问题
I am walking a directory that contains eggs to add those eggs to the sys.path. If there are two versions of the same .egg in the directory, I want to add only the latest one.
I have a regular expression r\"^(?P<eggName>\\w+)-(?P<eggVersion>[\\d\\.]+)-.+\\.egg$ to extract the name and version from the filename. The problem is comparing the version number, which is a string like 2.3.1.
Since I\'m comparing strings, 2 sorts above 10, but that\'s not correct for versions.
>>> \"2.3.1\" > \"10.1.1\"
True
I could do some splitting, parsing, casting to int, etc., and I would eventually get a workaround. But this is Python, not Java. Is there an elegant way to compare version strings?
回答1:
Use packaging.version.parse.
>>> from packaging import version
>>> version.parse("2.3.1") < version.parse("10.1.2")
True
>>> version.parse("1.3.a4") < version.parse("10.1.2")
True
>>> isinstance(version.parse("1.3.a4"), version.Version)
True
>>> isinstance(version.parse("1.3.xy123"), version.LegacyVersion)
True
>>> version.Version("1.3.xy123")
Traceback (most recent call last):
...
packaging.version.InvalidVersion: Invalid version: '1.3.xy123'
packaging.version.parse is a third-party utility but is used by setuptools (so you probably already have it installed) and is conformant to the current PEP 440; it will return a packaging.version.Version if the version is compliant and a packaging.version.LegacyVersion if not. The latter will always sort before valid versions.
An ancient alternative still used by a lot of software is distutils.version, built in but undocumented and conformant only to the superseded PEP 386;
>>> from distutils.version import LooseVersion, StrictVersion
>>> LooseVersion("2.3.1") < LooseVersion("10.1.2")
True
>>> StrictVersion("2.3.1") < StrictVersion("10.1.2")
True
>>> StrictVersion("1.3.a4")
Traceback (most recent call last):
...
ValueError: invalid version number '1.3.a4'
As you can see it sees valid PEP 440 versions as “not strict” and therefore doesn’t match modern Python’s notion of what a valid version is.
As distutils.version is undocumented, here's the relevant docstrings.
回答2:
setuptools defines parse_version(). This implements PEP 0440 -- Version Identification and is also able to parse versions that don't follow the PEP. This function is used by easy_install and pip to handle version comparison. From the docs:
Parsed a project's version string as defined by PEP 440. The returned value will be an object that represents the version. These objects may be compared to each other and sorted. The sorting algorithm is as defined by PEP 440 with the addition that any version which is not a valid PEP 440 version will be considered less than any valid PEP 440 version and the invalid versions will continue sorting using the original algorithm.
The "original algorithm" referenced was defined in older versions of the docs, before PEP 440 existed.
Semantically, the format is a rough cross between distutils'
StrictVersionandLooseVersionclasses; if you give it versions that would work withStrictVersion, then they will compare the same way. Otherwise, comparisons are more like a "smarter" form ofLooseVersion. It is possible to create pathological version coding schemes that will fool this parser, but they should be very rare in practice.
The documentation provides some examples:
If you want to be certain that your chosen numbering scheme works the way you think it will, you can use the
pkg_resources.parse_version()function to compare different version numbers:>>> from pkg_resources import parse_version >>> parse_version('1.9.a.dev') == parse_version('1.9a0dev') True >>> parse_version('2.1-rc2') < parse_version('2.1') True >>> parse_version('0.6a9dev-r41475') < parse_version('0.6a9') True
If you're not using setuptools, the packaging project splits this and other packaging-related functionality into a separate library.
from packaging import version
version.parse('1.0.3.dev')
from pkg_resources import parse_version
parse_version('1.0.3.dev')
回答3:
def versiontuple(v):
return tuple(map(int, (v.split("."))))
>>> versiontuple("2.3.1") > versiontuple("10.1.1")
False
回答4:
What's wrong with transforming the version string into a tuple and going from there? Seems elegant enough for me
>>> (2,3,1) < (10,1,1)
True
>>> (2,3,1) < (10,1,1,1)
True
>>> (2,3,1,10) < (10,1,1,1)
True
>>> (10,3,1,10) < (10,1,1,1)
False
>>> (10,3,1,10) < (10,4,1,1)
True
@kindall's solution is a quick example of how good the code would look.
回答5:
There is packaging package available, which will allow you to compare versions as per PEP-440, as well as legacy versions.
>>> from packaging.version import Version, LegacyVersion
>>> Version('1.1') < Version('1.2')
True
>>> Version('1.2.dev4+deadbeef') < Version('1.2')
True
>>> Version('1.2.8.5') <= Version('1.2')
False
>>> Version('1.2.8.5') <= Version('1.2.8.6')
True
Legacy version support:
>>> LegacyVersion('1.2.8.5-5-gdeadbeef')
<LegacyVersion('1.2.8.5-5-gdeadbeef')>
Comparing legacy version with PEP-440 version.
>>> LegacyVersion('1.2.8.5-5-gdeadbeef') < Version('1.2.8.6')
True
回答6:
You can use the semver package to determine if a version satisfies a semantic version requirement. This is not the same as comparing two actual versions, but is a type of comparison.
For example, version 3.6.0+1234 should be the same as 3.6.0.
import semver
semver.match('3.6.0+1234', '==3.6.0')
# True
from packaging import version
version.parse('3.6.0+1234') == version.parse('3.6.0')
# False
from distutils.version import LooseVersion
LooseVersion('3.6.0+1234') == LooseVersion('3.6.0')
# False
回答7:
Posting my full function based on Kindall's solution. I was able to support any alphanumeric characters mixed in with the numbers by padding each version section with leading zeros.
While certainly not as pretty as his one-liner function, it seems to work well with alpha-numeric version numbers. (Just be sure to set the zfill(#) value appropriately if you have long strings in your versioning system.)
def versiontuple(v):
filled = []
for point in v.split("."):
filled.append(point.zfill(8))
return tuple(filled)
.
>>> versiontuple("10a.4.5.23-alpha") > versiontuple("2a.4.5.23-alpha")
True
>>> "10a.4.5.23-alpha" > "2a.4.5.23-alpha"
False
回答8:
The way that setuptools does it, it uses the pkg_resources.parse_version function. It should be PEP440 compliant.
Example:
#! /usr/bin/python
# -*- coding: utf-8 -*-
"""Example comparing two PEP440 formatted versions
"""
import pkg_resources
VERSION_A = pkg_resources.parse_version("1.0.1-beta.1")
VERSION_B = pkg_resources.parse_version("v2.67-rc")
VERSION_C = pkg_resources.parse_version("2.67rc")
VERSION_D = pkg_resources.parse_version("2.67rc1")
VERSION_E = pkg_resources.parse_version("1.0.0")
print(VERSION_A)
print(VERSION_B)
print(VERSION_C)
print(VERSION_D)
print(VERSION_A==VERSION_B) #FALSE
print(VERSION_B==VERSION_C) #TRUE
print(VERSION_C==VERSION_D) #FALSE
print(VERSION_A==VERSION_E) #FALSE
回答9:
I was looking for a solution which wouldn't add any new dependencies. Check out the following (Python 3) solution:
class VersionManager:
@staticmethod
def compare_version_tuples(
major_a, minor_a, bugfix_a,
major_b, minor_b, bugfix_b,
):
"""
Compare two versions a and b, each consisting of 3 integers
(compare these as tuples)
version_a: major_a, minor_a, bugfix_a
version_b: major_b, minor_b, bugfix_b
:param major_a: first part of a
:param minor_a: second part of a
:param bugfix_a: third part of a
:param major_b: first part of b
:param minor_b: second part of b
:param bugfix_b: third part of b
:return: 1 if a > b
0 if a == b
-1 if a < b
"""
tuple_a = major_a, minor_a, bugfix_a
tuple_b = major_b, minor_b, bugfix_b
if tuple_a > tuple_b:
return 1
if tuple_b > tuple_a:
return -1
return 0
@staticmethod
def compare_version_integers(
major_a, minor_a, bugfix_a,
major_b, minor_b, bugfix_b,
):
"""
Compare two versions a and b, each consisting of 3 integers
(compare these as integers)
version_a: major_a, minor_a, bugfix_a
version_b: major_b, minor_b, bugfix_b
:param major_a: first part of a
:param minor_a: second part of a
:param bugfix_a: third part of a
:param major_b: first part of b
:param minor_b: second part of b
:param bugfix_b: third part of b
:return: 1 if a > b
0 if a == b
-1 if a < b
"""
# --
if major_a > major_b:
return 1
if major_b > major_a:
return -1
# --
if minor_a > minor_b:
return 1
if minor_b > minor_a:
return -1
# --
if bugfix_a > bugfix_b:
return 1
if bugfix_b > bugfix_a:
return -1
# --
return 0
@staticmethod
def test_compare_versions():
functions = [
(VersionManager.compare_version_tuples, "VersionManager.compare_version_tuples"),
(VersionManager.compare_version_integers, "VersionManager.compare_version_integers"),
]
data = [
# expected result, version a, version b
(1, 1, 0, 0, 0, 0, 1),
(1, 1, 5, 5, 0, 5, 5),
(1, 1, 0, 5, 0, 0, 5),
(1, 0, 2, 0, 0, 1, 1),
(1, 2, 0, 0, 1, 1, 0),
(0, 0, 0, 0, 0, 0, 0),
(0, -1, -1, -1, -1, -1, -1), # works even with negative version numbers :)
(0, 2, 2, 2, 2, 2, 2),
(-1, 5, 5, 0, 6, 5, 0),
(-1, 5, 5, 0, 5, 9, 0),
(-1, 5, 5, 5, 5, 5, 6),
(-1, 2, 5, 7, 2, 5, 8),
]
count = len(data)
index = 1
for expected_result, major_a, minor_a, bugfix_a, major_b, minor_b, bugfix_b in data:
for function_callback, function_name in functions:
actual_result = function_callback(
major_a=major_a, minor_a=minor_a, bugfix_a=bugfix_a,
major_b=major_b, minor_b=minor_b, bugfix_b=bugfix_b,
)
outcome = expected_result == actual_result
message = "{}/{}: {}: {}: a={}.{}.{} b={}.{}.{} expected={} actual={}".format(
index, count,
"ok" if outcome is True else "fail",
function_name,
major_a, minor_a, bugfix_a,
major_b, minor_b, bugfix_b,
expected_result, actual_result
)
print(message)
assert outcome is True
index += 1
# test passed!
if __name__ == '__main__':
VersionManager.test_compare_versions()
EDIT: added variant with tuple comparison. Of course the variant with tuple comparison is nicer, but I was looking for the variant with integer comparison
来源：https://stackoverflow.com/questions/11887762/how-do-i-compare-version-numbers-in-python