Repository navigation
Expand file tree
/
Copy pathMakefile
More file actions
129 lines (106 loc) · 5.81 KB
/
Copy pathMakefile
File metadata and controls
129 lines (106 loc) · 5.81 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
.PHONY: help demo review summarize market lint fix test dev serve kill review-deepseek review-agnes review-kimi review-minimax _auto-summarize
VENV_PY := .venv/bin/python
PY := $(shell command -v $(VENV_PY) 2>/dev/null || echo "python3")
CLI := $(PY) cli.py
# Web 服务绑定地址：默认仅本机；对外暴露须 WEB_BIND_HOST=0.0.0.0 且先设 WEB_AUTH_TOKEN
WEB_BIND_HOST ?= 127.0.0.1
help: ## 显示帮助
@echo "autogen-pse 可用命令:"
@grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | awk 'BEGIN {FS = ":.*?## "}; {printf " make %-18s %s\n", $$1, $$2}'
test: ## 运行测试
$(PY) -m pytest tests/ -v
lint: ## 代码检查
$(PY) -m ruff check src/ tasks/ tests/
$(PY) -m ruff format --check src/ tasks/ tests/
fix: ## 自动修复代码风格
$(PY) -m ruff check src/ tasks/ tests/ --fix
$(PY) -m ruff format src/ tasks/ tests/
demo: ## 运行 PSE 代码交付 Demo
$(CLI) run demo
summarize: ## 生成投资结构化摘要（零 LLM 成本）
$(CLI) prepare portfolio-review
market: ## 查看最新市场指数行情
$(PY) tasks/portfolio-review/prepare_market.py
review: review-deepseek ## 默认用 DeepSeek V4 Pro（等价于 make review-deepseek）
# 自动检测：比对 portfolio_review_prompt.md 日期 vs asset-lens / money-csv 最新数据
_auto-summarize:
@ASSET_DIR=$(shell grep '^ASSET_LENS_DIR=' .env | cut -d= -f2); \
MONEY_DIR=$(shell grep '^MONEY_CSV_DIR=' .env | cut -d= -f2); \
PROMPT_DATE=$$(sed -n 's/.*截止 \([0-9]\{4\}\)年\([0-9]\{2\}\)月\([0-9]\{2\}\)日.*/\1\2\3/p' tasks/portfolio-review/output/portfolio_review_prompt.md 2>/dev/null | head -1); \
[ -z "$$PROMPT_DATE" ] && PROMPT_DATE="00000000"; \
ASSET_DATE=$$(ls -t "$$ASSET_DIR"/output/投资收益率分析_*.json 2>/dev/null | head -1 | grep -o '[0-9]\{8\}' || echo "00000000"); \
MONEY_DATE=$$(ls -td "$$MONEY_DIR"/money_csv_* 2>/dev/null | head -1 | grep -o '[0-9]\{8\}' || echo "00000000"); \
if [ "$$ASSET_DATE" -gt "$$PROMPT_DATE" ] 2>/dev/null || [ "$$MONEY_DATE" -gt "$$PROMPT_DATE" ] 2>/dev/null; then \
echo "🔍 检测到新数据 (prompt: $$PROMPT_DATE asset: $$ASSET_DATE money: $$MONEY_DATE)，自动 summarize..."; \
$(MAKE) summarize; \
else \
echo "✅ 数据已是最新 (prompt: $$PROMPT_DATE asset: $$ASSET_DATE money: $$MONEY_DATE)"; \
fi
# 开发模式：通过 MODEL 环境变量切换模型
# MODEL=your-model-name make dev-review # DeepSeek Flash
# MODEL=llama3.1:8b make dev-review # 本地 Ollama
dev-review: _auto-summarize
OPENAI_BASE_URL=$(shell grep '^OPENAI_BASE_URL=' .env | cut -d= -f2) \
OPENAI_MODEL=$(MODEL) $(PY) tasks/portfolio-review/run.py
local-review: _auto-summarize ## 开发模式：使用本地 Ollama llama3.1:8b
OPENAI_BASE_URL=http://localhost:11434/v1 \
OPENAI_MODEL=llama3.1:8b \
OPENAI_API_KEY=ollama \
$(PY) tasks/portfolio-review/run.py
flash-review: _auto-summarize ## 开发模式：DeepSeek Flash（更快）
OPENAI_API_KEY=$(shell grep '^OPENAI_API_KEY=' .env | cut -d= -f2) \
OPENAI_BASE_URL=$(shell grep '^OPENAI_BASE_URL=' .env | cut -d= -f2) \
OPENAI_MODEL=$(MODEL) \
$(PY) tasks/portfolio-review/run.py
# ── 多模型快捷命令 ──
review-deepseek: _auto-summarize ## 用 DeepSeek V4 Pro 跑周报（流式，稳定）
OPENAI_API_KEY=$(shell grep '^OPENAI_API_KEY=' .env | cut -d= -f2) \
OPENAI_BASE_URL=$(shell grep '^OPENAI_BASE_URL=' .env | cut -d= -f2) \
OPENAI_MODEL=$(shell grep '^OPENAI_MODEL=' .env | cut -d= -f2) \
PSE_MODEL_STREAM=true \
PSE_TIMEOUT=300 \
$(PY) tasks/portfolio-review/run.py
review-agnes: _auto-summarize ## 用 Agnes 2.0 Flash 跑周报（非流式，免费，放宽循环参数）
OPENAI_API_KEY=$(shell grep '^AGNES_KEY=' .env | cut -d= -f2) \
OPENAI_BASE_URL=$(shell grep '^AGNES_BASE_URL=' .env | cut -d= -f2) \
OPENAI_MODEL=$(shell grep '^AGNES_MODEL=' .env | cut -d= -f2) \
PSE_MODEL_STREAM=false \
PSE_TURNS_PER_CYCLE=25 \
PSE_MAX_PARTIAL_RETRIES=5 \
PSE_TIMEOUT=300 \
$(PY) tasks/portfolio-review/run.py
review-kimi: _auto-summarize ## ⚠️ Kimi-K2.6 推理模型输出在 reasoning_content 中，周报质量不佳，建议用 MiniMax 替代
OPENAI_API_KEY=$(shell sed -n 's/^SCNET_KEY=//p' .env) \
OPENAI_BASE_URL=$(shell sed -n 's/^SCNET_BASE_URL=//p' .env) \
OPENAI_MODEL=$(shell sed -n 's/^SCNET_KIMI_MODEL=//p' .env) \
PSE_MODEL_STREAM=false \
PSE_TURNS_PER_CYCLE=25 \
PSE_MAX_PARTIAL_RETRIES=3 \
PSE_TIMEOUT=300 \
$(PY) tasks/portfolio-review/run.py
review-minimax: _auto-summarize ## 用 MiniMax M2.5 跑周报（SCNet，非流式）
OPENAI_API_KEY=$(shell sed -n 's/^SCNET_KEY=//p' .env) \
OPENAI_BASE_URL=$(shell sed -n 's/^SCNET_BASE_URL=//p' .env) \
OPENAI_MODEL=$(shell sed -n 's/^SCNET_MINIMAX_MODEL=//p' .env) \
PSE_MODEL_STREAM=false \
PSE_TURNS_PER_CYCLE=25 \
PSE_MAX_PARTIAL_RETRIES=3 \
PSE_TIMEOUT=300 \
$(PY) tasks/portfolio-review/run.py
# 绑定守卫：仅在对外暴露（0.0.0.0）且未设 WEB_AUTH_TOKEN 时拒绝启动
guard-bind:
@if [ "$(WEB_BIND_HOST)" = "0.0.0.0" ] && ! grep -q '^WEB_AUTH_TOKEN=' .env 2>/dev/null; then \
echo "❌ 拒绝绑定 0.0.0.0：未设置 WEB_AUTH_TOKEN。本机访问用 make serve，或先设 WEB_AUTH_TOKEN 再 make serve WEB_BIND_HOST=0.0.0.0"; exit 1; \
fi
serve: guard-bind serve-web ## 启动 Web Dashboard（默认仅本机 127.0.0.1）
serve-web: web-build
$(PY) -m uvicorn web_server:app --host $(WEB_BIND_HOST) --port 8080
dev: guard-bind kill ## 开发模式：先杀旧进程，再启动 FastAPI + Vite
$(PY) -m uvicorn web_server:app --host $(WEB_BIND_HOST) --port 8080 &
@sleep 2
cd web && npm run dev
web-build: ## 构建前端
cd web && npm install --silent && npm run build
kill: ## 杀掉开发服务器进程（8080 / 5173）
@lsof -ti:8080 | xargs kill -9 2>/dev/null && echo "✅ 已杀掉 :8080" || echo " :8080 无进程"
@lsof -ti:5173 | xargs kill -9 2>/dev/null && echo "✅ 已杀掉 :5173" || echo " :5173 无进程"