Source code for bokeh.models.tickers

#-----------------------------------------------------------------------------
# Copyright (c) Anaconda, Inc., and Bokeh Contributors.
# All rights reserved.
#
# The full license is in the file LICENSE.txt, distributed with this software.
#-----------------------------------------------------------------------------
''' Models for computing good tick locations on different kinds
of plots.

'''

#-----------------------------------------------------------------------------
# Boilerplate
#-----------------------------------------------------------------------------
from __future__ import annotations

import logging # isort:skip
log = logging.getLogger(__name__)

#-----------------------------------------------------------------------------
# Imports
#-----------------------------------------------------------------------------

# Bokeh imports
from ..core.enums import LatLon
from ..core.has_props import abstract
from ..core.properties import (
    Auto,
    Either,
    Enum,
    Float,
    Instance,
    Int,
    NonEmpty,
    Nullable,
    Override,
    Required,
    Seq,
)
from ..core.validation import error
from ..core.validation.errors import MISSING_MERCATOR_DIMENSION
from ..model import Model
from .mappers import ScanningColorMapper

#-----------------------------------------------------------------------------
# Globals and constants
#-----------------------------------------------------------------------------

__all__ = (
    'Ticker',
    'BinnedTicker',
    'ContinuousTicker',
    'FixedTicker',
    'AdaptiveTicker',
    'CompositeTicker',
    'SingleIntervalTicker',
    'DaysTicker',
    'MonthsTicker',
    'YearsTicker',
    'BasicTicker',
    'LogTicker',
    'MercatorTicker',
    'CategoricalTicker',
    'DatetimeTicker',
)

#-----------------------------------------------------------------------------
# General API
#-----------------------------------------------------------------------------


[docs]
@abstract
class Ticker(Model):
    ''' A base class for all ticker types.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)



[docs]
@abstract
class ContinuousTicker(Ticker):
    ''' A base class for non-categorical ticker types.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)

    num_minor_ticks = Int(5, help="""
    The number of minor tick positions to generate between
    adjacent major tick values.
    """)

    desired_num_ticks = Int(6, help="""
    A desired target number of major tick positions to generate across
    the plot range.

    .. note:
        This value is a suggestion, and ticker subclasses may ignore
        it entirely, or use it only as an ideal goal to approach as well
        as can be, in the context of a specific ticking strategy.
    """)



[docs]
class FixedTicker(ContinuousTicker):
    ''' Generate ticks at fixed, explicitly supplied locations.

    .. note::
        The ``desired_num_ticks`` property is ignored by this Ticker.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)

    ticks = Seq(Float, default=[], help="""
    List of major tick locations.
    """)

    minor_ticks = Seq(Float, default=[], help="""
    List of minor tick locations.
    """)



[docs]
class AdaptiveTicker(ContinuousTicker):
    ''' Generate "nice" round ticks at any magnitude.

    Creates ticks that are "base" multiples of a set of given
    mantissas. For example, with ``base=10`` and
    ``mantissas=[1, 2, 5]``, the ticker will generate the sequence::

        ..., 0.1, 0.2, 0.5, 1, 2, 5, 10, 20, 50, 100, ...

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)

    base = Float(10.0, help="""
    The multiplier to use for scaling mantissas.
    """)

    mantissas = Seq(Float, default=[1, 2, 5], help="""
    The acceptable list numbers to generate multiples of.
    """)

    min_interval = Float(0.0, help="""
    The smallest allowable interval between two adjacent ticks.
    """)

    max_interval = Nullable(Float, help="""
    The largest allowable interval between two adjacent ticks.

    .. note::
        To specify an unbounded interval, set to ``None``.
    """)



[docs]
class CompositeTicker(ContinuousTicker):
    ''' Combine different tickers at different scales.

    Uses the ``min_interval`` and ``max_interval`` interval attributes
    of the tickers to select the appropriate ticker at different
    scales.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)

    tickers = NonEmpty(Seq(Instance(Ticker)), help="""
    A list of ``Ticker`` objects to combine at different scales in order
    to generate tick values. The supplied tickers should be in order.
    Specifically, if S comes before T, then it should be the case that:

    .. code-block:: javascript

        S.get_max_interval() < T.get_min_interval()

    """)


class BaseSingleIntervalTicker(ContinuousTicker):
    ''' Base class for single interval tickers. '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)


[docs]
class SingleIntervalTicker(BaseSingleIntervalTicker):
    ''' Generate evenly spaced ticks at a fixed interval regardless of
    scale.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)

    interval = Required(Float, help="""
    The interval between adjacent ticks.
    """)



[docs]
class DaysTicker(BaseSingleIntervalTicker):
    ''' Generate ticks spaced apart by specific, even multiples of days.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)

    days = Seq(Int, default=[], help="""
    The intervals of days to use.
    """)

    num_minor_ticks = Override(default=0)



[docs]
class MonthsTicker(BaseSingleIntervalTicker):
    ''' Generate ticks spaced apart by specific, even multiples of months.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)

    months = Seq(Int, default=[], help="""
    The intervals of months to use.
    """)



[docs]
class YearsTicker(BaseSingleIntervalTicker):
    ''' Generate ticks spaced apart even numbers of years.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)



[docs]
class BasicTicker(AdaptiveTicker):
    ''' Generate ticks on a linear scale.

    .. note::
        This class may be renamed to ``LinearTicker`` in the future.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)



[docs]
class LogTicker(AdaptiveTicker):
    ''' Generate ticks on a log scale.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)

    mantissas = Override(default=[1, 5])




[docs]
class MercatorTicker(BasicTicker):
    ''' Generate nice lat/lon ticks form underlying WebMercator coordinates.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)

    dimension = Nullable(Enum(LatLon), help="""
    Specify whether to generate ticks for Latitude or Longitude.

    Projected coordinates are not separable, computing Latitude and Longitude
    tick locations from Web Mercator requires considering coordinates from
    both dimensions together. Use this property to specify which result should
    be returned.

    Typically, if the ticker is for an x-axis, then dimension should be
    ``"lon"`` and if the ticker is for a y-axis, then the dimension
    should be `"lat"``.

    In order to prevent hard to debug errors, there is no default value for
    dimension. Using an un-configured ``MercatorTicker`` will result in a
    validation error and a JavaScript console error.
    """)

    @error(MISSING_MERCATOR_DIMENSION)
    def _check_missing_dimension(self):
        if self.dimension is None:
            return str(self)



[docs]
class CategoricalTicker(Ticker):
    ''' Generate ticks for categorical ranges.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)


ONE_MILLI = 1.0
ONE_SECOND = 1000.0
ONE_MINUTE = 60.0 * ONE_SECOND
ONE_HOUR = 60 * ONE_MINUTE
ONE_DAY = 24 * ONE_HOUR
ONE_MONTH = 30 * ONE_DAY # An approximation, obviously.
ONE_YEAR = 365 * ONE_DAY


[docs]
class DatetimeTicker(CompositeTicker):
    ''' Generate nice ticks across different date and time scales.

    '''

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)

    num_minor_ticks = Override(default=0)

    # TODO: (bev) InstanceDefault for this, someday
    tickers = Override(default=lambda: [
        AdaptiveTicker(
            mantissas=[1, 2, 5],
            base=10,
            min_interval=0,
            max_interval=500*ONE_MILLI,
            num_minor_ticks=0,
        ),
        AdaptiveTicker(
            mantissas=[1, 2, 5, 10, 15, 20, 30],
            base=60,
            min_interval=ONE_SECOND,
            max_interval=30*ONE_MINUTE,
            num_minor_ticks=0,
        ),
        AdaptiveTicker(
            mantissas=[1, 2, 4, 6, 8, 12],
            base=24,
            min_interval=ONE_HOUR,
            max_interval=12*ONE_HOUR,
            num_minor_ticks=0,
        ),
        DaysTicker(days=list(range(1, 32))),
        DaysTicker(days=list(range(1, 31, 3))),
        DaysTicker(days=[1, 8, 15, 22]),
        DaysTicker(days=[1, 15]),

        MonthsTicker(months=list(range(0, 12, 1))),
        MonthsTicker(months=list(range(0, 12, 2))),
        MonthsTicker(months=list(range(0, 12, 4))),
        MonthsTicker(months=list(range(0, 12, 6))),

        YearsTicker(),
    ])



[docs]
class BinnedTicker(Ticker):
    """ Ticker that aligns ticks exactly at bin boundaries of a scanning color mapper.

    """

    # explicit __init__ to support Init signatures
    def __init__(self, *args, **kwargs) -> None:
        super().__init__(*args, **kwargs)

    mapper = Instance(ScanningColorMapper, help="""
    A scanning color mapper (e.g. ``EqHistColorMapper``) to use.
    """)

    num_major_ticks = Either(Int, Auto, default=8, help="""
    The number of major tick positions to show or "auto" to use the
    number of bins provided by the mapper.
    """)


#-----------------------------------------------------------------------------
# Dev API
#-----------------------------------------------------------------------------

#-----------------------------------------------------------------------------
# Private API
#-----------------------------------------------------------------------------

#-----------------------------------------------------------------------------
# Code
#-----------------------------------------------------------------------------