Source code for matplotlib.cbook

"""
A collection of utility functions and classes.  Originally, many
(but not all) were from the Python Cookbook -- hence the name cbook.

This module is safe to import from anywhere within Matplotlib;
it imports Matplotlib only at runtime.
"""

import collections
import collections.abc
import contextlib
import functools
import gzip
import itertools
import operator
import os
from pathlib import Path
import re
import shlex
import subprocess
import sys
import time
import traceback
import types
import warnings
import weakref

import numpy as np

import matplotlib
from .deprecation import (
    deprecated, warn_deprecated,
    _rename_parameter, _delete_parameter, _make_keyword_only,
    _deprecate_method_override, _deprecate_privatize_attribute,
    _suppress_matplotlib_deprecation_warning,
    MatplotlibDeprecationWarning, mplDeprecation)


def _get_running_interactive_framework():
    """
    Return the interactive framework whose event loop is currently running, if
    any, or "headless" if no event loop can be started, or None.

    Returns
    -------
    Optional[str]
        One of the following values: "qt5", "qt4", "gtk3", "wx", "tk",
        "macosx", "headless", ``None``.
    """
    QtWidgets = (sys.modules.get("PyQt5.QtWidgets")
                 or sys.modules.get("PySide2.QtWidgets"))
    if QtWidgets and QtWidgets.QApplication.instance():
        return "qt5"
    QtGui = (sys.modules.get("PyQt4.QtGui")
             or sys.modules.get("PySide.QtGui"))
    if QtGui and QtGui.QApplication.instance():
        return "qt4"
    Gtk = sys.modules.get("gi.repository.Gtk")
    if Gtk and Gtk.main_level():
        return "gtk3"
    wx = sys.modules.get("wx")
    if wx and wx.GetApp():
        return "wx"
    tkinter = sys.modules.get("tkinter")
    if tkinter:
        for frame in sys._current_frames().values():
            while frame:
                if frame.f_code == tkinter.mainloop.__code__:
                    return "tk"
                frame = frame.f_back
    if 'matplotlib.backends._macosx' in sys.modules:
        if sys.modules["matplotlib.backends._macosx"].event_loop_is_running():
            return "macosx"
    if sys.platform.startswith("linux") and not os.environ.get("DISPLAY"):
        return "headless"
    return None


def _exception_printer(exc):
    if _get_running_interactive_framework() in ["headless", None]:
        raise exc
    else:
        traceback.print_exc()


class _StrongRef:
    """
    Wrapper similar to a weakref, but keeping a strong reference to the object.
    """

    def __init__(self, obj):
        self._obj = obj

    def __call__(self):
        return self._obj

    def __eq__(self, other):
        return isinstance(other, _StrongRef) and self._obj == other._obj

    def __hash__(self):
        return hash(self._obj)


[docs]class CallbackRegistry: """ Handle registering and disconnecting for a set of signals and callbacks: >>> def oneat(x): ... print('eat', x) >>> def ondrink(x): ... print('drink', x) >>> from matplotlib.cbook import CallbackRegistry >>> callbacks = CallbackRegistry() >>> id_eat = callbacks.connect('eat', oneat) >>> id_drink = callbacks.connect('drink', ondrink) >>> callbacks.process('drink', 123) drink 123 >>> callbacks.process('eat', 456) eat 456 >>> callbacks.process('be merry', 456) # nothing will be called >>> callbacks.disconnect(id_eat) >>> callbacks.process('eat', 456) # nothing will be called In practice, one should always disconnect all callbacks when they are no longer needed to avoid dangling references (and thus memory leaks). However, real code in Matplotlib rarely does so, and due to its design, it is rather difficult to place this kind of code. To get around this, and prevent this class of memory leaks, we instead store weak references to bound methods only, so when the destination object needs to die, the CallbackRegistry won't keep it alive. Parameters ---------- exception_handler : callable, optional If provided must have signature :: def handler(exc: Exception) -> None: If not None this function will be called with any `Exception` subclass raised by the callbacks in `CallbackRegistry.process`. The handler may either consume the exception or re-raise. The callable must be pickle-able. The default handler is :: def h(exc): traceback.print_exc() """ # We maintain two mappings: # callbacks: signal -> {cid -> weakref-to-callback} # _func_cid_map: signal -> {weakref-to-callback -> cid} def __init__(self, exception_handler=_exception_printer): self.exception_handler = exception_handler self.callbacks = {} self._cid_gen = itertools.count() self._func_cid_map = {} def __getstate__(self): # In general, callbacks may not be pickled, so we just drop them. return {**vars(self), "callbacks": {}, "_func_cid_map": {}}
[docs] def connect(self, s, func): """Register *func* to be called when signal *s* is generated.""" self._func_cid_map.setdefault(s, {}) try: proxy = weakref.WeakMethod(func, self._remove_proxy) except TypeError: proxy = _StrongRef(func) if proxy in self._func_cid_map[s]: return self._func_cid_map[s][proxy] cid = next(self._cid_gen) self._func_cid_map[s][proxy] = cid self.callbacks.setdefault(s, {}) self.callbacks[s][cid] = proxy return cid
# Keep a reference to sys.is_finalizing, as sys may have been cleared out # at that point. def _remove_proxy(self, proxy, *, _is_finalizing=sys.is_finalizing): if _is_finalizing(): # Weakrefs can't be properly torn down at that point anymore. return for signal, proxies in list(self._func_cid_map.items()): try: del self.callbacks[signal][proxies[proxy]] except KeyError: pass if len(self.callbacks[signal]) == 0: del self.callbacks[signal] del self._func_cid_map[signal]
[docs] def disconnect(self, cid): """Disconnect the callback registered with callback id *cid*.""" for eventname, callbackd in list(self.callbacks.items()): try: del callbackd[cid] except KeyError: continue else: for signal, functions in list(self._func_cid_map.items()): for function, value in list(functions.items()): if value == cid: del functions[function] return
[docs] def process(self, s, *args, **kwargs): """ Process signal *s*. All of the functions registered to receive callbacks on *s* will be called with ``*args`` and ``**kwargs``. """ for cid, ref in list(self.callbacks.get(s, {}).items()): func = ref() if func is not None: try: func(*args, **kwargs) # this does not capture KeyboardInterrupt, SystemExit, # and GeneratorExit except Exception as exc: if self.exception_handler is not None: self.exception_handler(exc) else: raise
[docs]class silent_list(list): """ A list with a short ``repr()``. This is meant to be used for a homogeneous list of artists, so that they don't cause long, meaningless output. Instead of :: [<matplotlib.lines.Line2D object at 0x7f5749fed3c8>, <matplotlib.lines.Line2D object at 0x7f5749fed4e0>, <matplotlib.lines.Line2D object at 0x7f5758016550>] one will get :: <a list of 3 Line2D objects> """ def __init__(self, type, seq=None): self.type = type if seq is not None: self.extend(seq) def __repr__(self): return '<a list of %d %s objects>' % (len(self), self.type)
[docs]@deprecated("3.3") class IgnoredKeywordWarning(UserWarning): """ A class for issuing warnings about keyword arguments that will be ignored by Matplotlib. """ pass
[docs]@deprecated("3.3", alternative="normalize_kwargs") def local_over_kwdict(local_var, kwargs, *keys): """ Enforces the priority of a local variable over potentially conflicting argument(s) from a kwargs dict. The following possible output values are considered in order of priority:: local_var > kwargs[keys[0]] > ... > kwargs[keys[-1]] The first of these whose value is not None will be returned. If all are None then None will be returned. Each key in keys will be removed from the kwargs dict in place. Parameters ---------- local_var : any object The local variable (highest priority). kwargs : dict Dictionary of keyword arguments; modified in place. keys : str(s) Name(s) of keyword arguments to process, in descending order of priority. Returns ------- any object Either local_var or one of kwargs[key] for key in keys. Raises ------ IgnoredKeywordWarning For each key in keys that is removed from kwargs but not used as the output value. """ return _local_over_kwdict(local_var, kwargs, *keys, IgnoredKeywordWarning)
def _local_over_kwdict( local_var, kwargs, *keys, warning_cls=MatplotlibDeprecationWarning): out = local_var for key in keys: kwarg_val = kwargs.pop(key, None) if kwarg_val is not None: if out is None: out = kwarg_val else: _warn_external('"%s" keyword argument will be ignored' % key, warning_cls) return out
[docs]def strip_math(s): """ Remove latex formatting from mathtext. Only handles fully math and fully non-math strings. """ if len(s) >= 2 and s[0] == s[-1] == "$": s = s[1:-1] for tex, plain in [ (r"\times", "x"), # Specifically for Formatter support. (r"\mathdefault", ""), (r"\rm", ""), (r"\cal", ""), (r"\tt", ""), (r"\it", ""), ("\\", ""), ("{", ""), ("}", ""), ]: s = s.replace(tex, plain) return s
[docs]def is_writable_file_like(obj): """Return whether *obj* looks like a file object with a *write* method.""" return callable(getattr(obj, 'write', None))
[docs]def file_requires_unicode(x): """ Return whether the given writable file-like object requires Unicode to be written to it. """ try: x.write(b'') except TypeError: return True else: return False
[docs]def to_filehandle(fname, flag='r', return_opened=False, encoding=None): """ Convert a path to an open file handle or pass-through a file-like object. Consider using `open_file_cm` instead, as it allows one to properly close newly created file objects more easily. Parameters ---------- fname : str or path-like or file-like If `str` or `os.PathLike`, the file is opened using the flags specified by *flag* and *encoding*. If a file-like object, it is passed through. flag : str, default 'r' Passed as the *mode* argument to `open` when *fname* is `str` or `os.PathLike`; ignored if *fname* is file-like. return_opened : bool, default False If True, return both the file object and a boolean indicating whether this was a new file (that the caller needs to close). If False, return only the new file. encoding : str or None, default None Passed as the *mode* argument to `open` when *fname* is `str` or `os.PathLike`; ignored if *fname* is file-like. Returns ------- fh : file-like opened : bool *opened* is only returned if *return_opened* is True. """ if isinstance(fname, os.PathLike): fname = os.fspath(fname) if "U" in flag: warn_deprecated("3.3", message="Passing a flag containing 'U' to " "to_filehandle() is deprecated since %(since)s and " "will be removed %(removal)s.") flag = flag.replace("U", "") if isinstance(fname, str): if fname.endswith('.gz'): fh = gzip.open(fname, flag) elif fname.endswith('.bz2'): # python may not be complied with bz2 support, # bury import until we need it import bz2 fh = bz2.BZ2File(fname, flag) else: fh = open(fname, flag, encoding=encoding) opened = True elif hasattr(fname, 'seek'): fh = fname opened = False else: raise ValueError('fname must be a PathLike or file handle') if return_opened: return fh, opened return fh
[docs]@contextlib.contextmanager def open_file_cm(path_or_file, mode="r", encoding=None): r"""Pass through file objects and context-manage path-likes.""" fh, opened = to_filehandle(path_or_file, mode, True, encoding) if opened: with fh: yield fh else: yield fh
[docs]def is_scalar_or_string(val): """Return whether the given object is a scalar or string like.""" return isinstance(val, str) or not np.iterable(val)
[docs]def get_sample_data(fname, asfileobj=True, *, np_load=False): """ Return a sample data file. *fname* is a path relative to the :file:`mpl-data/sample_data` directory. If *asfileobj* is `True` return a file object, otherwise just a file path. Sample data files are stored in the 'mpl-data/sample_data' directory within the Matplotlib package. If the filename ends in .gz, the file is implicitly ungzipped. If the filename ends with .npy or .npz, *asfileobj* is True, and *np_load* is True, the file is loaded with `numpy.load`. *np_load* currently defaults to False but will default to True in a future release. """ path = _get_data_path('sample_data', fname) if asfileobj: suffix = path.suffix.lower() if suffix == '.gz': return gzip.open(path) elif suffix in ['.npy', '.npz']: if np_load: return np.load(path) else: warn_deprecated( "3.3", message="In a future release, get_sample_data " "will automatically load numpy arrays. Set np_load to " "True to get the array and suppress this warning. Set " "asfileobj to False to get the path to the data file and " "suppress this warning.") return path.open('rb') elif suffix in ['.csv', '.xrc', '.txt']: return path.open('r') else: return path.open('rb') else: return str(path)
def _get_data_path(*args): """ Return the `Path` to a resource file provided by Matplotlib. ``*args`` specify a path relative to the base data path. """ return Path(matplotlib.get_data_path(), *args)
[docs]def flatten(seq, scalarp=is_scalar_or_string): """ Return a generator of flattened nested containers. For example: >>> from matplotlib.cbook import flatten >>> l = (('John', ['Hunter']), (1, 23), [[([42, (5, 23)], )]]) >>> print(list(flatten(l))) ['John', 'Hunter', 1, 23, 42, 5, 23] By: Composite of Holger Krekel and Luther Blissett From: https://code.activestate.com/recipes/121294/ and Recipe 1.12 in cookbook """ for item in seq: if scalarp(item) or item is None: yield item else: yield from flatten(item, scalarp)
[docs]@deprecated("3.3", alternative="os.path.realpath and os.stat") @functools.lru_cache() def get_realpath_and_stat(path): realpath = os.path.realpath(path) stat = os.stat(realpath) stat_key = (stat.st_ino, stat.st_dev) return realpath, stat_key
# A regular expression used to determine the amount of space to # remove. It looks for the first sequence of spaces immediately # following the first newline, or at the beginning of the string. _find_dedent_regex = re.compile(r"(?:(?:\n\r?)|^)( *)\S") # A cache to hold the regexs that actually remove the indent. _dedent_regex = {}
[docs]class maxdict(dict): """ A dictionary with a maximum size. Notes ----- This doesn't override all the relevant methods to constrain the size, just ``__setitem__``, so use with caution. """ def __init__(self, maxsize): dict.__init__(self) self.maxsize = maxsize self._killkeys = [] def __setitem__(self, k, v): if k not in self: if len(self) >= self.maxsize: del self[self._killkeys[0]] del self._killkeys[0] self._killkeys.append(k) dict.__setitem__(self, k, v)
[docs]class Stack: """ Stack of elements with a movable cursor. Mimics home/back/forward in a web browser. """ def __init__(self, default=None): self.clear() self._default = default def __call__(self): """Return the current element, or None.""" if not self._elements: return self._default else: return self._elements[self._pos] def __len__(self): return len(self._elements) def __getitem__(self, ind): return self._elements[ind]
[docs] def forward(self): """Move the position forward and return the current element.""" self._pos = min(self._pos + 1, len(self._elements) - 1) return self()
[docs] def back(self): """Move the position back and return the current element.""" if self._pos > 0: self._pos -= 1 return self()
[docs] def push(self, o): """ Push *o* to the stack at current position. Discard all later elements. *o* is returned. """ self._elements = self._elements[:self._pos + 1] + [o] self._pos = len(self._elements) - 1 return self()
[docs] def home(self): """ Push the first element onto the top of the stack. The first element is returned. """ if not self._elements: return self.push(self._elements[0]) return self()
[docs] def empty(self): """Return whether the stack is empty.""" return len(self._elements) == 0
[docs] def clear(self): """Empty the stack.""" self._pos = -1 self._elements = []
[docs] def bubble(self, o): """ Raise all references of *o* to the top of the stack, and return it. Raises ------ ValueError If *o* is not in the stack. """ if o not in self._elements: raise ValueError('Given element not contained in the stack') old_elements = self._elements.copy() self.clear() top_elements = [] for elem in old_elements: if elem == o: top_elements.append(elem) else: self.push(elem) for _ in top_elements: self.push(o) return o
[docs] def remove(self, o): """ Remove *o* from the stack. Raises ------ ValueError If *o* is not in the stack. """ if o not in self._elements: raise ValueError('Given element not contained in the stack') old_elements = self._elements.copy() self.clear() for elem in old_elements: if elem != o: self.push(elem)
[docs]def report_memory(i=0): # argument may go away """Return the memory consumed by the process.""" def call(command, os_name): try: return subprocess.check_output(command) except subprocess.CalledProcessError as err: raise NotImplementedError( "report_memory works on %s only if " "the '%s' program is found" % (os_name, command[0]) ) from err pid = os.getpid() if sys.platform == 'sunos5': lines = call(['ps', '-p', '%d' % pid, '-o', 'osz'], 'Sun OS') mem = int(lines[-1].strip()) elif sys.platform == 'linux': lines = call(['ps', '-p', '%d' % pid, '-o', 'rss,sz'], 'Linux') mem = int(lines[1].split()[1]) elif sys.platform == 'darwin': lines = call(['ps', '-p', '%d' % pid, '-o', 'rss,vsz'], 'Mac OS') mem = int(lines[1].split()[0]) elif sys.platform == 'win32': lines = call(["tasklist", "/nh", "/fi", "pid eq %d" % pid], 'Windows') mem = int(lines.strip().split()[-2].replace(',', '')) else: raise NotImplementedError( "We don't have a memory monitor for %s" % sys.platform) return mem
[docs]def safe_masked_invalid(x, copy=False): x = np.array(x, subok=True, copy=copy) if not x.dtype.isnative: # If we have already made a copy, do the byteswap in place, else make a # copy with the byte order swapped. x = x.byteswap(inplace=copy).newbyteorder('N') # Swap to native order. try: xm = np.ma.masked_invalid(x, copy=False) xm.shrink_mask() except TypeError: return x return xm
[docs]class Grouper: """ A disjoint-set data structure. Objects can be joined using :meth:`join`, tested for connectedness using :meth:`joined`, and all disjoint sets can be retrieved by using the object as an iterator. The objects being joined must be hashable and weak-referenceable. Examples -------- >>> from matplotlib.cbook import Grouper >>> class Foo: ... def __init__(self, s): ... self.s = s ... def __repr__(self): ... return self.s ... >>> a, b, c, d, e, f = [Foo(x) for x in 'abcdef'] >>> grp = Grouper() >>> grp.join(a, b) >>> grp.join(b, c) >>> grp.join(d, e) >>> sorted(map(tuple, grp)) [(a, b, c), (d, e)] >>> grp.joined(a, b) True >>> grp.joined(a, c) True >>> grp.joined(a, d) False """ def __init__(self, init=()): self._mapping = {weakref.ref(x): [weakref.ref(x)] for x in init} def __contains__(self, item): return weakref.ref(item) in self._mapping
[docs] def clean(self): """Clean dead weak references from the dictionary.""" mapping = self._mapping to_drop = [key for key in mapping if key() is None] for key in to_drop: val = mapping.pop(key) val.remove(key)
[docs] def join(self, a, *args): """ Join given arguments into the same set. Accepts one or more arguments. """ mapping = self._mapping set_a = mapping.setdefault(weakref.ref(a), [weakref.ref(a)]) for arg in args: set_b = mapping.get(weakref.ref(arg), [weakref.ref(arg)]) if set_b is not set_a: if len(set_b) > len(set_a): set_a, set_b = set_b, set_a set_a.extend(set_b) for elem in set_b: mapping[elem] = set_a self.clean()
[docs] def joined(self, a, b): """Return whether *a* and *b* are members of the same set.""" self.clean() return (self._mapping.get(weakref.ref(a), object()) is self._mapping.get(weakref.ref(b)))
[docs] def remove(self, a): self.clean() set_a = self._mapping.pop(weakref.ref(a), None) if set_a: set_a.remove(weakref.ref(a))
def __iter__(self): """ Iterate over each of the disjoint sets as a list. The iterator is invalid if interleaved with calls to join(). """ self.clean() unique_groups = {id(group): group for group in self._mapping.values()} for group in unique_groups.values(): yield [x() for x in group]
[docs] def get_siblings(self, a): """Return all of the items joined with *a*, including itself.""" self.clean() siblings = self._mapping.get(weakref.ref(a), [weakref.ref(a)]) return [x() for x in siblings]
[docs]def simple_linear_interpolation(a, steps): """ Resample an array with ``steps - 1`` points between original point pairs. Along each column of *a*, ``(steps - 1)`` points are introduced between each original values; the values are linearly interpolated. Parameters ---------- a : array, shape (n, ...) steps : int Returns ------- array shape ``((n - 1) * steps + 1, ...)`` """ fps = a.reshape((len(a), -1)) xp = np.arange(len(a)) * steps x = np.arange((len(a) - 1) * steps + 1) return (np.column_stack([np.interp(x, xp, fp) for fp in fps.T]) .reshape((len(x),) + a.shape[1:]))
[docs]def delete_masked_points(*args): """ Find all masked and/or non-finite points in a set of arguments, and return the arguments with only the unmasked points remaining. Arguments can be in any of 5 categories: 1) 1-D masked arrays 2) 1-D ndarrays 3) ndarrays with more than one dimension 4) other non-string iterables 5) anything else The first argument must be in one of the first four categories; any argument with a length differing from that of the first argument (and hence anything in category 5) then will be passed through unchanged. Masks are obtained from all arguments of the correct length in categories 1, 2, and 4; a point is bad if masked in a masked array or if it is a nan or inf. No attempt is made to extract a mask from categories 2, 3, and 4 if `numpy.isfinite` does not yield a Boolean array. All input arguments that are not passed unchanged are returned as ndarrays after removing the points or rows corresponding to masks in any of the arguments. A vastly simpler version of this function was originally written as a helper for Axes.scatter(). """ if not len(args): return () if is_scalar_or_string(args[0]): raise ValueError("First argument must be a sequence") nrecs = len(args[0]) margs = [] seqlist = [False] * len(args) for i, x in enumerate(args): if not isinstance(x, str) and np.iterable(x) and len(x) == nrecs: seqlist[i] = True if isinstance(x, np.ma.MaskedArray): if x.ndim > 1: raise ValueError("Masked arrays must be 1-D") else: x = np.asarray(x) margs.append(x) masks = [] # list of masks that are True where good for i, x in enumerate(margs): if seqlist[i]: if x.ndim > 1: continue # Don't try to get nan locations unless 1-D. if isinstance(x, np.ma.MaskedArray): masks.append(~np.ma.getmaskarray(x)) # invert the mask xd = x.data else: xd = x try: mask = np.isfinite(xd) if isinstance(mask, np.ndarray): masks.append(mask) except Exception: # Fixme: put in tuple of possible exceptions? pass if len(masks): mask = np.logical_and.reduce(masks) igood = mask.nonzero()[0] if len(igood) < nrecs: for i, x in enumerate(margs): if seqlist[i]: margs[i] = x[igood] for i, x in enumerate(margs): if seqlist[i] and isinstance(x, np.ma.MaskedArray): margs[i] = x.filled() return margs
def _combine_masks(*args): """ Find all masked and/or non-finite points in a set of arguments, and return the arguments as masked arrays with a common mask. Arguments can be in any of 5 categories: 1) 1-D masked arrays 2) 1-D ndarrays 3) ndarrays with more than one dimension 4) other non-string iterables 5) anything else The first argument must be in one of the first four categories; any argument with a length differing from that of the first argument (and hence anything in category 5) then will be passed through unchanged. Masks are obtained from all arguments of the correct length in categories 1, 2, and 4; a point is bad if masked in a masked array or if it is a nan or inf. No attempt is made to extract a mask from categories 2 and 4 if :meth:`np.isfinite` does not yield a Boolean array. Category 3 is included to support RGB or RGBA ndarrays, which are assumed to have only valid values and which are passed through unchanged. All input arguments that are not passed unchanged are returned as masked arrays if any masked points are found, otherwise as ndarrays. """ if not len(args): return () if is_scalar_or_string(args[0]): raise ValueError("First argument must be a sequence") nrecs = len(args[0]) margs = [] # Output args; some may be modified. seqlist = [False] * len(args) # Flags: True if output will be masked. masks = [] # List of masks. for i, x in enumerate(args): if is_scalar_or_string(x) or len(x) != nrecs: margs.append(x) # Leave it unmodified. else: if isinstance(x, np.ma.MaskedArray) and x.ndim > 1: raise ValueError("Masked arrays must be 1-D") try: x = np.asanyarray(x) except (np.VisibleDeprecationWarning, ValueError): # NumPy 1.19 raises a warning about ragged arrays, but we want # to accept basically anything here. x = np.asanyarray(x, dtype=object) if x.ndim == 1: x = safe_masked_invalid(x) seqlist[i] = True if np.ma.is_masked(x): masks.append(np.ma.getmaskarray(x)) margs.append(x) # Possibly modified. if len(masks): mask = np.logical_or.reduce(masks) for i, x in enumerate(margs): if seqlist[i]: margs[i] = np.ma.array(x, mask=mask) return margs
[docs]def boxplot_stats(X, whis=1.5, bootstrap=None, labels=None, autorange=False): r""" Return a list of dictionaries of statistics used to draw a series of box and whisker plots using `~.Axes.bxp`. Parameters ---------- X : array-like Data that will be represented in the boxplots. Should have 2 or fewer dimensions. whis : float or (float, float), default: 1.5 The position of the whiskers. If a float, the lower whisker is at the lowest datum above ``Q1 - whis*(Q3-Q1)``, and the upper whisker at the highest datum below ``Q3 + whis*(Q3-Q1)``, where Q1 and Q3 are the first and third quartiles. The default value of ``whis = 1.5`` corresponds to Tukey's original definition of boxplots. If a pair of floats, they indicate the percentiles at which to draw the whiskers (e.g., (5, 95)). In particular, setting this to (0, 100) results in whiskers covering the whole range of the data. "range" is a deprecated synonym for (0, 100). In the edge case where ``Q1 == Q3``, *whis* is automatically set to (0, 100) (cover the whole range of the data) if *autorange* is True. Beyond the whiskers, data are considered outliers and are plotted as individual points. bootstrap : int, optional Number of times the confidence intervals around the median should be bootstrapped (percentile method). labels : array-like, optional Labels for each dataset. Length must be compatible with dimensions of *X*. autorange : bool, optional (False) When `True` and the data are distributed such that the 25th and 75th percentiles are equal, ``whis`` is set to (0, 100) such that the whisker ends are at the minimum and maximum of the data. Returns ------- list of dict A list of dictionaries containing the results for each column of data. Keys of each dictionary are the following: ======== =================================== Key Value Description ======== =================================== label tick label for the boxplot mean arithmetic mean value med 50th percentile q1 first quartile (25th percentile) q3 third quartile (75th percentile) cilo lower notch around the median cihi upper notch around the median whislo end of the lower whisker whishi end of the upper whisker fliers outliers ======== =================================== Notes ----- Non-bootstrapping approach to confidence interval uses Gaussian-based asymptotic approximation: .. math:: \mathrm{med} \pm 1.57 \times \frac{\mathrm{iqr}}{\sqrt{N}} General approach from: McGill, R., Tukey, J.W., and Larsen, W.A. (1978) "Variations of Boxplots", The American Statistician, 32:12-16. """ def _bootstrap_median(data, N=5000): # determine 95% confidence intervals of the median M = len(data) percentiles = [2.5, 97.5] bs_index = np.random.randint(M, size=(N, M)) bsData = data[bs_index] estimate = np.median(bsData, axis=1, overwrite_input=True) CI = np.percentile(estimate, percentiles) return CI def _compute_conf_interval(data, med, iqr, bootstrap): if bootstrap is not None: # Do a bootstrap estimate of notch locations. # get conf. intervals around median CI = _bootstrap_median(data, N=bootstrap) notch_min = CI[0] notch_max = CI[1] else: N = len(data) notch_min = med - 1.57 * iqr / np.sqrt(N) notch_max = med + 1.57 * iqr / np.sqrt(N) return notch_min, notch_max # output is a list of dicts bxpstats = [] # convert X to a list of lists X = _reshape_2D(X, "X") ncols = len(X) if labels is None: labels = itertools.repeat(None) elif len(labels) != ncols: raise ValueError("Dimensions of labels and X must be compatible") input_whis = whis for ii, (x, label) in enumerate(zip(X, labels)): # empty dict stats = {} if label is not None: stats['label'] = label # restore whis to the input values in case it got changed in the loop whis = input_whis # note tricksiness, append up here and then mutate below bxpstats.append(stats) # if empty, bail if len(x) == 0: stats['fliers'] = np.array([]) stats['mean'] = np.nan stats['med'] = np.nan stats['q1'] = np.nan stats['q3'] = np.nan stats['cilo'] = np.nan stats['cihi'] = np.nan stats['whislo'] = np.nan stats['whishi'] = np.nan stats['med'] = np.nan continue # up-convert to an array, just to be safe x = np.asarray(x) # arithmetic mean stats['mean'] = np.mean(x) # medians and quartiles q1, med, q3 = np.percentile(x, [25, 50, 75]) # interquartile range stats['iqr'] = q3 - q1 if stats['iqr'] == 0 and autorange: whis = (0, 100) # conf. interval around median stats['cilo'], stats['cihi'] = _compute_conf_interval( x, med, stats['iqr'], bootstrap ) # lowest/highest non-outliers if np.isscalar(whis): if np.isreal(whis): loval = q1 - whis * stats['iqr'] hival = q3 + whis * stats['iqr'] elif whis in ['range', 'limit', 'limits', 'min/max']: warn_deprecated( "3.2", message=f"Setting whis to {whis!r} is deprecated " "since %(since)s and support for it will be removed " "%(removal)s; set it to [0, 100] to achieve the same " "effect.") loval = np.min(x) hival = np.max(x) else: raise ValueError('whis must be a float or list of percentiles') else: loval, hival = np.percentile(x, whis) # get high extreme wiskhi = x[x <= hival] if len(wiskhi) == 0 or np.max(wiskhi) < q3: stats['whishi'] = q3 else: stats['whishi'] = np.max(wiskhi) # get low extreme wisklo = x[x >= loval] if len(wisklo) == 0 or np.min(wisklo) > q1: stats['whislo'] = q1 else: stats['whislo'] = np.min(wisklo) # compute a single array of outliers stats['fliers'] = np.hstack([ x[x < stats['whislo']], x[x > stats['whishi']], ]) # add in the remaining stats stats['q1'], stats['med'], stats['q3'] = q1, med, q3 return bxpstats
# The ls_mapper maps short codes for line style to their full name used by # backends; the reverse mapper is for mapping full names to short ones. ls_mapper = {'-': 'solid', '--': 'dashed', '-.': 'dashdot', ':': 'dotted'} ls_mapper_r = {v: k for k, v in ls_mapper.items()}
[docs]def contiguous_regions(mask): """ Return a list of (ind0, ind1) such that ``mask[ind0:ind1].all()`` is True and we cover all such regions. """ mask = np.asarray(mask, dtype=bool) if not mask.size: return [] # Find the indices of region changes, and correct offset idx, = np.nonzero(mask[:-1] != mask[1:]) idx += 1 # List operations are faster for moderately sized arrays idx = idx.tolist() # Add first and/or last index if needed if mask[0]: idx = [0] + idx if mask[-1]: idx.append(len(mask)) return list(zip(idx[::2], idx[1::2]))
[docs]def is_math_text(s): """ Return whether the string *s* contains math expressions. This is done by checking whether *s* contains an even number of non-escaped dollar signs. """ s = str(s) dollar_count = s.count(r'$') - s.count(r'\$') even_dollars = (dollar_count > 0 and dollar_count % 2 == 0) return even_dollars
def _to_unmasked_float_array(x): """ Convert a sequence to a float array; if input was a masked array, masked values are converted to nans. """ if hasattr(x, 'mask'): return np.ma.asarray(x, float).filled(np.nan) else: return np.asarray(x, float) def _check_1d(x): """Convert scalars to 1d arrays; pass-through arrays as is.""" if not hasattr(x, 'shape') or len(x.shape) < 1: return np.atleast_1d(x) else: try: # work around # https://github.com/pandas-dev/pandas/issues/27775 which # means the shape of multi-dimensional slicing is not as # expected. That this ever worked was an unintentional # quirk of pandas and will raise an exception in the # future. This slicing warns in pandas >= 1.0rc0 via # https://github.com/pandas-dev/pandas/pull/30588 # # < 1.0rc0 : x[:, None].ndim == 1, no warning, custom type # >= 1.0rc1 : x[:, None].ndim == 2, warns, numpy array # future : x[:, None] -> raises # # This code should correctly identify and coerce to a # numpy array all pandas versions. with warnings.catch_warnings(record=True) as w: warnings.filterwarnings( "always", category=Warning, message='Support for multi-dimensional indexing') ndim = x[:, None].ndim # we have definitely hit a pandas index or series object # cast to a numpy array. if len(w) > 0: return np.asanyarray(x) # We have likely hit a pandas object, or at least # something where 2D slicing does not result in a 2D # object. if ndim < 2: return np.atleast_1d(x) return x # In pandas 1.1.0, multidimensional indexing leads to an # AssertionError for some Series objects, but should be # IndexError as described in # https://github.com/pandas-dev/pandas/issues/35527 except (AssertionError, IndexError, TypeError): return np.atleast_1d(x) def _reshape_2D(X, name): """ Use Fortran ordering to convert ndarrays and lists of iterables to lists of 1D arrays. Lists of iterables are converted by applying `np.asanyarray` to each of their elements. 1D ndarrays are returned in a singleton list containing them. 2D ndarrays are converted to the list of their *columns*. *name* is used to generate the error message for invalid inputs. """ # Iterate over columns for ndarrays. if isinstance(X, np.ndarray): X = X.T if len(X) == 0: return [[]] elif X.ndim == 1 and np.ndim(X[0]) == 0: # 1D array of scalars: directly return it. return [X] elif X.ndim in [1, 2]: # 2D array, or 1D array of iterables: flatten them first. return [np.reshape(x, -1) for x in X] else: raise ValueError(f'{name} must have 2 or fewer dimensions') # Iterate over list of iterables. if len(X) == 0: return [[]] result = [] is_1d = True for xi in X: # check if this is iterable, except for strings which we # treat as singletons. if (isinstance(xi, collections.abc.Iterable) and not isinstance(xi, str)): is_1d = False xi = np.asanyarray(xi) nd = np.ndim(xi) if nd > 1: raise ValueError(f'{name} must have 2 or fewer dimensions') result.append(xi.reshape(-1)) if is_1d: # 1D array of scalars: directly return it. return [np.reshape(result, -1)] else: # 2D array, or 1D array of iterables: use flattened version. return result
[docs]def violin_stats(X, method, points=100, quantiles=None): """ Return a list of dictionaries of data which can be used to draw a series of violin plots. See the ``Returns`` section below to view the required keys of the dictionary. Users can skip this function and pass a user-defined set of dictionaries with the same keys to `~.axes.Axes.violinplot` instead of using Matplotlib to do the calculations. See the *Returns* section below for the keys that must be present in the dictionaries. Parameters ---------- X : array-like Sample data that will be used to produce the gaussian kernel density estimates. Must have 2 or fewer dimensions. method : callable The method used to calculate the kernel density estimate for each column of data. When called via ``method(v, coords)``, it should return a vector of the values of the KDE evaluated at the values specified in coords. points : int, default: 100 Defines the number of points to evaluate each of the gaussian kernel density estimates at. quantiles : array-like, default: None Defines (if not None) a list of floats in interval [0, 1] for each column of data, which represents the quantiles that will be rendered for that column of data. Must have 2 or fewer dimensions. 1D array will be treated as a singleton list containing them. Returns ------- list of dict A list of dictionaries containing the results for each column of data. The dictionaries contain at least the following: - coords: A list of scalars containing the coordinates this particular kernel density estimate was evaluated at. - vals: A list of scalars containing the values of the kernel density estimate at each of the coordinates given in *coords*. - mean: The mean value for this column of data. - median: The median value for this column of data. - min: The minimum value for this column of data. - max: The maximum value for this column of data. - quantiles: The quantile values for this column of data. """ # List of dictionaries describing each of the violins. vpstats = [] # Want X to be a list of data sequences X = _reshape_2D(X, "X") # Want quantiles to be as the same shape as data sequences if quantiles is not None and len(quantiles) != 0: quantiles = _reshape_2D(quantiles, "quantiles") # Else, mock quantiles if is none or empty else: quantiles = [[]] * len(X) # quantiles should has the same size as dataset if len(X) != len(quantiles): raise ValueError("List of violinplot statistics and quantiles values" " must have the same length") # Zip x and quantiles for (x, q) in zip(X, quantiles): # Dictionary of results for this distribution stats = {} # Calculate basic stats for the distribution min_val = np.min(x) max_val = np.max(x) quantile_val = np.percentile(x, 100 * q) # Evaluate the kernel density estimate coords = np.linspace(min_val, max_val, points) stats['vals'] = method(x, coords) stats['coords'] = coords # Store additional statistics for this distribution stats['mean'] = np.mean(x) stats['median'] = np.median(x) stats['min'] = min_val stats['max'] = max_val stats['quantiles'] = np.atleast_1d(quantile_val) # Append to output vpstats.append(stats) return vpstats
[docs]def pts_to_prestep(x, *args): """ Convert continuous line to pre-steps. Given a set of ``N`` points, convert to ``2N - 1`` points, which when connected linearly give a step function which changes values at the beginning of the intervals. Parameters ---------- x : array The x location of the steps. May be empty. y1, ..., yp : array y arrays to be turned into steps; all must be the same length as ``x``. Returns ------- array The x and y values converted to steps in the same order as the input; can be unpacked as ``x_out, y1_out, ..., yp_out``. If the input is length ``N``, each of these arrays will be length ``2N + 1``. For ``N=0``, the length will be 0. Examples -------- >>> x_s, y1_s, y2_s = pts_to_prestep(x, y1, y2) """ steps = np.zeros((1 + len(args), max(2 * len(x) - 1, 0))) # In all `pts_to_*step` functions, only assign once using *x* and *args*, # as converting to an array may be expensive. steps[0, 0::2] = x steps[0, 1::2] = steps[0, 0:-2:2] steps[1:, 0::2] = args steps[1:, 1::2] = steps[1:, 2::2] return steps
[docs]def pts_to_poststep(x, *args): """ Convert continuous line to post-steps. Given a set of ``N`` points convert to ``2N + 1`` points, which when connected linearly give a step function which changes values at the end of the intervals. Parameters ---------- x : array The x location of the steps. May be empty. y1, ..., yp : array y arrays to be turned into steps; all must be the same length as ``x``. Returns ------- array The x and y values converted to steps in the same order as the input; can be unpacked as ``x_out, y1_out, ..., yp_out``. If the input is length ``N``, each of these arrays will be length ``2N + 1``. For ``N=0``, the length will be 0. Examples -------- >>> x_s, y1_s, y2_s = pts_to_poststep(x, y1, y2) """ steps = np.zeros((1 + len(args), max(2 * len(x) - 1, 0))) steps[0, 0::2] = x steps[0, 1::2] = steps[0, 2::2] steps[1:, 0::2] = args steps[1:, 1::2] = steps[1:, 0:-2:2] return steps
[docs]def pts_to_midstep(x, *args): """ Convert continuous line to mid-steps. Given a set of ``N`` points convert to ``2N`` points which when connected linearly give a step function which changes values at the middle of the intervals. Parameters ---------- x : array The x location of the steps. May be empty. y1, ..., yp : array y arrays to be turned into steps; all must be the same length as ``x``. Returns ------- array The x and y values converted to steps in the same order as the input; can be unpacked as ``x_out, y1_out, ..., yp_out``. If the input is length ``N``, each of these arrays will be length ``2N``. Examples -------- >>> x_s, y1_s, y2_s = pts_to_midstep(x, y1, y2) """ steps = np.zeros((1 + len(args), 2 * len(x))) x = np.asanyarray(x) steps[0, 1:-1:2] = steps[0, 2::2] = (x[:-1] + x[1:]) / 2 steps[0, :1] = x[:1] # Also works for zero-sized input. steps[0, -1:] = x[-1:] steps[1:, 0::2] = args steps[1:, 1::2] = steps[1:, 0::2] return steps
STEP_LOOKUP_MAP = {'default': lambda x, y: (x, y), 'steps': pts_to_prestep, 'steps-pre': pts_to_prestep, 'steps-post': pts_to_poststep, 'steps-mid': pts_to_midstep}
[docs]def index_of(y): """ A helper function to create reasonable x values for the given *y*. This is used for plotting (x, y) if x values are not explicitly given. First try ``y.index`` (assuming *y* is a `pandas.Series`), if that fails, use ``range(len(y))``. This will be extended in the future to deal with more types of labeled data. Parameters ---------- y : float or array-like Returns ------- x, y : ndarray The x and y values to plot. """ try: return y.index.values, y.values except AttributeError: pass try: y = _check_1d(y) except (np.VisibleDeprecationWarning, ValueError): # NumPy 1.19 will warn on ragged input, and we can't actually use it. pass else: return np.arange(y.shape[0], dtype=float), y raise ValueError('Input could not be cast to an at-least-1D NumPy array')
[docs]def safe_first_element(obj): """ Return the first element in *obj*. This is an type-independent way of obtaining the first element, supporting both index access and the iterator protocol. """ if isinstance(obj, collections.abc.Iterator): # needed to accept `array.flat` as input. # np.flatiter reports as an instance of collections.Iterator # but can still be indexed via []. # This has the side effect of re-setting the iterator, but # that is acceptable. try: return obj[0] except TypeError: pass raise RuntimeError("matplotlib does not support generators " "as input") return next(iter(obj))
[docs]def sanitize_sequence(data): """ Convert dictview objects to list. Other inputs are returned unchanged. """ return (list(data) if isinstance(data, collections.abc.MappingView) else data)
[docs]@_delete_parameter("3.3", "required") @_delete_parameter("3.3", "forbidden") @_delete_parameter("3.3", "allowed") def normalize_kwargs(kw, alias_mapping=None, required=(), forbidden=(), allowed=None): """ Helper function to normalize kwarg inputs. The order they are resolved are: 1. aliasing 2. required 3. forbidden 4. allowed This order means that only the canonical names need appear in *allowed*, *forbidden*, *required*. Parameters ---------- kw : dict A dict of keyword arguments. alias_mapping : dict or Artist subclass or Artist instance, optional A mapping between a canonical name to a list of aliases, in order of precedence from lowest to highest. If the canonical value is not in the list it is assumed to have the highest priority. If an Artist subclass or instance is passed, use its properties alias mapping. required : list of str, optional A list of keys that must be in *kws*. This parameter is deprecated. forbidden : list of str, optional A list of keys which may not be in *kw*. This parameter is deprecated. allowed : list of str, optional A list of allowed fields. If this not None, then raise if *kw* contains any keys not in the union of *required* and *allowed*. To allow only the required fields pass in an empty tuple ``allowed=()``. This parameter is deprecated. Raises ------ TypeError To match what python raises if invalid args/kwargs are passed to a callable. """ from matplotlib.artist import Artist # deal with default value of alias_mapping if alias_mapping is None: alias_mapping = dict() elif (isinstance(alias_mapping, type) and issubclass(alias_mapping, Artist) or isinstance(alias_mapping, Artist)): alias_mapping = getattr(alias_mapping, "_alias_map", {}) to_canonical = {alias: canonical for canonical, alias_list in alias_mapping.items() for alias in alias_list} canonical_to_seen = {} ret = {} # output dictionary for k, v in kw.items(): canonical = to_canonical.get(k, k) if canonical in canonical_to_seen: raise TypeError(f"Got both {canonical_to_seen[canonical]!r} and " f"{k!r}, which are aliases of one another") canonical_to_seen[canonical] = k ret[canonical] = v fail_keys = [k for k in required if k not in ret] if fail_keys: raise TypeError("The required keys {keys!r} " "are not in kwargs".format(keys=fail_keys)) fail_keys = [k for k in forbidden if k in ret] if fail_keys: raise TypeError("The forbidden keys {keys!r} " "are in kwargs".format(keys=fail_keys)) if allowed is not None: allowed_set = {*required, *allowed} fail_keys = [k for k in ret if k not in allowed_set] if fail_keys: raise TypeError( "kwargs contains {keys!r} which are not in the required " "{req!r} or allowed {allow!r} keys".format( keys=fail_keys, req=required, allow=allowed)) return ret
@contextlib.contextmanager def _lock_path(path): """ Context manager for locking a path. Usage:: with _lock_path(path): ... Another thread or process that attempts to lock the same path will wait until this context manager is exited. The lock is implemented by creating a temporary file in the parent directory, so that directory must exist and be writable. """ path = Path(path) lock_path = path.with_name(path.name + ".matplotlib-lock") retries = 50 sleeptime = 0.1 for _ in range(retries): try: with lock_path.open("xb"): break except FileExistsError: time.sleep(sleeptime) else: raise TimeoutError("""\ Lock error: Matplotlib failed to acquire the following lock file: {} This maybe due to another process holding this lock file. If you are sure no other Matplotlib process is running, remove this file and try again.""".format( lock_path)) try: yield finally: lock_path.unlink() def _topmost_artist( artists, _cached_max=functools.partial(max, key=operator.attrgetter("zorder"))): """ Get the topmost artist of a list. In case of a tie, return the *last* of the tied artists, as it will be drawn on top of the others. `max` returns the first maximum in case of ties, so we need to iterate over the list in reverse order. """ return _cached_max(reversed(artists)) def _str_equal(obj, s): """ Return whether *obj* is a string equal to string *s*. This helper solely exists to handle the case where *obj* is a numpy array, because in such cases, a naive ``obj == s`` would yield an array, which cannot be used in a boolean context. """ return isinstance(obj, str) and obj == s def _str_lower_equal(obj, s): """ Return whether *obj* is a string equal, when lowercased, to string *s*. This helper solely exists to handle the case where *obj* is a numpy array, because in such cases, a naive ``obj == s`` would yield an array, which cannot be used in a boolean context. """ return isinstance(obj, str) and obj.lower() == s def _define_aliases(alias_d, cls=None): """ Class decorator for defining property aliases. Use as :: @cbook._define_aliases({"property": ["alias", ...], ...}) class C: ... For each property, if the corresponding ``get_property`` is defined in the class so far, an alias named ``get_alias`` will be defined; the same will be done for setters. If neither the getter nor the setter exists, an exception will be raised. The alias map is stored as the ``_alias_map`` attribute on the class and can be used by `~.normalize_kwargs` (which assumes that higher priority aliases come last). """ if cls is None: # Return the actual class decorator. return functools.partial(_define_aliases, alias_d) def make_alias(name): # Enforce a closure over *name*. @functools.wraps(getattr(cls, name)) def method(self, *args, **kwargs): return getattr(self, name)(*args, **kwargs) return method for prop, aliases in alias_d.items(): exists = False for prefix in ["get_", "set_"]: if prefix + prop in vars(cls): exists = True for alias in aliases: method = make_alias(prefix + prop) method.__name__ = prefix + alias method.__doc__ = "Alias for `{}`.".format(prefix + prop) setattr(cls, prefix + alias, method) if not exists: raise ValueError( "Neither getter nor setter exists for {!r}".format(prop)) def get_aliased_and_aliases(d): return {*d, *(alias for aliases in d.values() for alias in aliases)} preexisting_aliases = getattr(cls, "_alias_map", {}) conflicting = (get_aliased_and_aliases(preexisting_aliases) & get_aliased_and_aliases(alias_d)) if conflicting: # Need to decide on conflict resolution policy. raise NotImplementedError( f"Parent class already defines conflicting aliases: {conflicting}") cls._alias_map = {**preexisting_aliases, **alias_d} return cls def _array_perimeter(arr): """ Get the elements on the perimeter of *arr*. Parameters ---------- arr : ndarray, shape (M, N) The input array. Returns ------- ndarray, shape (2*(M - 1) + 2*(N - 1),) The elements on the perimeter of the array:: [arr[0, 0], ..., arr[0, -1], ..., arr[-1, -1], ..., arr[-1, 0], ...] Examples -------- >>> i, j = np.ogrid[:3,:4] >>> a = i*10 + j >>> a array([[ 0, 1, 2, 3], [10, 11, 12, 13], [20, 21, 22, 23]]) >>> _array_perimeter(a) array([ 0, 1, 2, 3, 13, 23, 22, 21, 20, 10]) """ # note we use Python's half-open ranges to avoid repeating # the corners forward = np.s_[0:-1] # [0 ... -1) backward = np.s_[-1:0:-1] # [-1 ... 0) return np.concatenate(( arr[0, forward], arr[forward, -1], arr[-1, backward], arr[backward, 0], )) def _unfold(arr, axis, size, step): """ Append an extra dimension containing sliding windows along *axis*. All windows are of size *size* and begin with every *step* elements. Parameters ---------- arr : ndarray, shape (N_1, ..., N_k) The input array axis : int Axis along which the windows are extracted size : int Size of the windows step : int Stride between first elements of subsequent windows. Returns ------- ndarray, shape (N_1, ..., 1 + (N_axis-size)/step, ..., N_k, size) Examples -------- >>> i, j = np.ogrid[:3,:7] >>> a = i*10 + j >>> a array([[ 0, 1, 2, 3, 4, 5, 6], [10, 11, 12, 13, 14, 15, 16], [20, 21, 22, 23, 24, 25, 26]]) >>> _unfold(a, axis=1, size=3, step=2) array([[[ 0, 1, 2], [ 2, 3, 4], [ 4, 5, 6]], [[10, 11, 12], [12, 13, 14], [14, 15, 16]], [[20, 21, 22], [22, 23, 24], [24, 25, 26]]]) """ new_shape = [*arr.shape, size] new_strides = [*arr.strides, arr.strides[axis]] new_shape[axis] = (new_shape[axis] - size) // step + 1 new_strides[axis] = new_strides[axis] * step return np.lib.stride_tricks.as_strided(arr, shape=new_shape, strides=new_strides, writeable=False) def _array_patch_perimeters(x, rstride, cstride): """ Extract perimeters of patches from *arr*. Extracted patches are of size (*rstride* + 1) x (*cstride* + 1) and share perimeters with their neighbors. The ordering of the vertices matches that returned by ``_array_perimeter``. Parameters ---------- x : ndarray, shape (N, M) Input array rstride : int Vertical (row) stride between corresponding elements of each patch cstride : int Horizontal (column) stride between corresponding elements of each patch Returns ------- ndarray, shape (N/rstride * M/cstride, 2 * (rstride + cstride)) """ assert rstride > 0 and cstride > 0 assert (x.shape[0] - 1) % rstride == 0 assert (x.shape[1] - 1) % cstride == 0 # We build up each perimeter from four half-open intervals. Here is an # illustrated explanation for rstride == cstride == 3 # # T T T R # L R # L R # L B B B # # where T means that this element will be in the top array, R for right, # B for bottom and L for left. Each of the arrays below has a shape of: # # (number of perimeters that can be extracted vertically, # number of perimeters that can be extracted horizontally, # cstride for top and bottom and rstride for left and right) # # Note that _unfold doesn't incur any memory copies, so the only costly # operation here is the np.concatenate. top = _unfold(x[:-1:rstride, :-1], 1, cstride, cstride) bottom = _unfold(x[rstride::rstride, 1:], 1, cstride, cstride)[..., ::-1] right = _unfold(x[:-1, cstride::cstride], 0, rstride, rstride) left = _unfold(x[1:, :-1:cstride], 0, rstride, rstride)[..., ::-1] return (np.concatenate((top, right, bottom, left), axis=2) .reshape(-1, 2 * (rstride + cstride))) @contextlib.contextmanager def _setattr_cm(obj, **kwargs): """ Temporarily set some attributes; restore original state at context exit. """ sentinel = object() origs = {} for attr in kwargs: orig = getattr(obj, attr, sentinel) if attr in obj.__dict__ or orig is sentinel: # if we are pulling from the instance dict or the object # does not have this attribute we can trust the above origs[attr] = orig else: # if the attribute is not in the instance dict it must be # from the class level cls_orig = getattr(type(obj), attr) # if we are dealing with a property (but not a general descriptor) # we want to set the original value back. if isinstance(cls_orig, property): origs[attr] = orig # otherwise this is _something_ we are going to shadow at # the instance dict level from higher up in the MRO. We # are going to assume we can delattr(obj, attr) to clean # up after ourselves. It is possible that this code will # fail if used with a non-property custom descriptor which # implements __set__ (and __delete__ does not act like a # stack). However, this is an internal tool and we do not # currently have any custom descriptors. else: origs[attr] = sentinel try: for attr, val in kwargs.items(): setattr(obj, attr, val) yield finally: for attr, orig in origs.items(): if orig is sentinel: delattr(obj, attr) else: setattr(obj, attr, orig) def _warn_external(message, category=None): """ `warnings.warn` wrapper that sets *stacklevel* to "outside Matplotlib". The original emitter of the warning can be obtained by patching this function back to `warnings.warn`, i.e. ``cbook._warn_external = warnings.warn`` (or ``functools.partial(warnings.warn, stacklevel=2)``, etc.). """ frame = sys._getframe() for stacklevel in itertools.count(1): # lgtm[py/unused-loop-variable] if frame is None: # when called in embedded context may hit frame is None break if not re.match(r"\A(matplotlib|mpl_toolkits)(\Z|\.(?!tests\.))", # Work around sphinx-gallery not setting __name__. frame.f_globals.get("__name__", "")): break frame = frame.f_back warnings.warn(message, category, stacklevel) class _OrderedSet(collections.abc.MutableSet): def __init__(self): self._od = collections.OrderedDict() def __contains__(self, key): return key in self._od def __iter__(self): return iter(self._od) def __len__(self): return len(self._od) def add(self, key): self._od.pop(key, None) self._od[key] = None def discard(self, key): self._od.pop(key, None) # Agg's buffers are unmultiplied RGBA8888, which neither PyQt4 nor cairo # support; however, both do support premultiplied ARGB32. def _premultiplied_argb32_to_unmultiplied_rgba8888(buf): """ Convert a premultiplied ARGB32 buffer to an unmultiplied RGBA8888 buffer. """ rgba = np.take( # .take() ensures C-contiguity of the result. buf, [2, 1, 0, 3] if sys.byteorder == "little" else [1, 2, 3, 0], axis=2) rgb = rgba[..., :-1] alpha = rgba[..., -1] # Un-premultiply alpha. The formula is the same as in cairo-png.c. mask = alpha != 0 for channel in np.rollaxis(rgb, -1): channel[mask] = ( (channel[mask].astype(int) * 255 + alpha[mask] // 2) // alpha[mask]) return rgba def _unmultiplied_rgba8888_to_premultiplied_argb32(rgba8888): """ Convert an unmultiplied RGBA8888 buffer to a premultiplied ARGB32 buffer. """ if sys.byteorder == "little": argb32 = np.take(rgba8888, [2, 1, 0, 3], axis=2) rgb24 = argb32[..., :-1] alpha8 = argb32[..., -1:] else: argb32 = np.take(rgba8888, [3, 0, 1, 2], axis=2) alpha8 = argb32[..., :1] rgb24 = argb32[..., 1:] # Only bother premultiplying when the alpha channel is not fully opaque, # as the cost is not negligible. The unsafe cast is needed to do the # multiplication in-place in an integer buffer. if alpha8.min() != 0xff: np.multiply(rgb24, alpha8 / 0xff, out=rgb24, casting="unsafe") return argb32 def _pformat_subprocess(command): """Pretty-format a subprocess command for printing/logging purposes.""" return (command if isinstance(command, str) else " ".join(shlex.quote(os.fspath(arg)) for arg in command)) def _check_and_log_subprocess(command, logger, **kwargs): """ Run *command*, returning its stdout output if it succeeds. If it fails (exits with nonzero return code), raise an exception whose text includes the failed command and captured stdout and stderr output. Regardless of the return code, the command is logged at DEBUG level on *logger*. In case of success, the output is likewise logged. """ logger.debug('%s', _pformat_subprocess(command)) proc = subprocess.run( command, stdout=subprocess.PIPE, stderr=subprocess.PIPE, **kwargs) if proc.returncode: stdout = proc.stdout if isinstance(stdout, bytes): stdout = stdout.decode() stderr = proc.stderr if isinstance(stderr, bytes): stderr = stderr.decode() raise RuntimeError( f"The command\n" f" {_pformat_subprocess(command)}\n" f"failed and generated the following output:\n" f"{stdout}\n" f"and the following error:\n" f"{stderr}") if proc.stdout: logger.debug("stdout:\n%s", proc.stdout) if proc.stderr: logger.debug("stderr:\n%s", proc.stderr) return proc.stdout # In the following _check_foo functions, the first parameter starts with an # underscore because it is intended to be positional-only (e.g., so that # `_check_isinstance([...], types=foo)` doesn't fail. def _check_isinstance(_types, **kwargs): """ For each *key, value* pair in *kwargs*, check that *value* is an instance of one of *_types*; if not, raise an appropriate TypeError. As a special case, a ``None`` entry in *_types* is treated as NoneType. Examples -------- >>> cbook._check_isinstance((SomeClass, None), arg=arg) """ types = _types if isinstance(types, type) or types is None: types = (types,) none_allowed = None in types types = tuple(tp for tp in types if tp is not None) def type_name(tp): return (tp.__qualname__ if tp.__module__ == "builtins" else f"{tp.__module__}.{tp.__qualname__}") names = [*map(type_name, types)] if none_allowed: types = (*types, type(None)) names.append("None") for k, v in kwargs.items(): if not isinstance(v, types): raise TypeError( "{!r} must be an instance of {}, not a {}".format( k, ", ".join(names[:-1]) + " or " + names[-1] if len(names) > 1 else names[0], type_name(type(v)))) def _check_in_list(_values, **kwargs): """ For each *key, value* pair in *kwargs*, check that *value* is in *_values*; if not, raise an appropriate ValueError. Examples -------- >>> cbook._check_in_list(["foo", "bar"], arg=arg, other_arg=other_arg) """ values = _values for k, v in kwargs.items(): if v not in values: raise ValueError( "{!r} is not a valid value for {}; supported values are {}" .format(v, k, ', '.join(map(repr, values)))) def _check_shape(_shape, **kwargs): """ For each *key, value* pair in *kwargs*, check that *value* has the shape *_shape*, if not, raise an appropriate ValueError. *None* in the shape is treated as a "free" size that can have any length. e.g. (None, 2) -> (N, 2) The values checked must be numpy arrays. Examples -------- To check for (N, 2) shaped arrays >>> cbook._check_in_list((None, 2), arg=arg, other_arg=other_arg) """ target_shape = _shape for k, v in kwargs.items(): data_shape = v.shape if len(target_shape) != len(data_shape) or any( t not in [s, None] for t, s in zip(target_shape, data_shape) ): dim_labels = iter(itertools.chain( 'MNLIJKLH', (f"D{i}" for i in itertools.count()))) text_shape = ", ".join((str(n) if n is not None else next(dim_labels) for n in target_shape)) raise ValueError( f"{k!r} must be {len(target_shape)}D " f"with shape ({text_shape}). " f"Your input has shape {v.shape}." ) def _check_getitem(_mapping, **kwargs): """ *kwargs* must consist of a single *key, value* pair. If *key* is in *_mapping*, return ``_mapping[value]``; else, raise an appropriate ValueError. Examples -------- >>> cbook._check_getitem({"foo": "bar"}, arg=arg) """ mapping = _mapping if len(kwargs) != 1: raise ValueError("_check_getitem takes a single keyword argument") (k, v), = kwargs.items() try: return mapping[v] except KeyError: raise ValueError( "{!r} is not a valid value for {}; supported values are {}" .format(v, k, ', '.join(map(repr, mapping)))) from None class _classproperty: """ Like `property`, but also triggers on access via the class, and it is the *class* that's passed as argument. Examples -------- :: class C: @classproperty def foo(cls): return cls.__name__ assert C.foo == "C" """ def __init__(self, fget): self._fget = fget def __get__(self, instance, owner): return self._fget(owner) def _backend_module_name(name): """ Convert a backend name (either a standard backend -- "Agg", "TkAgg", ... -- or a custom backend -- "module://...") to the corresponding module name). """ return (name[9:] if name.startswith("module://") else "matplotlib.backends.backend_{}".format(name.lower()))