Add python venv

This commit is contained in:
Isaac Shoebottom
2022-10-31 10:10:52 -03:00
parent fb1a0435c1
commit a50f49d2c8
913 changed files with 287881 additions and 0 deletions

View File

@ -0,0 +1,4 @@
from .more import * # noqa
from .recipes import * # noqa
__version__ = '8.8.0'

File diff suppressed because it is too large Load Diff

View File

@ -0,0 +1,620 @@
"""Imported from the recipes section of the itertools documentation.
All functions taken from the recipes section of the itertools library docs
[1]_.
Some backward-compatible usability improvements have been made.
.. [1] http://docs.python.org/library/itertools.html#recipes
"""
import warnings
from collections import deque
from itertools import (
chain,
combinations,
count,
cycle,
groupby,
islice,
repeat,
starmap,
tee,
zip_longest,
)
import operator
from random import randrange, sample, choice
__all__ = [
'all_equal',
'consume',
'convolve',
'dotproduct',
'first_true',
'flatten',
'grouper',
'iter_except',
'ncycles',
'nth',
'nth_combination',
'padnone',
'pad_none',
'pairwise',
'partition',
'powerset',
'prepend',
'quantify',
'random_combination_with_replacement',
'random_combination',
'random_permutation',
'random_product',
'repeatfunc',
'roundrobin',
'tabulate',
'tail',
'take',
'unique_everseen',
'unique_justseen',
]
def take(n, iterable):
"""Return first *n* items of the iterable as a list.
>>> take(3, range(10))
[0, 1, 2]
If there are fewer than *n* items in the iterable, all of them are
returned.
>>> take(10, range(3))
[0, 1, 2]
"""
return list(islice(iterable, n))
def tabulate(function, start=0):
"""Return an iterator over the results of ``func(start)``,
``func(start + 1)``, ``func(start + 2)``...
*func* should be a function that accepts one integer argument.
If *start* is not specified it defaults to 0. It will be incremented each
time the iterator is advanced.
>>> square = lambda x: x ** 2
>>> iterator = tabulate(square, -3)
>>> take(4, iterator)
[9, 4, 1, 0]
"""
return map(function, count(start))
def tail(n, iterable):
"""Return an iterator over the last *n* items of *iterable*.
>>> t = tail(3, 'ABCDEFG')
>>> list(t)
['E', 'F', 'G']
"""
return iter(deque(iterable, maxlen=n))
def consume(iterator, n=None):
"""Advance *iterable* by *n* steps. If *n* is ``None``, consume it
entirely.
Efficiently exhausts an iterator without returning values. Defaults to
consuming the whole iterator, but an optional second argument may be
provided to limit consumption.
>>> i = (x for x in range(10))
>>> next(i)
0
>>> consume(i, 3)
>>> next(i)
4
>>> consume(i)
>>> next(i)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
StopIteration
If the iterator has fewer items remaining than the provided limit, the
whole iterator will be consumed.
>>> i = (x for x in range(3))
>>> consume(i, 5)
>>> next(i)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
StopIteration
"""
# Use functions that consume iterators at C speed.
if n is None:
# feed the entire iterator into a zero-length deque
deque(iterator, maxlen=0)
else:
# advance to the empty slice starting at position n
next(islice(iterator, n, n), None)
def nth(iterable, n, default=None):
"""Returns the nth item or a default value.
>>> l = range(10)
>>> nth(l, 3)
3
>>> nth(l, 20, "zebra")
'zebra'
"""
return next(islice(iterable, n, None), default)
def all_equal(iterable):
"""
Returns ``True`` if all the elements are equal to each other.
>>> all_equal('aaaa')
True
>>> all_equal('aaab')
False
"""
g = groupby(iterable)
return next(g, True) and not next(g, False)
def quantify(iterable, pred=bool):
"""Return the how many times the predicate is true.
>>> quantify([True, False, True])
2
"""
return sum(map(pred, iterable))
def pad_none(iterable):
"""Returns the sequence of elements and then returns ``None`` indefinitely.
>>> take(5, pad_none(range(3)))
[0, 1, 2, None, None]
Useful for emulating the behavior of the built-in :func:`map` function.
See also :func:`padded`.
"""
return chain(iterable, repeat(None))
padnone = pad_none
def ncycles(iterable, n):
"""Returns the sequence elements *n* times
>>> list(ncycles(["a", "b"], 3))
['a', 'b', 'a', 'b', 'a', 'b']
"""
return chain.from_iterable(repeat(tuple(iterable), n))
def dotproduct(vec1, vec2):
"""Returns the dot product of the two iterables.
>>> dotproduct([10, 10], [20, 20])
400
"""
return sum(map(operator.mul, vec1, vec2))
def flatten(listOfLists):
"""Return an iterator flattening one level of nesting in a list of lists.
>>> list(flatten([[0, 1], [2, 3]]))
[0, 1, 2, 3]
See also :func:`collapse`, which can flatten multiple levels of nesting.
"""
return chain.from_iterable(listOfLists)
def repeatfunc(func, times=None, *args):
"""Call *func* with *args* repeatedly, returning an iterable over the
results.
If *times* is specified, the iterable will terminate after that many
repetitions:
>>> from operator import add
>>> times = 4
>>> args = 3, 5
>>> list(repeatfunc(add, times, *args))
[8, 8, 8, 8]
If *times* is ``None`` the iterable will not terminate:
>>> from random import randrange
>>> times = None
>>> args = 1, 11
>>> take(6, repeatfunc(randrange, times, *args)) # doctest:+SKIP
[2, 4, 8, 1, 8, 4]
"""
if times is None:
return starmap(func, repeat(args))
return starmap(func, repeat(args, times))
def _pairwise(iterable):
"""Returns an iterator of paired items, overlapping, from the original
>>> take(4, pairwise(count()))
[(0, 1), (1, 2), (2, 3), (3, 4)]
On Python 3.10 and above, this is an alias for :func:`itertools.pairwise`.
"""
a, b = tee(iterable)
next(b, None)
yield from zip(a, b)
try:
from itertools import pairwise as itertools_pairwise
except ImportError:
pairwise = _pairwise
else:
def pairwise(iterable):
yield from itertools_pairwise(iterable)
pairwise.__doc__ = _pairwise.__doc__
def grouper(iterable, n, fillvalue=None):
"""Collect data into fixed-length chunks or blocks.
>>> list(grouper('ABCDEFG', 3, 'x'))
[('A', 'B', 'C'), ('D', 'E', 'F'), ('G', 'x', 'x')]
"""
if isinstance(iterable, int):
warnings.warn(
"grouper expects iterable as first parameter", DeprecationWarning
)
n, iterable = iterable, n
args = [iter(iterable)] * n
return zip_longest(fillvalue=fillvalue, *args)
def roundrobin(*iterables):
"""Yields an item from each iterable, alternating between them.
>>> list(roundrobin('ABC', 'D', 'EF'))
['A', 'D', 'E', 'B', 'F', 'C']
This function produces the same output as :func:`interleave_longest`, but
may perform better for some inputs (in particular when the number of
iterables is small).
"""
# Recipe credited to George Sakkis
pending = len(iterables)
nexts = cycle(iter(it).__next__ for it in iterables)
while pending:
try:
for next in nexts:
yield next()
except StopIteration:
pending -= 1
nexts = cycle(islice(nexts, pending))
def partition(pred, iterable):
"""
Returns a 2-tuple of iterables derived from the input iterable.
The first yields the items that have ``pred(item) == False``.
The second yields the items that have ``pred(item) == True``.
>>> is_odd = lambda x: x % 2 != 0
>>> iterable = range(10)
>>> even_items, odd_items = partition(is_odd, iterable)
>>> list(even_items), list(odd_items)
([0, 2, 4, 6, 8], [1, 3, 5, 7, 9])
If *pred* is None, :func:`bool` is used.
>>> iterable = [0, 1, False, True, '', ' ']
>>> false_items, true_items = partition(None, iterable)
>>> list(false_items), list(true_items)
([0, False, ''], [1, True, ' '])
"""
if pred is None:
pred = bool
evaluations = ((pred(x), x) for x in iterable)
t1, t2 = tee(evaluations)
return (
(x for (cond, x) in t1 if not cond),
(x for (cond, x) in t2 if cond),
)
def powerset(iterable):
"""Yields all possible subsets of the iterable.
>>> list(powerset([1, 2, 3]))
[(), (1,), (2,), (3,), (1, 2), (1, 3), (2, 3), (1, 2, 3)]
:func:`powerset` will operate on iterables that aren't :class:`set`
instances, so repeated elements in the input will produce repeated elements
in the output. Use :func:`unique_everseen` on the input to avoid generating
duplicates:
>>> seq = [1, 1, 0]
>>> list(powerset(seq))
[(), (1,), (1,), (0,), (1, 1), (1, 0), (1, 0), (1, 1, 0)]
>>> from more_itertools import unique_everseen
>>> list(powerset(unique_everseen(seq)))
[(), (1,), (0,), (1, 0)]
"""
s = list(iterable)
return chain.from_iterable(combinations(s, r) for r in range(len(s) + 1))
def unique_everseen(iterable, key=None):
"""
Yield unique elements, preserving order.
>>> list(unique_everseen('AAAABBBCCDAABBB'))
['A', 'B', 'C', 'D']
>>> list(unique_everseen('ABBCcAD', str.lower))
['A', 'B', 'C', 'D']
Sequences with a mix of hashable and unhashable items can be used.
The function will be slower (i.e., `O(n^2)`) for unhashable items.
Remember that ``list`` objects are unhashable - you can use the *key*
parameter to transform the list to a tuple (which is hashable) to
avoid a slowdown.
>>> iterable = ([1, 2], [2, 3], [1, 2])
>>> list(unique_everseen(iterable)) # Slow
[[1, 2], [2, 3]]
>>> list(unique_everseen(iterable, key=tuple)) # Faster
[[1, 2], [2, 3]]
Similary, you may want to convert unhashable ``set`` objects with
``key=frozenset``. For ``dict`` objects,
``key=lambda x: frozenset(x.items())`` can be used.
"""
seenset = set()
seenset_add = seenset.add
seenlist = []
seenlist_add = seenlist.append
use_key = key is not None
for element in iterable:
k = key(element) if use_key else element
try:
if k not in seenset:
seenset_add(k)
yield element
except TypeError:
if k not in seenlist:
seenlist_add(k)
yield element
def unique_justseen(iterable, key=None):
"""Yields elements in order, ignoring serial duplicates
>>> list(unique_justseen('AAAABBBCCDAABBB'))
['A', 'B', 'C', 'D', 'A', 'B']
>>> list(unique_justseen('ABBCcAD', str.lower))
['A', 'B', 'C', 'A', 'D']
"""
return map(next, map(operator.itemgetter(1), groupby(iterable, key)))
def iter_except(func, exception, first=None):
"""Yields results from a function repeatedly until an exception is raised.
Converts a call-until-exception interface to an iterator interface.
Like ``iter(func, sentinel)``, but uses an exception instead of a sentinel
to end the loop.
>>> l = [0, 1, 2]
>>> list(iter_except(l.pop, IndexError))
[2, 1, 0]
"""
try:
if first is not None:
yield first()
while 1:
yield func()
except exception:
pass
def first_true(iterable, default=None, pred=None):
"""
Returns the first true value in the iterable.
If no true value is found, returns *default*
If *pred* is not None, returns the first item for which
``pred(item) == True`` .
>>> first_true(range(10))
1
>>> first_true(range(10), pred=lambda x: x > 5)
6
>>> first_true(range(10), default='missing', pred=lambda x: x > 9)
'missing'
"""
return next(filter(pred, iterable), default)
def random_product(*args, repeat=1):
"""Draw an item at random from each of the input iterables.
>>> random_product('abc', range(4), 'XYZ') # doctest:+SKIP
('c', 3, 'Z')
If *repeat* is provided as a keyword argument, that many items will be
drawn from each iterable.
>>> random_product('abcd', range(4), repeat=2) # doctest:+SKIP
('a', 2, 'd', 3)
This equivalent to taking a random selection from
``itertools.product(*args, **kwarg)``.
"""
pools = [tuple(pool) for pool in args] * repeat
return tuple(choice(pool) for pool in pools)
def random_permutation(iterable, r=None):
"""Return a random *r* length permutation of the elements in *iterable*.
If *r* is not specified or is ``None``, then *r* defaults to the length of
*iterable*.
>>> random_permutation(range(5)) # doctest:+SKIP
(3, 4, 0, 1, 2)
This equivalent to taking a random selection from
``itertools.permutations(iterable, r)``.
"""
pool = tuple(iterable)
r = len(pool) if r is None else r
return tuple(sample(pool, r))
def random_combination(iterable, r):
"""Return a random *r* length subsequence of the elements in *iterable*.
>>> random_combination(range(5), 3) # doctest:+SKIP
(2, 3, 4)
This equivalent to taking a random selection from
``itertools.combinations(iterable, r)``.
"""
pool = tuple(iterable)
n = len(pool)
indices = sorted(sample(range(n), r))
return tuple(pool[i] for i in indices)
def random_combination_with_replacement(iterable, r):
"""Return a random *r* length subsequence of elements in *iterable*,
allowing individual elements to be repeated.
>>> random_combination_with_replacement(range(3), 5) # doctest:+SKIP
(0, 0, 1, 2, 2)
This equivalent to taking a random selection from
``itertools.combinations_with_replacement(iterable, r)``.
"""
pool = tuple(iterable)
n = len(pool)
indices = sorted(randrange(n) for i in range(r))
return tuple(pool[i] for i in indices)
def nth_combination(iterable, r, index):
"""Equivalent to ``list(combinations(iterable, r))[index]``.
The subsequences of *iterable* that are of length *r* can be ordered
lexicographically. :func:`nth_combination` computes the subsequence at
sort position *index* directly, without computing the previous
subsequences.
>>> nth_combination(range(5), 3, 5)
(0, 3, 4)
``ValueError`` will be raised If *r* is negative or greater than the length
of *iterable*.
``IndexError`` will be raised if the given *index* is invalid.
"""
pool = tuple(iterable)
n = len(pool)
if (r < 0) or (r > n):
raise ValueError
c = 1
k = min(r, n - r)
for i in range(1, k + 1):
c = c * (n - k + i) // i
if index < 0:
index += c
if (index < 0) or (index >= c):
raise IndexError
result = []
while r:
c, n, r = c * r // n, n - 1, r - 1
while index >= c:
index -= c
c, n = c * (n - r) // n, n - 1
result.append(pool[-1 - n])
return tuple(result)
def prepend(value, iterator):
"""Yield *value*, followed by the elements in *iterator*.
>>> value = '0'
>>> iterator = ['1', '2', '3']
>>> list(prepend(value, iterator))
['0', '1', '2', '3']
To prepend multiple values, see :func:`itertools.chain`
or :func:`value_chain`.
"""
return chain([value], iterator)
def convolve(signal, kernel):
"""Convolve the iterable *signal* with the iterable *kernel*.
>>> signal = (1, 2, 3, 4, 5)
>>> kernel = [3, 2, 1]
>>> list(convolve(signal, kernel))
[3, 8, 14, 20, 26, 14, 5]
Note: the input arguments are not interchangeable, as the *kernel*
is immediately consumed and stored.
"""
kernel = tuple(kernel)[::-1]
n = len(kernel)
window = deque([0], maxlen=n) * n
for x in chain(signal, repeat(0, n - 1)):
window.append(x)
yield sum(map(operator.mul, kernel, window))

View File

@ -0,0 +1,488 @@
"""
An OrderedSet is a custom MutableSet that remembers its order, so that every
entry has an index that can be looked up.
Based on a recipe originally posted to ActiveState Recipes by Raymond Hettiger,
and released under the MIT license.
"""
import itertools as it
from collections import deque
try:
# Python 3
from collections.abc import MutableSet, Sequence
except ImportError:
# Python 2.7
from collections import MutableSet, Sequence
SLICE_ALL = slice(None)
__version__ = "3.1"
def is_iterable(obj):
"""
Are we being asked to look up a list of things, instead of a single thing?
We check for the `__iter__` attribute so that this can cover types that
don't have to be known by this module, such as NumPy arrays.
Strings, however, should be considered as atomic values to look up, not
iterables. The same goes for tuples, since they are immutable and therefore
valid entries.
We don't need to check for the Python 2 `unicode` type, because it doesn't
have an `__iter__` attribute anyway.
"""
return (
hasattr(obj, "__iter__")
and not isinstance(obj, str)
and not isinstance(obj, tuple)
)
class OrderedSet(MutableSet, Sequence):
"""
An OrderedSet is a custom MutableSet that remembers its order, so that
every entry has an index that can be looked up.
Example:
>>> OrderedSet([1, 1, 2, 3, 2])
OrderedSet([1, 2, 3])
"""
def __init__(self, iterable=None):
self.items = []
self.map = {}
if iterable is not None:
self |= iterable
def __len__(self):
"""
Returns the number of unique elements in the ordered set
Example:
>>> len(OrderedSet([]))
0
>>> len(OrderedSet([1, 2]))
2
"""
return len(self.items)
def __getitem__(self, index):
"""
Get the item at a given index.
If `index` is a slice, you will get back that slice of items, as a
new OrderedSet.
If `index` is a list or a similar iterable, you'll get a list of
items corresponding to those indices. This is similar to NumPy's
"fancy indexing". The result is not an OrderedSet because you may ask
for duplicate indices, and the number of elements returned should be
the number of elements asked for.
Example:
>>> oset = OrderedSet([1, 2, 3])
>>> oset[1]
2
"""
if isinstance(index, slice) and index == SLICE_ALL:
return self.copy()
elif is_iterable(index):
return [self.items[i] for i in index]
elif hasattr(index, "__index__") or isinstance(index, slice):
result = self.items[index]
if isinstance(result, list):
return self.__class__(result)
else:
return result
else:
raise TypeError("Don't know how to index an OrderedSet by %r" % index)
def copy(self):
"""
Return a shallow copy of this object.
Example:
>>> this = OrderedSet([1, 2, 3])
>>> other = this.copy()
>>> this == other
True
>>> this is other
False
"""
return self.__class__(self)
def __getstate__(self):
if len(self) == 0:
# The state can't be an empty list.
# We need to return a truthy value, or else __setstate__ won't be run.
#
# This could have been done more gracefully by always putting the state
# in a tuple, but this way is backwards- and forwards- compatible with
# previous versions of OrderedSet.
return (None,)
else:
return list(self)
def __setstate__(self, state):
if state == (None,):
self.__init__([])
else:
self.__init__(state)
def __contains__(self, key):
"""
Test if the item is in this ordered set
Example:
>>> 1 in OrderedSet([1, 3, 2])
True
>>> 5 in OrderedSet([1, 3, 2])
False
"""
return key in self.map
def add(self, key):
"""
Add `key` as an item to this OrderedSet, then return its index.
If `key` is already in the OrderedSet, return the index it already
had.
Example:
>>> oset = OrderedSet()
>>> oset.append(3)
0
>>> print(oset)
OrderedSet([3])
"""
if key not in self.map:
self.map[key] = len(self.items)
self.items.append(key)
return self.map[key]
append = add
def update(self, sequence):
"""
Update the set with the given iterable sequence, then return the index
of the last element inserted.
Example:
>>> oset = OrderedSet([1, 2, 3])
>>> oset.update([3, 1, 5, 1, 4])
4
>>> print(oset)
OrderedSet([1, 2, 3, 5, 4])
"""
item_index = None
try:
for item in sequence:
item_index = self.add(item)
except TypeError:
raise ValueError(
"Argument needs to be an iterable, got %s" % type(sequence)
)
return item_index
def index(self, key):
"""
Get the index of a given entry, raising an IndexError if it's not
present.
`key` can be an iterable of entries that is not a string, in which case
this returns a list of indices.
Example:
>>> oset = OrderedSet([1, 2, 3])
>>> oset.index(2)
1
"""
if is_iterable(key):
return [self.index(subkey) for subkey in key]
return self.map[key]
# Provide some compatibility with pd.Index
get_loc = index
get_indexer = index
def pop(self):
"""
Remove and return the last element from the set.
Raises KeyError if the set is empty.
Example:
>>> oset = OrderedSet([1, 2, 3])
>>> oset.pop()
3
"""
if not self.items:
raise KeyError("Set is empty")
elem = self.items[-1]
del self.items[-1]
del self.map[elem]
return elem
def discard(self, key):
"""
Remove an element. Do not raise an exception if absent.
The MutableSet mixin uses this to implement the .remove() method, which
*does* raise an error when asked to remove a non-existent item.
Example:
>>> oset = OrderedSet([1, 2, 3])
>>> oset.discard(2)
>>> print(oset)
OrderedSet([1, 3])
>>> oset.discard(2)
>>> print(oset)
OrderedSet([1, 3])
"""
if key in self:
i = self.map[key]
del self.items[i]
del self.map[key]
for k, v in self.map.items():
if v >= i:
self.map[k] = v - 1
def clear(self):
"""
Remove all items from this OrderedSet.
"""
del self.items[:]
self.map.clear()
def __iter__(self):
"""
Example:
>>> list(iter(OrderedSet([1, 2, 3])))
[1, 2, 3]
"""
return iter(self.items)
def __reversed__(self):
"""
Example:
>>> list(reversed(OrderedSet([1, 2, 3])))
[3, 2, 1]
"""
return reversed(self.items)
def __repr__(self):
if not self:
return "%s()" % (self.__class__.__name__,)
return "%s(%r)" % (self.__class__.__name__, list(self))
def __eq__(self, other):
"""
Returns true if the containers have the same items. If `other` is a
Sequence, then order is checked, otherwise it is ignored.
Example:
>>> oset = OrderedSet([1, 3, 2])
>>> oset == [1, 3, 2]
True
>>> oset == [1, 2, 3]
False
>>> oset == [2, 3]
False
>>> oset == OrderedSet([3, 2, 1])
False
"""
# In Python 2 deque is not a Sequence, so treat it as one for
# consistent behavior with Python 3.
if isinstance(other, (Sequence, deque)):
# Check that this OrderedSet contains the same elements, in the
# same order, as the other object.
return list(self) == list(other)
try:
other_as_set = set(other)
except TypeError:
# If `other` can't be converted into a set, it's not equal.
return False
else:
return set(self) == other_as_set
def union(self, *sets):
"""
Combines all unique items.
Each items order is defined by its first appearance.
Example:
>>> oset = OrderedSet.union(OrderedSet([3, 1, 4, 1, 5]), [1, 3], [2, 0])
>>> print(oset)
OrderedSet([3, 1, 4, 5, 2, 0])
>>> oset.union([8, 9])
OrderedSet([3, 1, 4, 5, 2, 0, 8, 9])
>>> oset | {10}
OrderedSet([3, 1, 4, 5, 2, 0, 10])
"""
cls = self.__class__ if isinstance(self, OrderedSet) else OrderedSet
containers = map(list, it.chain([self], sets))
items = it.chain.from_iterable(containers)
return cls(items)
def __and__(self, other):
# the parent implementation of this is backwards
return self.intersection(other)
def intersection(self, *sets):
"""
Returns elements in common between all sets. Order is defined only
by the first set.
Example:
>>> oset = OrderedSet.intersection(OrderedSet([0, 1, 2, 3]), [1, 2, 3])
>>> print(oset)
OrderedSet([1, 2, 3])
>>> oset.intersection([2, 4, 5], [1, 2, 3, 4])
OrderedSet([2])
>>> oset.intersection()
OrderedSet([1, 2, 3])
"""
cls = self.__class__ if isinstance(self, OrderedSet) else OrderedSet
if sets:
common = set.intersection(*map(set, sets))
items = (item for item in self if item in common)
else:
items = self
return cls(items)
def difference(self, *sets):
"""
Returns all elements that are in this set but not the others.
Example:
>>> OrderedSet([1, 2, 3]).difference(OrderedSet([2]))
OrderedSet([1, 3])
>>> OrderedSet([1, 2, 3]).difference(OrderedSet([2]), OrderedSet([3]))
OrderedSet([1])
>>> OrderedSet([1, 2, 3]) - OrderedSet([2])
OrderedSet([1, 3])
>>> OrderedSet([1, 2, 3]).difference()
OrderedSet([1, 2, 3])
"""
cls = self.__class__
if sets:
other = set.union(*map(set, sets))
items = (item for item in self if item not in other)
else:
items = self
return cls(items)
def issubset(self, other):
"""
Report whether another set contains this set.
Example:
>>> OrderedSet([1, 2, 3]).issubset({1, 2})
False
>>> OrderedSet([1, 2, 3]).issubset({1, 2, 3, 4})
True
>>> OrderedSet([1, 2, 3]).issubset({1, 4, 3, 5})
False
"""
if len(self) > len(other): # Fast check for obvious cases
return False
return all(item in other for item in self)
def issuperset(self, other):
"""
Report whether this set contains another set.
Example:
>>> OrderedSet([1, 2]).issuperset([1, 2, 3])
False
>>> OrderedSet([1, 2, 3, 4]).issuperset({1, 2, 3})
True
>>> OrderedSet([1, 4, 3, 5]).issuperset({1, 2, 3})
False
"""
if len(self) < len(other): # Fast check for obvious cases
return False
return all(item in self for item in other)
def symmetric_difference(self, other):
"""
Return the symmetric difference of two OrderedSets as a new set.
That is, the new set will contain all elements that are in exactly
one of the sets.
Their order will be preserved, with elements from `self` preceding
elements from `other`.
Example:
>>> this = OrderedSet([1, 4, 3, 5, 7])
>>> other = OrderedSet([9, 7, 1, 3, 2])
>>> this.symmetric_difference(other)
OrderedSet([4, 5, 9, 2])
"""
cls = self.__class__ if isinstance(self, OrderedSet) else OrderedSet
diff1 = cls(self).difference(other)
diff2 = cls(other).difference(self)
return diff1.union(diff2)
def _update_items(self, items):
"""
Replace the 'items' list of this OrderedSet with a new one, updating
self.map accordingly.
"""
self.items = items
self.map = {item: idx for (idx, item) in enumerate(items)}
def difference_update(self, *sets):
"""
Update this OrderedSet to remove items from one or more other sets.
Example:
>>> this = OrderedSet([1, 2, 3])
>>> this.difference_update(OrderedSet([2, 4]))
>>> print(this)
OrderedSet([1, 3])
>>> this = OrderedSet([1, 2, 3, 4, 5])
>>> this.difference_update(OrderedSet([2, 4]), OrderedSet([1, 4, 6]))
>>> print(this)
OrderedSet([3, 5])
"""
items_to_remove = set()
for other in sets:
items_to_remove |= set(other)
self._update_items([item for item in self.items if item not in items_to_remove])
def intersection_update(self, other):
"""
Update this OrderedSet to keep only items in another set, preserving
their order in this set.
Example:
>>> this = OrderedSet([1, 4, 3, 5, 7])
>>> other = OrderedSet([9, 7, 1, 3, 2])
>>> this.intersection_update(other)
>>> print(this)
OrderedSet([1, 3, 7])
"""
other = set(other)
self._update_items([item for item in self.items if item in other])
def symmetric_difference_update(self, other):
"""
Update this OrderedSet to remove items from another set, then
add items from the other set that were not present in this set.
Example:
>>> this = OrderedSet([1, 4, 3, 5, 7])
>>> other = OrderedSet([9, 7, 1, 3, 2])
>>> this.symmetric_difference_update(other)
>>> print(this)
OrderedSet([4, 5, 9, 2])
"""
items_to_add = [item for item in other if item not in self]
items_to_remove = set(other)
self._update_items(
[item for item in self.items if item not in items_to_remove] + items_to_add
)

View File

@ -0,0 +1,26 @@
# This file is dual licensed under the terms of the Apache License, Version
# 2.0, and the BSD License. See the LICENSE file in the root of this repository
# for complete details.
__all__ = [
"__title__",
"__summary__",
"__uri__",
"__version__",
"__author__",
"__email__",
"__license__",
"__copyright__",
]
__title__ = "packaging"
__summary__ = "Core utilities for Python packages"
__uri__ = "https://github.com/pypa/packaging"
__version__ = "21.2"
__author__ = "Donald Stufft and individual contributors"
__email__ = "donald@stufft.io"
__license__ = "BSD-2-Clause or Apache-2.0"
__copyright__ = "2014-2019 %s" % __author__

View File

@ -0,0 +1,25 @@
# This file is dual licensed under the terms of the Apache License, Version
# 2.0, and the BSD License. See the LICENSE file in the root of this repository
# for complete details.
from .__about__ import (
__author__,
__copyright__,
__email__,
__license__,
__summary__,
__title__,
__uri__,
__version__,
)
__all__ = [
"__title__",
"__summary__",
"__uri__",
"__version__",
"__author__",
"__email__",
"__license__",
"__copyright__",
]

View File

@ -0,0 +1,301 @@
import collections
import functools
import os
import re
import struct
import sys
import warnings
from typing import IO, Dict, Iterator, NamedTuple, Optional, Tuple
# Python does not provide platform information at sufficient granularity to
# identify the architecture of the running executable in some cases, so we
# determine it dynamically by reading the information from the running
# process. This only applies on Linux, which uses the ELF format.
class _ELFFileHeader:
# https://en.wikipedia.org/wiki/Executable_and_Linkable_Format#File_header
class _InvalidELFFileHeader(ValueError):
"""
An invalid ELF file header was found.
"""
ELF_MAGIC_NUMBER = 0x7F454C46
ELFCLASS32 = 1
ELFCLASS64 = 2
ELFDATA2LSB = 1
ELFDATA2MSB = 2
EM_386 = 3
EM_S390 = 22
EM_ARM = 40
EM_X86_64 = 62
EF_ARM_ABIMASK = 0xFF000000
EF_ARM_ABI_VER5 = 0x05000000
EF_ARM_ABI_FLOAT_HARD = 0x00000400
def __init__(self, file: IO[bytes]) -> None:
def unpack(fmt: str) -> int:
try:
data = file.read(struct.calcsize(fmt))
result: Tuple[int, ...] = struct.unpack(fmt, data)
except struct.error:
raise _ELFFileHeader._InvalidELFFileHeader()
return result[0]
self.e_ident_magic = unpack(">I")
if self.e_ident_magic != self.ELF_MAGIC_NUMBER:
raise _ELFFileHeader._InvalidELFFileHeader()
self.e_ident_class = unpack("B")
if self.e_ident_class not in {self.ELFCLASS32, self.ELFCLASS64}:
raise _ELFFileHeader._InvalidELFFileHeader()
self.e_ident_data = unpack("B")
if self.e_ident_data not in {self.ELFDATA2LSB, self.ELFDATA2MSB}:
raise _ELFFileHeader._InvalidELFFileHeader()
self.e_ident_version = unpack("B")
self.e_ident_osabi = unpack("B")
self.e_ident_abiversion = unpack("B")
self.e_ident_pad = file.read(7)
format_h = "<H" if self.e_ident_data == self.ELFDATA2LSB else ">H"
format_i = "<I" if self.e_ident_data == self.ELFDATA2LSB else ">I"
format_q = "<Q" if self.e_ident_data == self.ELFDATA2LSB else ">Q"
format_p = format_i if self.e_ident_class == self.ELFCLASS32 else format_q
self.e_type = unpack(format_h)
self.e_machine = unpack(format_h)
self.e_version = unpack(format_i)
self.e_entry = unpack(format_p)
self.e_phoff = unpack(format_p)
self.e_shoff = unpack(format_p)
self.e_flags = unpack(format_i)
self.e_ehsize = unpack(format_h)
self.e_phentsize = unpack(format_h)
self.e_phnum = unpack(format_h)
self.e_shentsize = unpack(format_h)
self.e_shnum = unpack(format_h)
self.e_shstrndx = unpack(format_h)
def _get_elf_header() -> Optional[_ELFFileHeader]:
try:
with open(sys.executable, "rb") as f:
elf_header = _ELFFileHeader(f)
except (OSError, TypeError, _ELFFileHeader._InvalidELFFileHeader):
return None
return elf_header
def _is_linux_armhf() -> bool:
# hard-float ABI can be detected from the ELF header of the running
# process
# https://static.docs.arm.com/ihi0044/g/aaelf32.pdf
elf_header = _get_elf_header()
if elf_header is None:
return False
result = elf_header.e_ident_class == elf_header.ELFCLASS32
result &= elf_header.e_ident_data == elf_header.ELFDATA2LSB
result &= elf_header.e_machine == elf_header.EM_ARM
result &= (
elf_header.e_flags & elf_header.EF_ARM_ABIMASK
) == elf_header.EF_ARM_ABI_VER5
result &= (
elf_header.e_flags & elf_header.EF_ARM_ABI_FLOAT_HARD
) == elf_header.EF_ARM_ABI_FLOAT_HARD
return result
def _is_linux_i686() -> bool:
elf_header = _get_elf_header()
if elf_header is None:
return False
result = elf_header.e_ident_class == elf_header.ELFCLASS32
result &= elf_header.e_ident_data == elf_header.ELFDATA2LSB
result &= elf_header.e_machine == elf_header.EM_386
return result
def _have_compatible_abi(arch: str) -> bool:
if arch == "armv7l":
return _is_linux_armhf()
if arch == "i686":
return _is_linux_i686()
return arch in {"x86_64", "aarch64", "ppc64", "ppc64le", "s390x"}
# If glibc ever changes its major version, we need to know what the last
# minor version was, so we can build the complete list of all versions.
# For now, guess what the highest minor version might be, assume it will
# be 50 for testing. Once this actually happens, update the dictionary
# with the actual value.
_LAST_GLIBC_MINOR: Dict[int, int] = collections.defaultdict(lambda: 50)
class _GLibCVersion(NamedTuple):
major: int
minor: int
def _glibc_version_string_confstr() -> Optional[str]:
"""
Primary implementation of glibc_version_string using os.confstr.
"""
# os.confstr is quite a bit faster than ctypes.DLL. It's also less likely
# to be broken or missing. This strategy is used in the standard library
# platform module.
# https://github.com/python/cpython/blob/fcf1d003bf4f0100c/Lib/platform.py#L175-L183
try:
# os.confstr("CS_GNU_LIBC_VERSION") returns a string like "glibc 2.17".
version_string = os.confstr("CS_GNU_LIBC_VERSION")
assert version_string is not None
_, version = version_string.split()
except (AssertionError, AttributeError, OSError, ValueError):
# os.confstr() or CS_GNU_LIBC_VERSION not available (or a bad value)...
return None
return version
def _glibc_version_string_ctypes() -> Optional[str]:
"""
Fallback implementation of glibc_version_string using ctypes.
"""
try:
import ctypes
except ImportError:
return None
# ctypes.CDLL(None) internally calls dlopen(NULL), and as the dlopen
# manpage says, "If filename is NULL, then the returned handle is for the
# main program". This way we can let the linker do the work to figure out
# which libc our process is actually using.
#
# We must also handle the special case where the executable is not a
# dynamically linked executable. This can occur when using musl libc,
# for example. In this situation, dlopen() will error, leading to an
# OSError. Interestingly, at least in the case of musl, there is no
# errno set on the OSError. The single string argument used to construct
# OSError comes from libc itself and is therefore not portable to
# hard code here. In any case, failure to call dlopen() means we
# can proceed, so we bail on our attempt.
try:
process_namespace = ctypes.CDLL(None)
except OSError:
return None
try:
gnu_get_libc_version = process_namespace.gnu_get_libc_version
except AttributeError:
# Symbol doesn't exist -> therefore, we are not linked to
# glibc.
return None
# Call gnu_get_libc_version, which returns a string like "2.5"
gnu_get_libc_version.restype = ctypes.c_char_p
version_str: str = gnu_get_libc_version()
# py2 / py3 compatibility:
if not isinstance(version_str, str):
version_str = version_str.decode("ascii")
return version_str
def _glibc_version_string() -> Optional[str]:
"""Returns glibc version string, or None if not using glibc."""
return _glibc_version_string_confstr() or _glibc_version_string_ctypes()
def _parse_glibc_version(version_str: str) -> Tuple[int, int]:
"""Parse glibc version.
We use a regexp instead of str.split because we want to discard any
random junk that might come after the minor version -- this might happen
in patched/forked versions of glibc (e.g. Linaro's version of glibc
uses version strings like "2.20-2014.11"). See gh-3588.
"""
m = re.match(r"(?P<major>[0-9]+)\.(?P<minor>[0-9]+)", version_str)
if not m:
warnings.warn(
"Expected glibc version with 2 components major.minor,"
" got: %s" % version_str,
RuntimeWarning,
)
return -1, -1
return int(m.group("major")), int(m.group("minor"))
@functools.lru_cache()
def _get_glibc_version() -> Tuple[int, int]:
version_str = _glibc_version_string()
if version_str is None:
return (-1, -1)
return _parse_glibc_version(version_str)
# From PEP 513, PEP 600
def _is_compatible(name: str, arch: str, version: _GLibCVersion) -> bool:
sys_glibc = _get_glibc_version()
if sys_glibc < version:
return False
# Check for presence of _manylinux module.
try:
import _manylinux # noqa
except ImportError:
return True
if hasattr(_manylinux, "manylinux_compatible"):
result = _manylinux.manylinux_compatible(version[0], version[1], arch)
if result is not None:
return bool(result)
return True
if version == _GLibCVersion(2, 5):
if hasattr(_manylinux, "manylinux1_compatible"):
return bool(_manylinux.manylinux1_compatible)
if version == _GLibCVersion(2, 12):
if hasattr(_manylinux, "manylinux2010_compatible"):
return bool(_manylinux.manylinux2010_compatible)
if version == _GLibCVersion(2, 17):
if hasattr(_manylinux, "manylinux2014_compatible"):
return bool(_manylinux.manylinux2014_compatible)
return True
_LEGACY_MANYLINUX_MAP = {
# CentOS 7 w/ glibc 2.17 (PEP 599)
(2, 17): "manylinux2014",
# CentOS 6 w/ glibc 2.12 (PEP 571)
(2, 12): "manylinux2010",
# CentOS 5 w/ glibc 2.5 (PEP 513)
(2, 5): "manylinux1",
}
def platform_tags(linux: str, arch: str) -> Iterator[str]:
if not _have_compatible_abi(arch):
return
# Oldest glibc to be supported regardless of architecture is (2, 17).
too_old_glibc2 = _GLibCVersion(2, 16)
if arch in {"x86_64", "i686"}:
# On x86/i686 also oldest glibc to be supported is (2, 5).
too_old_glibc2 = _GLibCVersion(2, 4)
current_glibc = _GLibCVersion(*_get_glibc_version())
glibc_max_list = [current_glibc]
# We can assume compatibility across glibc major versions.
# https://sourceware.org/bugzilla/show_bug.cgi?id=24636
#
# Build a list of maximum glibc versions so that we can
# output the canonical list of all glibc from current_glibc
# down to too_old_glibc2, including all intermediary versions.
for glibc_major in range(current_glibc.major - 1, 1, -1):
glibc_minor = _LAST_GLIBC_MINOR[glibc_major]
glibc_max_list.append(_GLibCVersion(glibc_major, glibc_minor))
for glibc_max in glibc_max_list:
if glibc_max.major == too_old_glibc2.major:
min_minor = too_old_glibc2.minor
else:
# For other glibc major versions oldest supported is (x, 0).
min_minor = -1
for glibc_minor in range(glibc_max.minor, min_minor, -1):
glibc_version = _GLibCVersion(glibc_max.major, glibc_minor)
tag = "manylinux_{}_{}".format(*glibc_version)
if _is_compatible(tag, arch, glibc_version):
yield linux.replace("linux", tag)
# Handle the legacy manylinux1, manylinux2010, manylinux2014 tags.
if glibc_version in _LEGACY_MANYLINUX_MAP:
legacy_tag = _LEGACY_MANYLINUX_MAP[glibc_version]
if _is_compatible(legacy_tag, arch, glibc_version):
yield linux.replace("linux", legacy_tag)

View File

@ -0,0 +1,136 @@
"""PEP 656 support.
This module implements logic to detect if the currently running Python is
linked against musl, and what musl version is used.
"""
import contextlib
import functools
import operator
import os
import re
import struct
import subprocess
import sys
from typing import IO, Iterator, NamedTuple, Optional, Tuple
def _read_unpacked(f: IO[bytes], fmt: str) -> Tuple[int, ...]:
return struct.unpack(fmt, f.read(struct.calcsize(fmt)))
def _parse_ld_musl_from_elf(f: IO[bytes]) -> Optional[str]:
"""Detect musl libc location by parsing the Python executable.
Based on: https://gist.github.com/lyssdod/f51579ae8d93c8657a5564aefc2ffbca
ELF header: https://refspecs.linuxfoundation.org/elf/gabi4+/ch4.eheader.html
"""
f.seek(0)
try:
ident = _read_unpacked(f, "16B")
except struct.error:
return None
if ident[:4] != tuple(b"\x7fELF"): # Invalid magic, not ELF.
return None
f.seek(struct.calcsize("HHI"), 1) # Skip file type, machine, and version.
try:
# e_fmt: Format for program header.
# p_fmt: Format for section header.
# p_idx: Indexes to find p_type, p_offset, and p_filesz.
e_fmt, p_fmt, p_idx = {
1: ("IIIIHHH", "IIIIIIII", (0, 1, 4)), # 32-bit.
2: ("QQQIHHH", "IIQQQQQQ", (0, 2, 5)), # 64-bit.
}[ident[4]]
except KeyError:
return None
else:
p_get = operator.itemgetter(*p_idx)
# Find the interpreter section and return its content.
try:
_, e_phoff, _, _, _, e_phentsize, e_phnum = _read_unpacked(f, e_fmt)
except struct.error:
return None
for i in range(e_phnum + 1):
f.seek(e_phoff + e_phentsize * i)
try:
p_type, p_offset, p_filesz = p_get(_read_unpacked(f, p_fmt))
except struct.error:
return None
if p_type != 3: # Not PT_INTERP.
continue
f.seek(p_offset)
interpreter = os.fsdecode(f.read(p_filesz)).strip("\0")
if "musl" not in interpreter:
return None
return interpreter
return None
class _MuslVersion(NamedTuple):
major: int
minor: int
def _parse_musl_version(output: str) -> Optional[_MuslVersion]:
lines = [n for n in (n.strip() for n in output.splitlines()) if n]
if len(lines) < 2 or lines[0][:4] != "musl":
return None
m = re.match(r"Version (\d+)\.(\d+)", lines[1])
if not m:
return None
return _MuslVersion(major=int(m.group(1)), minor=int(m.group(2)))
@functools.lru_cache()
def _get_musl_version(executable: str) -> Optional[_MuslVersion]:
"""Detect currently-running musl runtime version.
This is done by checking the specified executable's dynamic linking
information, and invoking the loader to parse its output for a version
string. If the loader is musl, the output would be something like::
musl libc (x86_64)
Version 1.2.2
Dynamic Program Loader
"""
with contextlib.ExitStack() as stack:
try:
f = stack.enter_context(open(executable, "rb"))
except IOError:
return None
ld = _parse_ld_musl_from_elf(f)
if not ld:
return None
proc = subprocess.run([ld], stderr=subprocess.PIPE, universal_newlines=True)
return _parse_musl_version(proc.stderr)
def platform_tags(arch: str) -> Iterator[str]:
"""Generate musllinux tags compatible to the current platform.
:param arch: Should be the part of platform tag after the ``linux_``
prefix, e.g. ``x86_64``. The ``linux_`` prefix is assumed as a
prerequisite for the current platform to be musllinux-compatible.
:returns: An iterator of compatible musllinux tags.
"""
sys_musl = _get_musl_version(sys.executable)
if sys_musl is None: # Python not dynamically linked against musl.
return
for minor in range(sys_musl.minor, -1, -1):
yield f"musllinux_{sys_musl.major}_{minor}_{arch}"
if __name__ == "__main__": # pragma: no cover
import sysconfig
plat = sysconfig.get_platform()
assert plat.startswith("linux-"), "not linux"
print("plat:", plat)
print("musl:", _get_musl_version(sys.executable))
print("tags:", end=" ")
for t in platform_tags(re.sub(r"[.-]", "_", plat.split("-", 1)[-1])):
print(t, end="\n ")

View File

@ -0,0 +1,67 @@
# This file is dual licensed under the terms of the Apache License, Version
# 2.0, and the BSD License. See the LICENSE file in the root of this repository
# for complete details.
class InfinityType:
def __repr__(self) -> str:
return "Infinity"
def __hash__(self) -> int:
return hash(repr(self))
def __lt__(self, other: object) -> bool:
return False
def __le__(self, other: object) -> bool:
return False
def __eq__(self, other: object) -> bool:
return isinstance(other, self.__class__)
def __ne__(self, other: object) -> bool:
return not isinstance(other, self.__class__)
def __gt__(self, other: object) -> bool:
return True
def __ge__(self, other: object) -> bool:
return True
def __neg__(self: object) -> "NegativeInfinityType":
return NegativeInfinity
Infinity = InfinityType()
class NegativeInfinityType:
def __repr__(self) -> str:
return "-Infinity"
def __hash__(self) -> int:
return hash(repr(self))
def __lt__(self, other: object) -> bool:
return True
def __le__(self, other: object) -> bool:
return True
def __eq__(self, other: object) -> bool:
return isinstance(other, self.__class__)
def __ne__(self, other: object) -> bool:
return not isinstance(other, self.__class__)
def __gt__(self, other: object) -> bool:
return False
def __ge__(self, other: object) -> bool:
return False
def __neg__(self: object) -> InfinityType:
return Infinity
NegativeInfinity = NegativeInfinityType()

View File

@ -0,0 +1,304 @@
# This file is dual licensed under the terms of the Apache License, Version
# 2.0, and the BSD License. See the LICENSE file in the root of this repository
# for complete details.
import operator
import os
import platform
import sys
from typing import Any, Callable, Dict, List, Optional, Tuple, Union
from setuptools.extern.pyparsing import ( # noqa: N817
Forward,
Group,
Literal as L,
ParseException,
ParseResults,
QuotedString,
ZeroOrMore,
stringEnd,
stringStart,
)
from .specifiers import InvalidSpecifier, Specifier
__all__ = [
"InvalidMarker",
"UndefinedComparison",
"UndefinedEnvironmentName",
"Marker",
"default_environment",
]
Operator = Callable[[str, str], bool]
class InvalidMarker(ValueError):
"""
An invalid marker was found, users should refer to PEP 508.
"""
class UndefinedComparison(ValueError):
"""
An invalid operation was attempted on a value that doesn't support it.
"""
class UndefinedEnvironmentName(ValueError):
"""
A name was attempted to be used that does not exist inside of the
environment.
"""
class Node:
def __init__(self, value: Any) -> None:
self.value = value
def __str__(self) -> str:
return str(self.value)
def __repr__(self) -> str:
return f"<{self.__class__.__name__}('{self}')>"
def serialize(self) -> str:
raise NotImplementedError
class Variable(Node):
def serialize(self) -> str:
return str(self)
class Value(Node):
def serialize(self) -> str:
return f'"{self}"'
class Op(Node):
def serialize(self) -> str:
return str(self)
VARIABLE = (
L("implementation_version")
| L("platform_python_implementation")
| L("implementation_name")
| L("python_full_version")
| L("platform_release")
| L("platform_version")
| L("platform_machine")
| L("platform_system")
| L("python_version")
| L("sys_platform")
| L("os_name")
| L("os.name") # PEP-345
| L("sys.platform") # PEP-345
| L("platform.version") # PEP-345
| L("platform.machine") # PEP-345
| L("platform.python_implementation") # PEP-345
| L("python_implementation") # undocumented setuptools legacy
| L("extra") # PEP-508
)
ALIASES = {
"os.name": "os_name",
"sys.platform": "sys_platform",
"platform.version": "platform_version",
"platform.machine": "platform_machine",
"platform.python_implementation": "platform_python_implementation",
"python_implementation": "platform_python_implementation",
}
VARIABLE.setParseAction(lambda s, l, t: Variable(ALIASES.get(t[0], t[0])))
VERSION_CMP = (
L("===") | L("==") | L(">=") | L("<=") | L("!=") | L("~=") | L(">") | L("<")
)
MARKER_OP = VERSION_CMP | L("not in") | L("in")
MARKER_OP.setParseAction(lambda s, l, t: Op(t[0]))
MARKER_VALUE = QuotedString("'") | QuotedString('"')
MARKER_VALUE.setParseAction(lambda s, l, t: Value(t[0]))
BOOLOP = L("and") | L("or")
MARKER_VAR = VARIABLE | MARKER_VALUE
MARKER_ITEM = Group(MARKER_VAR + MARKER_OP + MARKER_VAR)
MARKER_ITEM.setParseAction(lambda s, l, t: tuple(t[0]))
LPAREN = L("(").suppress()
RPAREN = L(")").suppress()
MARKER_EXPR = Forward()
MARKER_ATOM = MARKER_ITEM | Group(LPAREN + MARKER_EXPR + RPAREN)
MARKER_EXPR << MARKER_ATOM + ZeroOrMore(BOOLOP + MARKER_EXPR)
MARKER = stringStart + MARKER_EXPR + stringEnd
def _coerce_parse_result(results: Union[ParseResults, List[Any]]) -> List[Any]:
if isinstance(results, ParseResults):
return [_coerce_parse_result(i) for i in results]
else:
return results
def _format_marker(
marker: Union[List[str], Tuple[Node, ...], str], first: Optional[bool] = True
) -> str:
assert isinstance(marker, (list, tuple, str))
# Sometimes we have a structure like [[...]] which is a single item list
# where the single item is itself it's own list. In that case we want skip
# the rest of this function so that we don't get extraneous () on the
# outside.
if (
isinstance(marker, list)
and len(marker) == 1
and isinstance(marker[0], (list, tuple))
):
return _format_marker(marker[0])
if isinstance(marker, list):
inner = (_format_marker(m, first=False) for m in marker)
if first:
return " ".join(inner)
else:
return "(" + " ".join(inner) + ")"
elif isinstance(marker, tuple):
return " ".join([m.serialize() for m in marker])
else:
return marker
_operators: Dict[str, Operator] = {
"in": lambda lhs, rhs: lhs in rhs,
"not in": lambda lhs, rhs: lhs not in rhs,
"<": operator.lt,
"<=": operator.le,
"==": operator.eq,
"!=": operator.ne,
">=": operator.ge,
">": operator.gt,
}
def _eval_op(lhs: str, op: Op, rhs: str) -> bool:
try:
spec = Specifier("".join([op.serialize(), rhs]))
except InvalidSpecifier:
pass
else:
return spec.contains(lhs)
oper: Optional[Operator] = _operators.get(op.serialize())
if oper is None:
raise UndefinedComparison(f"Undefined {op!r} on {lhs!r} and {rhs!r}.")
return oper(lhs, rhs)
class Undefined:
pass
_undefined = Undefined()
def _get_env(environment: Dict[str, str], name: str) -> str:
value: Union[str, Undefined] = environment.get(name, _undefined)
if isinstance(value, Undefined):
raise UndefinedEnvironmentName(
f"{name!r} does not exist in evaluation environment."
)
return value
def _evaluate_markers(markers: List[Any], environment: Dict[str, str]) -> bool:
groups: List[List[bool]] = [[]]
for marker in markers:
assert isinstance(marker, (list, tuple, str))
if isinstance(marker, list):
groups[-1].append(_evaluate_markers(marker, environment))
elif isinstance(marker, tuple):
lhs, op, rhs = marker
if isinstance(lhs, Variable):
lhs_value = _get_env(environment, lhs.value)
rhs_value = rhs.value
else:
lhs_value = lhs.value
rhs_value = _get_env(environment, rhs.value)
groups[-1].append(_eval_op(lhs_value, op, rhs_value))
else:
assert marker in ["and", "or"]
if marker == "or":
groups.append([])
return any(all(item) for item in groups)
def format_full_version(info: "sys._version_info") -> str:
version = "{0.major}.{0.minor}.{0.micro}".format(info)
kind = info.releaselevel
if kind != "final":
version += kind[0] + str(info.serial)
return version
def default_environment() -> Dict[str, str]:
iver = format_full_version(sys.implementation.version)
implementation_name = sys.implementation.name
return {
"implementation_name": implementation_name,
"implementation_version": iver,
"os_name": os.name,
"platform_machine": platform.machine(),
"platform_release": platform.release(),
"platform_system": platform.system(),
"platform_version": platform.version(),
"python_full_version": platform.python_version(),
"platform_python_implementation": platform.python_implementation(),
"python_version": ".".join(platform.python_version_tuple()[:2]),
"sys_platform": sys.platform,
}
class Marker:
def __init__(self, marker: str) -> None:
try:
self._markers = _coerce_parse_result(MARKER.parseString(marker))
except ParseException as e:
raise InvalidMarker(
f"Invalid marker: {marker!r}, parse error at "
f"{marker[e.loc : e.loc + 8]!r}"
)
def __str__(self) -> str:
return _format_marker(self._markers)
def __repr__(self) -> str:
return f"<Marker('{self}')>"
def evaluate(self, environment: Optional[Dict[str, str]] = None) -> bool:
"""Evaluate a marker.
Return the boolean from evaluating the given marker against the
environment. environment is an optional argument to override all or
part of the determined environment.
The environment is determined from the current Python process.
"""
current_environment = default_environment()
if environment is not None:
current_environment.update(environment)
return _evaluate_markers(self._markers, current_environment)

View File

@ -0,0 +1,146 @@
# This file is dual licensed under the terms of the Apache License, Version
# 2.0, and the BSD License. See the LICENSE file in the root of this repository
# for complete details.
import re
import string
import urllib.parse
from typing import List, Optional as TOptional, Set
from setuptools.extern.pyparsing import ( # noqa
Combine,
Literal as L,
Optional,
ParseException,
Regex,
Word,
ZeroOrMore,
originalTextFor,
stringEnd,
stringStart,
)
from .markers import MARKER_EXPR, Marker
from .specifiers import LegacySpecifier, Specifier, SpecifierSet
class InvalidRequirement(ValueError):
"""
An invalid requirement was found, users should refer to PEP 508.
"""
ALPHANUM = Word(string.ascii_letters + string.digits)
LBRACKET = L("[").suppress()
RBRACKET = L("]").suppress()
LPAREN = L("(").suppress()
RPAREN = L(")").suppress()
COMMA = L(",").suppress()
SEMICOLON = L(";").suppress()
AT = L("@").suppress()
PUNCTUATION = Word("-_.")
IDENTIFIER_END = ALPHANUM | (ZeroOrMore(PUNCTUATION) + ALPHANUM)
IDENTIFIER = Combine(ALPHANUM + ZeroOrMore(IDENTIFIER_END))
NAME = IDENTIFIER("name")
EXTRA = IDENTIFIER
URI = Regex(r"[^ ]+")("url")
URL = AT + URI
EXTRAS_LIST = EXTRA + ZeroOrMore(COMMA + EXTRA)
EXTRAS = (LBRACKET + Optional(EXTRAS_LIST) + RBRACKET)("extras")
VERSION_PEP440 = Regex(Specifier._regex_str, re.VERBOSE | re.IGNORECASE)
VERSION_LEGACY = Regex(LegacySpecifier._regex_str, re.VERBOSE | re.IGNORECASE)
VERSION_ONE = VERSION_PEP440 ^ VERSION_LEGACY
VERSION_MANY = Combine(
VERSION_ONE + ZeroOrMore(COMMA + VERSION_ONE), joinString=",", adjacent=False
)("_raw_spec")
_VERSION_SPEC = Optional((LPAREN + VERSION_MANY + RPAREN) | VERSION_MANY)
_VERSION_SPEC.setParseAction(lambda s, l, t: t._raw_spec or "")
VERSION_SPEC = originalTextFor(_VERSION_SPEC)("specifier")
VERSION_SPEC.setParseAction(lambda s, l, t: t[1])
MARKER_EXPR = originalTextFor(MARKER_EXPR())("marker")
MARKER_EXPR.setParseAction(
lambda s, l, t: Marker(s[t._original_start : t._original_end])
)
MARKER_SEPARATOR = SEMICOLON
MARKER = MARKER_SEPARATOR + MARKER_EXPR
VERSION_AND_MARKER = VERSION_SPEC + Optional(MARKER)
URL_AND_MARKER = URL + Optional(MARKER)
NAMED_REQUIREMENT = NAME + Optional(EXTRAS) + (URL_AND_MARKER | VERSION_AND_MARKER)
REQUIREMENT = stringStart + NAMED_REQUIREMENT + stringEnd
# setuptools.extern.pyparsing isn't thread safe during initialization, so we do it eagerly, see
# issue #104
REQUIREMENT.parseString("x[]")
class Requirement:
"""Parse a requirement.
Parse a given requirement string into its parts, such as name, specifier,
URL, and extras. Raises InvalidRequirement on a badly-formed requirement
string.
"""
# TODO: Can we test whether something is contained within a requirement?
# If so how do we do that? Do we need to test against the _name_ of
# the thing as well as the version? What about the markers?
# TODO: Can we normalize the name and extra name?
def __init__(self, requirement_string: str) -> None:
try:
req = REQUIREMENT.parseString(requirement_string)
except ParseException as e:
raise InvalidRequirement(
f'Parse error at "{ requirement_string[e.loc : e.loc + 8]!r}": {e.msg}'
)
self.name: str = req.name
if req.url:
parsed_url = urllib.parse.urlparse(req.url)
if parsed_url.scheme == "file":
if urllib.parse.urlunparse(parsed_url) != req.url:
raise InvalidRequirement("Invalid URL given")
elif not (parsed_url.scheme and parsed_url.netloc) or (
not parsed_url.scheme and not parsed_url.netloc
):
raise InvalidRequirement(f"Invalid URL: {req.url}")
self.url: TOptional[str] = req.url
else:
self.url = None
self.extras: Set[str] = set(req.extras.asList() if req.extras else [])
self.specifier: SpecifierSet = SpecifierSet(req.specifier)
self.marker: TOptional[Marker] = req.marker if req.marker else None
def __str__(self) -> str:
parts: List[str] = [self.name]
if self.extras:
formatted_extras = ",".join(sorted(self.extras))
parts.append(f"[{formatted_extras}]")
if self.specifier:
parts.append(str(self.specifier))
if self.url:
parts.append(f"@ {self.url}")
if self.marker:
parts.append(" ")
if self.marker:
parts.append(f"; {self.marker}")
return "".join(parts)
def __repr__(self) -> str:
return f"<Requirement('{self}')>"

View File

@ -0,0 +1,828 @@
# This file is dual licensed under the terms of the Apache License, Version
# 2.0, and the BSD License. See the LICENSE file in the root of this repository
# for complete details.
import abc
import functools
import itertools
import re
import warnings
from typing import (
Callable,
Dict,
Iterable,
Iterator,
List,
Optional,
Pattern,
Set,
Tuple,
TypeVar,
Union,
)
from .utils import canonicalize_version
from .version import LegacyVersion, Version, parse
ParsedVersion = Union[Version, LegacyVersion]
UnparsedVersion = Union[Version, LegacyVersion, str]
VersionTypeVar = TypeVar("VersionTypeVar", bound=UnparsedVersion)
CallableOperator = Callable[[ParsedVersion, str], bool]
class InvalidSpecifier(ValueError):
"""
An invalid specifier was found, users should refer to PEP 440.
"""
class BaseSpecifier(metaclass=abc.ABCMeta):
@abc.abstractmethod
def __str__(self) -> str:
"""
Returns the str representation of this Specifier like object. This
should be representative of the Specifier itself.
"""
@abc.abstractmethod
def __hash__(self) -> int:
"""
Returns a hash value for this Specifier like object.
"""
@abc.abstractmethod
def __eq__(self, other: object) -> bool:
"""
Returns a boolean representing whether or not the two Specifier like
objects are equal.
"""
@abc.abstractmethod
def __ne__(self, other: object) -> bool:
"""
Returns a boolean representing whether or not the two Specifier like
objects are not equal.
"""
@abc.abstractproperty
def prereleases(self) -> Optional[bool]:
"""
Returns whether or not pre-releases as a whole are allowed by this
specifier.
"""
@prereleases.setter
def prereleases(self, value: bool) -> None:
"""
Sets whether or not pre-releases as a whole are allowed by this
specifier.
"""
@abc.abstractmethod
def contains(self, item: str, prereleases: Optional[bool] = None) -> bool:
"""
Determines if the given item is contained within this specifier.
"""
@abc.abstractmethod
def filter(
self, iterable: Iterable[VersionTypeVar], prereleases: Optional[bool] = None
) -> Iterable[VersionTypeVar]:
"""
Takes an iterable of items and filters them so that only items which
are contained within this specifier are allowed in it.
"""
class _IndividualSpecifier(BaseSpecifier):
_operators: Dict[str, str] = {}
_regex: Pattern[str]
def __init__(self, spec: str = "", prereleases: Optional[bool] = None) -> None:
match = self._regex.search(spec)
if not match:
raise InvalidSpecifier(f"Invalid specifier: '{spec}'")
self._spec: Tuple[str, str] = (
match.group("operator").strip(),
match.group("version").strip(),
)
# Store whether or not this Specifier should accept prereleases
self._prereleases = prereleases
def __repr__(self) -> str:
pre = (
f", prereleases={self.prereleases!r}"
if self._prereleases is not None
else ""
)
return "<{}({!r}{})>".format(self.__class__.__name__, str(self), pre)
def __str__(self) -> str:
return "{}{}".format(*self._spec)
@property
def _canonical_spec(self) -> Tuple[str, str]:
return self._spec[0], canonicalize_version(self._spec[1])
def __hash__(self) -> int:
return hash(self._canonical_spec)
def __eq__(self, other: object) -> bool:
if isinstance(other, str):
try:
other = self.__class__(str(other))
except InvalidSpecifier:
return NotImplemented
elif not isinstance(other, self.__class__):
return NotImplemented
return self._canonical_spec == other._canonical_spec
def __ne__(self, other: object) -> bool:
if isinstance(other, str):
try:
other = self.__class__(str(other))
except InvalidSpecifier:
return NotImplemented
elif not isinstance(other, self.__class__):
return NotImplemented
return self._spec != other._spec
def _get_operator(self, op: str) -> CallableOperator:
operator_callable: CallableOperator = getattr(
self, f"_compare_{self._operators[op]}"
)
return operator_callable
def _coerce_version(self, version: UnparsedVersion) -> ParsedVersion:
if not isinstance(version, (LegacyVersion, Version)):
version = parse(version)
return version
@property
def operator(self) -> str:
return self._spec[0]
@property
def version(self) -> str:
return self._spec[1]
@property
def prereleases(self) -> Optional[bool]:
return self._prereleases
@prereleases.setter
def prereleases(self, value: bool) -> None:
self._prereleases = value
def __contains__(self, item: str) -> bool:
return self.contains(item)
def contains(
self, item: UnparsedVersion, prereleases: Optional[bool] = None
) -> bool:
# Determine if prereleases are to be allowed or not.
if prereleases is None:
prereleases = self.prereleases
# Normalize item to a Version or LegacyVersion, this allows us to have
# a shortcut for ``"2.0" in Specifier(">=2")
normalized_item = self._coerce_version(item)
# Determine if we should be supporting prereleases in this specifier
# or not, if we do not support prereleases than we can short circuit
# logic if this version is a prereleases.
if normalized_item.is_prerelease and not prereleases:
return False
# Actually do the comparison to determine if this item is contained
# within this Specifier or not.
operator_callable: CallableOperator = self._get_operator(self.operator)
return operator_callable(normalized_item, self.version)
def filter(
self, iterable: Iterable[VersionTypeVar], prereleases: Optional[bool] = None
) -> Iterable[VersionTypeVar]:
yielded = False
found_prereleases = []
kw = {"prereleases": prereleases if prereleases is not None else True}
# Attempt to iterate over all the values in the iterable and if any of
# them match, yield them.
for version in iterable:
parsed_version = self._coerce_version(version)
if self.contains(parsed_version, **kw):
# If our version is a prerelease, and we were not set to allow
# prereleases, then we'll store it for later in case nothing
# else matches this specifier.
if parsed_version.is_prerelease and not (
prereleases or self.prereleases
):
found_prereleases.append(version)
# Either this is not a prerelease, or we should have been
# accepting prereleases from the beginning.
else:
yielded = True
yield version
# Now that we've iterated over everything, determine if we've yielded
# any values, and if we have not and we have any prereleases stored up
# then we will go ahead and yield the prereleases.
if not yielded and found_prereleases:
for version in found_prereleases:
yield version
class LegacySpecifier(_IndividualSpecifier):
_regex_str = r"""
(?P<operator>(==|!=|<=|>=|<|>))
\s*
(?P<version>
[^,;\s)]* # Since this is a "legacy" specifier, and the version
# string can be just about anything, we match everything
# except for whitespace, a semi-colon for marker support,
# a closing paren since versions can be enclosed in
# them, and a comma since it's a version separator.
)
"""
_regex = re.compile(r"^\s*" + _regex_str + r"\s*$", re.VERBOSE | re.IGNORECASE)
_operators = {
"==": "equal",
"!=": "not_equal",
"<=": "less_than_equal",
">=": "greater_than_equal",
"<": "less_than",
">": "greater_than",
}
def __init__(self, spec: str = "", prereleases: Optional[bool] = None) -> None:
super().__init__(spec, prereleases)
warnings.warn(
"Creating a LegacyVersion has been deprecated and will be "
"removed in the next major release",
DeprecationWarning,
)
def _coerce_version(self, version: UnparsedVersion) -> LegacyVersion:
if not isinstance(version, LegacyVersion):
version = LegacyVersion(str(version))
return version
def _compare_equal(self, prospective: LegacyVersion, spec: str) -> bool:
return prospective == self._coerce_version(spec)
def _compare_not_equal(self, prospective: LegacyVersion, spec: str) -> bool:
return prospective != self._coerce_version(spec)
def _compare_less_than_equal(self, prospective: LegacyVersion, spec: str) -> bool:
return prospective <= self._coerce_version(spec)
def _compare_greater_than_equal(
self, prospective: LegacyVersion, spec: str
) -> bool:
return prospective >= self._coerce_version(spec)
def _compare_less_than(self, prospective: LegacyVersion, spec: str) -> bool:
return prospective < self._coerce_version(spec)
def _compare_greater_than(self, prospective: LegacyVersion, spec: str) -> bool:
return prospective > self._coerce_version(spec)
def _require_version_compare(
fn: Callable[["Specifier", ParsedVersion, str], bool]
) -> Callable[["Specifier", ParsedVersion, str], bool]:
@functools.wraps(fn)
def wrapped(self: "Specifier", prospective: ParsedVersion, spec: str) -> bool:
if not isinstance(prospective, Version):
return False
return fn(self, prospective, spec)
return wrapped
class Specifier(_IndividualSpecifier):
_regex_str = r"""
(?P<operator>(~=|==|!=|<=|>=|<|>|===))
(?P<version>
(?:
# The identity operators allow for an escape hatch that will
# do an exact string match of the version you wish to install.
# This will not be parsed by PEP 440 and we cannot determine
# any semantic meaning from it. This operator is discouraged
# but included entirely as an escape hatch.
(?<====) # Only match for the identity operator
\s*
[^\s]* # We just match everything, except for whitespace
# since we are only testing for strict identity.
)
|
(?:
# The (non)equality operators allow for wild card and local
# versions to be specified so we have to define these two
# operators separately to enable that.
(?<===|!=) # Only match for equals and not equals
\s*
v?
(?:[0-9]+!)? # epoch
[0-9]+(?:\.[0-9]+)* # release
(?: # pre release
[-_\.]?
(a|b|c|rc|alpha|beta|pre|preview)
[-_\.]?
[0-9]*
)?
(?: # post release
(?:-[0-9]+)|(?:[-_\.]?(post|rev|r)[-_\.]?[0-9]*)
)?
# You cannot use a wild card and a dev or local version
# together so group them with a | and make them optional.
(?:
(?:[-_\.]?dev[-_\.]?[0-9]*)? # dev release
(?:\+[a-z0-9]+(?:[-_\.][a-z0-9]+)*)? # local
|
\.\* # Wild card syntax of .*
)?
)
|
(?:
# The compatible operator requires at least two digits in the
# release segment.
(?<=~=) # Only match for the compatible operator
\s*
v?
(?:[0-9]+!)? # epoch
[0-9]+(?:\.[0-9]+)+ # release (We have a + instead of a *)
(?: # pre release
[-_\.]?
(a|b|c|rc|alpha|beta|pre|preview)
[-_\.]?
[0-9]*
)?
(?: # post release
(?:-[0-9]+)|(?:[-_\.]?(post|rev|r)[-_\.]?[0-9]*)
)?
(?:[-_\.]?dev[-_\.]?[0-9]*)? # dev release
)
|
(?:
# All other operators only allow a sub set of what the
# (non)equality operators do. Specifically they do not allow
# local versions to be specified nor do they allow the prefix
# matching wild cards.
(?<!==|!=|~=) # We have special cases for these
# operators so we want to make sure they
# don't match here.
\s*
v?
(?:[0-9]+!)? # epoch
[0-9]+(?:\.[0-9]+)* # release
(?: # pre release
[-_\.]?
(a|b|c|rc|alpha|beta|pre|preview)
[-_\.]?
[0-9]*
)?
(?: # post release
(?:-[0-9]+)|(?:[-_\.]?(post|rev|r)[-_\.]?[0-9]*)
)?
(?:[-_\.]?dev[-_\.]?[0-9]*)? # dev release
)
)
"""
_regex = re.compile(r"^\s*" + _regex_str + r"\s*$", re.VERBOSE | re.IGNORECASE)
_operators = {
"~=": "compatible",
"==": "equal",
"!=": "not_equal",
"<=": "less_than_equal",
">=": "greater_than_equal",
"<": "less_than",
">": "greater_than",
"===": "arbitrary",
}
@_require_version_compare
def _compare_compatible(self, prospective: ParsedVersion, spec: str) -> bool:
# Compatible releases have an equivalent combination of >= and ==. That
# is that ~=2.2 is equivalent to >=2.2,==2.*. This allows us to
# implement this in terms of the other specifiers instead of
# implementing it ourselves. The only thing we need to do is construct
# the other specifiers.
# We want everything but the last item in the version, but we want to
# ignore suffix segments.
prefix = ".".join(
list(itertools.takewhile(_is_not_suffix, _version_split(spec)))[:-1]
)
# Add the prefix notation to the end of our string
prefix += ".*"
return self._get_operator(">=")(prospective, spec) and self._get_operator("==")(
prospective, prefix
)
@_require_version_compare
def _compare_equal(self, prospective: ParsedVersion, spec: str) -> bool:
# We need special logic to handle prefix matching
if spec.endswith(".*"):
# In the case of prefix matching we want to ignore local segment.
prospective = Version(prospective.public)
# Split the spec out by dots, and pretend that there is an implicit
# dot in between a release segment and a pre-release segment.
split_spec = _version_split(spec[:-2]) # Remove the trailing .*
# Split the prospective version out by dots, and pretend that there
# is an implicit dot in between a release segment and a pre-release
# segment.
split_prospective = _version_split(str(prospective))
# Shorten the prospective version to be the same length as the spec
# so that we can determine if the specifier is a prefix of the
# prospective version or not.
shortened_prospective = split_prospective[: len(split_spec)]
# Pad out our two sides with zeros so that they both equal the same
# length.
padded_spec, padded_prospective = _pad_version(
split_spec, shortened_prospective
)
return padded_prospective == padded_spec
else:
# Convert our spec string into a Version
spec_version = Version(spec)
# If the specifier does not have a local segment, then we want to
# act as if the prospective version also does not have a local
# segment.
if not spec_version.local:
prospective = Version(prospective.public)
return prospective == spec_version
@_require_version_compare
def _compare_not_equal(self, prospective: ParsedVersion, spec: str) -> bool:
return not self._compare_equal(prospective, spec)
@_require_version_compare
def _compare_less_than_equal(self, prospective: ParsedVersion, spec: str) -> bool:
# NB: Local version identifiers are NOT permitted in the version
# specifier, so local version labels can be universally removed from
# the prospective version.
return Version(prospective.public) <= Version(spec)
@_require_version_compare
def _compare_greater_than_equal(
self, prospective: ParsedVersion, spec: str
) -> bool:
# NB: Local version identifiers are NOT permitted in the version
# specifier, so local version labels can be universally removed from
# the prospective version.
return Version(prospective.public) >= Version(spec)
@_require_version_compare
def _compare_less_than(self, prospective: ParsedVersion, spec_str: str) -> bool:
# Convert our spec to a Version instance, since we'll want to work with
# it as a version.
spec = Version(spec_str)
# Check to see if the prospective version is less than the spec
# version. If it's not we can short circuit and just return False now
# instead of doing extra unneeded work.
if not prospective < spec:
return False
# This special case is here so that, unless the specifier itself
# includes is a pre-release version, that we do not accept pre-release
# versions for the version mentioned in the specifier (e.g. <3.1 should
# not match 3.1.dev0, but should match 3.0.dev0).
if not spec.is_prerelease and prospective.is_prerelease:
if Version(prospective.base_version) == Version(spec.base_version):
return False
# If we've gotten to here, it means that prospective version is both
# less than the spec version *and* it's not a pre-release of the same
# version in the spec.
return True
@_require_version_compare
def _compare_greater_than(self, prospective: ParsedVersion, spec_str: str) -> bool:
# Convert our spec to a Version instance, since we'll want to work with
# it as a version.
spec = Version(spec_str)
# Check to see if the prospective version is greater than the spec
# version. If it's not we can short circuit and just return False now
# instead of doing extra unneeded work.
if not prospective > spec:
return False
# This special case is here so that, unless the specifier itself
# includes is a post-release version, that we do not accept
# post-release versions for the version mentioned in the specifier
# (e.g. >3.1 should not match 3.0.post0, but should match 3.2.post0).
if not spec.is_postrelease and prospective.is_postrelease:
if Version(prospective.base_version) == Version(spec.base_version):
return False
# Ensure that we do not allow a local version of the version mentioned
# in the specifier, which is technically greater than, to match.
if prospective.local is not None:
if Version(prospective.base_version) == Version(spec.base_version):
return False
# If we've gotten to here, it means that prospective version is both
# greater than the spec version *and* it's not a pre-release of the
# same version in the spec.
return True
def _compare_arbitrary(self, prospective: Version, spec: str) -> bool:
return str(prospective).lower() == str(spec).lower()
@property
def prereleases(self) -> bool:
# If there is an explicit prereleases set for this, then we'll just
# blindly use that.
if self._prereleases is not None:
return self._prereleases
# Look at all of our specifiers and determine if they are inclusive
# operators, and if they are if they are including an explicit
# prerelease.
operator, version = self._spec
if operator in ["==", ">=", "<=", "~=", "==="]:
# The == specifier can include a trailing .*, if it does we
# want to remove before parsing.
if operator == "==" and version.endswith(".*"):
version = version[:-2]
# Parse the version, and if it is a pre-release than this
# specifier allows pre-releases.
if parse(version).is_prerelease:
return True
return False
@prereleases.setter
def prereleases(self, value: bool) -> None:
self._prereleases = value
_prefix_regex = re.compile(r"^([0-9]+)((?:a|b|c|rc)[0-9]+)$")
def _version_split(version: str) -> List[str]:
result: List[str] = []
for item in version.split("."):
match = _prefix_regex.search(item)
if match:
result.extend(match.groups())
else:
result.append(item)
return result
def _is_not_suffix(segment: str) -> bool:
return not any(
segment.startswith(prefix) for prefix in ("dev", "a", "b", "rc", "post")
)
def _pad_version(left: List[str], right: List[str]) -> Tuple[List[str], List[str]]:
left_split, right_split = [], []
# Get the release segment of our versions
left_split.append(list(itertools.takewhile(lambda x: x.isdigit(), left)))
right_split.append(list(itertools.takewhile(lambda x: x.isdigit(), right)))
# Get the rest of our versions
left_split.append(left[len(left_split[0]) :])
right_split.append(right[len(right_split[0]) :])
# Insert our padding
left_split.insert(1, ["0"] * max(0, len(right_split[0]) - len(left_split[0])))
right_split.insert(1, ["0"] * max(0, len(left_split[0]) - len(right_split[0])))
return (list(itertools.chain(*left_split)), list(itertools.chain(*right_split)))
class SpecifierSet(BaseSpecifier):
def __init__(
self, specifiers: str = "", prereleases: Optional[bool] = None
) -> None:
# Split on , to break each individual specifier into it's own item, and
# strip each item to remove leading/trailing whitespace.
split_specifiers = [s.strip() for s in specifiers.split(",") if s.strip()]
# Parsed each individual specifier, attempting first to make it a
# Specifier and falling back to a LegacySpecifier.
parsed: Set[_IndividualSpecifier] = set()
for specifier in split_specifiers:
try:
parsed.add(Specifier(specifier))
except InvalidSpecifier:
parsed.add(LegacySpecifier(specifier))
# Turn our parsed specifiers into a frozen set and save them for later.
self._specs = frozenset(parsed)
# Store our prereleases value so we can use it later to determine if
# we accept prereleases or not.
self._prereleases = prereleases
def __repr__(self) -> str:
pre = (
f", prereleases={self.prereleases!r}"
if self._prereleases is not None
else ""
)
return "<SpecifierSet({!r}{})>".format(str(self), pre)
def __str__(self) -> str:
return ",".join(sorted(str(s) for s in self._specs))
def __hash__(self) -> int:
return hash(self._specs)
def __and__(self, other: Union["SpecifierSet", str]) -> "SpecifierSet":
if isinstance(other, str):
other = SpecifierSet(other)
elif not isinstance(other, SpecifierSet):
return NotImplemented
specifier = SpecifierSet()
specifier._specs = frozenset(self._specs | other._specs)
if self._prereleases is None and other._prereleases is not None:
specifier._prereleases = other._prereleases
elif self._prereleases is not None and other._prereleases is None:
specifier._prereleases = self._prereleases
elif self._prereleases == other._prereleases:
specifier._prereleases = self._prereleases
else:
raise ValueError(
"Cannot combine SpecifierSets with True and False prerelease "
"overrides."
)
return specifier
def __eq__(self, other: object) -> bool:
if isinstance(other, (str, _IndividualSpecifier)):
other = SpecifierSet(str(other))
elif not isinstance(other, SpecifierSet):
return NotImplemented
return self._specs == other._specs
def __ne__(self, other: object) -> bool:
if isinstance(other, (str, _IndividualSpecifier)):
other = SpecifierSet(str(other))
elif not isinstance(other, SpecifierSet):
return NotImplemented
return self._specs != other._specs
def __len__(self) -> int:
return len(self._specs)
def __iter__(self) -> Iterator[_IndividualSpecifier]:
return iter(self._specs)
@property
def prereleases(self) -> Optional[bool]:
# If we have been given an explicit prerelease modifier, then we'll
# pass that through here.
if self._prereleases is not None:
return self._prereleases
# If we don't have any specifiers, and we don't have a forced value,
# then we'll just return None since we don't know if this should have
# pre-releases or not.
if not self._specs:
return None
# Otherwise we'll see if any of the given specifiers accept
# prereleases, if any of them do we'll return True, otherwise False.
return any(s.prereleases for s in self._specs)
@prereleases.setter
def prereleases(self, value: bool) -> None:
self._prereleases = value
def __contains__(self, item: UnparsedVersion) -> bool:
return self.contains(item)
def contains(
self, item: UnparsedVersion, prereleases: Optional[bool] = None
) -> bool:
# Ensure that our item is a Version or LegacyVersion instance.
if not isinstance(item, (LegacyVersion, Version)):
item = parse(item)
# Determine if we're forcing a prerelease or not, if we're not forcing
# one for this particular filter call, then we'll use whatever the
# SpecifierSet thinks for whether or not we should support prereleases.
if prereleases is None:
prereleases = self.prereleases
# We can determine if we're going to allow pre-releases by looking to
# see if any of the underlying items supports them. If none of them do
# and this item is a pre-release then we do not allow it and we can
# short circuit that here.
# Note: This means that 1.0.dev1 would not be contained in something
# like >=1.0.devabc however it would be in >=1.0.debabc,>0.0.dev0
if not prereleases and item.is_prerelease:
return False
# We simply dispatch to the underlying specs here to make sure that the
# given version is contained within all of them.
# Note: This use of all() here means that an empty set of specifiers
# will always return True, this is an explicit design decision.
return all(s.contains(item, prereleases=prereleases) for s in self._specs)
def filter(
self, iterable: Iterable[VersionTypeVar], prereleases: Optional[bool] = None
) -> Iterable[VersionTypeVar]:
# Determine if we're forcing a prerelease or not, if we're not forcing
# one for this particular filter call, then we'll use whatever the
# SpecifierSet thinks for whether or not we should support prereleases.
if prereleases is None:
prereleases = self.prereleases
# If we have any specifiers, then we want to wrap our iterable in the
# filter method for each one, this will act as a logical AND amongst
# each specifier.
if self._specs:
for spec in self._specs:
iterable = spec.filter(iterable, prereleases=bool(prereleases))
return iterable
# If we do not have any specifiers, then we need to have a rough filter
# which will filter out any pre-releases, unless there are no final
# releases, and which will filter out LegacyVersion in general.
else:
filtered: List[VersionTypeVar] = []
found_prereleases: List[VersionTypeVar] = []
item: UnparsedVersion
parsed_version: Union[Version, LegacyVersion]
for item in iterable:
# Ensure that we some kind of Version class for this item.
if not isinstance(item, (LegacyVersion, Version)):
parsed_version = parse(item)
else:
parsed_version = item
# Filter out any item which is parsed as a LegacyVersion
if isinstance(parsed_version, LegacyVersion):
continue
# Store any item which is a pre-release for later unless we've
# already found a final version or we are accepting prereleases
if parsed_version.is_prerelease and not prereleases:
if not filtered:
found_prereleases.append(item)
else:
filtered.append(item)
# If we've found no items except for pre-releases, then we'll go
# ahead and use the pre-releases
if not filtered and found_prereleases and prereleases is None:
return found_prereleases
return filtered

View File

@ -0,0 +1,484 @@
# This file is dual licensed under the terms of the Apache License, Version
# 2.0, and the BSD License. See the LICENSE file in the root of this repository
# for complete details.
import logging
import platform
import sys
import sysconfig
from importlib.machinery import EXTENSION_SUFFIXES
from typing import (
Dict,
FrozenSet,
Iterable,
Iterator,
List,
Optional,
Sequence,
Tuple,
Union,
cast,
)
from . import _manylinux, _musllinux
logger = logging.getLogger(__name__)
PythonVersion = Sequence[int]
MacVersion = Tuple[int, int]
INTERPRETER_SHORT_NAMES: Dict[str, str] = {
"python": "py", # Generic.
"cpython": "cp",
"pypy": "pp",
"ironpython": "ip",
"jython": "jy",
}
_32_BIT_INTERPRETER = sys.maxsize <= 2 ** 32
class Tag:
"""
A representation of the tag triple for a wheel.
Instances are considered immutable and thus are hashable. Equality checking
is also supported.
"""
__slots__ = ["_interpreter", "_abi", "_platform", "_hash"]
def __init__(self, interpreter: str, abi: str, platform: str) -> None:
self._interpreter = interpreter.lower()
self._abi = abi.lower()
self._platform = platform.lower()
# The __hash__ of every single element in a Set[Tag] will be evaluated each time
# that a set calls its `.disjoint()` method, which may be called hundreds of
# times when scanning a page of links for packages with tags matching that
# Set[Tag]. Pre-computing the value here produces significant speedups for
# downstream consumers.
self._hash = hash((self._interpreter, self._abi, self._platform))
@property
def interpreter(self) -> str:
return self._interpreter
@property
def abi(self) -> str:
return self._abi
@property
def platform(self) -> str:
return self._platform
def __eq__(self, other: object) -> bool:
if not isinstance(other, Tag):
return NotImplemented
return (
(self._hash == other._hash) # Short-circuit ASAP for perf reasons.
and (self._platform == other._platform)
and (self._abi == other._abi)
and (self._interpreter == other._interpreter)
)
def __hash__(self) -> int:
return self._hash
def __str__(self) -> str:
return f"{self._interpreter}-{self._abi}-{self._platform}"
def __repr__(self) -> str:
return "<{self} @ {self_id}>".format(self=self, self_id=id(self))
def parse_tag(tag: str) -> FrozenSet[Tag]:
"""
Parses the provided tag (e.g. `py3-none-any`) into a frozenset of Tag instances.
Returning a set is required due to the possibility that the tag is a
compressed tag set.
"""
tags = set()
interpreters, abis, platforms = tag.split("-")
for interpreter in interpreters.split("."):
for abi in abis.split("."):
for platform_ in platforms.split("."):
tags.add(Tag(interpreter, abi, platform_))
return frozenset(tags)
def _get_config_var(name: str, warn: bool = False) -> Union[int, str, None]:
value = sysconfig.get_config_var(name)
if value is None and warn:
logger.debug(
"Config variable '%s' is unset, Python ABI tag may be incorrect", name
)
return value
def _normalize_string(string: str) -> str:
return string.replace(".", "_").replace("-", "_")
def _abi3_applies(python_version: PythonVersion) -> bool:
"""
Determine if the Python version supports abi3.
PEP 384 was first implemented in Python 3.2.
"""
return len(python_version) > 1 and tuple(python_version) >= (3, 2)
def _cpython_abis(py_version: PythonVersion, warn: bool = False) -> List[str]:
py_version = tuple(py_version) # To allow for version comparison.
abis = []
version = _version_nodot(py_version[:2])
debug = pymalloc = ucs4 = ""
with_debug = _get_config_var("Py_DEBUG", warn)
has_refcount = hasattr(sys, "gettotalrefcount")
# Windows doesn't set Py_DEBUG, so checking for support of debug-compiled
# extension modules is the best option.
# https://github.com/pypa/pip/issues/3383#issuecomment-173267692
has_ext = "_d.pyd" in EXTENSION_SUFFIXES
if with_debug or (with_debug is None and (has_refcount or has_ext)):
debug = "d"
if py_version < (3, 8):
with_pymalloc = _get_config_var("WITH_PYMALLOC", warn)
if with_pymalloc or with_pymalloc is None:
pymalloc = "m"
if py_version < (3, 3):
unicode_size = _get_config_var("Py_UNICODE_SIZE", warn)
if unicode_size == 4 or (
unicode_size is None and sys.maxunicode == 0x10FFFF
):
ucs4 = "u"
elif debug:
# Debug builds can also load "normal" extension modules.
# We can also assume no UCS-4 or pymalloc requirement.
abis.append(f"cp{version}")
abis.insert(
0,
"cp{version}{debug}{pymalloc}{ucs4}".format(
version=version, debug=debug, pymalloc=pymalloc, ucs4=ucs4
),
)
return abis
def cpython_tags(
python_version: Optional[PythonVersion] = None,
abis: Optional[Iterable[str]] = None,
platforms: Optional[Iterable[str]] = None,
*,
warn: bool = False,
) -> Iterator[Tag]:
"""
Yields the tags for a CPython interpreter.
The tags consist of:
- cp<python_version>-<abi>-<platform>
- cp<python_version>-abi3-<platform>
- cp<python_version>-none-<platform>
- cp<less than python_version>-abi3-<platform> # Older Python versions down to 3.2.
If python_version only specifies a major version then user-provided ABIs and
the 'none' ABItag will be used.
If 'abi3' or 'none' are specified in 'abis' then they will be yielded at
their normal position and not at the beginning.
"""
if not python_version:
python_version = sys.version_info[:2]
interpreter = "cp{}".format(_version_nodot(python_version[:2]))
if abis is None:
if len(python_version) > 1:
abis = _cpython_abis(python_version, warn)
else:
abis = []
abis = list(abis)
# 'abi3' and 'none' are explicitly handled later.
for explicit_abi in ("abi3", "none"):
try:
abis.remove(explicit_abi)
except ValueError:
pass
platforms = list(platforms or platform_tags())
for abi in abis:
for platform_ in platforms:
yield Tag(interpreter, abi, platform_)
if _abi3_applies(python_version):
yield from (Tag(interpreter, "abi3", platform_) for platform_ in platforms)
yield from (Tag(interpreter, "none", platform_) for platform_ in platforms)
if _abi3_applies(python_version):
for minor_version in range(python_version[1] - 1, 1, -1):
for platform_ in platforms:
interpreter = "cp{version}".format(
version=_version_nodot((python_version[0], minor_version))
)
yield Tag(interpreter, "abi3", platform_)
def _generic_abi() -> Iterator[str]:
abi = sysconfig.get_config_var("SOABI")
if abi:
yield _normalize_string(abi)
def generic_tags(
interpreter: Optional[str] = None,
abis: Optional[Iterable[str]] = None,
platforms: Optional[Iterable[str]] = None,
*,
warn: bool = False,
) -> Iterator[Tag]:
"""
Yields the tags for a generic interpreter.
The tags consist of:
- <interpreter>-<abi>-<platform>
The "none" ABI will be added if it was not explicitly provided.
"""
if not interpreter:
interp_name = interpreter_name()
interp_version = interpreter_version(warn=warn)
interpreter = "".join([interp_name, interp_version])
if abis is None:
abis = _generic_abi()
platforms = list(platforms or platform_tags())
abis = list(abis)
if "none" not in abis:
abis.append("none")
for abi in abis:
for platform_ in platforms:
yield Tag(interpreter, abi, platform_)
def _py_interpreter_range(py_version: PythonVersion) -> Iterator[str]:
"""
Yields Python versions in descending order.
After the latest version, the major-only version will be yielded, and then
all previous versions of that major version.
"""
if len(py_version) > 1:
yield "py{version}".format(version=_version_nodot(py_version[:2]))
yield "py{major}".format(major=py_version[0])
if len(py_version) > 1:
for minor in range(py_version[1] - 1, -1, -1):
yield "py{version}".format(version=_version_nodot((py_version[0], minor)))
def compatible_tags(
python_version: Optional[PythonVersion] = None,
interpreter: Optional[str] = None,
platforms: Optional[Iterable[str]] = None,
) -> Iterator[Tag]:
"""
Yields the sequence of tags that are compatible with a specific version of Python.
The tags consist of:
- py*-none-<platform>
- <interpreter>-none-any # ... if `interpreter` is provided.
- py*-none-any
"""
if not python_version:
python_version = sys.version_info[:2]
platforms = list(platforms or platform_tags())
for version in _py_interpreter_range(python_version):
for platform_ in platforms:
yield Tag(version, "none", platform_)
if interpreter:
yield Tag(interpreter, "none", "any")
for version in _py_interpreter_range(python_version):
yield Tag(version, "none", "any")
def _mac_arch(arch: str, is_32bit: bool = _32_BIT_INTERPRETER) -> str:
if not is_32bit:
return arch
if arch.startswith("ppc"):
return "ppc"
return "i386"
def _mac_binary_formats(version: MacVersion, cpu_arch: str) -> List[str]:
formats = [cpu_arch]
if cpu_arch == "x86_64":
if version < (10, 4):
return []
formats.extend(["intel", "fat64", "fat32"])
elif cpu_arch == "i386":
if version < (10, 4):
return []
formats.extend(["intel", "fat32", "fat"])
elif cpu_arch == "ppc64":
# TODO: Need to care about 32-bit PPC for ppc64 through 10.2?
if version > (10, 5) or version < (10, 4):
return []
formats.append("fat64")
elif cpu_arch == "ppc":
if version > (10, 6):
return []
formats.extend(["fat32", "fat"])
if cpu_arch in {"arm64", "x86_64"}:
formats.append("universal2")
if cpu_arch in {"x86_64", "i386", "ppc64", "ppc", "intel"}:
formats.append("universal")
return formats
def mac_platforms(
version: Optional[MacVersion] = None, arch: Optional[str] = None
) -> Iterator[str]:
"""
Yields the platform tags for a macOS system.
The `version` parameter is a two-item tuple specifying the macOS version to
generate platform tags for. The `arch` parameter is the CPU architecture to
generate platform tags for. Both parameters default to the appropriate value
for the current system.
"""
version_str, _, cpu_arch = platform.mac_ver()
if version is None:
version = cast("MacVersion", tuple(map(int, version_str.split(".")[:2])))
else:
version = version
if arch is None:
arch = _mac_arch(cpu_arch)
else:
arch = arch
if (10, 0) <= version and version < (11, 0):
# Prior to Mac OS 11, each yearly release of Mac OS bumped the
# "minor" version number. The major version was always 10.
for minor_version in range(version[1], -1, -1):
compat_version = 10, minor_version
binary_formats = _mac_binary_formats(compat_version, arch)
for binary_format in binary_formats:
yield "macosx_{major}_{minor}_{binary_format}".format(
major=10, minor=minor_version, binary_format=binary_format
)
if version >= (11, 0):
# Starting with Mac OS 11, each yearly release bumps the major version
# number. The minor versions are now the midyear updates.
for major_version in range(version[0], 10, -1):
compat_version = major_version, 0
binary_formats = _mac_binary_formats(compat_version, arch)
for binary_format in binary_formats:
yield "macosx_{major}_{minor}_{binary_format}".format(
major=major_version, minor=0, binary_format=binary_format
)
if version >= (11, 0):
# Mac OS 11 on x86_64 is compatible with binaries from previous releases.
# Arm64 support was introduced in 11.0, so no Arm binaries from previous
# releases exist.
#
# However, the "universal2" binary format can have a
# macOS version earlier than 11.0 when the x86_64 part of the binary supports
# that version of macOS.
if arch == "x86_64":
for minor_version in range(16, 3, -1):
compat_version = 10, minor_version
binary_formats = _mac_binary_formats(compat_version, arch)
for binary_format in binary_formats:
yield "macosx_{major}_{minor}_{binary_format}".format(
major=compat_version[0],
minor=compat_version[1],
binary_format=binary_format,
)
else:
for minor_version in range(16, 3, -1):
compat_version = 10, minor_version
binary_format = "universal2"
yield "macosx_{major}_{minor}_{binary_format}".format(
major=compat_version[0],
minor=compat_version[1],
binary_format=binary_format,
)
def _linux_platforms(is_32bit: bool = _32_BIT_INTERPRETER) -> Iterator[str]:
linux = _normalize_string(sysconfig.get_platform())
if is_32bit:
if linux == "linux_x86_64":
linux = "linux_i686"
elif linux == "linux_aarch64":
linux = "linux_armv7l"
_, arch = linux.split("_", 1)
yield from _manylinux.platform_tags(linux, arch)
yield from _musllinux.platform_tags(arch)
yield linux
def _generic_platforms() -> Iterator[str]:
yield _normalize_string(sysconfig.get_platform())
def platform_tags() -> Iterator[str]:
"""
Provides the platform tags for this installation.
"""
if platform.system() == "Darwin":
return mac_platforms()
elif platform.system() == "Linux":
return _linux_platforms()
else:
return _generic_platforms()
def interpreter_name() -> str:
"""
Returns the name of the running interpreter.
"""
name = sys.implementation.name
return INTERPRETER_SHORT_NAMES.get(name) or name
def interpreter_version(*, warn: bool = False) -> str:
"""
Returns the version of the running interpreter.
"""
version = _get_config_var("py_version_nodot", warn=warn)
if version:
version = str(version)
else:
version = _version_nodot(sys.version_info[:2])
return version
def _version_nodot(version: PythonVersion) -> str:
return "".join(map(str, version))
def sys_tags(*, warn: bool = False) -> Iterator[Tag]:
"""
Returns the sequence of tag triples for the running interpreter.
The order of the sequence corresponds to priority order for the
interpreter, from most to least important.
"""
interp_name = interpreter_name()
if interp_name == "cp":
yield from cpython_tags(warn=warn)
else:
yield from generic_tags()
yield from compatible_tags()

View File

@ -0,0 +1,136 @@
# This file is dual licensed under the terms of the Apache License, Version
# 2.0, and the BSD License. See the LICENSE file in the root of this repository
# for complete details.
import re
from typing import FrozenSet, NewType, Tuple, Union, cast
from .tags import Tag, parse_tag
from .version import InvalidVersion, Version
BuildTag = Union[Tuple[()], Tuple[int, str]]
NormalizedName = NewType("NormalizedName", str)
class InvalidWheelFilename(ValueError):
"""
An invalid wheel filename was found, users should refer to PEP 427.
"""
class InvalidSdistFilename(ValueError):
"""
An invalid sdist filename was found, users should refer to the packaging user guide.
"""
_canonicalize_regex = re.compile(r"[-_.]+")
# PEP 427: The build number must start with a digit.
_build_tag_regex = re.compile(r"(\d+)(.*)")
def canonicalize_name(name: str) -> NormalizedName:
# This is taken from PEP 503.
value = _canonicalize_regex.sub("-", name).lower()
return cast(NormalizedName, value)
def canonicalize_version(version: Union[Version, str]) -> str:
"""
This is very similar to Version.__str__, but has one subtle difference
with the way it handles the release segment.
"""
if isinstance(version, str):
try:
parsed = Version(version)
except InvalidVersion:
# Legacy versions cannot be normalized
return version
else:
parsed = version
parts = []
# Epoch
if parsed.epoch != 0:
parts.append(f"{parsed.epoch}!")
# Release segment
# NB: This strips trailing '.0's to normalize
parts.append(re.sub(r"(\.0)+$", "", ".".join(str(x) for x in parsed.release)))
# Pre-release
if parsed.pre is not None:
parts.append("".join(str(x) for x in parsed.pre))
# Post-release
if parsed.post is not None:
parts.append(f".post{parsed.post}")
# Development release
if parsed.dev is not None:
parts.append(f".dev{parsed.dev}")
# Local version segment
if parsed.local is not None:
parts.append(f"+{parsed.local}")
return "".join(parts)
def parse_wheel_filename(
filename: str,
) -> Tuple[NormalizedName, Version, BuildTag, FrozenSet[Tag]]:
if not filename.endswith(".whl"):
raise InvalidWheelFilename(
f"Invalid wheel filename (extension must be '.whl'): {filename}"
)
filename = filename[:-4]
dashes = filename.count("-")
if dashes not in (4, 5):
raise InvalidWheelFilename(
f"Invalid wheel filename (wrong number of parts): {filename}"
)
parts = filename.split("-", dashes - 2)
name_part = parts[0]
# See PEP 427 for the rules on escaping the project name
if "__" in name_part or re.match(r"^[\w\d._]*$", name_part, re.UNICODE) is None:
raise InvalidWheelFilename(f"Invalid project name: {filename}")
name = canonicalize_name(name_part)
version = Version(parts[1])
if dashes == 5:
build_part = parts[2]
build_match = _build_tag_regex.match(build_part)
if build_match is None:
raise InvalidWheelFilename(
f"Invalid build number: {build_part} in '{filename}'"
)
build = cast(BuildTag, (int(build_match.group(1)), build_match.group(2)))
else:
build = ()
tags = parse_tag(parts[-1])
return (name, version, build, tags)
def parse_sdist_filename(filename: str) -> Tuple[NormalizedName, Version]:
if filename.endswith(".tar.gz"):
file_stem = filename[: -len(".tar.gz")]
elif filename.endswith(".zip"):
file_stem = filename[: -len(".zip")]
else:
raise InvalidSdistFilename(
f"Invalid sdist filename (extension must be '.tar.gz' or '.zip'):"
f" {filename}"
)
# We are requiring a PEP 440 version, which cannot contain dashes,
# so we split on the last dash.
name_part, sep, version_part = file_stem.rpartition("-")
if not sep:
raise InvalidSdistFilename(f"Invalid sdist filename: {filename}")
name = canonicalize_name(name_part)
version = Version(version_part)
return (name, version)

View File

@ -0,0 +1,504 @@
# This file is dual licensed under the terms of the Apache License, Version
# 2.0, and the BSD License. See the LICENSE file in the root of this repository
# for complete details.
import collections
import itertools
import re
import warnings
from typing import Callable, Iterator, List, Optional, SupportsInt, Tuple, Union
from ._structures import Infinity, InfinityType, NegativeInfinity, NegativeInfinityType
__all__ = ["parse", "Version", "LegacyVersion", "InvalidVersion", "VERSION_PATTERN"]
InfiniteTypes = Union[InfinityType, NegativeInfinityType]
PrePostDevType = Union[InfiniteTypes, Tuple[str, int]]
SubLocalType = Union[InfiniteTypes, int, str]
LocalType = Union[
NegativeInfinityType,
Tuple[
Union[
SubLocalType,
Tuple[SubLocalType, str],
Tuple[NegativeInfinityType, SubLocalType],
],
...,
],
]
CmpKey = Tuple[
int, Tuple[int, ...], PrePostDevType, PrePostDevType, PrePostDevType, LocalType
]
LegacyCmpKey = Tuple[int, Tuple[str, ...]]
VersionComparisonMethod = Callable[
[Union[CmpKey, LegacyCmpKey], Union[CmpKey, LegacyCmpKey]], bool
]
_Version = collections.namedtuple(
"_Version", ["epoch", "release", "dev", "pre", "post", "local"]
)
def parse(version: str) -> Union["LegacyVersion", "Version"]:
"""
Parse the given version string and return either a :class:`Version` object
or a :class:`LegacyVersion` object depending on if the given version is
a valid PEP 440 version or a legacy version.
"""
try:
return Version(version)
except InvalidVersion:
return LegacyVersion(version)
class InvalidVersion(ValueError):
"""
An invalid version was found, users should refer to PEP 440.
"""
class _BaseVersion:
_key: Union[CmpKey, LegacyCmpKey]
def __hash__(self) -> int:
return hash(self._key)
# Please keep the duplicated `isinstance` check
# in the six comparisons hereunder
# unless you find a way to avoid adding overhead function calls.
def __lt__(self, other: "_BaseVersion") -> bool:
if not isinstance(other, _BaseVersion):
return NotImplemented
return self._key < other._key
def __le__(self, other: "_BaseVersion") -> bool:
if not isinstance(other, _BaseVersion):
return NotImplemented
return self._key <= other._key
def __eq__(self, other: object) -> bool:
if not isinstance(other, _BaseVersion):
return NotImplemented
return self._key == other._key
def __ge__(self, other: "_BaseVersion") -> bool:
if not isinstance(other, _BaseVersion):
return NotImplemented
return self._key >= other._key
def __gt__(self, other: "_BaseVersion") -> bool:
if not isinstance(other, _BaseVersion):
return NotImplemented
return self._key > other._key
def __ne__(self, other: object) -> bool:
if not isinstance(other, _BaseVersion):
return NotImplemented
return self._key != other._key
class LegacyVersion(_BaseVersion):
def __init__(self, version: str) -> None:
self._version = str(version)
self._key = _legacy_cmpkey(self._version)
warnings.warn(
"Creating a LegacyVersion has been deprecated and will be "
"removed in the next major release",
DeprecationWarning,
)
def __str__(self) -> str:
return self._version
def __repr__(self) -> str:
return f"<LegacyVersion('{self}')>"
@property
def public(self) -> str:
return self._version
@property
def base_version(self) -> str:
return self._version
@property
def epoch(self) -> int:
return -1
@property
def release(self) -> None:
return None
@property
def pre(self) -> None:
return None
@property
def post(self) -> None:
return None
@property
def dev(self) -> None:
return None
@property
def local(self) -> None:
return None
@property
def is_prerelease(self) -> bool:
return False
@property
def is_postrelease(self) -> bool:
return False
@property
def is_devrelease(self) -> bool:
return False
_legacy_version_component_re = re.compile(r"(\d+ | [a-z]+ | \.| -)", re.VERBOSE)
_legacy_version_replacement_map = {
"pre": "c",
"preview": "c",
"-": "final-",
"rc": "c",
"dev": "@",
}
def _parse_version_parts(s: str) -> Iterator[str]:
for part in _legacy_version_component_re.split(s):
part = _legacy_version_replacement_map.get(part, part)
if not part or part == ".":
continue
if part[:1] in "0123456789":
# pad for numeric comparison
yield part.zfill(8)
else:
yield "*" + part
# ensure that alpha/beta/candidate are before final
yield "*final"
def _legacy_cmpkey(version: str) -> LegacyCmpKey:
# We hardcode an epoch of -1 here. A PEP 440 version can only have a epoch
# greater than or equal to 0. This will effectively put the LegacyVersion,
# which uses the defacto standard originally implemented by setuptools,
# as before all PEP 440 versions.
epoch = -1
# This scheme is taken from pkg_resources.parse_version setuptools prior to
# it's adoption of the packaging library.
parts: List[str] = []
for part in _parse_version_parts(version.lower()):
if part.startswith("*"):
# remove "-" before a prerelease tag
if part < "*final":
while parts and parts[-1] == "*final-":
parts.pop()
# remove trailing zeros from each series of numeric parts
while parts and parts[-1] == "00000000":
parts.pop()
parts.append(part)
return epoch, tuple(parts)
# Deliberately not anchored to the start and end of the string, to make it
# easier for 3rd party code to reuse
VERSION_PATTERN = r"""
v?
(?:
(?:(?P<epoch>[0-9]+)!)? # epoch
(?P<release>[0-9]+(?:\.[0-9]+)*) # release segment
(?P<pre> # pre-release
[-_\.]?
(?P<pre_l>(a|b|c|rc|alpha|beta|pre|preview))
[-_\.]?
(?P<pre_n>[0-9]+)?
)?
(?P<post> # post release
(?:-(?P<post_n1>[0-9]+))
|
(?:
[-_\.]?
(?P<post_l>post|rev|r)
[-_\.]?
(?P<post_n2>[0-9]+)?
)
)?
(?P<dev> # dev release
[-_\.]?
(?P<dev_l>dev)
[-_\.]?
(?P<dev_n>[0-9]+)?
)?
)
(?:\+(?P<local>[a-z0-9]+(?:[-_\.][a-z0-9]+)*))? # local version
"""
class Version(_BaseVersion):
_regex = re.compile(r"^\s*" + VERSION_PATTERN + r"\s*$", re.VERBOSE | re.IGNORECASE)
def __init__(self, version: str) -> None:
# Validate the version and parse it into pieces
match = self._regex.search(version)
if not match:
raise InvalidVersion(f"Invalid version: '{version}'")
# Store the parsed out pieces of the version
self._version = _Version(
epoch=int(match.group("epoch")) if match.group("epoch") else 0,
release=tuple(int(i) for i in match.group("release").split(".")),
pre=_parse_letter_version(match.group("pre_l"), match.group("pre_n")),
post=_parse_letter_version(
match.group("post_l"), match.group("post_n1") or match.group("post_n2")
),
dev=_parse_letter_version(match.group("dev_l"), match.group("dev_n")),
local=_parse_local_version(match.group("local")),
)
# Generate a key which will be used for sorting
self._key = _cmpkey(
self._version.epoch,
self._version.release,
self._version.pre,
self._version.post,
self._version.dev,
self._version.local,
)
def __repr__(self) -> str:
return f"<Version('{self}')>"
def __str__(self) -> str:
parts = []
# Epoch
if self.epoch != 0:
parts.append(f"{self.epoch}!")
# Release segment
parts.append(".".join(str(x) for x in self.release))
# Pre-release
if self.pre is not None:
parts.append("".join(str(x) for x in self.pre))
# Post-release
if self.post is not None:
parts.append(f".post{self.post}")
# Development release
if self.dev is not None:
parts.append(f".dev{self.dev}")
# Local version segment
if self.local is not None:
parts.append(f"+{self.local}")
return "".join(parts)
@property
def epoch(self) -> int:
_epoch: int = self._version.epoch
return _epoch
@property
def release(self) -> Tuple[int, ...]:
_release: Tuple[int, ...] = self._version.release
return _release
@property
def pre(self) -> Optional[Tuple[str, int]]:
_pre: Optional[Tuple[str, int]] = self._version.pre
return _pre
@property
def post(self) -> Optional[int]:
return self._version.post[1] if self._version.post else None
@property
def dev(self) -> Optional[int]:
return self._version.dev[1] if self._version.dev else None
@property
def local(self) -> Optional[str]:
if self._version.local:
return ".".join(str(x) for x in self._version.local)
else:
return None
@property
def public(self) -> str:
return str(self).split("+", 1)[0]
@property
def base_version(self) -> str:
parts = []
# Epoch
if self.epoch != 0:
parts.append(f"{self.epoch}!")
# Release segment
parts.append(".".join(str(x) for x in self.release))
return "".join(parts)
@property
def is_prerelease(self) -> bool:
return self.dev is not None or self.pre is not None
@property
def is_postrelease(self) -> bool:
return self.post is not None
@property
def is_devrelease(self) -> bool:
return self.dev is not None
@property
def major(self) -> int:
return self.release[0] if len(self.release) >= 1 else 0
@property
def minor(self) -> int:
return self.release[1] if len(self.release) >= 2 else 0
@property
def micro(self) -> int:
return self.release[2] if len(self.release) >= 3 else 0
def _parse_letter_version(
letter: str, number: Union[str, bytes, SupportsInt]
) -> Optional[Tuple[str, int]]:
if letter:
# We consider there to be an implicit 0 in a pre-release if there is
# not a numeral associated with it.
if number is None:
number = 0
# We normalize any letters to their lower case form
letter = letter.lower()
# We consider some words to be alternate spellings of other words and
# in those cases we want to normalize the spellings to our preferred
# spelling.
if letter == "alpha":
letter = "a"
elif letter == "beta":
letter = "b"
elif letter in ["c", "pre", "preview"]:
letter = "rc"
elif letter in ["rev", "r"]:
letter = "post"
return letter, int(number)
if not letter and number:
# We assume if we are given a number, but we are not given a letter
# then this is using the implicit post release syntax (e.g. 1.0-1)
letter = "post"
return letter, int(number)
return None
_local_version_separators = re.compile(r"[\._-]")
def _parse_local_version(local: str) -> Optional[LocalType]:
"""
Takes a string like abc.1.twelve and turns it into ("abc", 1, "twelve").
"""
if local is not None:
return tuple(
part.lower() if not part.isdigit() else int(part)
for part in _local_version_separators.split(local)
)
return None
def _cmpkey(
epoch: int,
release: Tuple[int, ...],
pre: Optional[Tuple[str, int]],
post: Optional[Tuple[str, int]],
dev: Optional[Tuple[str, int]],
local: Optional[Tuple[SubLocalType]],
) -> CmpKey:
# When we compare a release version, we want to compare it with all of the
# trailing zeros removed. So we'll use a reverse the list, drop all the now
# leading zeros until we come to something non zero, then take the rest
# re-reverse it back into the correct order and make it a tuple and use
# that for our sorting key.
_release = tuple(
reversed(list(itertools.dropwhile(lambda x: x == 0, reversed(release))))
)
# We need to "trick" the sorting algorithm to put 1.0.dev0 before 1.0a0.
# We'll do this by abusing the pre segment, but we _only_ want to do this
# if there is not a pre or a post segment. If we have one of those then
# the normal sorting rules will handle this case correctly.
if pre is None and post is None and dev is not None:
_pre: PrePostDevType = NegativeInfinity
# Versions without a pre-release (except as noted above) should sort after
# those with one.
elif pre is None:
_pre = Infinity
else:
_pre = pre
# Versions without a post segment should sort before those with one.
if post is None:
_post: PrePostDevType = NegativeInfinity
else:
_post = post
# Versions without a development segment should sort after those with one.
if dev is None:
_dev: PrePostDevType = Infinity
else:
_dev = dev
if local is None:
# Versions without a local segment should sort before those with one.
_local: LocalType = NegativeInfinity
else:
# Versions with a local segment need that segment parsed to implement
# the sorting rules in PEP440.
# - Alpha numeric segments sort before numeric segments
# - Alpha numeric segments sort lexicographically
# - Numeric segments sort numerically
# - Shorter versions sort before longer versions when the prefixes
# match exactly
_local = tuple(
(i, "") if isinstance(i, int) else (NegativeInfinity, i) for i in local
)
return epoch, _release, _pre, _post, _dev, _local

File diff suppressed because it is too large Load Diff