harry
commited on
Commit
·
d01fa51
1
Parent(s):
5d13bbd
add tencent translator
Browse files- deep_translator/__init__.py +2 -0
- deep_translator/constants.py +23 -0
- deep_translator/exceptions.py +13 -0
- deep_translator/tencent.py +140 -0
- tests/test_tencent.py +37 -0
deep_translator/__init__.py
CHANGED
|
@@ -13,6 +13,7 @@ from deep_translator.mymemory import MyMemoryTranslator
|
|
| 13 |
from deep_translator.papago import PapagoTranslator
|
| 14 |
from deep_translator.pons import PonsTranslator
|
| 15 |
from deep_translator.qcri import QcriTranslator
|
|
|
|
| 16 |
from deep_translator.yandex import YandexTranslator
|
| 17 |
|
| 18 |
__author__ = """Nidhal Baccouri"""
|
|
@@ -31,6 +32,7 @@ __all__ = [
|
|
| 31 |
"LibreTranslator",
|
| 32 |
"PapagoTranslator",
|
| 33 |
"ChatGptTranslator",
|
|
|
|
| 34 |
"single_detection",
|
| 35 |
"batch_detection",
|
| 36 |
]
|
|
|
|
| 13 |
from deep_translator.papago import PapagoTranslator
|
| 14 |
from deep_translator.pons import PonsTranslator
|
| 15 |
from deep_translator.qcri import QcriTranslator
|
| 16 |
+
from deep_translator.tencent import TencentTranslator
|
| 17 |
from deep_translator.yandex import YandexTranslator
|
| 18 |
|
| 19 |
__author__ = """Nidhal Baccouri"""
|
|
|
|
| 32 |
"LibreTranslator",
|
| 33 |
"PapagoTranslator",
|
| 34 |
"ChatGptTranslator",
|
| 35 |
+
"TencentTranslator",
|
| 36 |
"single_detection",
|
| 37 |
"batch_detection",
|
| 38 |
]
|
deep_translator/constants.py
CHANGED
|
@@ -7,6 +7,8 @@ LIBRE_ENV_VAR = "LIBRE_API_KEY"
|
|
| 7 |
MSFT_ENV_VAR = "MICROSOFT_API_KEY"
|
| 8 |
QCRI_ENV_VAR = "QCRI_API_KEY"
|
| 9 |
YANDEX_ENV_VAR = "YANDEX_API_KEY"
|
|
|
|
|
|
|
| 10 |
|
| 11 |
|
| 12 |
BASE_URLS = {
|
|
@@ -23,6 +25,7 @@ BASE_URLS = {
|
|
| 23 |
"PAPAGO_API": "https://openapi.naver.com/v1/papago/n2mt",
|
| 24 |
"LIBRE": "https://libretranslate.com/",
|
| 25 |
"LIBRE_FREE": "https://libretranslate.de/",
|
|
|
|
| 26 |
}
|
| 27 |
|
| 28 |
GOOGLE_LANGUAGES_TO_CODES = {
|
|
@@ -280,3 +283,23 @@ LIBRE_LANGUAGES_TO_CODES = {
|
|
| 280 |
"Turkish": "tr",
|
| 281 |
"Vietnamese": "vi",
|
| 282 |
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
MSFT_ENV_VAR = "MICROSOFT_API_KEY"
|
| 8 |
QCRI_ENV_VAR = "QCRI_API_KEY"
|
| 9 |
YANDEX_ENV_VAR = "YANDEX_API_KEY"
|
| 10 |
+
TENCENT_ID_VAR = "TENCENT_ID"
|
| 11 |
+
TENCENT_KEY_VAR = "TENCENT_KEY"
|
| 12 |
|
| 13 |
|
| 14 |
BASE_URLS = {
|
|
|
|
| 25 |
"PAPAGO_API": "https://openapi.naver.com/v1/papago/n2mt",
|
| 26 |
"LIBRE": "https://libretranslate.com/",
|
| 27 |
"LIBRE_FREE": "https://libretranslate.de/",
|
| 28 |
+
"TENENT": "https://tmt.tencentcloudapi.com",
|
| 29 |
}
|
| 30 |
|
| 31 |
GOOGLE_LANGUAGES_TO_CODES = {
|
|
|
|
| 283 |
"Turkish": "tr",
|
| 284 |
"Vietnamese": "vi",
|
| 285 |
}
|
| 286 |
+
|
| 287 |
+
TENCENT_LANGUAGE_TO_CODE = {
|
| 288 |
+
"arabic": "ar",
|
| 289 |
+
"chinese (simplified)": "zh",
|
| 290 |
+
"chinese (traditional)": "zh-TW",
|
| 291 |
+
"english": "en",
|
| 292 |
+
"french": "fr",
|
| 293 |
+
"german": "de",
|
| 294 |
+
"hindi": "hi",
|
| 295 |
+
"indonesian": "id",
|
| 296 |
+
"japanese": "ja",
|
| 297 |
+
"korean": "ko",
|
| 298 |
+
"malay": "ms",
|
| 299 |
+
"portuguese": "pt",
|
| 300 |
+
"russian": "ru",
|
| 301 |
+
"spanish": "es",
|
| 302 |
+
"thai": "th",
|
| 303 |
+
"turkish": "tr",
|
| 304 |
+
"vietnamese": "vi",
|
| 305 |
+
}
|
deep_translator/exceptions.py
CHANGED
|
@@ -180,3 +180,16 @@ class AuthorizationException(Exception):
|
|
| 180 |
def __init__(self, api_key, *args):
|
| 181 |
msg = "Unauthorized access with the api key " + api_key
|
| 182 |
super().__init__(msg, *args)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 180 |
def __init__(self, api_key, *args):
|
| 181 |
msg = "Unauthorized access with the api key " + api_key
|
| 182 |
super().__init__(msg, *args)
|
| 183 |
+
|
| 184 |
+
|
| 185 |
+
class TencentAPIerror(Exception):
|
| 186 |
+
"""
|
| 187 |
+
exception thrown if Tencent API returns one of its errors
|
| 188 |
+
"""
|
| 189 |
+
|
| 190 |
+
def __init__(self, api_message):
|
| 191 |
+
self.api_message = str(api_message)
|
| 192 |
+
self.message = "Tencent API returned the following error"
|
| 193 |
+
|
| 194 |
+
def __str__(self):
|
| 195 |
+
return "{}: {}".format(self.message, self.api_message)
|
deep_translator/tencent.py
ADDED
|
@@ -0,0 +1,140 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"""
|
| 2 |
+
tencent translator API
|
| 3 |
+
"""
|
| 4 |
+
|
| 5 |
+
__copyright__ = "Copyright (C) 2020 Nidhal Baccouri"
|
| 6 |
+
|
| 7 |
+
import base64
|
| 8 |
+
import hashlib
|
| 9 |
+
import hmac
|
| 10 |
+
import os
|
| 11 |
+
import time
|
| 12 |
+
from typing import List, Optional
|
| 13 |
+
|
| 14 |
+
import requests
|
| 15 |
+
|
| 16 |
+
from deep_translator.base import BaseTranslator
|
| 17 |
+
from deep_translator.constants import (
|
| 18 |
+
BASE_URLS,
|
| 19 |
+
TENCENT_ID_VAR,
|
| 20 |
+
TENCENT_KEY_VAR,
|
| 21 |
+
TENCENT_LANGUAGE_TO_CODE,
|
| 22 |
+
)
|
| 23 |
+
from deep_translator.exceptions import (
|
| 24 |
+
ApiKeyException,
|
| 25 |
+
AuthorizationException,
|
| 26 |
+
ServerException,
|
| 27 |
+
TencentAPIerror,
|
| 28 |
+
TranslationNotFound,
|
| 29 |
+
)
|
| 30 |
+
from deep_translator.validate import is_empty, is_input_valid
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
class TencentTranslator(BaseTranslator):
|
| 34 |
+
"""
|
| 35 |
+
class that wraps functions, which use the TentCentTranslator translator
|
| 36 |
+
under the hood to translate word(s)
|
| 37 |
+
"""
|
| 38 |
+
|
| 39 |
+
def __init__(
|
| 40 |
+
self,
|
| 41 |
+
source: str = "en",
|
| 42 |
+
target: str = "zh",
|
| 43 |
+
secret_id: Optional[str] = os.getenv(TENCENT_ID_VAR, None),
|
| 44 |
+
secret_key: Optional[str] = os.getenv(TENCENT_KEY_VAR, None),
|
| 45 |
+
**kwargs
|
| 46 |
+
):
|
| 47 |
+
"""
|
| 48 |
+
@param secret_id: your tencent cloud api secret id.
|
| 49 |
+
Get one here: https://console.cloud.tencent.com/capi
|
| 50 |
+
@param secret_key: your tencent cloud api secret key.
|
| 51 |
+
@param source: source language
|
| 52 |
+
@param target: target language
|
| 53 |
+
"""
|
| 54 |
+
if not secret_id:
|
| 55 |
+
raise ApiKeyException(env_var=TENCENT_ID_VAR)
|
| 56 |
+
|
| 57 |
+
if not secret_key:
|
| 58 |
+
raise ApiKeyException(env_var=TENCENT_KEY_VAR)
|
| 59 |
+
|
| 60 |
+
self.secret_id = secret_id
|
| 61 |
+
self.secret_key = secret_key
|
| 62 |
+
url = BASE_URLS.get("TENENT")
|
| 63 |
+
super().__init__(
|
| 64 |
+
base_url=url,
|
| 65 |
+
source=source,
|
| 66 |
+
target=target,
|
| 67 |
+
languages=TENCENT_LANGUAGE_TO_CODE,
|
| 68 |
+
**kwargs
|
| 69 |
+
)
|
| 70 |
+
|
| 71 |
+
def translate(self, text: str, **kwargs) -> str:
|
| 72 |
+
"""
|
| 73 |
+
@param text: text to translate
|
| 74 |
+
@return: translated text
|
| 75 |
+
"""
|
| 76 |
+
if is_input_valid(text):
|
| 77 |
+
if self._same_source_target() or is_empty(text):
|
| 78 |
+
return text
|
| 79 |
+
|
| 80 |
+
# Create the request parameters.
|
| 81 |
+
translate_endpoint = "tmt.tencentcloudapi.com"
|
| 82 |
+
params = {
|
| 83 |
+
"Action": "TextTranslate",
|
| 84 |
+
"Nonce": 11886,
|
| 85 |
+
"ProjectId": 0,
|
| 86 |
+
"Region": "ap-guangzhou",
|
| 87 |
+
"SecretId": self.secret_id,
|
| 88 |
+
"Source": self.source,
|
| 89 |
+
"SourceText": text,
|
| 90 |
+
"Target": self.target,
|
| 91 |
+
"Timestamp": int(time.time()), # int(time.time())
|
| 92 |
+
"Version": "2018-03-21",
|
| 93 |
+
}
|
| 94 |
+
s = "GET" + translate_endpoint + "/?"
|
| 95 |
+
query_str = "&".join(
|
| 96 |
+
"%s=%s" % (k, params[k]) for k in sorted(params)
|
| 97 |
+
)
|
| 98 |
+
hmac_str = hmac.new(
|
| 99 |
+
self.secret_key.encode("utf8"),
|
| 100 |
+
(s + query_str).encode("utf8"),
|
| 101 |
+
hashlib.sha1,
|
| 102 |
+
).digest()
|
| 103 |
+
params["Signature"] = base64.b64encode(hmac_str)
|
| 104 |
+
|
| 105 |
+
# Do the request and check the connection.
|
| 106 |
+
try:
|
| 107 |
+
response = requests.get(self._base_url, params=params)
|
| 108 |
+
except ConnectionError:
|
| 109 |
+
raise ServerException(503)
|
| 110 |
+
# If the answer is not success, raise server exception.
|
| 111 |
+
if response.status_code == 403:
|
| 112 |
+
raise AuthorizationException(self.secret_id)
|
| 113 |
+
elif response.status_code != 200:
|
| 114 |
+
raise ServerException(response.status_code)
|
| 115 |
+
# Get the response and check is not empty.
|
| 116 |
+
res = response.json()
|
| 117 |
+
if not res:
|
| 118 |
+
raise TranslationNotFound(text)
|
| 119 |
+
# Process and return the response.
|
| 120 |
+
if "Error" in res["Response"]:
|
| 121 |
+
raise TencentAPIerror(res["Response"]["Error"]["Code"])
|
| 122 |
+
return res["Response"]["TargetText"]
|
| 123 |
+
|
| 124 |
+
def translate_file(self, path: str, **kwargs) -> str:
|
| 125 |
+
return self._translate_file(path, **kwargs)
|
| 126 |
+
|
| 127 |
+
def translate_batch(self, batch: List[str], **kwargs) -> List[str]:
|
| 128 |
+
"""
|
| 129 |
+
@param batch: list of texts to translate
|
| 130 |
+
@return: list of translations
|
| 131 |
+
"""
|
| 132 |
+
return self._translate_batch(batch, **kwargs)
|
| 133 |
+
|
| 134 |
+
|
| 135 |
+
if __name__ == "__main__":
|
| 136 |
+
d = TencentTranslator(
|
| 137 |
+
target="zh", secret_id="some-id", secret_key="some-key"
|
| 138 |
+
)
|
| 139 |
+
t = d.translate("Ich habe keine ahnung")
|
| 140 |
+
print("text: ", t)
|
tests/test_tencent.py
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from unittest.mock import Mock, patch
|
| 2 |
+
|
| 3 |
+
import pytest
|
| 4 |
+
|
| 5 |
+
from deep_translator import TencentTranslator
|
| 6 |
+
from deep_translator.exceptions import AuthorizationException
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
@patch("deep_translator.tencent.requests")
|
| 10 |
+
def test_simple_translation(mock_requests):
|
| 11 |
+
translator = TencentTranslator(
|
| 12 |
+
secret_id="this-is-an-valid-api-id",
|
| 13 |
+
source="en",
|
| 14 |
+
target="zh",
|
| 15 |
+
secret_key="this-is-an-valid-api-key",
|
| 16 |
+
)
|
| 17 |
+
# Set the request response mock.
|
| 18 |
+
mock_response = Mock()
|
| 19 |
+
mock_response.status_code = 200
|
| 20 |
+
mock_response.json.return_value = {"translations": [{"text": "hola"}]}
|
| 21 |
+
mock_requests.get.return_value = mock_response
|
| 22 |
+
translation = translator.translate("hello")
|
| 23 |
+
assert translation == "你好"
|
| 24 |
+
|
| 25 |
+
|
| 26 |
+
@patch("deep_translator.tencent.requests.get")
|
| 27 |
+
def test_wrong_api_key(mock_requests):
|
| 28 |
+
translator = TencentTranslator(
|
| 29 |
+
secret_id="this-is-a-wrong-api-id",
|
| 30 |
+
source="en",
|
| 31 |
+
target="zh",
|
| 32 |
+
secret_key="this-is-a-wrong-api-key",
|
| 33 |
+
)
|
| 34 |
+
# Set the response status_code only.
|
| 35 |
+
mock_requests.return_value = Mock(status_code=403)
|
| 36 |
+
with pytest.raises(AuthorizationException):
|
| 37 |
+
translator.translate("Hello")
|