feat: 初始化零工后端代码

This commit is contained in:
Daniel
2026-04-01 14:19:25 +08:00
parent c6fabe262c
commit 84f8be7c0e
41 changed files with 2813 additions and 147 deletions

View File

@@ -0,0 +1,114 @@
#!/usr/bin/env sh
set -eu
API_BASE="${API_BASE:-http://127.0.0.1:8000}"
BOOTSTRAP_ON_RUN="${BOOTSTRAP_ON_RUN:-true}"
TMP_DIR="$(mktemp -d)"
trap 'rm -rf "$TMP_DIR"' EXIT
step() {
echo "[ACCEPTANCE] $1"
}
fail() {
echo "[ACCEPTANCE][FAIL] $1" >&2
exit 1
}
parse_json() {
PY_EXPR="$1"
INPUT_FILE="$2"
python3 - "$PY_EXPR" "$INPUT_FILE" <<'PY'
import json
import sys
expr = sys.argv[1]
path = sys.argv[2]
data = json.loads(open(path, "r", encoding="utf-8").read())
safe_builtins = {"bool": bool, "len": len, "str": str}
value = eval(expr, {"__builtins__": safe_builtins}, {"data": data})
if isinstance(value, (dict, list)):
print(json.dumps(value, ensure_ascii=False))
elif value is None:
print("")
else:
print(str(value))
PY
}
step "健康检查"
curl -fsS "${API_BASE}/health" >"$TMP_DIR/health.json" || fail "health 接口不可用"
if [ "$BOOTSTRAP_ON_RUN" = "true" ]; then
step "执行 bootstrap"
curl -fsS -X POST "${API_BASE}/poc/ingest/bootstrap" >"$TMP_DIR/bootstrap.json" || fail "bootstrap 失败"
fi
step "抽取岗位"
curl -fsS -X POST "${API_BASE}/poc/extract/job" \
-H "Content-Type: application/json" \
-d '{"text":"明天下午南山会展中心需要2个签到协助5小时150/人,女生优先"}' \
>"$TMP_DIR/extract_job.json" || fail "岗位抽取调用失败"
[ "$(parse_json "bool(data.get('success'))" "$TMP_DIR/extract_job.json")" = "True" ] || fail "岗位抽取失败"
parse_json "data.get('data')" "$TMP_DIR/extract_job.json" >"$TMP_DIR/job.json"
JOB_ID="$(parse_json "data.get('data', {}).get('job_id')" "$TMP_DIR/extract_job.json")"
[ -n "$JOB_ID" ] || fail "岗位抽取缺少 job_id"
step "岗位入库"
python3 - "$TMP_DIR/job.json" >"$TMP_DIR/ingest_job_payload.json" <<'PY'
import json
import sys
job = json.loads(open(sys.argv[1], "r", encoding="utf-8").read())
print(json.dumps({"job": job}, ensure_ascii=False))
PY
curl -fsS -X POST "${API_BASE}/poc/ingest/job" \
-H "Content-Type: application/json" \
--data @"$TMP_DIR/ingest_job_payload.json" \
>"$TMP_DIR/ingest_job.json" || fail "岗位入库失败"
step "岗位匹配工人"
curl -fsS -X POST "${API_BASE}/poc/match/workers" \
-H "Content-Type: application/json" \
-d "{\"job_id\":\"${JOB_ID}\",\"top_n\":3}" \
>"$TMP_DIR/match_workers.json" || fail "岗位匹配工人失败"
MATCH_ID_1="$(parse_json "((data.get('items') or [{}])[0]).get('match_id')" "$TMP_DIR/match_workers.json")"
[ -n "$MATCH_ID_1" ] || fail "岗位匹配工人未返回 match_id"
step "解释匹配(岗位->工人)"
curl -fsS "${API_BASE}/poc/match/explain/${MATCH_ID_1}" >"$TMP_DIR/explain_1.json" || fail "匹配解释失败(岗位->工人)"
step "抽取工人"
curl -fsS -X POST "${API_BASE}/poc/extract/worker" \
-H "Content-Type: application/json" \
-d '{"text":"我做过商场促销和活动签到,也能做登记和引导,周末都能接,福田南山都方便。"}' \
>"$TMP_DIR/extract_worker.json" || fail "工人抽取调用失败"
[ "$(parse_json "bool(data.get('success'))" "$TMP_DIR/extract_worker.json")" = "True" ] || fail "工人抽取失败"
parse_json "data.get('data')" "$TMP_DIR/extract_worker.json" >"$TMP_DIR/worker.json"
WORKER_ID="$(parse_json "data.get('data', {}).get('worker_id')" "$TMP_DIR/extract_worker.json")"
[ -n "$WORKER_ID" ] || fail "工人抽取缺少 worker_id"
step "工人入库"
python3 - "$TMP_DIR/worker.json" >"$TMP_DIR/ingest_worker_payload.json" <<'PY'
import json
import sys
worker = json.loads(open(sys.argv[1], "r", encoding="utf-8").read())
print(json.dumps({"worker": worker}, ensure_ascii=False))
PY
curl -fsS -X POST "${API_BASE}/poc/ingest/worker" \
-H "Content-Type: application/json" \
--data @"$TMP_DIR/ingest_worker_payload.json" \
>"$TMP_DIR/ingest_worker.json" || fail "工人入库失败"
step "工人匹配岗位"
curl -fsS -X POST "${API_BASE}/poc/match/jobs" \
-H "Content-Type: application/json" \
-d "{\"worker_id\":\"${WORKER_ID}\",\"top_n\":3}" \
>"$TMP_DIR/match_jobs.json" || fail "工人匹配岗位失败"
MATCH_ID_2="$(parse_json "((data.get('items') or [{}])[0]).get('match_id')" "$TMP_DIR/match_jobs.json")"
[ -n "$MATCH_ID_2" ] || fail "工人匹配岗位未返回 match_id"
step "解释匹配(工人->岗位)"
curl -fsS "${API_BASE}/poc/match/explain/${MATCH_ID_2}" >"$TMP_DIR/explain_2.json" || fail "匹配解释失败(工人->岗位)"
step "链路验收通过:抽取 -> 入库 -> 匹配 -> 解释"

View File

@@ -3,6 +3,10 @@ set -eu
SCRIPT_DIR=$(CDPATH= cd -- "$(dirname "$0")" && pwd)
INFRA_DIR=$(CDPATH= cd -- "$SCRIPT_DIR/.." && pwd)
PROJECT_DIR=$(CDPATH= cd -- "$INFRA_DIR/.." && pwd)
RUN_ACCEPTANCE_ON_UP="${RUN_ACCEPTANCE_ON_UP:-true}"
EXPORT_OPENAPI_ON_UP="${EXPORT_OPENAPI_ON_UP:-true}"
RUN_BASELINE_ON_UP="${RUN_BASELINE_ON_UP:-false}"
cd "$INFRA_DIR"
docker compose -f docker-compose.yml up --build -d
@@ -13,4 +17,20 @@ done
until curl -fsS -X POST http://127.0.0.1:8000/poc/ingest/bootstrap >/dev/null 2>&1; do
sleep 3
done
if [ "$RUN_ACCEPTANCE_ON_UP" = "true" ]; then
echo "执行一键闭环验收脚本..."
sh "$SCRIPT_DIR/acceptance-e2e.sh"
fi
if [ "$EXPORT_OPENAPI_ON_UP" = "true" ]; then
echo "导出 OpenAPI 固化产物到 docs/openapi.json ..."
sh "$SCRIPT_DIR/export-openapi.sh" "$PROJECT_DIR/docs/openapi.json"
fi
if [ "$RUN_BASELINE_ON_UP" = "true" ]; then
echo "执行容量基线压测..."
sh "$SCRIPT_DIR/load-baseline.sh" "$PROJECT_DIR/docs/CAPACITY_BASELINE.md"
fi
echo "本地环境已启动。Web: http://127.0.0.1:5173 API: http://127.0.0.1:8000/docs"

View File

@@ -0,0 +1,11 @@
#!/usr/bin/env sh
set -eu
SCRIPT_DIR=$(CDPATH= cd -- "$(dirname "$0")" && pwd)
PROJECT_DIR=$(CDPATH= cd -- "$SCRIPT_DIR/../.." && pwd)
API_PORT="${API_PORT:-8000}"
OUTPUT_PATH="${1:-$PROJECT_DIR/docs/openapi.json}"
mkdir -p "$(dirname "$OUTPUT_PATH")"
curl -fsS "http://127.0.0.1:${API_PORT}/openapi.json" -o "$OUTPUT_PATH"
echo "OpenAPI 已导出到: $OUTPUT_PATH"

View File

@@ -0,0 +1,37 @@
#!/usr/bin/env sh
set -eu
SCRIPT_DIR=$(CDPATH= cd -- "$(dirname "$0")" && pwd)
PROJECT_DIR=$(CDPATH= cd -- "$SCRIPT_DIR/../.." && pwd)
OUTPUT_PATH="${1:-$PROJECT_DIR/docs/openapi.json}"
API_PORT="${API_PORT:-8000}"
mkdir -p "$(dirname "$OUTPUT_PATH")"
if PYTHONPATH="$PROJECT_DIR/apps/api" python3 - "$OUTPUT_PATH" <<'PY'
import json
import sys
from app.main import app
output = sys.argv[1]
spec = app.openapi()
with open(output, "w", encoding="utf-8") as f:
json.dump(spec, f, ensure_ascii=False, indent=2)
f.write("\n")
print(f"OpenAPI 已固化到: {output}")
PY
then
exit 0
fi
echo "本机缺少 API 依赖,尝试从已运行 API 导出..."
if curl -fsS "http://127.0.0.1:${API_PORT}/openapi.json" -o "$OUTPUT_PATH"; then
echo "OpenAPI 已固化到: $OUTPUT_PATH"
exit 0
fi
echo "本机 API 端口不可用,尝试通过 Docker 运行 API 镜像离线导出..."
docker compose -f "$PROJECT_DIR/infrastructure/docker-compose.yml" run --rm api \
python -c "import json; from app.main import app; print(json.dumps(app.openapi(), ensure_ascii=False, indent=2))" \
> "$OUTPUT_PATH"
echo "OpenAPI 已固化到: $OUTPUT_PATH"

View File

@@ -0,0 +1,137 @@
#!/usr/bin/env sh
set -eu
API_BASE="${API_BASE:-http://127.0.0.1:8000}"
TOTAL_REQUESTS="${TOTAL_REQUESTS:-400}"
CONCURRENCY="${CONCURRENCY:-40}"
OUTPUT_PATH="${1:-$(CDPATH= cd -- "$(dirname "$0")/../.." && pwd)/docs/CAPACITY_BASELINE.md}"
TMP_DIR="$(mktemp -d)"
trap 'rm -rf "$TMP_DIR"' EXIT
echo "[BASELINE] health check"
curl -fsS "$API_BASE/health" >/dev/null
echo "[BASELINE] ensure bootstrap data"
curl -fsS -X POST "$API_BASE/poc/ingest/bootstrap" >/dev/null
JOB_ID="$(curl -fsS "$API_BASE/poc/jobs" | python3 -c 'import json,sys; data=json.load(sys.stdin); print((data.get("items") or [{}])[0].get("job_id",""))')"
WORKER_ID="$(curl -fsS "$API_BASE/poc/workers" | python3 -c 'import json,sys; data=json.load(sys.stdin); print((data.get("items") or [{}])[0].get("worker_id",""))')"
[ -n "$JOB_ID" ] || { echo "no job id found"; exit 1; }
[ -n "$WORKER_ID" ] || { echo "no worker id found"; exit 1; }
run_case() {
NAME="$1"
METHOD="$2"
URL="$3"
BODY_FILE="$4"
OUT_FILE="$5"
python3 - "$METHOD" "$URL" "$BODY_FILE" "$TOTAL_REQUESTS" "$CONCURRENCY" "$OUT_FILE" <<'PY'
import json
import sys
import time
import urllib.request
from concurrent.futures import ThreadPoolExecutor, as_completed
method, url, body_file, total, concurrency, out_file = sys.argv[1:]
total = int(total)
concurrency = int(concurrency)
payload = None
if body_file != "-":
payload = open(body_file, "rb").read()
durations = []
success = 0
fail = 0
def once():
start = time.perf_counter()
req = urllib.request.Request(url=url, method=method)
req.add_header("Content-Type", "application/json")
try:
if payload is None:
with urllib.request.urlopen(req, timeout=20) as resp:
code = resp.getcode()
else:
with urllib.request.urlopen(req, data=payload, timeout=20) as resp:
code = resp.getcode()
ok = 200 <= code < 400
except Exception:
ok = False
ms = (time.perf_counter() - start) * 1000
return ok, ms
bench_start = time.perf_counter()
with ThreadPoolExecutor(max_workers=concurrency) as ex:
futures = [ex.submit(once) for _ in range(total)]
for f in as_completed(futures):
ok, ms = f.result()
durations.append(ms)
if ok:
success += 1
else:
fail += 1
elapsed = time.perf_counter() - bench_start
durations.sort()
def pct(p):
if not durations:
return 0.0
idx = min(len(durations) - 1, int(len(durations) * p))
return round(durations[idx], 2)
result = {
"total": total,
"success": success,
"fail": fail,
"success_rate": round(success / total, 4) if total else 0.0,
"rps": round(total / elapsed, 2) if elapsed > 0 else 0.0,
"latency_ms_avg": round(sum(durations) / len(durations), 2) if durations else 0.0,
"latency_ms_p95": pct(0.95),
"latency_ms_p99": pct(0.99),
}
with open(out_file, "w", encoding="utf-8") as f:
json.dump(result, f, ensure_ascii=False, indent=2)
PY
echo "[BASELINE] done $NAME"
}
printf '{"job_id":"%s","top_n":10}\n' "$JOB_ID" >"$TMP_DIR/match_workers.json"
printf '{"worker_id":"%s","top_n":10}\n' "$WORKER_ID" >"$TMP_DIR/match_jobs.json"
run_case "health" "GET" "$API_BASE/health" "-" "$TMP_DIR/health.result.json"
run_case "jobs_list" "GET" "$API_BASE/poc/jobs" "-" "$TMP_DIR/jobs.result.json"
run_case "match_workers" "POST" "$API_BASE/poc/match/workers" "$TMP_DIR/match_workers.json" "$TMP_DIR/match_workers.result.json"
run_case "match_jobs" "POST" "$API_BASE/poc/match/jobs" "$TMP_DIR/match_jobs.json" "$TMP_DIR/match_jobs.result.json"
run_case "match_workers_cached" "POST" "$API_BASE/poc/match/workers" "$TMP_DIR/match_workers.json" "$TMP_DIR/match_workers_cached.result.json"
run_case "match_jobs_cached" "POST" "$API_BASE/poc/match/jobs" "$TMP_DIR/match_jobs.json" "$TMP_DIR/match_jobs_cached.result.json"
run_case "match_workers_async_enqueue" "POST" "$API_BASE/poc/match/workers/async" "$TMP_DIR/match_workers.json" "$TMP_DIR/match_workers_async.result.json"
run_case "match_jobs_async_enqueue" "POST" "$API_BASE/poc/match/jobs/async" "$TMP_DIR/match_jobs.json" "$TMP_DIR/match_jobs_async.result.json"
NOW="$(date '+%Y-%m-%d %H:%M:%S %z')"
mkdir -p "$(dirname "$OUTPUT_PATH")"
{
echo "# 容量基线(自动生成)"
echo
echo "- 生成时间: $NOW"
echo "- API_BASE: $API_BASE"
echo "- TOTAL_REQUESTS: $TOTAL_REQUESTS"
echo "- CONCURRENCY: $CONCURRENCY"
echo
echo "| 场景 | 成功率 | RPS | 平均延迟(ms) | P95(ms) | P99(ms) |"
echo "| --- | --- | --- | --- | --- | --- |"
for case in health jobs match_workers match_jobs match_workers_cached match_jobs_cached match_workers_async match_jobs_async; do
FILE="$TMP_DIR/${case}.result.json"
python3 - "$case" "$FILE" <<'PY'
import json
import sys
case, path = sys.argv[1], sys.argv[2]
data = json.loads(open(path, "r", encoding="utf-8").read())
print(f"| {case} | {data['success_rate']} | {data['rps']} | {data['latency_ms_avg']} | {data['latency_ms_p95']} | {data['latency_ms_p99']} |")
PY
done
echo
echo "> 建议:该基线仅代表当前单机/当前数据量下表现,发布前请在目标环境按 2x/5x 峰值复测。"
} >"$OUTPUT_PATH"
echo "[BASELINE] report generated at $OUTPUT_PATH"

View File

@@ -0,0 +1,8 @@
#!/usr/bin/env sh
set -eu
SCRIPT_DIR=$(CDPATH= cd -- "$(dirname "$0")" && pwd)
INFRA_DIR=$(CDPATH= cd -- "$SCRIPT_DIR/.." && pwd)
cd "$INFRA_DIR"
docker compose -f docker-compose.prod.yml down

View File

@@ -3,7 +3,21 @@ set -eu
SCRIPT_DIR=$(CDPATH= cd -- "$(dirname "$0")" && pwd)
INFRA_DIR=$(CDPATH= cd -- "$SCRIPT_DIR/.." && pwd)
API_PORT="${API_PORT:-8000}"
BOOTSTRAP_ON_UP="${BOOTSTRAP_ON_UP:-true}"
cd "$INFRA_DIR"
docker compose -f docker-compose.prod.yml up --build -d
echo "生产部署容器已启动。请按实际域名或端口访问 Web。"
echo "等待生产 API 健康检查..."
until curl -fsS "http://127.0.0.1:${API_PORT}/health" >/dev/null 2>&1; do
sleep 3
done
if [ "$BOOTSTRAP_ON_UP" = "true" ]; then
echo "执行 bootstrap 样本初始化..."
until curl -fsS -X POST "http://127.0.0.1:${API_PORT}/poc/ingest/bootstrap" >/dev/null 2>&1; do
sleep 3
done
fi
echo "生产环境已启动。Web: http://127.0.0.1:${WEB_PORT:-80} API: http://127.0.0.1:${API_PORT}/docs"