translator.py

import hashlib, hmac, json, os, random, time
from datetime import datetime
from typing import List, Sequence, Optional
import requests
import threading

APP_TENCENT_SECRET_ID  = os.environ.get("TENCENT_SECRET_ID")
APP_TENCENT_SECRET_KEY = os.environ.get("TENCENT_SECRET_KEY")
APP_BAIDU_CREDENTIALS  = json.loads(os.environ.get("BAIDU_CREDENTIALS_JSON", "[]"))
APP_ACCESS_KEY         = os.environ.get("TRANSLATOR_ACCESS_KEY")

TENCENT_TRANSLATE_URL = os.environ.get("TENCENT_TRANSLATE_URL", "https://tmt.tencentcloudapi.com")
BAIDU_TRANSLATE_URL = os.environ.get("BAIDU_TRANSLATE_URL", "https://fanyi-api.baidu.com/api/trans/vip/translate")

_user_keys = threading.local()

def set_user_provided_keys(tencent_id, tencent_key, baidu_json_str):
    _user_keys.tencent_id = tencent_id if tencent_id else None
    _user_keys.tencent_key = tencent_key if tencent_key else None
    try:
        _user_keys.baidu_credentials = json.loads(baidu_json_str) if baidu_json_str else []
    except json.JSONDecodeError:
        _user_keys.baidu_credentials = []

def clear_user_provided_keys():
    if hasattr(_user_keys, 'tencent_id'): del _user_keys.tencent_id
    if hasattr(_user_keys, 'tencent_key'): del _user_keys.tencent_key
    if hasattr(_user_keys, 'baidu_credentials'): del _user_keys.baidu_credentials

def _get_tencent_credentials():
    user_id = getattr(_user_keys, 'tencent_id', None)
    user_key = getattr(_user_keys, 'tencent_key', None)
    if user_id and user_key:
        # print("DEBUG: Using user-provided Tencent keys.")
        return user_id, user_key
    # print("DEBUG: Using app-level Tencent keys.")
    return APP_TENCENT_SECRET_ID, APP_TENCENT_SECRET_KEY

def _get_baidu_credentials():
    user_creds = getattr(_user_keys, 'baidu_credentials', [])
    if user_creds:
        # print("DEBUG: Using user-provided Baidu credentials.")
        return user_creds
    # print("DEBUG: Using app-level Baidu credentials.")
    return APP_BAIDU_CREDENTIALS
_baidu_idx = 0
def _next_baidu_cred():
    global _baidu_idx
    creds_list = _get_baidu_credentials()
    if not creds_list:
        return None
    cred = creds_list[_baidu_idx]
    _baidu_idx = (_baidu_idx + 1) % len(creds_list)
    return cred

def _sign(key: bytes, msg: str) -> bytes:
    return hmac.new(key, msg.encode("utf-8"), hashlib.sha256).digest()

def _tc3_signature(secret_key: str, date: str, service: str, string_to_sign: str) -> str:
    secret_date     = _sign(("TC3" + secret_key).encode(), date)
    secret_service  = _sign(secret_date, service)
    secret_signing  = _sign(secret_service, "tc3_request")
    return hmac.new(secret_signing, string_to_sign.encode("utf-8"), hashlib.sha256).hexdigest()

def _translate_with_tencent(texts: Sequence[str], src="auto", tgt="zh") -> Optional[List[str]]:
    secret_id, secret_key = _get_tencent_credentials()
    if not (secret_id and secret_key):
        return None

    service, host, action, version, region = "tmt", "tmt.tencentcloudapi.com", "TextTranslate", "2018-03-21", "ap-beijing"
    ts = int(time.time())
    date = datetime.utcfromtimestamp(ts).strftime("%Y-%m-%d")
    algorithm = "TC3-HMAC-SHA256"

    payload = {"SourceText": "\n".join(texts), "Source": src, "Target": tgt, "ProjectId": 0}
    payload_str = json.dumps(payload, ensure_ascii=False)
    hashed_request_payload = hashlib.sha256(payload_str.encode()).hexdigest()

    canonical_headers = f"content-type:application/json; charset=utf-8\nhost:{host}\nx-tc-action:{action.lower()}\n"
    signed_headers = "content-type;host;x-tc-action"
    canonical_request = f"POST\n/\n\n{canonical_headers}\n{signed_headers}\n{hashed_request_payload}"
    
    credential_scope = f"{date}/{service}/tc3_request"
    string_to_sign = f"{algorithm}\n{ts}\n{credential_scope}\n{hashlib.sha256(canonical_request.encode()).hexdigest()}"
    
    signature = _tc3_signature(secret_key, date, service, string_to_sign)
    authorization = f"{algorithm} Credential={secret_id}/{credential_scope}, SignedHeaders={signed_headers}, Signature={signature}"

    headers = {
        "Authorization": authorization, "Content-Type": "application/json; charset=utf-8",
        "Host": host, "X-TC-Action": action, "X-TC-Timestamp": str(ts),
        "X-TC-Version": version, "X-TC-Region": region,
    }

    try:
        resp = requests.post(TENCENT_TRANSLATE_URL, headers=headers, data=payload_str, timeout=8)
        resp.raise_for_status()
        data = resp.json()
        if "Error" in data.get("Response", {}):
            print(f"[translator] Tencent API returned error: {data['Response']['Error']['Message']}")
            return None
        return data["Response"]["TargetText"].split("\n")
    except Exception as e:
        print(f"[translator] Tencent API request failed → {e}")
        return None

def _translate_with_baidu(texts: Sequence[str], src="auto", tgt="zh") -> Optional[List[str]]:
    creds = _next_baidu_cred()
    if not creds:
        return None
    app_id, secret_key = creds["app_id"], creds["secret_key"]
    salt = random.randint(32768, 65536)
    query = "\n".join(texts)
    sign = hashlib.md5((app_id + query + str(salt) + secret_key).encode()).hexdigest()
    params = {"q": query, "from": src, "to": tgt, "appid": app_id, "salt": salt, "sign": sign}
    
    try:
        resp = requests.get(BAIDU_TRANSLATE_URL, params=params, timeout=8)
        resp.raise_for_status()
        data = resp.json()
        if "error_code" in data:
            print(f"[translator] Baidu API returned error: {data['error_msg']} (code: {data['error_code']})")
            return None
        return [item["dst"] for item in data["trans_result"]]
    except Exception as e:
        print(f"[translator] Baidu API request failed → {e}")
        return None

def translate_texts(texts: Sequence[str],
                    src_lang: str = "auto",
                    tgt_lang: str = "zh",
                    access_key: str = None) -> List[str]:
    if not texts:
        return []

    has_user_tencent = getattr(_user_keys, 'tencent_id', None) and getattr(_user_keys, 'tencent_key', None)
    has_user_baidu = bool(getattr(_user_keys, 'baidu_credentials', []))

    can_use_app_keys = False
    if not (has_user_tencent or has_user_baidu):
        if not APP_ACCESS_KEY:
            can_use_app_keys = True
        elif access_key and access_key == APP_ACCESS_KEY:
            can_use_app_keys = True
    if has_user_tencent or has_user_baidu or can_use_app_keys:
        out = _translate_with_tencent(texts, src_lang, tgt_lang)
        if out is None:
            out = _translate_with_baidu(texts, src_lang, tgt_lang)
        if out is not None:
            return out
    return list(texts)