Files
office_translator/services/providers/deepseek_provider.py
sepehr e6e1678b1d
Some checks failed
Deploy to Homelab / Deploy Wordly to 192.168.1.151 (push) Has been cancelled
Deploy to Homelab / Deploy Monitoring (if configured) (push) Has been cancelled
feat: add DeepSeek and Minimax (m2.7) translation providers
New providers:
- DeepSeek: direct API with deepseek-chat model, very cost-effective
- Minimax: MiniMax-M1 model via OpenAI-compatible API, supports m2.7

Changes:
- Full provider implementations with retry, health check, batch support
- Provider config with env vars (DEEPSEEK_*, MINIMAX_*)
- Auto-registration in provider registry
- Updated fallback chain to include new providers
- Updated setup-env.sh wizard with options 6 (deepseek) and 7 (minimax)
- Updated manage-keys.sh with new menu entries and provider switching
- Updated docker-compose.yml with new env vars

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-10 12:30:36 +02:00

245 lines
9.7 KiB
Python

"""
DeepSeek Provider - Cloud LLM translation via DeepSeek API.
DeepSeek uses an OpenAI-compatible Chat Completions API.
"""
import threading
import time
from datetime import datetime, timezone
from typing import Any, Dict, List, Optional
import requests
from requests.exceptions import Timeout, ConnectionError as RequestsConnectionError
from core.logging import get_logger
from .base import TranslationProvider
from .schemas import ProviderHealthStatus, TranslationRequest, TranslationResponse
logger = get_logger(__name__)
DEEPSEEK_RATE_LIMITED = "DEEPSEEK_RATE_LIMITED"
DEEPSEEK_INVALID_KEY = "DEEPSEEK_INVALID_KEY"
DEEPSEEK_TIMEOUT = "DEEPSEEK_TIMEOUT"
DEEPSEEK_SERVICE_ERROR = "DEEPSEEK_SERVICE_ERROR"
_RETRYABLE_ERRORS = {DEEPSEEK_RATE_LIMITED, DEEPSEEK_TIMEOUT, DEEPSEEK_SERVICE_ERROR}
DEFAULT_TRANSLATION_PROMPT = """You are a professional translator. Translate the following text from {source_lang} to {target_lang}.
Rules:
- Translate ONLY the text, do not add explanations or notes
- Preserve the original formatting, line breaks, and structure
- Maintain the original tone and style
- For technical terms, use the standard translation in the target language
- If the text contains proper nouns or brand names, keep them unchanged unless there's a well-known translation"""
def _get_language_name(code: str) -> str:
language_names = {
"en": "English", "fr": "French", "es": "Spanish", "de": "German",
"it": "Italian", "pt": "Portuguese", "nl": "Dutch", "ru": "Russian",
"zh": "Chinese", "ja": "Japanese", "ko": "Korean", "ar": "Arabic",
"hi": "Hindi", "tr": "Turkish", "pl": "Polish", "vi": "Vietnamese",
"th": "Thai", "uk": "Ukrainian", "cs": "Czech", "sv": "Swedish",
"ro": "Romanian", "hu": "Hungarian", "el": "Greek", "he": "Hebrew",
}
return language_names.get(code.split("-")[0].lower(), code)
class DeepSeekProviderError(Exception):
def __init__(self, code: str, message: str, details: Optional[Dict[str, Any]] = None):
self.code = code
self.message = message
self.details = details or {}
super().__init__(message)
class DeepSeekTranslationProvider(TranslationProvider):
"""
DeepSeek translation provider using OpenAI-compatible API.
Supports DeepSeek Chat and DeepSeek Reasoner models.
"""
def __init__(
self,
api_key: str,
model: str = "deepseek-chat",
timeout: int = 60,
max_retries: int = 3,
retry_delay: float = 1.0,
base_url: str = "https://api.deepseek.com/v1",
):
if not api_key or not api_key.strip():
raise ValueError("DeepSeek API key cannot be empty")
self._api_key = api_key
self._model = model
self._base_url = base_url.rstrip("/")
self._provider_name = "deepseek"
self._timeout = timeout
self._max_retries = max_retries
self._retry_delay = retry_delay
self._health_cache: Dict[str, Any] = {}
self._health_cache_ttl = 60
self._health_cache_lock = threading.Lock()
def _make_api_request(self, text: str, system_prompt: str) -> tuple:
if not text or not text.strip():
return text, {}
url = f"{self._base_url}/chat/completions"
headers = {
"Authorization": f"Bearer {self._api_key}",
"Content-Type": "application/json",
}
payload = {
"model": self._model,
"messages": [
{"role": "system", "content": system_prompt},
{"role": "user", "content": text},
],
"temperature": 0.3,
"max_tokens": 4096,
}
try:
response = requests.post(url, headers=headers, json=payload, timeout=self._timeout)
if response.status_code == 401:
raise DeepSeekProviderError(DEEPSEEK_INVALID_KEY, "Cle API DeepSeek invalide.")
if response.status_code == 429:
raise DeepSeekProviderError(DEEPSEEK_RATE_LIMITED, "Limite de requetes DeepSeek atteinte.")
if response.status_code >= 500:
raise DeepSeekProviderError(DEEPSEEK_SERVICE_ERROR, "Service DeepSeek temporairement indisponible.")
if response.status_code != 200:
raise DeepSeekProviderError(DEEPSEEK_SERVICE_ERROR, f"Erreur DeepSeek: {response.text[:200]}")
data = response.json()
choices = data.get("choices", [])
if not choices:
raise DeepSeekProviderError(DEEPSEEK_SERVICE_ERROR, "Reponse DeepSeek vide")
content = choices[0].get("message", {}).get("content", "")
if not content:
raise DeepSeekProviderError(DEEPSEEK_SERVICE_ERROR, "Reponse DeepSeek vide")
return content.strip(), data.get("usage", {})
except Timeout:
raise DeepSeekProviderError(DEEPSEEK_TIMEOUT, "Delai d'attente DeepSeek depasse.")
except RequestsConnectionError:
raise DeepSeekProviderError(DEEPSEEK_SERVICE_ERROR, "Service DeepSeek indisponible.")
except DeepSeekProviderError:
raise
except Exception as e:
raise DeepSeekProviderError(DEEPSEEK_SERVICE_ERROR, f"Erreur DeepSeek: {str(e)[:100]}")
def get_name(self) -> str:
return self._provider_name
def is_available(self) -> bool:
try:
headers = {"Authorization": f"Bearer {self._api_key}"}
response = requests.get(f"{self._base_url}/models", headers=headers, timeout=5)
return response.status_code == 200
except Exception:
return False
def translate_text(self, request: TranslationRequest) -> TranslationResponse:
text = request.text
target_language = request.target_language
source_language = request.source_language or "auto"
if not text or not text.strip():
return TranslationResponse(translated_text=text, provider_name=self._provider_name, from_cache=False)
source_lang_name = _get_language_name(source_language)
target_lang_name = _get_language_name(target_language)
custom_prompt = request.metadata.get("custom_prompt") if request.metadata else None
system_prompt = custom_prompt or DEFAULT_TRANSLATION_PROMPT.format(
source_lang=source_lang_name, target_lang=target_lang_name
)
last_error = None
for attempt in range(self._max_retries + 1):
try:
start_time = time.time()
result, usage = self._make_api_request(text, system_prompt)
latency = time.time() - start_time
logger.info("deepseek_translation_success",
chars=len(text), source_lang=source_language, target_lang=target_language,
model=self._model, latency_ms=round(latency * 1000, 2), retries=attempt)
return TranslationResponse(
translated_text=result, provider_name=self._provider_name,
from_cache=False, source_language=source_language)
except DeepSeekProviderError as e:
last_error = e
if e.code not in _RETRYABLE_ERRORS or attempt >= self._max_retries:
break
delay = self._retry_delay * (2 ** attempt)
time.sleep(delay)
return TranslationResponse(
translated_text=text, provider_name=self._provider_name, from_cache=False,
error=last_error.message if last_error else "Unknown error",
error_code=last_error.code if last_error else DEEPSEEK_SERVICE_ERROR)
def translate_batch(self, requests: List[TranslationRequest]) -> List[TranslationResponse]:
return [self.translate_text(req) for req in requests]
def health_check(self) -> ProviderHealthStatus:
start_time = time.time()
try:
headers = {"Authorization": f"Bearer {self._api_key}"}
response = requests.get(f"{self._base_url}/models", headers=headers, timeout=5)
latency_ms = (time.time() - start_time) * 1000
return ProviderHealthStatus(
name=self._provider_name, available=response.status_code == 200,
latency_ms=round(latency_ms, 2), last_check=datetime.now(timezone.utc).isoformat(),
model=self._model)
except Exception as e:
return ProviderHealthStatus(
name=self._provider_name, available=False,
latency_ms=round((time.time() - start_time) * 1000, 2),
error=str(e)[:100], last_check=datetime.now(timezone.utc).isoformat(),
model=self._model)
_provider_instance: Optional[DeepSeekTranslationProvider] = None
_provider_lock = threading.Lock()
def get_deepseek_provider() -> DeepSeekTranslationProvider:
global _provider_instance
if _provider_instance is None:
with _provider_lock:
if _provider_instance is None:
from .config import ProvidersConfig
_provider_instance = DeepSeekTranslationProvider(
api_key=ProvidersConfig.DEEPSEEK_API_KEY,
model=ProvidersConfig.DEEPSEEK_MODEL,
timeout=ProvidersConfig.DEEPSEEK_TIMEOUT,
max_retries=ProvidersConfig.DEEPSEEK_MAX_RETRIES,
retry_delay=ProvidersConfig.DEEPSEEK_RETRY_DELAY,
base_url=ProvidersConfig.DEEPSEEK_BASE_URL,
)
return _provider_instance
def register_deepseek_provider():
from .registry import registry
provider = get_deepseek_provider()
registry.register("deepseek", provider)
return provider
def reset_deepseek_provider():
global _provider_instance
with _provider_lock:
_provider_instance = None