ReMe/tests4/integration/test_auto_dream.py
jinliyl 83831ec90c
feat(core): enhance reme4 (#281)
### 1. Agent Wrapper(统一 Agent 后端抽象)
- **`base_agent_wrapper.py`**:`reply()` 返回值从 `tuple[str, Any]` 改为 `dict`(含 `session_id` / `last_message` / `result` / 可选 `structured_output`);`reply_stream()` 改为产出统一的 `StreamChunk`。废弃 `add_tools()`,改为 `add_job_tools(names: list[str])`(按名解析 BaseJob)与 `add_skills()`;新增 `_resolve_job_tools()`、`_merged_kwargs()`、`_chunk()` 辅助方法及 `project_path` / `project_skills_root` 属性。
- **`as_agent_wrapper.py`(AgentScope 后端)**:
  - 会话持久化重写:`session_path` 落地到 `<vault>/<session_dir>/agentscope/`,`_load_state` 支持 `resume` / `session_id` / `fork_session`,并做 UUID 校验(`_validate_session_id`);`_cleanup_expired_sessions` 按天数清理过期会话。
  - 新增内置工具集(`BypassAnalysisBash` + Edit/Glob/Grep/Read/Write),`BypassAnalysisBash` 绕过 AgentScope 自带 Bash 静态分析以让 permission_mode 生效;`_resolve_skills()` 把配置的 skill 暴露给后端,`_load_tool_env()` 注入项目 `.env`。
  - `_event_to_chunk()` 把 20+ 种 AgentScope 事件(Reply/Text/Thinking/Data/ToolCall/ToolResult/ModelCall/ExceedMaxIters)归一化为 `StreamChunk`。
- **`cc_agent_wrapper.py`(Claude Code SDK 后端,+551 行)**:
  - 新增 `_CcFileSessionStore`:基于 vault 的文件型会话存储,实现 append(按 uuid 去重)/ load / list / delete / list_subkeys,并对路径做 `_safe_parts` + `resolve()` 防越界校验。
  - `_build_options()`:统一构建 `ClaudeAgentOptions`,处理 skills、disallowed_tools(默认禁 `WebSearch`)、`.env` 注入、Claude Code 的 API 凭据解析(`_claude_code_api_env`,多级 base_url/api_key 回退)、`CLAUDE_CONFIG_DIR` 设置、skill 目录软链接(`_ensure_claude_skill_dir`)。
  - `_raw_event_to_chunk()` / `_message_content_to_chunks()`:把 Anthropic 流式事件(message_start/delta/stop、content_block_*)与 SDK 消息块(AssistantMessage/UserMessage/ResultMessage/RateLimitEvent)转换为统一 `StreamChunk`;跟踪 block_id/block_type/tool_call_name 做关联;处理尾部 `"success"` 误报异常的吞掉逻辑。

### 2. 统一流式协议(StreamChunk / ChunkEnum)
- **`stream_chunk.py`**:`StreamChunk` 扩展为承载 AS + CC 双后端完整信息的统一结构,新增 `session_id` / `block_id` / `tool_call_id` / `tool_call_name` / `media_type` / `input_tokens` / `output_tokens` 等字段,纯文本流仍保持轻量。
- **`chunk_enum.py`**:补全生命周期标记 `REPLY_START` / `REPLY_END`,并文档化两套后端事件 → ChunkEnum 的映射。

### 3. Index 模块重构(变化批次化 + dispatch)
- 新增 `_change_batch.py`:`coalesce_changes()` 把同路径多次事件折叠为最终状态(结合 path 存在性判定),`bucket_changes()` 按 watchfiles.Change 分桶。
- 新增 `init_changes.py`(`InitChangesStep`):一次性扫描,对比 file_store / file_catalog 已索引节点计算 added/modified/deleted,写入 `context["changes"]` 后 dispatch。
- 新增 `update_changes.py`:抽象基类 `ChangeApplyStep` 统一 added/modified/deleted 处理与错误收集;`UpdateCatalogStep`(写 file_catalog)、`UpdateIndexStep`(写 file_store,含按后缀解析 chunker)。
- **`watch_changes.py`**:改用 `dispatch_step_specs`(基类提供的 `dispatch_steps()`),每批先 `coalesce_changes` 再 dispatch;默认参数调整(debounce 5000ms / step 1000ms / poll 5000ms)并暴露常量。
- 删除旧步骤:`clear_and_scan` / `foreach_dispatch` / `scan_changes` / `update_catalog`(旧) / `update_index`(旧);`clear_store.py` 取代 clear_and_scan。

### 4. Evolve / Dream 模块(拆分为多步 pipeline)
- 删除旧的单体 `auto_dream.py` / `dream.py` / `dream.yaml`,新增 `dream/` 子包,按 5 个步骤组织:
  - **`extract.py`**:扫描当日 day-index + daily 笔记,对比 file_catalog 找出 changed/deleted,调用 LLM 全局抽取 `units`(procedure/personal/wiki 三桶)与 `topics`,路径与桶做清洗/路由。
  - **`integrate.py`**:逐个 unit 调用 LLM 写入 digest,结构化输出 `IntegrateOutcome`(CREATE/CORROBORATE/REFINE/CORRECT),失败 unit/路径收集回写。
  - **`topics.py`**:写 `daily/<date>/interests.yaml`,结合当天已有 + 近 N 天做去重(`normalize_topic`),可走 LLM 或纯规则去重两条路径。
  - **`proactive.py`**:读取当日 `interests.yaml`,作为主动推荐话题的入口。
  - **`finish.py`**:把变更路径落盘到 dream file_catalog(checkpoint),渲染最终汇总摘要。
- 新增 `schema.py`(`DreamState` 等跨步骤共享状态与结构化输出模型)与 `utils.py`(状态存取、扫描打包、YAML 读写、结构化回复解析等公共函数)。
- `evolve/__init__.py` 导出全部新 step。

### 5. auto_memory / auto_resource(适配新 Agent API)
- **`auto_memory.py`**:会话路径迁移到 `<session_dir>/dialog/<session_id>.jsonl`;改用 `job_tools`;新增 `source_conversation` frontmatter 反向链接(`_session_link`);执行后刷新 day 索引(`refresh_day_index`),并对 session_id 做合法性校验。
- **`auto_resource.py`**:资源改用「同名 daily note」方案(`_compute_note_stem` 取文件 stem);批量处理 `changes: list[dict]`(`_handle_change` 逐项处理,返回逐项结果摘要);agent 会话 id 用稳定的 `uuid5`;同样刷新 day 索引。

### 6. BaseStep 基类增强
- 新增 `dispatch_steps` / `dispatch_step_specs` 机制:`_resolve_dispatch_step()` 支持字符串或 dict 形式的 step spec,`dispatch_steps()` 复用当前 context 调用下游 step。
- 新增 `config_value()`:按 key 取 app config,缺失时回退 `ApplicationConfig` 默认值。
- 小幅清理:`language` 初始化、`copy()`、`Ref.__init__` 签名精简。

### 7. Components 改动
- **`file_store/local_file_store.py`**:持久化改用 zstd 压缩(`.jsonl.zst`,通过新 `utils/jsonl_zst.py`);upsert 时先删除旧 chunk 的 keyword 文档;embedding 复用改为 `(text, embedding)` 键控,要求文本一致才复用;新增 `_matches_search_filter()` 对 vector/keyword 搜索做 path/path_prefix/metadata 的统一后过滤。
- **`keyword_index/bm25_index.py`**:索引文件名加入组件名 + tokenizer 指纹(sha256 前 12 位),快照/恢复时校验指纹防配置漂移;空索引 dump 时删除文件,加载失败抛错而非静默。
- **`file_chunker/markdown_file_chunker.py`**:弃用 `python-frontmatter`,改用内置 YAML 解析(非法 YAML 不阻断正文索引),并修正因 frontmatter 占用行号导致的 AST 行号偏移(`line_offset`)。
- **`cron_job.py`**:大幅简化(-187 行),由原来「dispatch 外部 job/step + 多种调度模式」改为「在自身 steps 上跑 cron 表达式」;`Application` 启动顺序随之调整为 base > stream > background > cron。
- 其余小调整:service(base/http/mcp)、file_graph、file_catalog、as_llm、as_embedding、tokenizer、prompt_handler、base_component 的签名/接口微调。

### 8. Application 生命周期
- `_start()` 启动顺序明确为 components → base → stream → background → cron,启动失败会触发 `_close()` 回滚并 re-raise(不再吞异常)。
- 启动时创建 `session_dir` 目录;新增 `update_component()`(按类型/名就地更新已存在组件,不存在则报错)。

### 9. File IO / 路径安全
- **`_path.py`**:`resolve_path` 增加 vault 越界防护(`is_relative_to` 校验),禁止 `.` / `..` 路径分量,支持 `allow_empty`。
- **`read.py`**:大文件(超过 `MAX_FILE_READ_BYTES`)走按行读取 `read_file_lines_safe`,避免一次性载入内存。
- **`_file_io.py` / `_daily_index.py` / `_path.py`** 等支持函数补齐(如 `refresh_day_index`、`read_file_lines_safe`)。
- **`env_utils.py`**:新增 `parse_env_file()`,`load_env()` 返回加载到的键值、支持 `override`、对无路径调用做幂等缓存。

### 10. Config
- `ApplicationConfig` 新增 `session_dir`(默认 `reme_session`)。
- `config_parser.py`:环境变量展开后做类型转换(`_convert_value`)、dot-notation 与 key=value 参数校验更严格、配置文件路径支持相对 `_CONFIG_DIR` 查找、根非 dict 报错。
- `default.yaml`:作业编排改用 `init_changes_step` + `dispatch_steps`(index/resource/digest 三个 watch loop 与 reindex);新增 `auto_dream`(4 步)、`proactive` 作业,移除旧 `dream`;file_catalog 增配 `resource` / `digest` / `dream` 实例;LLM 默认值与 Claude Code 凭据配置调整(tool_result_limit 50000、thinking_enable=false 等)。

### 11. 其它
- 新增 `steps/common/add.py`(`AddStep` 算术 demo)、`channel/__init__.py` 与 common `__init__` 导出整理。
- 新增 4 篇文档:`docs4/auto_dream_logic_and_step_refactor.md`、`docs4/watch_loop_step_refactor_plan.md`、`docs4/todo.md`,以及 `reme_design.md` 更新。
**
2026-06-19 01:35:31 +08:00

229 lines
10 KiB
Python

"""Integration test for the 4-step auto_dream job and proactive reader.
Runs against a real LLM. The test seeds a dream vault, runs ``auto_dream`` for
2026-05-28, verifies digest/interests/catalog effects, then runs ``proactive``.
Agent messages and generated markdown/yaml/jsonl artifacts are copied to
``tests4/integration/logs/auto_dream_latest/`` for manual inspection.
"""
import asyncio
import json
import shutil
import sys
from pathlib import Path
import yaml
INTEGRATION_DIR = Path(__file__).resolve().parent
ARTIFACT_DIR = INTEGRATION_DIR / "logs" / "auto_dream_latest"
sys.path.insert(0, str(INTEGRATION_DIR))
# pylint: disable=wrong-import-position
from _vault_fixture import DREAM_INPUT_PATH, vault_env # noqa: E402
DREAM_DATE = "2026-05-28"
def _read_text(path: Path) -> str:
return path.read_text(encoding="utf-8")
def _print_text_file(label: str, path: Path) -> str:
text = _read_text(path)
print("\n" + "=" * 70)
print(f"[{label}] {path} ({len(text)} bytes)")
print(text)
print("=" * 70)
return text
def _reset_artifacts() -> None:
if ARTIFACT_DIR.exists():
shutil.rmtree(ARTIFACT_DIR)
ARTIFACT_DIR.mkdir(parents=True, exist_ok=True)
def _copy_artifact(src: Path, label: str) -> Path | None:
if not src.exists():
return None
rel = Path(label) / src.name if src.is_file() else Path(label)
dst = ARTIFACT_DIR / rel
dst.parent.mkdir(parents=True, exist_ok=True)
if src.is_dir():
shutil.copytree(src, dst, dirs_exist_ok=True)
else:
shutil.copy2(src, dst)
return dst
def _copy_outputs(env, message_files: list[Path]) -> list[Path]:
copied: list[Path] = []
for path in message_files:
if dst := _copy_artifact(path, "messages"):
copied.append(dst)
for root in ("daily", "digest", "reme_metadata", "reme_session", "agent_logs"):
if dst := _copy_artifact(env.vault_dir / root, root):
copied.append(dst)
return copied
def _print_message_files(paths: list[Path]) -> None:
for idx, path in enumerate(paths, 1):
text = _read_text(path)
print("\n" + "=" * 70)
print(f"[messages] {idx}: {path} ({len(text)} bytes)")
print(text[:6000])
if len(text) > 6000:
print("\n[truncated]")
print("=" * 70)
def _all_digest_text(env) -> str:
return "\n\n".join(_read_text(p) for p in env.digest_files())
def _file_graph_links(env) -> dict[str, list[dict]]:
"""Map ``path -> links`` from every ``reme_metadata/file_graph/*.jsonl``."""
out: dict[str, list[dict]] = {}
graph_dir = env.vault_dir / "reme_metadata" / "file_graph"
if not graph_dir.is_dir():
return out
for graph_path in sorted(graph_dir.glob("*.jsonl")):
for line in graph_path.read_text(encoding="utf-8").splitlines():
if not line.strip():
continue
node = json.loads(line)
out[node.get("path", "")] = node.get("links") or []
return out
def test_auto_dream_and_proactive():
"""Run auto_dream end to end, save transcripts/results, then read interests via proactive."""
async def run():
_reset_artifacts()
with vault_env() as env:
seeded = env.seed_dream_vault()
app = await env.make_app()
message_files: list[Path] = []
try:
print("\n" + "=" * 70)
print("[setup] vault_root =", env.vault_dir)
print("[setup] date =", DREAM_DATE)
print("[setup] seeded =", json.dumps(seeded, ensure_ascii=False, indent=2))
print("=" * 70)
before_digest = _all_digest_text(env)
with env.record_agents(prefix="agent_dream") as recorder:
response = await app.run_job(
"auto_dream",
date=DREAM_DATE,
hint="Integration test: preserve SOC2, JWT kid, Redis current_kid, and small-PR facts.",
topic_count=3,
topic_diversity_days=7,
)
dumped = await recorder.dump()
session_jsonl = sorted((env.vault_dir / "reme_session" / "agentscope").glob("*.jsonl"))
message_files = [*dumped, *session_jsonl]
_print_message_files(message_files)
assert response.success is True, f"auto_dream failed: {response.answer!r}\n{response.metadata!r}"
dream = (response.metadata or {}).get("dream") or {}
assert dream.get("date") == DREAM_DATE
assert dream.get("files_changed", 0) >= 1, dream
assert dream.get("units"), f"extract produced no units: {dream!r}"
assert dream.get("integrate_results"), f"integrate produced no results: {dream!r}"
assert dream.get("checkpoint_paths"), f"finish did not checkpoint changed paths: {dream!r}"
day_index = env.vault_dir / "daily" / f"{DREAM_DATE}.md"
changed_note = env.vault_dir / DREAM_INPUT_PATH
interests = env.vault_dir / "daily" / DREAM_DATE / "interests.yaml"
catalog = env.vault_dir / "reme_metadata" / "file_catalog" / "dream.jsonl"
assert changed_note.is_file(), f"changed note missing: {changed_note}"
assert interests.is_file(), f"interests.yaml missing: {interests}"
assert catalog.is_file(), f"dream catalog missing: {catalog}"
after_digest = _all_digest_text(env)
new_signal = [
needle
for needle in ("SOC2", "24", "current_kid", "Redis", "300", "next steps", "kid")
if needle in after_digest and needle not in before_digest
]
print(f"[dream] new digest signals: {new_signal}")
assert len(new_signal) >= 2, f"digest missed expected new signal\n--- digest ---\n{after_digest}"
# wikilink: the seeded daily note cites existing digest nodes, so it
# should have outbound wikilink edges in the file_graph, and the dream
# integrate agents should emit provenance and digest↔digest wikilinks
# in the markdown they just created or updated.
graph_links = _file_graph_links(env)
note_links = [
link.get("target_path") for link in graph_links.get(DREAM_INPUT_PATH, []) if isinstance(link, dict)
]
print(f"[wikilink] {DREAM_INPUT_PATH} -> {note_links}")
assert note_links, (
f"seeded daily note produced no wikilink out-edges in file_graph\n"
f"file_graph links: {graph_links}"
)
assert any(
str(target).lstrip("!").startswith("digest/") for target in note_links
), f"daily note did not link out to any digest node: {note_links}"
target_paths = [
str(result.get("target_path") or "")
for result in dream.get("integrate_results", [])
if result.get("target_path")
]
target_texts = {
rel: _read_text(env.vault_dir / rel) for rel in target_paths if (env.vault_dir / rel).is_file()
}
digest_wikilinks = [rel for rel, text in target_texts.items() if "[[digest/" in text]
provenance_links = [
rel for rel, text in target_texts.items() if f"derived_from:: [[{DREAM_INPUT_PATH}]]" in text
]
print(f"[wikilink] integrated targets: {target_paths}")
print(f"[wikilink] integrated targets with [[digest/...]] links: {digest_wikilinks}")
print(f"[wikilink] integrated targets with derived_from source links: {provenance_links}")
assert target_texts, f"no integrated target files found: {target_paths}"
assert provenance_links, (
"no derived_from wikilink back to the changed daily note in integrated targets\n"
f"targets: {target_paths}"
)
assert digest_wikilinks, (
"no digest↔digest wikilink found in integrated target markdown\n" f"targets: {target_paths}"
)
interests_text = _print_text_file("interests.yaml", interests)
interests_data = yaml.safe_load(interests_text) or {}
topics = interests_data.get("topics") or []
assert isinstance(topics, list) and topics, f"no topics in interests.yaml\n{interests_text}"
proactive = await app.run_job("proactive", date=DREAM_DATE, include_content=True)
assert proactive.success is True, f"proactive failed: {proactive.answer!r}"
assert proactive.metadata.get("path") == f"daily/{DREAM_DATE}/interests.yaml"
assert proactive.metadata.get("topics"), f"proactive returned no topics: {proactive.metadata!r}"
if day_index.is_file():
_print_text_file("day_index.md", day_index)
else:
print(f"[day_index] skipped: no direct daily notes, so {day_index} was not created")
_print_text_file("changed input.md", changed_note)
for path in env.digest_files():
_print_text_file(f"digest {path.relative_to(env.vault_dir)}", path)
_print_text_file("dream catalog.jsonl", catalog)
finally:
copied = _copy_outputs(env, message_files)
print("\n" + "=" * 70)
print(f"[artifacts] copied {len(copied)} item(s) to {ARTIFACT_DIR}")
for path in copied:
print(f"[artifacts] {path}")
print("=" * 70)
await env.close_all()
asyncio.run(run())
if __name__ == "__main__":
print("=== auto_dream integration test ===")
test_auto_dream_and_proactive()
print("\nIntegration test passed!")