utils

This module contains convenience functions for any purposes.

Timer([info, verbose])	Provide a simple time profiling utility
concat_each_time_coordinate(*datasets)	Concatenate xarray datasets along each time coordinate
date2num(dates, units[, calendar])	Convert an array of integer into datetime objects.
deprecated([func, new_name, message])	Decorator which can be used to mark functions as deprecated.
extract_block_diag(M, n)	Extract diagonal blocks from square Matrix.
get_time_coordinates(ds)	From a xarray dataset or dataarray, get coordinates with at least 1 time dimension
get_time_dimensions(ds)	From a xarray dataset or dataarray, get dimensions corresponding to time coordinates
get_xarray_groups(dataset[, only_names])	Get pseudo groups from xarray dataset object (only direct under root)
image2mpeg(glob, outfile[, framerate, ...])	Combine image files to a video using ffmpeg.
num2date(times, units[, calendar])	Convert an array of integers into datetime objects.
path_append(dirname[, path])	Append a directory to environment path variable.
path_prepend(dirname[, path])	Prepend a directory to environment path variable.
path_remove(dirname[, path])	Remove a directory from environment path variable.
reraise_with_stack(func)	Make functions include the whole stack in raised exceptions
safe_eval(expr)	Safely evaluate string that may contain basic arithmetic
set_time_resolution(datetime_obj, resolution)	Set the resolution of a python datetime object.
split_units(value)	Splits a string into float number and potential unit
stack_xarray_repdim(da, **dims)	Like xarrays stack, but with partial support for repeated dimensions
to_datetime(obj)	Convert an object to a python datetime object.
to_timedelta(obj[, numbers_as])	Convert an object to a python datetime object.
undo_xarray_floatification(ds[, fields])	convert floats back to ints in xarray dataset where appropriate
unique(seq)	Remove duplicates from list whilst keeping the original order

utils.cache

Utilities related to caching and memoisation

mutable_cache([maxsize])

In-memory cache like functools.lru_cache but for any object