from __future__ import absolute_import, division, print_function, unicode_literals
import operator as op
import re
from .common.compat import string_types, zip, zip_longest
from .exceptions import CondaValueError, InvalidSpecError
# normalized_version() is needed by conda-env
# It is currently being pulled from resolve instead, but
# eventually it ought to come from here
def normalized_version(version):
return VersionOrder(version)
def ver_eval(vtest, spec):
return VersionSpec(spec).match(vtest)
version_check_re = re.compile(r'^[\*\.\+!_0-9a-z]+$')
version_split_re = re.compile('([0-9]+|[*]+|[^0-9*]+)')
version_cache = {}
class VersionOrder(object):
"""
This class implements an order relation between version strings.
Version strings can contain the usual alphanumeric characters
(A-Za-z0-9), separated into components by dots and underscores. Empty
segments (i.e. two consecutive dots, a leading/trailing underscore)
are not permitted. An optional epoch number - an integer
followed by '!' - can preceed the actual version string
(this is useful to indicate a change in the versioning
scheme itself). Version comparison is case-insensitive.
Conda supports six types of version strings:
* Release versions contain only integers, e.g. '1.0', '2.3.5'.
* Pre-release versions use additional letters such as 'a' or 'rc',
for example '1.0a1', '1.2.beta3', '2.3.5rc3'.
* Development versions are indicated by the string 'dev',
for example '1.0dev42', '2.3.5.dev12'.
* Post-release versions are indicated by the string 'post',
for example '1.0post1', '2.3.5.post2'.
* Tagged versions have a suffix that specifies a particular
property of interest, e.g. '1.1.parallel'. Tags can be added
to any of the preceding four types. As far as sorting is concerned,
tags are treated like strings in pre-release versions.
* An optional local version string separated by '+' can be appended
to the main (upstream) version string. It is only considered
in comparisons when the main versions are equal, but otherwise
handled in exactly the same manner.
To obtain a predictable version ordering, it is crucial to keep the
version number scheme of a given package consistent over time.
Specifically,
* version strings should always have the same number of components
(except for an optional tag suffix or local version string),
* letters/strings indicating non-release versions should always
occur at the same position.
Before comparison, version strings are parsed as follows:
* They are first split into epoch, version number, and local version
number at '!' and '+' respectively. If there is no '!', the epoch is
set to 0. If there is no '+', the local version is empty.
* The version part is then split into components at '.' and '_'.
* Each component is split again into runs of numerals and non-numerals
* Subcomponents containing only numerals are converted to integers.
* Strings are converted to lower case, with special treatment for 'dev'
and 'post'.
* When a component starts with a letter, the fillvalue 0 is inserted
to keep numbers and strings in phase, resulting in '1.1.a1' == 1.1.0a1'.
* The same is repeated for the local version part.
Examples:
1.2g.beta15.rc => [[0], [1], [2, 'g'], [0, 'beta', 15], [0, 'rc']]
1!2.15.1_ALPHA => [[1], [2], [15], [1, '_alpha']]
The resulting lists are compared lexicographically, where the following
rules are applied to each pair of corresponding subcomponents:
* integers are compared numerically
* strings are compared lexicographically, case-insensitive
* strings are smaller than integers, except
* 'dev' versions are smaller than all corresponding versions of other types
* 'post' versions are greater than all corresponding versions of other types
* if a subcomponent has no correspondent, the missing correspondent is
treated as integer 0 to ensure '1.1' == '1.1.0'.
The resulting order is:
0.4
< 0.4.0
< 0.4.1.rc
== 0.4.1.RC # case-insensitive comparison
< 0.4.1
< 0.5a1
< 0.5b3
< 0.5C1 # case-insensitive comparison
< 0.5
< 0.9.6
< 0.960923
< 1.0
< 1.1dev1 # special case 'dev'
< 1.1a1
< 1.1.0dev1 # special case 'dev'
== 1.1.dev1 # 0 is inserted before string
< 1.1.a1
< 1.1.0rc1
< 1.1.0
== 1.1
< 1.1.0post1 # special case 'post'
== 1.1.post1 # 0 is inserted before string
< 1.1post1 # special case 'post'
< 1996.07.12
< 1!0.4.1 # epoch increased
< 1!3.1.1.6
< 2!0.4.1 # epoch increased again
Some packages (most notably openssl) have incompatible version conventions.
In particular, openssl interprets letters as version counters rather than
pre-release identifiers. For openssl, the relation
1.0.1 < 1.0.1a => True # for openssl
holds, whereas conda packages use the opposite ordering. You can work-around
this problem by appending a dash to plain version numbers:
1.0.1a => 1.0.1post.a # ensure correct ordering for openssl
"""
def __new__(cls, version):
if isinstance(version, cls):
return version
self = version_cache.get(version)
if self is not None:
return self
self = version_cache[version] = object.__new__(cls)
# when fillvalue == 0 => 1.1 == 1.1.0
# when fillvalue == -1 => 1.1 < 1.1.0
self.fillvalue = 0
message = "Malformed version string '%s': " % version
# version comparison is case-insensitive
version = version.strip().rstrip().lower()
# basic validity checks
if version == '':
raise CondaValueError("Empty version string.")
invalid = not version_check_re.match(version)
if invalid and '-' in version and '_' not in version:
# Allow for dashes as long as there are no underscores
# as well, by converting the former to the latter.
version = version.replace('-', '_')
invalid = not version_check_re.match(version)
if invalid:
raise CondaValueError(message + "invalid character(s).")
self.norm_version = version
# find epoch
version = version.split('!')
if len(version) == 1:
# epoch not given => set it to '0'
epoch = ['0']
elif len(version) == 2:
# epoch given, must be an integer
if not version[0].isdigit():
raise CondaValueError(message + "epoch must be an integer.")
epoch = [version[0]]
else:
raise CondaValueError(message + "duplicated epoch separator '!'.")
# find local version string
version = version[-1].split('+')
if len(version) == 1:
# no local version
self.local = []
elif len(version) == 2:
# local version given
self.local = version[1].replace('_', '.').split('.')
else:
raise CondaValueError(message + "duplicated local version separator '+'.")
# split version
self.version = epoch + version[0].replace('_', '.').split('.')
# split components into runs of numerals and non-numerals,
# convert numerals to int, handle special strings
for v in (self.version, self.local):
for k in range(len(v)):
c = version_split_re.findall(v[k])
if not c:
raise CondaValueError(message + "empty version component.")
for j in range(len(c)):
if c[j].isdigit():
c[j] = int(c[j])
elif c[j] == 'post':
# ensure number < 'post' == infinity
c[j] = float('inf')
elif c[j] == 'dev':
# ensure '*' < 'DEV' < '_' < 'a' < number
# by upper-casing (all other strings are lower case)
c[j] = 'DEV'
if v[k][0].isdigit():
v[k] = c
else:
# components shall start with a number to keep numbers and
# strings in phase => prepend fillvalue
v[k] = [self.fillvalue] + c
return self
def __str__(self):
return self.norm_version
def _eq(self, t1, t2):
for v1, v2 in zip_longest(t1, t2, fillvalue=[]):
for c1, c2 in zip_longest(v1, v2, fillvalue=self.fillvalue):
if c1 != c2:
return False
return True
def __eq__(self, other):
return (self._eq(self.version, other.version) and
self._eq(self.local, other.local))
def startswith(self, other):
# Tests if the version lists match up to the last element in "other".
if other.local:
if not self._eq(self.version, other.version):
return False
t1 = self.local
t2 = other.local
elif other.version:
t1 = self.version
t2 = other.version
else:
return True
nt = len(t2) - 1
if not self._eq(t1[:nt], t2[:nt]):
return False
v1 = [] if len(t1) <= nt else t1[nt]
v2 = t2[nt]
nt = len(v2) - 1
if not self._eq([v1[:nt]], [v2[:nt]]):
return False
c1 = self.fillvalue if len(v1) <= nt else v1[nt]
c2 = v2[nt]
if isinstance(c2, string_types):
return isinstance(c1, string_types) and c1.startswith(c2)
return c1 == c2
def __ne__(self, other):
return not (self == other)
def __lt__(self, other):
for t1, t2 in zip([self.version, self.local], [other.version, other.local]):
for v1, v2 in zip_longest(t1, t2, fillvalue=[]):
for c1, c2 in zip_longest(v1, v2, fillvalue=self.fillvalue):
if c1 == c2:
continue
elif isinstance(c1, string_types):
if not isinstance(c2, string_types):
# str < int
return True
elif isinstance(c2, string_types):
# not (int < str)
return False
# c1 and c2 have the same type
return c1 < c2
# self == other
return False
def __gt__(self, other):
return other < self
def __le__(self, other):
return not (other < self)
def __ge__(self, other):
return not (self < other)
# This RE matches the operators '==', '!=', '<=', '>=', '<', '>'
# followed by a version string. It rejects expressions like
# '<= 1.2' (space after operator), '<>1.2' (unknown operator),
# and '<=!1.2' (nonsensical operator).
version_relation_re = re.compile(r'(==|!=|<=|>=|<|>)(?![=<>!])(\S+)$')
regex_split_re = re.compile(r'(\^\S+?\$)')
regex_split_converter = {
'|': 'any',
',': 'all',
}
opdict = {'==': op.__eq__, '!=': op.__ne__, '<=': op.__le__,
'>=': op.__ge__, '<': op.__lt__, '>': op.__gt__}
class VersionSpec(object):
def exact_match_(self, vspec):
return self.spec == vspec
def regex_match_(self, vspec):
return bool(self.regex.match(vspec))
def veval_match_(self, vspec):
return self.op(VersionOrder(vspec), self.cmp)
def all_match_(self, vspec):
return all(s.match(vspec) for s in self.spec[1])
def any_match_(self, vspec):
return any(s.match(vspec) for s in self.spec[1])
def triv_match_(self, vspec):
return True
def __new__(cls, spec):
if isinstance(spec, cls):
return spec
self = object.__new__(cls)
self.spec = spec
if isinstance(spec, tuple):
self.match = self.all_match_ if spec[0] == 'all' else self.any_match_
elif regex_split_re.match(spec):
m = regex_split_re.match(spec)
first = m.group()
operator = spec[m.end()] if len(spec) > m.end() else None
if operator is None:
self.spec = first
self.regex = re.compile(spec)
self.match = self.regex_match_
else:
return VersionSpec((regex_split_converter[operator],
tuple(VersionSpec(s) for s in (first, spec[m.end()+1:]))))
elif '|' in spec:
return VersionSpec(('any', tuple(VersionSpec(s) for s in spec.split('|'))))
elif ',' in spec:
return VersionSpec(('all', tuple(VersionSpec(s) for s in spec.split(','))))
elif spec.startswith(('=', '<', '>', '!')):
m = version_relation_re.match(spec)
if m is None:
raise InvalidSpecError(spec)
op, b = m.groups()
self.op = opdict[op]
self.cmp = VersionOrder(b)
self.match = self.veval_match_
elif spec == '*':
self.match = self.triv_match_
elif '*' in spec.rstrip('*'):
self.spec = spec
rx = spec.replace('.', r'\.')
rx = rx.replace('+', r'\+')
rx = rx.replace('*', r'.*')
rx = r'^(?:%s)$' % rx
self.regex = re.compile(rx)
self.match = self.regex_match_
elif spec.endswith('*'):
self.op = VersionOrder.startswith
self.cmp = VersionOrder(spec.rstrip('*').rstrip('.'))
self.match = self.veval_match_
else:
self.match = self.exact_match_
return self
def str(self, inand=False):
s = self.spec
if isinstance(s, tuple):
newand = not inand and s[0] == all
inand = inand and s[0] == any
s = (',' if s[0] == 'all' else '|').join(x.str(newand) for x in s[1])
if inand:
s = '(%s)' % s
return s
def is_exact(self):
return self.match == self.exact_match_
def __str__(self):
return self.str()
def __repr__(self):
return "VersionSpec('%s')" % self.str()
def __and__(self, other):
if not isinstance(other, VersionSpec):
other = VersionSpec(other)
return VersionSpec((all, (self, other)))
def __or__(self, other):
if not isinstance(other, VersionSpec):
other = VersionSpec(other)
return VersionSpec((any, (self, other)))