ mock_response.json.return_value = {"embeddings": {"float": [[0.1] * 1024]}}
mock_response.raise_for_status = MagicMock()
mock_client.post.return_value = mock_response
@@ -134,9 +130,7 @@ class TestCohereDenseEmbedder:
mock_client_class.return_value = mock_client
mock_response = MagicMock()
- mock_response.json.return_value = {
- "embeddings": {"float": [[0.1] * 1024]}
- }
+ mock_response.json.return_value = {"embeddings": {"float": [[0.1] * 1024]}}
mock_response.raise_for_status = MagicMock()
mock_client.post.return_value = mock_response
@@ -156,9 +150,7 @@ class TestCohereDenseEmbedder:
mock_client_class.return_value = mock_client
mock_response = MagicMock()
- mock_response.json.return_value = {
- "embeddings": {"float": [[0.1] * 1024, [0.2] * 1024]}
- }
+ mock_response.json.return_value = {"embeddings": {"float": [[0.1] * 1024, [0.2] * 1024]}}
mock_response.raise_for_status = MagicMock()
mock_client.post.return_value = mock_response
diff --git a/tests/unit/test_extra_headers_embedding.py b/tests/unit/test_extra_headers_embedding.py
index e5fe485d3..8075f7a06 100644
--- a/tests/unit/test_extra_headers_embedding.py
+++ b/tests/unit/test_extra_headers_embedding.py
@@ -143,9 +143,7 @@ class TestExtraHeadersViaFactory:
embedder = EmbeddingConfig(dense=cfg)._create_embedder("ollama", "dense", cfg)
with (
- patch(
- "openviking.models.embedder.openai_embedders.logger.warning"
- ) as mock_warning,
+ patch("openviking.models.embedder.openai_embedders.logger.warning") as mock_warning,
patch(
"openviking.models.embedder.base.time.monotonic",
side_effect=[0.0, 0.0, 0.0, 1.2],
diff --git a/uv.lock b/uv.lock
index 72bfd8620..9857cf5e2 100644
--- a/uv.lock
+++ b/uv.lock
@@ -3820,7 +3820,7 @@ requires-dist = [
{ name = "tree-sitter-typescript", specifier = ">=0.23.0" },
{ name = "typer", specifier = ">=0.12.0" },
{ name = "typing-extensions", specifier = ">=4.5.0" },
- { name = "urllib3", specifier = ">=2.6.3" },
+ { name = "urllib3", specifier = ">=2.7.0" },
{ name = "uvicorn", specifier = ">=0.39.0" },
{ name = "volcengine", specifier = ">=1.0.216" },
{ name = "volcengine-python-sdk", extras = ["ark"], specifier = ">=5.0.3" },
@@ -6381,11 +6381,11 @@ wheels = [
[[package]]
name = "urllib3"
-version = "2.6.3"
+version = "2.7.0"
source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/c7/24/5f1b3bdffd70275f6661c76461e25f024d5a38a46f04aaca912426a2b1d3/urllib3-2.6.3.tar.gz", hash = "sha256:1b62b6884944a57dbe321509ab94fd4d3b307075e0c2eae991ac71ee15ad38ed", size = 435556, upload-time = "2026-01-07T16:24:43.925Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/53/0c/06f8b233b8fd13b9e5ee11424ef85419ba0d8ba0b3138bf360be2ff56953/urllib3-2.7.0.tar.gz", hash = "sha256:231e0ec3b63ceb14667c67be60f2f2c40a518cb38b03af60abc813da26505f4c", size = 433602, upload-time = "2026-05-07T16:13:18.596Z" }
wheels = [
- { url = "https://files.pythonhosted.org/packages/39/08/aaaad47bc4e9dc8c725e68f9d04865dbcb2052843ff09c97b08904852d84/urllib3-2.6.3-py3-none-any.whl", hash = "sha256:bf272323e553dfb2e87d9bfd225ca7b0f467b919d7bbd355436d3fd37cb0acd4", size = 131584, upload-time = "2026-01-07T16:24:42.685Z" },
+ { url = "https://files.pythonhosted.org/packages/7f/3e/5db95bcf282c52709639744ca2a8b149baccf648e39c8cc87553df9eae0c/urllib3-2.7.0-py3-none-any.whl", hash = "sha256:9fb4c81ebbb1ce9531cce37674bbc6f1360472bc18ca9a553ede278ef7276897", size = 131087, upload-time = "2026-05-07T16:13:17.151Z" },
]
[[package]]