indextools

This module implements tools to determine time-related indices.

Module indextools implements the following members:

class hydpy.core.indextools.IndexerProperty(fget)[source]

Bases: BaseProperty

A property for handling time-related indices.

Some models (e.g. lland_dd) require time related index values. IndexerProperty provides some caching functionalities to avoid recalculating the same indices for different model instances over and over again. We illustrate this by taking property monthofyear as an example.

Generally, Indexer needs to know the relevant initialisation period before being able to calculate any time-related index values. If you forget to define one first, you get the following error message:

>>> from hydpy import print_vector, pub
>>> pub.indexer.monthofyear
Traceback (most recent call last):
...
hydpy.core.exceptiontools.AttributeNotReady: An Indexer object has been asked for an `monthofyear` array.  Such an array has neither been determined yet nor can it be determined automatically at the moment.   Either define an `monthofyear` array manually and pass it to the Indexer object, or make a proper Timegrids object available within the pub module.

For efficiency, repeated querying of monthofyear returns the same numpy array() object:

>>> pub.timegrids = "27.02.2004", "3.03.2004", "1d"
>>> monthofyear = pub.indexer.monthofyear
>>> print_vector(monthofyear)
1, 1, 1, 2, 2
>>> print_vector(pub.indexer.monthofyear)
1, 1, 1, 2, 2
>>> pub.indexer.monthofyear is monthofyear
True

When the Timegrids object handled by module pub changes, IndexerProperty calculates and returns a new index array:

>>> pub.timegrids.init.firstdate += "1d"
>>> print_vector(pub.indexer.monthofyear)
1, 1, 2, 2
>>> pub.indexer.monthofyear is monthofyear
False

When in doubt, you can manually delete the cached numpy ndarray and receive a freshly calculated index array afterwards:

>>> monthofyear = pub.indexer.monthofyear
>>> pub.indexer.monthofyear is monthofyear
True
>>> del pub.indexer.monthofyear
>>> print_vector(pub.indexer.monthofyear)
1, 1, 2, 2
>>> pub.indexer.monthofyear is monthofyear
False

You are allowed to define alternative values manually, which seems advisable only for testing purposes:

>>> pub.indexer.monthofyear = 0, 1, 2, 3
>>> print_vector(pub.indexer.monthofyear)
0, 1, 2, 3
>>> pub.timegrids.init.firstdate -= "1d"
>>> print_vector(pub.indexer.monthofyear)
1, 1, 1, 2, 2

When assigning inadequate data, you get errors like the following:

>>> pub.indexer.monthofyear = 1
Traceback (most recent call last):
...
TypeError: While trying to assign a new `monthofyear` index array to an Indexer object, the following error occurred: 'int' object is not subscriptable

>>> pub.indexer.monthofyear = [[0, 1, 2, 3], [4, 5, 6, 7]]
Traceback (most recent call last):
...
ValueError: The `monthofyear` index array of an Indexer object must be 1-dimensional.  However, the given value has interpreted as a 2-dimensional object.

>>> pub.indexer.monthofyear = 0, 1, 2, 3
Traceback (most recent call last):
...
ValueError: The `monthofyear` index array of an Indexer object must have a number of entries fitting to the initialization time period precisely.  However, the given value has been interpreted to be of length `4` and the length of the Timegrid object representing the actual initialisation period is `5`.
call_fget(obj) → ndarray[Any, dtype[float64]][source]

Method for implementing unique getter functionalities.

call_fset(obj, value)[source]

Method for implementing unique setter functionalities.

call_fdel(obj)[source]

Method for implementing unique deleter functionalities.

class hydpy.core.indextools.Indexer[source]

Bases: object

Handles different IndexerProperty objects defining time-related indices.

One can specify the index arrays manually, but they are usually determined automatically based on the Timegrids object made available through module pub.

monthofyear[source]

Index values, representing the month of the year.

The following example shows the month indices of the last days of February and the first days of March for a leap year:

>>> from hydpy import print_vector, pub
>>> pub.timegrids = "27.02.2004", "3.03.2004", "1d"
>>> monthofyear = pub.indexer.monthofyear
>>> print_vector(monthofyear)
1, 1, 1, 2, 2
dayofyear[source]

Index values, representing the month of the year.

For reasons of consistency between leap years and non-leap years, assuming a daily time step, index 59 is always associated with the 29th of February. Hence, it is missing in non-leap years:

>>> from hydpy import print_vector, pub
>>> from hydpy.core.indextools import Indexer
>>> pub.timegrids = "27.02.2004", "3.03.2004", "1d"
>>> print_vector(Indexer().dayofyear)
57, 58, 59, 60, 61
>>> pub.timegrids = "27.02.2005", "3.03.2005", "1d"
>>> print_vector(Indexer().dayofyear)
57, 58, 60, 61
timeofyear[source]

Index values, representing the time of the year.

Let us reconsider one of the examples of the documentation on property dayofyear:

>>> from hydpy import print_vector, pub, Timegrids, Timegrid
>>> from hydpy.core.indextools import Indexer
>>> pub.timegrids = "27.02.2005", "3.03.2005", "1d"

Due to the simulation step size of one day, the index arrays calculated by properties dayofyear and timeofyear are identical:

>>> print_vector(Indexer().dayofyear)
57, 58, 60, 61
>>> print_vector(Indexer().timeofyear)
57, 58, 60, 61

In the next example, we halve the step size:

>>> pub.timegrids = "27.02.2005", "3.03.2005", "12h"

Now two subsequent simulation steps associated are with the same day:

>>> print_vector(Indexer().dayofyear)
57, 57, 58, 58, 60, 60, 61, 61

However, the timeofyear array gives the index of the respective simulation steps of the actual year:

>>> print_vector(Indexer().timeofyear)
114, 115, 116, 117, 120, 121, 122, 123

Note the gap in the returned index array due to 2005 being not a leap year.

standardclocktime[source]

Standard clock time at the midpoints of the initialisation time steps in hours.

Note that the standard clock time is not usable as an index. Hence, we might later move property standardclocktime somewhere else or give class Indexer a more general purpose (and name) later.

The following examples demonstrate the calculation of the standard clock time for simulation step sizes of one day, one hour, one minute, and one second, respectively:

>>> from hydpy import pub, print_vector
>>> pub.timegrids = "27.02.2004", "3.03.2004", "1d"
>>> print_vector(pub.indexer.standardclocktime)
12.0, 12.0, 12.0, 12.0, 12.0

>>> pub.timegrids = "27.02.2004 21:00", "28.02.2004 03:00", "1h"
>>> print_vector(pub.indexer.standardclocktime)
21.5, 22.5, 23.5, 0.5, 1.5, 2.5

>>> pub.timegrids = "27.02.2004 23:57:0", "28.02.2004 00:03:00", "1m"
>>> print_vector(pub.indexer.standardclocktime)
23.958333, 23.975, 23.991667, 0.008333, 0.025, 0.041667

>>> pub.timegrids = "27.02.2004 23:59:57", "28.02.2004 00:00:03", "1s"
>>> print_vector(pub.indexer.standardclocktime)
23.999306, 23.999583, 23.999861, 0.000139, 0.000417, 0.000694