#-----------------------------------------------------------------------------
# Copyright (c) Anaconda, Inc., and Bokeh Contributors.
# All rights reserved.
#
# The full license is in the file LICENSE.txt, distributed with this software.
#-----------------------------------------------------------------------------
''' Control global configuration options with environment variables.
A global settings object that other parts of Bokeh can refer to.
Defined Settings
~~~~~~~~~~~~~~~~
Settings are accessible on the ``bokeh.settings.settings`` instance, via
accessor methods. For instance:
.. code-block:: python
settings.minified()
Bokeh provides the following defined settings:
.. bokeh-settings:: settings
:module: bokeh.settings
Precedence
~~~~~~~~~~
Setting values are always looked up in the following prescribed order:
immediately supplied values
These are values that are passed to the setting:
.. code-block:: python
settings.minified(minified_val)
If ``minified_val`` is not None, then it will be returned, as-is.
Otherwise, if None is passed, then the setting will continue to look
down the search order for a value. This is useful for passing optional
function parameters that are None by default. If the parameter is passed
to the function, then it will take precedence.
previously user-set values
If the value is set explicitly in code:
.. code-block:: python
settings.minified = False
Then this value will take precedence over other sources. Applications
may use this ability to set values supplied on the command line, so that
they take precedence over environment variables.
user-specified config override file
Values from a YAML configuration file that is explicitly loaded:
.. code-block:: python
settings.load_config("/path/to/bokeh.yaml)
Any values from ``bokeh.yaml`` will take precedence over the sources
below. Applications may offer command line arguments to load such a
file. e.g. ``bokeh serve --use-config myconf.yaml``
environment variable
Values found in the associated environment variables:
.. code-block:: sh
BOKEH_MINIFIED=no bokeh serve app.py
local user config file
Bokeh will look for a YAML configuration file in the current user's
home directory ``${HOME}/.bokeh/bokeh.yaml``.
global system configuration (not yet implemented)
Future support is planned to load Bokeh settings from global system
configurations.
local defaults
These are default values defined when accessing the setting:
.. code-block:: python
settings.resources(default="server")
Local defaults have lower precedence than every other setting mechanism
except global defaults.
global defaults
These are default values defined by the setting declarations. They have
lower precedence than every other setting mechanism.
If no value is obtained after searching all of these locations, then a
RuntimeError will be raised.
API
~~~
There are a few methods on the ``settings`` object:
.. autoclass:: Settings
:members:
'''
#-----------------------------------------------------------------------------
# Boilerplate
#-----------------------------------------------------------------------------
from __future__ import annotations
import logging # isort:skip
log = logging.getLogger(__name__)
#-----------------------------------------------------------------------------
# Imports
#-----------------------------------------------------------------------------
# Standard library imports
import os
from os.path import join
from pathlib import Path
from typing import (
TYPE_CHECKING,
Any,
Callable,
Generic,
Literal,
Sequence,
TypeAlias,
TypeVar,
cast,
)
# External imports
import yaml
# Bokeh imports
from .util.deprecation import deprecated
from .util.paths import bokehjs_path, server_path
if TYPE_CHECKING:
from .core.types import PathLike
from .resources import ResourcesMode
#-----------------------------------------------------------------------------
# Globals and constants
#-----------------------------------------------------------------------------
__all__ = (
'settings',
)
#-----------------------------------------------------------------------------
# General API
#-----------------------------------------------------------------------------
#-----------------------------------------------------------------------------
# Dev API
#-----------------------------------------------------------------------------
def convert_str(value: str) -> str:
''' Return a string as-is.
'''
return value
def convert_int(value: int | str) -> int:
''' Convert a string to an integer.
'''
return int(value)
def convert_bool(value: bool | str) -> bool:
''' Convert a string to True or False.
If a boolean is passed in, it is returned as-is. Otherwise the function
maps the following strings, ignoring case:
* "yes", "1", "on" -> True
* "no", "0", "off" -> False
Args:
value (str):
A string value to convert to bool
Returns:
bool
Raises:
ValueError
'''
if isinstance(value, bool):
return value
val = value.lower()
if val in ["yes", "1", "on", "true", "True"]:
return True
if val in ["no", "0", "off", "false", "False"]:
return False
raise ValueError(f"Cannot convert {value} to boolean value")
def convert_str_seq(value: list[str] | str) -> list[str]:
''' Convert a string to a list of strings.
If a list or tuple is passed in, it is returned as-is.
Args:
value (seq[str] or str) :
A string to convert to a list of strings
Returns
list[str]
'''
if isinstance(value, list | tuple):
return value
try:
return value.split(",")
except Exception:
raise ValueError(f"Cannot convert {value} to list value")
LogLevel: TypeAlias = Literal["trace", "debug", "info", "warn", "error", "fatal"]
PyLogLevel: TypeAlias = int | None
_log_levels = {
"CRITICAL" : logging.CRITICAL,
"ERROR" : logging.ERROR,
"WARNING" : logging.WARNING,
"INFO" : logging.INFO,
"DEBUG" : logging.DEBUG,
"TRACE" : 9, # Custom level hard-coded to avoid circular import
"NONE" : None,
}
def convert_logging(value: str | int) -> PyLogLevel:
'''Convert a string to a Python logging level
If a log level is passed in, it is returned as-is. Otherwise the function
understands the following strings, ignoring case:
* "critical"
* "error"
* "warning"
* "info"
* "debug"
* "trace"
* "none"
Args:
value (str):
A string value to convert to a logging level
Returns:
int or None
Raises:
ValueError
'''
if value is None or isinstance(value, int):
if value in set(_log_levels.values()):
return value
else:
value = value.upper()
if value in _log_levels:
return _log_levels[value]
raise ValueError(f"Cannot convert {value} to log level, valid values are: {', '.join(_log_levels)}")
ValidationLevel = Literal["none", "errors", "all"]
def convert_validation(value: str | ValidationLevel) -> ValidationLevel:
'''Convert a string to a validation level
If a validation level is passed in, it is returned as-is.
Args:
value (str):
A string value to convert to a validation level
Returns:
string
Raises:
ValueError
'''
VALID_LEVELS = {"none", "errors", "all"}
lowered = value.lower()
if lowered in VALID_LEVELS:
return cast(ValidationLevel, lowered)
raise ValueError(f"Cannot convert {value!r} to validation level, valid values are: {VALID_LEVELS!r}")
def convert_ico_path(value: str) -> str:
'''Convert a string to an .ico path
Args:
value (str):
A string value to convert to a .ico path
Returns:
string
Raises:
ValueError
'''
lowered = value.lower()
if lowered == "none":
return "none"
if lowered == "default":
return str(server_path() / "views" / "bokeh.ico")
# undocumented
if lowered == "default-dev":
return str(server_path() / "views" / "bokeh-dev.ico")
if not value.endswith(".ico"):
raise ValueError(f"Cannot convert {value!r} to valid .ico path")
return value
class _Unset: pass
T = TypeVar("T")
Unset: TypeAlias = T | type[_Unset]
def is_dev() -> bool:
return convert_bool(os.environ.get("BOKEH_DEV", False))
class PrioritizedSetting(Generic[T]):
''' Return a value for a global setting according to configuration precedence.
The following methods are searched in order for the setting:
7. immediately supplied values
6. previously user-set values (e.g. set from command line)
5. user-specified config override file
4. environment variable
3. local user config file
2. global system config file (not yet implemented)
1. local defaults
0. global defaults
Ref: https://stackoverflow.com/a/11077282/3406693
If a value cannot be determined, a ValueError is raised.
The ``env_var`` argument specifies the name of an environment to check for
setting values, e.g. ``"BOKEH_LOG_LEVEL"``.
The optional ``default`` argument specified an implicit default value for
the setting that is returned if no other methods provide a value.
A ``convert`` agument may be provided to convert values before they are
returned. For instance to concert log levels in environment variables
to ``logging`` module values.
'''
_parent: Settings | None
_user_value: Unset[str | T]
def __init__(self, name: str, env_var: str | None = None, default: Unset[T] = _Unset,
dev_default: Unset[T] = _Unset, convert: Callable[[T | str], T] | None = None, help: str = "") -> None:
self._convert = convert if convert else convert_str
self._default = default
self._dev_default = dev_default
self._env_var = env_var
self._help = help
self._name = name
self._parent = None
self._user_value = _Unset
def __call__(self, value: T | str | None = None, default: Unset[T] = _Unset) -> T:
'''Return the setting value according to the standard precedence.
Args:
value (any, optional):
An optional immediate value. If not None, the value will
be converted, then returned.
default (any, optional):
An optional default value that only takes precendence over
implicit default values specified on the property itself.
Returns:
str or int or float
Raises:
RuntimeError
'''
# 7. immediate values
if value is not None:
return self._convert(value)
# 6. previously user-set value
if self._user_value is not _Unset:
return self._convert(self._user_value)
# 5. user-named config file
if self._parent and self._name in self._parent.config_override:
return self._convert(self._parent.config_override[self._name])
# 4. environment variable
if self._env_var and self._env_var in os.environ:
return self._convert(os.environ[self._env_var])
# 3. local config file
if self._parent and self._name in self._parent.config_user:
return self._convert(self._parent.config_user[self._name])
# 2. global config file
if self._parent and self._name in self._parent.config_system:
return self._convert(self._parent.config_system[self._name])
# 1.5 (undocumented) dev defaults take precedence over other defaults
if is_dev() and self._dev_default is not _Unset:
return self._convert(self._dev_default)
# 1. local defaults
if default is not _Unset:
return self._convert(default)
# 0. global defaults
if self._default is not _Unset:
return self._convert(self._default)
raise RuntimeError(f"No configured value found for setting {self._name!r}")
def __get__(self, instance: Any, owner: type[Any]) -> PrioritizedSetting[T]:
return self
def __set__(self, instance: Any, value: str | T) -> None:
self.set_value(value)
def __delete__(self, instance: Any) -> None:
self.unset_value()
def set_value(self, value: str | T) -> None:
''' Specify a value for this setting programmatically.
A value set this way takes precedence over all other methods except
immediate values.
Args:
value (str or int or float):
A user-set value for this setting
Returns:
None
'''
# This triggers LGTMs "mutable descriptor" warning. Since descriptors
# are shared among all instances, it is usually not avised to store any
# data directly on them. But in our case we only ever have one single
# instance of a Settings object.
self._user_value = value # lgtm [py/mutable-descriptor]
def unset_value(self) -> None:
''' Unset the previous user value such that the priority is reset.
'''
self._user_value = _Unset
@property
def env_var(self) -> str | None:
return self._env_var
@property
def default(self) -> Unset[T]:
return self._default
@property
def dev_default(self) -> Unset[T]:
return self._dev_default
@property
def name(self) -> str:
return self._name
@property
def help(self) -> str:
return self._help
@property
def convert_type(self) -> str:
if self._convert is convert_str:
return "String"
if self._convert is convert_bool:
return "Bool"
if self._convert is convert_int:
return "Int"
if self._convert is convert_logging:
return "Log Level"
if self._convert is convert_str_seq:
return "List[String]"
if self._convert is convert_validation:
return "Validation Level"
if self._convert is convert_ico_path:
return "Ico Path"
raise RuntimeError("unreachable")
_config_user_locations: Sequence[Path] = (
Path.home() / ".bokeh" / "bokeh.yaml",
)
[docs]
class Settings:
'''
'''
_config_override: dict[str, Any]
_config_user: dict[str, Any]
_config_system: dict[str, Any]
def __init__(self) -> None:
self._config_override = {}
self._config_user = self._try_load_config(_config_user_locations)
self._config_system = {} # TODO (bev)
for x in self.__class__.__dict__.values():
if isinstance(x, PrioritizedSetting):
x._parent = self
@property
def config_system(self) -> dict[str, Any]:
return dict(self._config_system)
@property
def config_user(self) -> dict[str, Any]:
return dict(self._config_user)
@property
def config_override(self) -> dict[str, Any]:
return dict(self._config_override)
@property
def dev(self) -> bool:
return is_dev()
allowed_ws_origin: PrioritizedSetting[list[str]] = PrioritizedSetting("allowed_ws_origin", "BOKEH_ALLOW_WS_ORIGIN", default=[], convert=convert_str_seq, help="""
A comma-separated list of allowed websocket origins for Bokeh server applications.
""")
auth_module: PrioritizedSetting[str | None] = PrioritizedSetting("auth_module", "BOKEH_AUTH_MODULE", default=None, help="""
A path to a Python modules that implements user authentication functions for
the Bokeh server.
.. warning::
The contents of this module will be executed!
""")
browser: PrioritizedSetting[str | None] = PrioritizedSetting("browser", "BOKEH_BROWSER", default=None, dev_default="none", help="""
The default browser that Bokeh should use to show documents with.
Valid values are any of the predefined browser names understood by the
Python standard library :doc:`webbrowser <python:library/webbrowser>`
module.
""")
cdn_version: PrioritizedSetting[str | None] = PrioritizedSetting("version", "BOKEH_CDN_VERSION", default=None, help="""
What version of BokehJS to use with CDN resources.
See the :class:`~bokeh.resources.Resources` class reference for full details.
""")
chromedriver_path: PrioritizedSetting[str | None] = PrioritizedSetting("chromedriver_path", "BOKEH_CHROMEDRIVER_PATH", default=None, help="""
The name of or full path to chromedriver's executable.
This is used to allow ``bokeh.io.export`` to work on systems that use a
different name for ``chromedriver``, like ``chromedriver-binary`` or
``chromium.chromedriver`` (or its variant, which is used for example
by Snap package manager; see https://snapcraft.io/).
""")
cookie_secret: PrioritizedSetting[str | None] = PrioritizedSetting("cookie_secret", "BOKEH_COOKIE_SECRET", default=None, help="""
Configure the ``cookie_secret`` setting in Tornado. This value is required
if you use ``get_secure_cookie`` or ``set_secure_cookie``. It should be a
long, random sequence of bytes
""")
docs_cdn: PrioritizedSetting[str | None] = PrioritizedSetting("docs_cdn", "BOKEH_DOCS_CDN", default=None, help="""
The version of BokehJS that should be use for loading CDN resources when
building the docs.
To build and display the docs using a locally built BokehJS, use ``local``.
For example:
.. code-block:: sh
BOKEH_DOCS_CDN=local make clean serve
Will build a fresh copy of the docs using the locally built BokehJS and open
a new browser tab to view them.
Otherwise, the value is interpreted a version for CDN. For example:
.. code-block:: sh
BOKEH_DOCS_CDN=1.4.0rc1 make clean
will build docs that use BokehJS version ``1.4.0rc1`` from CDN.
""")
docs_version: PrioritizedSetting[str | None] = PrioritizedSetting("docs_version", "BOKEH_DOCS_VERSION", default=None, help="""
The Bokeh version to stipulate when building the docs.
This setting is necessary to re-deploy existing versions of docs with new
fixes or changes.
""")
ico_path: PrioritizedSetting[str] = PrioritizedSetting("ico_path", "BOKEH_ICO_PATH",
default="default", dev_default="default-dev", convert=convert_ico_path, help="""
Configure the file path to a .ico file for the Bokeh server to use as a
favicon.ico file.
The value should be the full path to a .ico file, or one the following
special values:
- ``default`` to use the default project .ico file
- ``none`` to turn off favicon.ico support entirely
""")
ignore_filename: PrioritizedSetting[bool] = PrioritizedSetting("ignore_filename", "BOKEH_IGNORE_FILENAME", default=False, convert=convert_bool, help="""
Whether to ignore the current script filename when saving Bokeh content.
""")
log_level: PrioritizedSetting[LogLevel] = PrioritizedSetting("log_level", "BOKEH_LOG_LEVEL", default="info", dev_default="debug", help="""
Set the log level for JavaScript BokehJS code.
Valid values are, in order of increasing severity:
- ``trace``
- ``debug``
- ``info``
- ``warn``
- ``error``
- ``fatal``
""")
minified: PrioritizedSetting[bool] = PrioritizedSetting("minified", "BOKEH_MINIFIED", convert=convert_bool, default=True, dev_default=False, help="""
Whether Bokeh should use minified BokehJS resources.
""")
nodejs_path: PrioritizedSetting[str | None] = PrioritizedSetting("nodejs_path", "BOKEH_NODEJS_PATH", default=None, help="""
Path to the Node executable.
NodeJS is an optional dependency that is required for PNG and SVG export,
and for compiling custom extensions. Bokeh will try to automatically locate
an installed Node executable. Use this environment variable to override the
location Bokeh finds, or to point to a non-standard location.
""")
perform_document_validation: PrioritizedSetting[bool] = PrioritizedSetting("validate_doc", "BOKEH_VALIDATE_DOC", convert=convert_bool, default=True, help="""
whether Bokeh should perform validation checks on documents.
Setting this value to False may afford a small performance improvement.
""")
pretty: PrioritizedSetting[bool] = PrioritizedSetting("pretty", "BOKEH_PRETTY", default=False, dev_default=True, help="""
Whether JSON strings should be pretty-printed.
""")
py_log_level: PrioritizedSetting[PyLogLevel] = PrioritizedSetting("py_log_level", "BOKEH_PY_LOG_LEVEL",
default="none", dev_default="debug", convert=convert_logging, help="""
The log level for Python Bokeh code.
Valid values are, in order of increasing severity:
- ``trace``
- ``debug``
- ``info``
- ``warn``
- ``error``
- ``fatal``
- ``none``
""")
resources: PrioritizedSetting[ResourcesMode] = PrioritizedSetting("resources", "BOKEH_RESOURCES", default="cdn", dev_default="server", help="""
What kind of BokehJS resources to configure, e.g ``inline`` or ``cdn``
See the :class:`~bokeh.resources.Resources` class reference for full details.
""")
rootdir: PrioritizedSetting[PathLike | None] = PrioritizedSetting("rootdir", "BOKEH_ROOTDIR", default=None, help="""
Root directory to use with ``relative`` resources
See the :class:`~bokeh.resources.Resources` class reference for full details.
""")
default_server_host: PrioritizedSetting[str] = PrioritizedSetting("default_server_host", "BOKEH_DEFAULT_SERVER_HOST", default="localhost", help="""
Allows to define the default host used by Bokeh's server and resources.
""")
default_server_port: PrioritizedSetting[int] = PrioritizedSetting("default_server_port", "BOKEH_DEFAULT_SERVER_PORT", default=5006, convert=convert_int, help="""
Allows to define the default port used by Bokeh's server and resources.
""")
secret_key: PrioritizedSetting[str | None] = PrioritizedSetting("secret_key", "BOKEH_SECRET_KEY", default=None, help="""
A long, cryptographically-random secret unique to a Bokeh deployment.
""")
serialize_include_defaults: PrioritizedSetting[bool] = \
PrioritizedSetting("serialize_include_defaults", "BOKEH_SERIALIZE_INCLUDE_DEFAULTS", default=False, help="""
Wether to include default values when serializing ``HasProps`` instances.
This is primarily useful for testing, debugging serialization/protocol and other internal purpose.
""")
sign_sessions: PrioritizedSetting[bool] = PrioritizedSetting("sign_sessions", "BOKEH_SIGN_SESSIONS", default=False, help="""
Whether the Bokeh server should only allow sessions signed with a secret key.
If True, ``BOKEH_SECRET_KEY`` must also be set.
""")
simple_ids: PrioritizedSetting[bool] = PrioritizedSetting("simple_ids", "BOKEH_SIMPLE_IDS", default=True, convert=convert_bool, help="""
Whether Bokeh should use simple integers for model IDs (starting at 1000).
If False, Bokeh will use UUIDs for object identifiers. This might be needed,
e.g., if multiple processes are contributing to a single Bokeh Document.
""")
ssl_certfile: PrioritizedSetting[str | None] = PrioritizedSetting("ssl_certfile", "BOKEH_SSL_CERTFILE", default=None, help="""
The path to a certificate file for SSL termination.
""")
ssl_keyfile: PrioritizedSetting[str | None] = PrioritizedSetting("ssl_keyfile", "BOKEH_SSL_KEYFILE", default=None, help="""
The path to a private key file for SSL termination.
""")
ssl_password: PrioritizedSetting[str | None] = PrioritizedSetting("ssl_password", "BOKEH_SSL_PASSWORD", default=None, help="""
A password to decrypt the SSL keyfile, if necessary.
""")
validation_level: PrioritizedSetting[ValidationLevel] = PrioritizedSetting("validation_level", "BOKEH_VALIDATION_LEVEL",
default="none", convert=convert_validation, help="""
Whether validation checks should log or raise exceptions on errors and warnings.
Valid values are:
- ``none``: no exceptions raised (default).
- ``errors``: exception raised on errors (but not on warnings)
- ``all``: exception raised on both errors and warnings
""")
xsrf_cookies: PrioritizedSetting[bool] = PrioritizedSetting("xsrf_cookies", "BOKEH_XSRF_COOKIES", default=False, convert=convert_bool, help="""
Whether to enable Tornado XSRF cookie protection on the Bokeh server. This
is only applicable when also using an auth module or custom handlers. See
https://www.tornadoweb.org/en/stable/guide/security.html#cross-site-request-forgery-protection
for more information about XSRF protection in Tornado. All PUT, POST, and
DELETE handlers will need to be appropriately instrumented when this setting
is active.
""")
# Non-settings methods
[docs]
def bokehjs_path(self) -> Path:
''' The location of the BokehJS source tree.
'''
return bokehjs_path(self.dev)
[docs]
def bokehjsdir(self) -> str:
''' The location of the BokehJS source tree.
.. deprecated:: 3.4.0
Use ``bokehjs_path()`` method instead.
'''
deprecated((3, 4, 0), "bokehjsdir()", "bokehjs_path() method")
return str(self.bokehjs_path())
[docs]
def css_files(self) -> list[str]:
''' The CSS files in the BokehJS directory.
'''
css_files: list[str] = []
for root, _, files in os.walk(self.bokehjs_path()):
for fname in files:
if fname.endswith(".css"):
css_files.append(join(root, fname))
return css_files
[docs]
def js_files(self) -> list[str]:
''' The JS files in the BokehJS directory.
'''
js_files: list[str] = []
for root, _, files in os.walk(self.bokehjs_path()):
for fname in files:
if fname.endswith(".js"):
js_files.append(join(root, fname))
return js_files
[docs]
def load_config(self, location: PathLike) -> None:
''' Load a user-specified override config file.
The file should be a YAML format with ``key: value`` lines.
'''
try:
with Path(location).absolute().open() as f:
self._config_override = yaml.load(f, Loader=yaml.SafeLoader)
except Exception:
raise RuntimeError(f"Could not load Bokeh config file: {location}")
[docs]
def secret_key_bytes(self) -> bytes | None:
''' Return the secret_key, converted to bytes and cached.
'''
if not hasattr(self, '_secret_key_bytes'):
key = self.secret_key()
if key is None:
self._secret_key_bytes = None
else:
self._secret_key_bytes = key.encode("utf-8")
return self._secret_key_bytes
def _try_load_config(self, locations: Sequence[Path]) -> dict[str, Any]:
for location in locations:
try:
with location.open() as f:
return yaml.load(f, Loader=yaml.SafeLoader)
except Exception:
pass
return {}
#-----------------------------------------------------------------------------
# Private API
#-----------------------------------------------------------------------------
#-----------------------------------------------------------------------------
# Code
#-----------------------------------------------------------------------------
settings = Settings()
_secret_key = settings.secret_key()
if _secret_key is not None and len(_secret_key) < 32:
from .util.warnings import warn
warn("BOKEH_SECRET_KEY is recommended to have at least 32 bytes of entropy chosen with a cryptographically-random algorithm")
del _secret_key
if settings.sign_sessions() and settings.secret_key() is None:
from .util.warnings import warn
warn("BOKEH_SECRET_KEY must be set if BOKEH_SIGN_SESSIONS is set to True")
