fix: Change user-agent string to be opt-in

JanEbbing · JanEbbing · commit 24ab267aa7c5 · 2023-02-15T16:48:25.000Z
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -8,6 +8,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ## [Unreleased]
 ### Added
 * Add [example](examples/json) to translate JSON inputs. 
+* Added platform and python version information to the user-agent string that is sent with API calls, along with an opt-out.
+* Added method for applications that use this library to identify themselves in API requests they make.
 
 
 ## [1.13.0] - 2023-01-26
@@ -19,7 +21,6 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
   Note: older library versions also support the new languages, this update only adds new code constants.
 ### Changed
-* Added system, python and request library version information to the user-agent string that is sent with API calls.
 ### Deprecated
 ### Removed
 ### Fixed
diff --git a/README.md b/README.md
@@ -434,6 +434,20 @@ Note that glossaries work for all target regional-variants: a glossary for the
 target language English (`"EN"`) supports translations to both American English
 (`"EN-US"`) and British English (`"EN-GB"`).
 
+### Writing a Plugin
+
+If you use this library in an application, please identify the application with
+`deepl.Translator.set_app_info`, which needs the name and version of the app:
+
+```python
+translator = deepl.Translator(...).set_app_info("sample_python_plugin", "1.0.2")
+```
+
+This information is passed along when the library makes calls to the DeepL API.
+Both name and version are required. Please note that setting the `User-Agent` header
+via `deepl.http_client.user_agent` will override this setting, if you need to use this,
+please manually identify your Application in the `User-Agent` header.
+
 ### Exceptions
 
 All module functions may raise `deepl.DeepLException` or one of its subclasses.
@@ -481,6 +495,21 @@ The proxy argument is passed to the underlying `requests` session, see the
 [documentation for requests][requests-proxy-docs]; a dictionary of schemes to
 proxy URLs is also accepted.
 
+#### Anonymous platform information
+
+By default, we send some basic information about the platform the client library is running on with each request, see [here for an explanation](https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/User-Agent). This data is completely anonymous and only used to improve our product, not track any individual users. If you do not wish to send this data, you can opt-out when creating your `deepl.Translator` object by setting the `send_platform_info` flag like so:
+
+```python
+translator = deepl.Translator(..., send_platform_info=False)
+```
+
+You can also customize the `user_agent` by setting its value explicitly before constructing your `deepl.Translator` object.
+
+```python
+deepl.http_client.user_agent = 'my custom user agent'
+translator = deepl.Translator(os.environ["DEEPL_AUTH_KEY"])
+```
+
 ## Command Line Interface
 
 The library can be run on the command line supporting all API functions. Use the
diff --git a/deepl/__main__.py b/deepl/__main__.py
@@ -207,6 +207,14 @@ def get_parser(prog_name):
         help="print additional information, can be supplied multiple times "
         "for more verbose output",
     )
+    parser.add_argument(
+        "--no-platform-info",
+        default=False,
+        action="store_true",
+        dest="noplatforminfo",
+        help="if this flag is enabled, do not send additional information "
+        "about the platform with API requests.",
+    )
 
     parser.add_argument(
         "--auth-key",
@@ -535,6 +543,7 @@ def main(args=None, prog_name=None):
             server_url=server_url,
             proxy=proxy_url,
             skip_language_check=True,
+            send_platform_info=not args.noplatforminfo,
         )
 
         if args.command == "text":
@@ -549,7 +558,13 @@ def main(args=None, prog_name=None):
                 sys.exit(1)
 
         # Remove global args so they are not unrecognised in action functions
-        del args.verbose, args.server_url, args.auth_key, args.proxy_url
+        del (
+            args.verbose,
+            args.server_url,
+            args.auth_key,
+            args.proxy_url,
+            args.noplatforminfo,
+        )
         args = vars(args)
         # Call action function corresponding to command with remaining args
         command = args.pop("command")
diff --git a/deepl/http_client.py b/deepl/http_client.py
@@ -9,14 +9,12 @@
 import random
 import requests
 import time
+from functools import lru_cache
 from typing import Dict, Optional, Tuple, Union
 from .util import log_info
 
 
-user_agent = (
-    f"deepl-python/{version.VERSION} ({platform.platform()}) "
-    f"python/{platform.python_version()} requests/{requests.__version__}"
-)
+user_agent = None
 max_network_retries = 5
 min_connection_timeout = 10.0
 
@@ -62,7 +60,11 @@ def sleep_until_deadline(self):
 
 
 class HttpClient:
-    def __init__(self, proxy: Union[Dict, str, None] = None):
+    def __init__(
+        self,
+        proxy: Union[Dict, str, None] = None,
+        send_platform_info: bool = True,
+    ):
         self._session = requests.Session()
         if proxy:
             if isinstance(proxy, str):
@@ -73,8 +75,14 @@ def __init__(self, proxy: Union[Dict, str, None] = None):
                     "containing URL strings for the http and https keys."
                 )
             self._session.proxies.update(proxy)
-        self._session.headers = {"User-Agent": user_agent}
-        pass
+        self._send_platform_info = send_platform_info
+        self._app_info_name = None
+        self._app_info_version = None
+
+    def set_app_info(self, app_info_name: str, app_info_version: str):
+        self._app_info_name = app_info_name
+        self._app_info_version = app_info_version
+        return self
 
     def close(self):
         self._session.close()
@@ -94,7 +102,15 @@ def request_with_backoff(
         backoff = _BackoffTimer()
 
         try:
-            headers.setdefault("User-Agent", user_agent)
+            headers.setdefault(
+                "User-Agent",
+                _generate_user_agent(
+                    user_agent,
+                    self._send_platform_info,
+                    self._app_info_name,
+                    self._app_info_version,
+                ),
+            )
             request = requests.Request(
                 method, url, data=data, headers=headers, **kwargs
             ).prepare()
@@ -151,7 +167,15 @@ def request(
         If no response is received will raise ConnectionException."""
 
         try:
-            headers.setdefault("User-Agent", user_agent)
+            headers.setdefault(
+                "User-Agent",
+                _generate_user_agent(
+                    user_agent,
+                    self._send_platform_info,
+                    self._app_info_name,
+                    self._app_info_version,
+                ),
+            )
             request = requests.Request(
                 method, url, data=data, headers=headers, **kwargs
             ).prepare()
@@ -206,3 +230,25 @@ def _should_retry(self, response, exception, num_retries):
         return status_code == http.HTTPStatus.TOO_MANY_REQUESTS or (
             status_code >= http.HTTPStatus.INTERNAL_SERVER_ERROR
         )
+
+
+@lru_cache(maxsize=4)
+def _generate_user_agent(
+    user_agent_str: Optional[str],
+    send_platform_info: bool,
+    app_info_name: Optional[str],
+    app_info_version: Optional[str],
+):
+    if user_agent_str:
+        library_info_str = user_agent_str
+    else:
+        library_info_str = f"deepl-python/{version.VERSION}"
+        if send_platform_info:
+            library_info_str += (
+                f" ({platform.platform()}) "
+                f"python/{platform.python_version()} "
+                f"requests/{requests.__version__}"
+            )
+    if app_info_name and app_info_version:
+        library_info_str += f" {app_info_name}/{app_info_version}"
+    return library_info_str
diff --git a/deepl/translator.py b/deepl/translator.py
@@ -454,6 +454,10 @@ class Translator:
         URL strings for the 'http' and 'https' keys. This is passed to the
         underlying requests session, see the requests proxy documentation for
         more information.
+    :param send_platform_info: (Optional) boolean that indicates if the client
+        library can send basic platform info (python version, OS, http library
+        version) to the DeepL API. True = send info, False = only send client
+        library version
     :param skip_language_check: Deprecated, and now has no effect as the
         corresponding internal functionality has been removed. This parameter
         will be removed in a future version.
@@ -475,6 +479,7 @@ def __init__(
         *,
         server_url: Optional[str] = None,
         proxy: Union[Dict, str, None] = None,
+        send_platform_info: bool = True,
         skip_language_check: bool = False,
     ):
         if not auth_key:
@@ -488,7 +493,7 @@ def __init__(
             )
 
         self._server_url = server_url
-        self._client = http_client.HttpClient(proxy)
+        self._client = http_client.HttpClient(proxy, send_platform_info)
         self.headers = {"Authorization": f"DeepL-Auth-Key {auth_key}"}
 
     def __del__(self):
@@ -701,6 +706,10 @@ def close(self):
         if hasattr(self, "_client"):
             self._client.close()
 
+    def set_app_info(self, app_info_name: str, app_info_version: str):
+        self._client.set_app_info(app_info_name, app_info_version)
+        return self
+
     @property
     def server_url(self):
         return self._server_url
diff --git a/tests/test_general.py b/tests/test_general.py
@@ -3,9 +3,12 @@
 # license that can be found in the LICENSE file.
 
 from .conftest import example_text, needs_mock_server, needs_mock_proxy_server
+from requests import Response
+from unittest.mock import patch, Mock
 import deepl
 import pathlib
 import pytest
+import os
 
 
 def test_version():
@@ -86,6 +89,85 @@ def test_server_url_selected_based_on_auth_key(server):
     assert translator_free.server_url == "https://api-free.deepl.com"
 
 
+@patch("requests.adapters.HTTPAdapter.send")
+def test_user_agent(mock_send):
+    mock_send.return_value = _build_test_response()
+    translator = deepl.Translator(os.environ["DEEPL_AUTH_KEY"])
+    translator.translate_text(example_text["EN"], target_lang="DA")
+    ua_header = mock_send.call_args[0][0].headers["User-agent"]
+    assert "requests/" in ua_header
+    assert " python/" in ua_header
+    assert "(" in ua_header
+
+
+@patch("requests.adapters.HTTPAdapter.send")
+def test_user_agent_opt_out(mock_send):
+    mock_send.return_value = _build_test_response()
+    translator = deepl.Translator(
+        os.environ["DEEPL_AUTH_KEY"], send_platform_info=False
+    )
+    translator.translate_text(example_text["EN"], target_lang="DA")
+    ua_header = mock_send.call_args[0][0].headers["User-agent"]
+    assert "requests/" not in ua_header
+    assert " python/" not in ua_header
+    assert "(" not in ua_header
+
+
+@patch("requests.adapters.HTTPAdapter.send")
+def test_custom_user_agent(mock_send):
+    mock_send.return_value = _build_test_response()
+    old_user_agent = deepl.http_client.user_agent
+    deepl.http_client.user_agent = "my custom user agent"
+    translator = deepl.Translator(os.environ["DEEPL_AUTH_KEY"])
+    translator.translate_text(example_text["EN"], target_lang="DA")
+    ua_header = mock_send.call_args[0][0].headers["User-agent"]
+    assert ua_header == "my custom user agent"
+    deepl.http_client.user_agent = old_user_agent
+
+
+@patch("requests.adapters.HTTPAdapter.send")
+def test_user_agent_with_app_info(mock_send):
+    mock_send.return_value = _build_test_response()
+    translator = deepl.Translator(
+        os.environ["DEEPL_AUTH_KEY"],
+    ).set_app_info("sample_python_plugin", "1.0.2")
+    translator.translate_text(example_text["EN"], target_lang="DA")
+    ua_header = mock_send.call_args[0][0].headers["User-agent"]
+    assert "requests/" in ua_header
+    assert " python/" in ua_header
+    assert "(" in ua_header
+    assert " sample_python_plugin/1.0.2" in ua_header
+
+
+@patch("requests.adapters.HTTPAdapter.send")
+def test_user_agent_opt_out_with_app_info(mock_send):
+    mock_send.return_value = _build_test_response()
+    translator = deepl.Translator(
+        os.environ["DEEPL_AUTH_KEY"],
+        send_platform_info=False,
+    ).set_app_info("sample_python_plugin", "1.0.2")
+    translator.translate_text(example_text["EN"], target_lang="DA")
+    ua_header = mock_send.call_args[0][0].headers["User-agent"]
+    assert "requests/" not in ua_header
+    assert " python/" not in ua_header
+    assert "(" not in ua_header
+    assert " sample_python_plugin/1.0.2" in ua_header
+
+
+@patch("requests.adapters.HTTPAdapter.send")
+def test_custom_user_agent_with_app_info(mock_send):
+    mock_send.return_value = _build_test_response()
+    old_user_agent = deepl.http_client.user_agent
+    deepl.http_client.user_agent = "my custom user agent"
+    translator = deepl.Translator(os.environ["DEEPL_AUTH_KEY"]).set_app_info(
+        "sample_python_plugin", "1.0.2"
+    )
+    translator.translate_text(example_text["EN"], target_lang="DA")
+    ua_header = mock_send.call_args[0][0].headers["User-agent"]
+    assert ua_header == "my custom user agent sample_python_plugin/1.0.2"
+    deepl.http_client.user_agent = old_user_agent
+
+
 @needs_mock_proxy_server
 def test_proxy_usage(
     server,
@@ -194,3 +276,26 @@ def test_usage_team_document_limit(
     assert not usage.document.limit_reached
     assert not usage.character.limit_reached
     assert usage.team_document.limit_reached
+
+
+def _build_test_response():
+    response = Mock(spec=Response)
+    response.status_code = 200
+    response.text = (
+        '{"translations": [{"detected_source_language": "EN", '
+        '"text": "protonstråle"}]}'
+    )
+    response.headers = {
+        "Content-Type": "application/json",
+        "Server": "nginx",
+        "Content-Length": str(len(response.text.encode("utf-8"))),
+        "Connection": "keep-alive",
+        "Access-Control-Allow-Origin": "*",
+    }
+    response.encoding = "utf-8"
+    response.history = None
+    response.raw = None
+    response.is_redirect = False
+    response.stream = False
+    response.url = "https://api.deepl.com/v2/translate"
+    return response
