Currently the Django Project supports 1.4, 1.7 and 1.8. In my setup.py I want to reflect these versions as being supported.
install_requires=['Django>=1.4.2,<1.8.99,!=1.5,!=1.6']
However this still allows 1.5.x and 1.6.x releases. How can I exclude a complete range?
Setuptools lists the following valid requirements as an example:
PickyThing<1.6,>1.9,!=1.9.6,<2.0a0,==2.4c1
However this doesn't work with pip (it should at least match 1.4.x / 1.5.x):
No matching distribution found for PickyThing!=1.9.6,<1.6,<2.0a0,==2.4c1,>1.9
Update with example
For example; I only want to include currently supported versions of Django. This would be 1.4.x, 1.7.x and 1.8.x. So I would write;
#setup.py
install_requires=['Django>=1.4.2,<1.4.99,>=1.7,<1.8.99']
However if I run pip install -e . on this project, it fails with;
Collecting Django<1.4.99,<1.8.99,>=1.4.2,>=1.7 (from ...)
Could not find a version that satisfies the requirement Django<1.4.99,<1.8.99,>=1.4.2,>=1.7 (from django-two-factor-auth==1.2.0) (from versions: 1.1.3, 1.1.4, 1.2, 1.2.1, 1.2.2, 1.2.3, 1.2.4, 1.2.5, 1.2.6, 1.2.7, 1.3, 1.3.1, 1.3.2, 1.3.3, 1.3.4, 1.3.5, 1.3.6, 1.3.7, 1.4, 1.4.1, 1.4.2, 1.4.3, 1.4.4, 1.4.5, 1.4.6, 1.4.7, 1.4.8, 1.4.9, 1.4.10, 1.4.11, 1.4.12, 1.4.13, 1.4.14, 1.4.15, 1.4.16, 1.4.17, 1.4.18, 1.4.19, 1.4.20, 1.5, 1.5.1, 1.5.2, 1.5.3, 1.5.4, 1.5.5, 1.5.6, 1.5.7, 1.5.8, 1.5.9, 1.5.10, 1.5.11, 1.5.12, 1.6, 1.6.1, 1.6.2, 1.6.3, 1.6.4, 1.6.5, 1.6.6, 1.6.7, 1.6.8, 1.6.9, 1.6.10, 1.6.11, 1.7, 1.7.1, 1.7.2, 1.7.3, 1.7.4, 1.7.5, 1.7.6, 1.7.7, 1.7.8, 1.8a1, 1.8b1, 1.8b2, 1.8rc1, 1.8, 1.8.1)
No matching distribution found for Django<1.4.99,<1.8.99,>=1.4.2,>=1.7 (from ...)
It is obvious that a version number 1.4.20 cannot satisfy >=1.7 and 1.8.1 cannot satisfy <1.4.99. However the documentation from Setuptools (see above) does suggest that something along these lines should be possible. However, this is non-obvious to me.
To use a specific version of setuptools it is necessary to have it in both locations - in pyproject. toml and at the beginning of install_requires of setup.py. The tool like pip will use the version from pyproject. toml to build the project.
In Fedora, our pip package Recommends setuptools. Practically that means: Majority of users who install pip will get setuptools by default. Users can explicitly uninstall setuptools after installing pip or exclude setuptools when installing pip.
The items listed in setup_requires get implicitly installed whenever you execute the setup.py but one of the common ways that the setup.py is executed is via another tool, such as pip , who is already managing dependencies.
You can use
Django>=1.4.2,<1.9,!=1.5.*,!=1.6.*
This is defined inside PEP440.
You can test this behavior with the packaging module that is vendored inside the last versions of setuptools and pip.
In [1]: from packaging import specifiers
In [2]: sp=specifiers.SpecifierSet(">=1.4.2,<1.9,!=1.5.*,!=1.6.*")
In [3]: sp.contains("1.4.2")
Out[3]: True
In [4]: sp.contains("1.6.4")
Out[4]: False
In [5]: sp.contains("1.8.2")
Out[5]: True
I've read some related code of pkg_resources. I think the document here is not accurate. Not only pip fails to find the right package version, python setup.py install, which actually uses setuptools, also fails.
Some of the related code:
pip/_vendor/packaging/specifiers.py
# 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=prereleases)
return iterable
You can see that in the comment, the author emphasized that this will cause an AND amongst each specifier, not OR. So if you do this:
PickyThing<1.6,>1.9,!=1.9.6,<2.0a0,==2.4c1
You will get nothing!
I tried with this code below:
import pkg_resources
a = ['1.4', '1.8', '1.9.2']
d = pkg_resources.Requirement.parse('PickyThing<1.6,>1.9,!=1.9.6')
r = d.specifier.filter(a)
print(list(r)) # Nothing, just an empty list []
You may want to file a bug to pip so they can fix it.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With