tinet-data-foundry-cli 0.1.0__py3-none-any.whl

tinet_data_foundry_cli/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.1.0"

tinet_data_foundry_cli/__main__.py

@@ -0,0 +1,4 @@
+from tinet_data_foundry_cli.cli import main
+
+if __name__ == "__main__":
+    raise SystemExit(main())

tinet_data_foundry_cli/auth.py

@@ -0,0 +1,55 @@
+"""HMAC-SHA256 签名，与 data-api SignatureVerifier 一致。"""
+
+from __future__ import annotations
+
+import base64
+import hashlib
+import hmac
+import os
+import time
+import uuid
+from typing import Tuple
+
+DEFAULT_ENDPOINT = "https://data-foundry-data-api.tinetcloud.com"
+
+
+def sha256_hex(body: bytes) -> str:
+    return hashlib.sha256(body or b"").hexdigest()
+
+
+def canonical(method: str, path: str, timestamp: str, nonce: str, body: bytes) -> str:
+    return "\n".join(
+        [
+            method.upper(),
+            path,
+            timestamp,
+            nonce,
+            sha256_hex(body),
+        ]
+    )
+
+
+def sign_base64(method: str, path: str, timestamp: str, nonce: str, body: bytes, secret: str) -> str:
+    raw = hmac.new(
+        secret.encode("utf-8"),
+        canonical(method, path, timestamp, nonce, body).encode("utf-8"),
+        hashlib.sha256,
+    ).digest()
+    return base64.b64encode(raw).decode("ascii")
+
+
+def new_nonce() -> str:
+    return uuid.uuid4().hex
+
+
+def now_ts() -> str:
+    return str(int(time.time()))
+
+
+def credentials(endpoint: str | None = None) -> Tuple[str, str, str]:
+    base = (endpoint or os.environ.get("DF_ENDPOINT") or DEFAULT_ENDPOINT).rstrip("/")
+    ak = (os.environ.get("DF_AK") or "").strip()
+    sk = (os.environ.get("DF_SK") or "").strip()
+    if not ak or not sk:
+        raise SystemExit("缺少 DF_AK / DF_SK 环境变量（不要写入本地配置文件）")
+    return base, ak, sk

tinet_data_foundry_cli/cli.py

@@ -0,0 +1,204 @@
+from __future__ import annotations
+
+import argparse
+import json
+import sys
+from typing import Any, List, Mapping, Optional
+
+from tinet_data_foundry_cli import __version__
+from tinet_data_foundry_cli.auth import credentials
+from tinet_data_foundry_cli.client import ApiError, request
+
+
+def _parse_json(raw: Optional[str], flag: str) -> Any:
+    if not raw:
+        return None
+    try:
+        return json.loads(raw)
+    except json.JSONDecodeError as ex:
+        raise SystemExit(f"{flag} 不是合法 JSON：{ex}") from ex
+
+
+def _print_json(data: Any) -> None:
+    json.dump(data, sys.stdout, ensure_ascii=False, indent=2)
+    sys.stdout.write("\n")
+
+
+def _cell(value: Any, width: int) -> str:
+    text = "—" if value is None or value == "" else str(value).replace("\n", " ")
+    if len(text) > width:
+        text = text[: max(0, width - 1)] + "…"
+    return text.ljust(width)
+
+
+def _print_catalog_table(rows: List[Mapping[str, Any]]) -> None:
+    if not rows:
+        print("当前凭证没有可查询的 Dataset。")
+        return
+    cols = [
+        ("id", "datasetId", 36),
+        ("name", "名称", 28),
+        ("tableComment", "中文描述", 24),
+        ("businessDomainName", "业务域", 12),
+        ("layer", "分层", 6),
+        ("schemaVersion", "版本", 8),
+        ("fieldCount", "列数", 6),
+        ("rlsColumn", "行级", 16),
+    ]
+    header = "  ".join(_cell(title, w) for _, title, w in cols)
+    print(header)
+    print("-" * len(header))
+    for row in rows:
+        print("  ".join(_cell(row.get(key), w) for key, _, w in cols))
+
+
+def _exit_api(err: ApiError) -> None:
+    payload = err.payload if isinstance(err.payload, dict) else {"message": str(err.payload)}
+    _print_json(payload)
+    code = err.http_status
+    if code == 401:
+        raise SystemExit(2)
+    if code == 403:
+        raise SystemExit(3)
+    if code == 422:
+        raise SystemExit(4)
+    raise SystemExit(1)
+
+
+def _ok_data(parsed: Any) -> Any:
+    if isinstance(parsed, dict) and parsed.get("code") not in (None, 0, "0"):
+        _print_json(parsed)
+        raise SystemExit(1)
+    return parsed.get("data") if isinstance(parsed, dict) else parsed
+
+
+def cmd_catalog(args: argparse.Namespace) -> int:
+    base, ak, sk = credentials(args.endpoint)
+    query = {
+        "keyword": args.keyword or "",
+        "layer": args.layer or "",
+        "domain": args.domain or "",
+    }
+    try:
+        _, parsed, _ = request(
+            "GET",
+            base,
+            "/data-api/v1/catalog",
+            ak,
+            sk,
+            query=query,
+            timeout=args.timeout,
+            insecure=args.insecure,
+            dry_run=args.dry_run,
+        )
+    except ApiError as ex:
+        _exit_api(ex)
+    if args.dry_run or args.format == "json":
+        _print_json(parsed)
+        return 0
+    data = _ok_data(parsed) or {}
+    rows = data.get("list") if isinstance(data, dict) else []
+    _print_catalog_table(list(rows or []))
+    return 0
+
+
+def cmd_query(args: argparse.Namespace) -> int:
+    base, ak, sk = credentials(args.endpoint)
+    payload: dict[str, Any] = {
+        "pageNo": args.page_no,
+        "pageSize": args.page_size,
+    }
+    if args.fields:
+        payload["fields"] = [p.strip() for p in args.fields.split(",") if p.strip()]
+    filt = _parse_json(args.filter, "--filter")
+    if filt is not None:
+        payload["filter"] = filt
+    sort = _parse_json(args.sort, "--sort")
+    if sort is not None:
+        payload["sort"] = sort
+    visitor = _parse_json(args.visitor, "--visitor")
+    if visitor is not None:
+        payload["visitor"] = visitor
+    body = json.dumps(payload, ensure_ascii=False, separators=(",", ":")).encode("utf-8")
+    path = f"/data-api/v1/datasets/{args.dataset}/query"
+    try:
+        _, parsed, _ = request(
+            "POST",
+            base,
+            path,
+            ak,
+            sk,
+            body=body,
+            timeout=args.timeout,
+            insecure=args.insecure,
+            dry_run=args.dry_run,
+        )
+    except ApiError as ex:
+        _exit_api(ex)
+    _print_json(parsed)
+    if isinstance(parsed, dict) and parsed.get("code") not in (None, 0, "0") and not args.dry_run:
+        return 1
+    return 0
+
+
+def cmd_schema(args: argparse.Namespace) -> int:
+    base, ak, sk = credentials(args.endpoint)
+    path = f"/data-api/v1/datasets/{args.dataset}/schema"
+    try:
+        _, parsed, _ = request(
+            "GET",
+            base,
+            path,
+            ak,
+            sk,
+            timeout=args.timeout,
+            insecure=args.insecure,
+            dry_run=args.dry_run,
+        )
+    except ApiError as ex:
+        _exit_api(ex)
+    _print_json(parsed)
+    if isinstance(parsed, dict) and parsed.get("code") not in (None, 0, "0") and not args.dry_run:
+        return 1
+    return 0
+
+
+def build_parser() -> argparse.ArgumentParser:
+    p = argparse.ArgumentParser(
+        prog="df",
+        description="Tinet Data Foundry 取数 CLI。凭证用环境变量 DF_ENDPOINT / DF_AK / DF_SK。",
+    )
+    p.add_argument("--version", action="version", version=f"tinet-data-foundry-cli {__version__}")
+    p.add_argument("--endpoint", help="覆盖 DF_ENDPOINT")
+    p.add_argument("--timeout", type=float, default=30)
+    p.add_argument("--insecure", action="store_true", help="跳过 TLS 校验（仅联调）")
+    p.add_argument("--dry-run", action="store_true", help="只打印将发送的请求，不访问网关")
+    sub = p.add_subparsers(dest="cmd", required=True)
+
+    c = sub.add_parser("catalog", help="列出当前凭证有权访问的数据目录")
+    c.add_argument("--keyword")
+    c.add_argument("--layer", choices=["ods", "dwd"])
+    c.add_argument("--domain", help="业务域 ID 或名称")
+    c.add_argument("--format", choices=["table", "json"], default="table")
+    c.set_defaults(func=cmd_catalog)
+
+    q = sub.add_parser("query", help="查询 Dataset")
+    q.add_argument("--dataset", required=True)
+    q.add_argument("--fields", help="逗号分隔列名")
+    q.add_argument("--filter", help="filter AST JSON")
+    q.add_argument("--sort", help="sort JSON 数组")
+    q.add_argument("--page-no", type=int, default=1)
+    q.add_argument("--page-size", type=int, default=10)
+    q.add_argument("--visitor", help="应用凭证 visitor JSON")
+    q.set_defaults(func=cmd_query)
+
+    s = sub.add_parser("schema", help="拉取 Dataset Schema")
+    s.add_argument("--dataset", required=True)
+    s.set_defaults(func=cmd_schema)
+    return p
+
+
+def main(argv: Optional[List[str]] = None) -> int:
+    parser = build_parser()
+    args = parser.parse_args(argv)
+    return int(args.func(args) or 0)

tinet_data_foundry_cli/client.py

@@ -0,0 +1,67 @@
+from __future__ import annotations
+
+import json
+import ssl
+import urllib.error
+import urllib.parse
+import urllib.request
+from typing import Any, Dict, Optional, Tuple
+
+from tinet_data_foundry_cli.auth import new_nonce, now_ts, sign_base64
+
+
+class ApiError(Exception):
+    def __init__(self, http_status: int, payload: Any):
+        self.http_status = http_status
+        self.payload = payload
+        super().__init__(f"HTTP {http_status}: {payload}")
+
+
+def request(
+    method: str,
+    endpoint: str,
+    path: str,
+    ak: str,
+    sk: str,
+    body: Optional[bytes] = None,
+    query: Optional[Dict[str, str]] = None,
+    timeout: float = 30,
+    insecure: bool = False,
+    dry_run: bool = False,
+) -> Tuple[int, Any, Dict[str, str]]:
+    payload = body or b""
+    ts = now_ts()
+    nonce = new_nonce()
+    signature = sign_base64(method, path, ts, nonce, payload, sk)
+    headers = {
+        "X-DF-Access-Key": ak,
+        "X-DF-Timestamp": ts,
+        "X-DF-Nonce": nonce,
+        "X-DF-Signature": signature,
+        "Accept": "application/json",
+    }
+    if method.upper() == "POST":
+        headers["Content-Type"] = "application/json; charset=utf-8"
+    url = endpoint.rstrip("/") + path
+    if query:
+        filtered = {k: v for k, v in query.items() if v}
+        if filtered:
+            url += "?" + urllib.parse.urlencode(filtered)
+    meta = {"method": method.upper(), "url": url, "path": path, "headers": dict(headers)}
+    if dry_run:
+        return 0, {"dryRun": True, **meta, "body": payload.decode("utf-8") if payload else ""}, headers
+
+    ctx = ssl._create_unverified_context() if insecure else None
+    req = urllib.request.Request(url, data=payload if method.upper() == "POST" else None, headers=headers, method=method.upper())
+    try:
+        with urllib.request.urlopen(req, timeout=timeout, context=ctx) as resp:
+            raw = resp.read()
+            parsed = json.loads(raw.decode("utf-8")) if raw else {}
+            return resp.status, parsed, headers
+    except urllib.error.HTTPError as ex:
+        raw = ex.read() if ex.fp else b""
+        try:
+            parsed = json.loads(raw.decode("utf-8")) if raw else {"message": str(ex)}
+        except json.JSONDecodeError:
+            parsed = {"message": raw.decode("utf-8", errors="replace")}
+        raise ApiError(ex.code, parsed) from ex

tinet_data_foundry_cli-0.1.0.dist-info/METADATA

@@ -0,0 +1,64 @@
+Metadata-Version: 2.4
+Name: tinet-data-foundry-cli
+Version: 0.1.0
+Summary: Tinet Data Foundry 取数 CLI（catalog / query / schema）
+Author: Tinet
+License: MIT
+Project-URL: Homepage, https://data-foundry.tinetcloud.com
+Project-URL: Documentation, https://data-foundry.tinetcloud.com
+Keywords: data-foundry,tinet,cli
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Environment :: Console
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+
+# tinet-data-foundry-cli
+
+Tinet Data Foundry 取数 CLI。包名 `tinet-data-foundry-cli`，命令 **`df`**（同包 **`tdf`**）。
+
+鉴权只用环境变量，**不要**把 AccessKey / SecretKey 写入本地文件。
+
+```bash
+pip install -U tinet-data-foundry-cli
+
+export DF_ENDPOINT="https://data-foundry-data-api.tinetcloud.com"
+export DF_AK="你的_AccessKey"
+export DF_SK="你的_SecretKey"
+
+df catalog
+df schema --dataset "DATASET_ID"
+df query \
+  --dataset "DATASET_ID" \
+  --fields "id" \
+  --filter '{"op":"eq","field":"id","value":"sample"}' \
+  --sort '[{"field":"id","direction":"DESC"}]' \
+  --page-no 1 \
+  --page-size 10
+```
+
+需要 Python **≥ 3.10**。
+
+| 命令 | 接口 |
+| --- | --- |
+| `df catalog` | `GET /data-api/v1/catalog` |
+| `df query` | `POST /data-api/v1/datasets/{id}/query` |
+| `df schema` | `GET /data-api/v1/datasets/{id}/schema` |
+
+签名：`canonical = METHOD + "\n" + PATH + "\n" + TIMESTAMP + "\n" + NONCE + "\n" + hex(SHA256(body))`，`X-DF-Signature = Base64(HMAC-SHA256(canonical, SK))`。GET 的 body 为空字节。PATH 不含域名、不含 query。
+
+## 发布到 PyPI
+
+版本号改 `pyproject.toml` 与 `src/tinet_data_foundry_cli/__init__.py`。
+
+```bash
+python -m pip install -U build twine
+python -m build
+python -m twine check dist/*
+python -m twine upload --repository testpypi dist/*
+python -m twine upload dist/*
+```
+
+已发布版本号不可覆盖。

tinet_data_foundry_cli-0.1.0.dist-info/RECORD

@@ -0,0 +1,10 @@
+tinet_data_foundry_cli/__init__.py,sha256=kUR5RAFc7HCeiqdlX36dZOHkUI5wI6V_43RpEcD8b-0,22
+tinet_data_foundry_cli/__main__.py,sha256=dK9X-vozIeT_0Vr07EYfal6Jr03NykcnXg5WNo0BUpM,101
+tinet_data_foundry_cli/auth.py,sha256=skVPoLI4wBD5ijD0xctQdmDQxgNJjadNtjp21FuTKNY,1447
+tinet_data_foundry_cli/cli.py,sha256=6fQHPGgIRgxLx52gNwmLspTZtOnzYOKEh9KQFfvqbCw,6642
+tinet_data_foundry_cli/client.py,sha256=ayiHM4u395mbihkqO8ZQI2acZH1YloIYXnoup2pJ-7s,2331
+tinet_data_foundry_cli-0.1.0.dist-info/METADATA,sha256=yH1Vm25z45z9QSUVoN8CGfC2sjrO-B1ItKz2dzoTnek,2089
+tinet_data_foundry_cli-0.1.0.dist-info/WHEEL,sha256=YVMoNqKzERt-wjUZwJ33xBGAwnFl-4cqbYkTtWa4itE,91
+tinet_data_foundry_cli-0.1.0.dist-info/entry_points.txt,sha256=TIpWTYGNUHt0QyPYaEj7um7QVe75RfzeObXkccwSM_s,93
+tinet_data_foundry_cli-0.1.0.dist-info/top_level.txt,sha256=GF3FZUCIXd8NjAICrx5DvsvTSkMDW4SLrhiLqd66Ezg,23
+tinet_data_foundry_cli-0.1.0.dist-info/RECORD,,

tinet_data_foundry_cli-0.1.0.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (84.0.0)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

tinet_data_foundry_cli-0.1.0.dist-info/entry_points.txt

@@ -0,0 +1,3 @@
+[console_scripts]
+df = tinet_data_foundry_cli.cli:main
+tdf = tinet_data_foundry_cli.cli:main

tinet_data_foundry_cli-0.1.0.dist-info/top_level.txt

@@ -0,0 +1 @@
+tinet_data_foundry_cli