File size: 9,344 Bytes
c8e7ce2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 |
# coding=utf-8
# Copyright 2022-present, the HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""Contains utilities to handle headers to send in calls to Huggingface Hub."""
from typing import Dict, Optional, Union
from .. import constants
from ._runtime import (
get_fastai_version,
get_fastcore_version,
get_hf_hub_version,
get_python_version,
get_tf_version,
get_torch_version,
is_fastai_available,
is_fastcore_available,
is_tf_available,
is_torch_available,
)
from ._token import get_token
from ._validators import validate_hf_hub_args
class LocalTokenNotFoundError(EnvironmentError):
"""Raised if local token is required but not found."""
@validate_hf_hub_args
def build_hf_headers(
*,
token: Optional[Union[bool, str]] = None,
is_write_action: bool = False,
library_name: Optional[str] = None,
library_version: Optional[str] = None,
user_agent: Union[Dict, str, None] = None,
) -> Dict[str, str]:
"""
Build headers dictionary to send in a HF Hub call.
By default, authorization token is always provided either from argument (explicit
use) or retrieved from the cache (implicit use). To explicitly avoid sending the
token to the Hub, set `token=False` or set the `HF_HUB_DISABLE_IMPLICIT_TOKEN`
environment variable.
In case of an API call that requires write access, an error is thrown if token is
`None` or token is an organization token (starting with `"api_org***"`).
In addition to the auth header, a user-agent is added to provide information about
the installed packages (versions of python, huggingface_hub, torch, tensorflow,
fastai and fastcore).
Args:
token (`str`, `bool`, *optional*):
The token to be sent in authorization header for the Hub call:
- if a string, it is used as the Hugging Face token
- if `True`, the token is read from the machine (cache or env variable)
- if `False`, authorization header is not set
- if `None`, the token is read from the machine only except if
`HF_HUB_DISABLE_IMPLICIT_TOKEN` env variable is set.
is_write_action (`bool`, default to `False`):
Set to True if the API call requires a write access. If `True`, the token
will be validated (cannot be `None`, cannot start by `"api_org***"`).
library_name (`str`, *optional*):
The name of the library that is making the HTTP request. Will be added to
the user-agent header.
library_version (`str`, *optional*):
The version of the library that is making the HTTP request. Will be added
to the user-agent header.
user_agent (`str`, `dict`, *optional*):
The user agent info in the form of a dictionary or a single string. It will
be completed with information about the installed packages.
Returns:
A `Dict` of headers to pass in your API call.
Example:
```py
>>> build_hf_headers(token="hf_***") # explicit token
{"authorization": "Bearer hf_***", "user-agent": ""}
>>> build_hf_headers(token=True) # explicitly use cached token
{"authorization": "Bearer hf_***",...}
>>> build_hf_headers(token=False) # explicitly don't use cached token
{"user-agent": ...}
>>> build_hf_headers() # implicit use of the cached token
{"authorization": "Bearer hf_***",...}
# HF_HUB_DISABLE_IMPLICIT_TOKEN=True # to set as env variable
>>> build_hf_headers() # token is not sent
{"user-agent": ...}
>>> build_hf_headers(token="api_org_***", is_write_action=True)
ValueError: You must use your personal account token for write-access methods.
>>> build_hf_headers(library_name="transformers", library_version="1.2.3")
{"authorization": ..., "user-agent": "transformers/1.2.3; hf_hub/0.10.2; python/3.10.4; tensorflow/1.55"}
```
Raises:
[`ValueError`](https://docs.python.org/3/library/exceptions.html#ValueError)
If organization token is passed and "write" access is required.
[`ValueError`](https://docs.python.org/3/library/exceptions.html#ValueError)
If "write" access is required but token is not passed and not saved locally.
[`EnvironmentError`](https://docs.python.org/3/library/exceptions.html#EnvironmentError)
If `token=True` but token is not saved locally.
"""
# Get auth token to send
token_to_send = get_token_to_send(token)
_validate_token_to_send(token_to_send, is_write_action=is_write_action)
# Combine headers
headers = {
"user-agent": _http_user_agent(
library_name=library_name,
library_version=library_version,
user_agent=user_agent,
)
}
if token_to_send is not None:
headers["authorization"] = f"Bearer {token_to_send}"
return headers
def get_token_to_send(token: Optional[Union[bool, str]]) -> Optional[str]:
"""Select the token to send from either `token` or the cache."""
# Case token is explicitly provided
if isinstance(token, str):
return token
# Case token is explicitly forbidden
if token is False:
return None
# Token is not provided: we get it from local cache
cached_token = get_token()
# Case token is explicitly required
if token is True:
if cached_token is None:
raise LocalTokenNotFoundError(
"Token is required (`token=True`), but no token found. You"
" need to provide a token or be logged in to Hugging Face with"
" `huggingface-cli login` or `huggingface_hub.login`. See"
" https://huggingface.co/settings/tokens."
)
return cached_token
# Case implicit use of the token is forbidden by env variable
if constants.HF_HUB_DISABLE_IMPLICIT_TOKEN:
return None
# Otherwise: we use the cached token as the user has not explicitly forbidden it
return cached_token
def _validate_token_to_send(token: Optional[str], is_write_action: bool) -> None:
if is_write_action:
if token is None:
raise ValueError(
"Token is required (write-access action) but no token found. You need"
" to provide a token or be logged in to Hugging Face with"
" `huggingface-cli login` or `huggingface_hub.login`. See"
" https://huggingface.co/settings/tokens."
)
if token.startswith("api_org"):
raise ValueError(
"You must use your personal account token for write-access methods. To"
" generate a write-access token, go to"
" https://huggingface.co/settings/tokens"
)
def _http_user_agent(
*,
library_name: Optional[str] = None,
library_version: Optional[str] = None,
user_agent: Union[Dict, str, None] = None,
) -> str:
"""Format a user-agent string containing information about the installed packages.
Args:
library_name (`str`, *optional*):
The name of the library that is making the HTTP request.
library_version (`str`, *optional*):
The version of the library that is making the HTTP request.
user_agent (`str`, `dict`, *optional*):
The user agent info in the form of a dictionary or a single string.
Returns:
The formatted user-agent string.
"""
if library_name is not None:
ua = f"{library_name}/{library_version}"
else:
ua = "unknown/None"
ua += f"; hf_hub/{get_hf_hub_version()}"
ua += f"; python/{get_python_version()}"
if not constants.HF_HUB_DISABLE_TELEMETRY:
if is_torch_available():
ua += f"; torch/{get_torch_version()}"
if is_tf_available():
ua += f"; tensorflow/{get_tf_version()}"
if is_fastai_available():
ua += f"; fastai/{get_fastai_version()}"
if is_fastcore_available():
ua += f"; fastcore/{get_fastcore_version()}"
if isinstance(user_agent, dict):
ua += "; " + "; ".join(f"{k}/{v}" for k, v in user_agent.items())
elif isinstance(user_agent, str):
ua += "; " + user_agent
return _deduplicate_user_agent(ua)
def _deduplicate_user_agent(user_agent: str) -> str:
"""Deduplicate redundant information in the generated user-agent."""
# Split around ";" > Strip whitespaces > Store as dict keys (ensure unicity) > format back as string
# Order is implicitly preserved by dictionary structure (see https://stackoverflow.com/a/53657523).
return "; ".join({key.strip(): None for key in user_agent.split(";")}.keys())
|