pip3 install error when I use virtulenv as general user - python-3.x

I found the following problem when I tried to install the package using the pip3 command in a virtual environment.
source /bin/activate ##activate virtualenv
pip3 install aiohttp ##
error: Could not install packages due to an EnvironmentError: [Errno
13] Permission denied:
'/home/yxs/venv/py34/lib64/python3.4/site-packages/multidict' Consider
using the --user option or check the permissions.
pip3 install --user aiohttp
error: Can not perform a '--user' install. User site-packages are not
visible in this virtualenv.
So, I tried sudo pip3 install aiohttp,When I checked with pip3 list, I found that aiohttp was installed in the system. Beacause
(py34) [yxs#yxs ~]$pip3 list ## no package named aiohttp
[yxs#yxs ~]$pip3 list ## found aiohttp
How can I install packages into virtualenv?
Edit:
The operating system is CentOS7.5, the default Python version is 2.7, I installed python3.4 through epel-realse, pip3 installed by get-pip.py.
I guess the problem is that I have to use command sudo to install, but this command will leave the virtualenv environment. Just like the following, but I don't know how to solve this problem. By the way, these operations are excuted in the tmux session.
(py34) [yxs#yxs ~]$su - root
Password:
Last login: Wed Sep 19 12:07:23 CST 2018 on pts/2
[root#yxs ~]#
All in all, I can only use root to install the package into virtualenv by command pip install

Source /bin/activate means you're using /bin root directory of unix/linux.
As you're using python 3. Why don't you use python's builtin venv module.
Add .env directory into your .gitignore file.
Usage
python3 -m venv .env
source .env/bin/activate
pip install django
pip freeze > requirements.txt

Related

Error when trying to install oct2py package

I am trying to install to oct2py package to import in my python script. However whenever I run the command pip install oct2py I get the following error:
Fatal error in launcher: Unable to create process using '"c:\python37\python.exe" "C:\Program Files\Python37\Scripts\pip.exe" ': file cannot be found.
How can I install oct2py?
Found the answer
Deinstalled python and reinstalled it. I checked the environment variable and changed it from this one : 'C:\Program Files\Python39' to this one 'C:\Users\lsee\AppData\Local\Programs\Python\Python39'.
Is unable to locate your pip installer, easy fix is unusually to either upgrade or test pip :
# to upgrade your pip
python -m pip install --upgrade pip
pip --version
# test you have pip in your machine
If your pip version is showing that means you have it installed, it could be a problem activating your virtual environnement, or your path.
It says the file cannot be locate, are you sure your path for Python and pip is correct ?
Try:
python -m pip install oct2py
This should work (as python -m pip install gave not errors)

How to uninstall all versions of pip on Mac Catalina OS

I got a new MacBook and transferred some apps over from my old MacBook with the Migration Assistant. I had pip installed on my old Mac. I didn't think it would be installed on the new Mac so I attempted to install in using the following command:
curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py
followed by:
python3 get-pip.py
When I tried to check if it was installed with pip3 --version I received the following warning:
WARNING: pip is being invoked by an old script wrapper. This will fail in a future version of pip.
Please see https://github.com/pypa/pip/issues/5599 for advice on fixing the underlying issue.
To avoid this problem you can invoke Python with '-m pip' instead of running pip directly.
I've tried uninstalling pip with python3 -m pip uninstall pip setuptools, but this only produces [Errno 13] Permission denied errors. I've tried using chmod 755 to change the permissions of the python files that show up when I run ls -la but it doesn't seem to change anything and I still get the [Errno 13] errors.
Is there a way to change all permission on my Mac? Or else is there a way to uninstall all versions of pip?
I also have Conda installed, but I dont think that is an issue here.
Thanks.

Can’t run pip3 due to import error after upgrade [duplicate]

Whenever I am trying to install any package using pip, I am getting this import error:
guru#guru-notebook:~$ pip3 install numpy
Traceback (most recent call last):
File "/usr/bin/pip3", line 9, in <module>
from pip import main
ImportError: cannot import name 'main'
guru#guru-notebook:~$ cat `which pip3`
#!/usr/bin/python3
# GENERATED BY DEBIAN
import sys
# Run the main entry point, similarly to how setuptools does it, but because
# we didn't install the actual entry point from setup.py, don't use the
# pkg_resources API.
from pip import main
if __name__ == '__main__':
sys.exit(main())
It was working fine earlier, I am not sure why it is throwing this error.
I have searched about this error, but can't find anything to fix it.
Please let me know if you need any further detail, I will update my question.
You must have inadvertently upgraded your system pip (probably through something like sudo pip install pip --upgrade)
pip 10.x adjusts where its internals are situated. The pip3 command you're seeing is one provided by your package maintainer (presumably debian based here?) and is not a file managed by pip.
You can read more about this on pip's issue tracker
You'll probably want to not upgrade your system pip and instead use a virtualenv.
To recover the pip3 binary you'll need to sudo python3 -m pip uninstall pip && sudo apt install python3-pip --reinstall.
If you want to continue in "unsupported territory" (upgrading a system package outside of the system package manager), you can probably get away with python3 -m pip ... instead of pip3.
We can clear the error by modifying the pip file.
Check the location of the file:
$ which pip
path -> /usr/bin/pip
Go to that location(/usr/bin/pip) and open terminal
Enter: $ sudo nano pip
You can see:
import sys
from pip import main
if __name__ == '__main__':
sys.exit(main())
Change to:
import sys
from pip import __main__
if __name__ == '__main__':
sys.exit(__main__._main())
then ctrl + o write the changes and exit
Hope this will do!!
For Ubuntu family, Debian, Linux Mint users
Thanks to Anthony's explanation above, you can retain your original system pip (in /usr/bin/ and dist-packages/) and remove the manually-installed pip (in ~/.local/) to resolve the conflict:
$ python3 -m pip uninstall pip
Ubuntu/Debian pip v8.1.1 (16.04) from python3-pip debian package (see$ pip3 -V) shows the same search results as the latest pip v10.0.1, and installs latest modules from PyPI just fine. It has a working pip command (already in the $PATH), plus the nice --user option patched-in by default since 2016. Looking at pip release notes, the newer versions are mostly about use-case specific bug fixes and certain new features, so not everyone has to rush upgrading pip just yet. And the new pip 10 can be deployed to Python virtualenvs, anyway.
But regardless of pips, your OS allows to quickly install common Python modules (including numpy) with APT, without the need for pip, for example:
$ sudo apt install python3-numpy python3-scipy (with system dependencies)
$ sudo apt install python3-pip (Debian-patched pip, slightly older but it doesn't matter)
Quick apt syntax reminder (please see man apt for details):
$ sudo apt update (to resync Ubuntu package index files from up-to-date sources)
$ apt search <python-package-name> (full text-search on all available packages)
$ apt show <python-package-name> (displays the detailed package description)
$ sudo apt install <python-package-name>
Package names prefixed with python- are for Python 2; and prefixed with python3- are for Python 3 (e.g. python3-pandas). There are thousands, and they undergo integration testing within Debian and Ubuntu. Unless you seek to install at per-user level (pip install --user option) or within virtualenv/venv, apt could be what you needed. These system packages are accessible from virtual envs too, as virtualenv will gracefully fall back to using system libs on import if your envs don't have given copies of modules.
Your custom-installed (with pip --user) per-user modules in ~/.local/lib will override them too.
Note, since this is a system-wide installation, you'd rarely need to remove them (need to be mindful about OS dependencies). This is convenient for packages with many system dependencies (such as with scipy or matplotlib), as APT will keep track and provide all required system libs and C extensions, while with pip you have no such guarantees.
In fact, for system-wide Python packages (in contrast to per-user, home dir level, or lower), Ubuntu expects using the APT package manager (rather than sudo pip) to avoid breaking OS: sudo pip3 targets the very same /usr/lib/python3/dist-packages directory where APT stores OS-sensitive modules. Recent Debian/Ubuntu releases depend heavily on Python 3, so its pre-installed modules are managed by apt and shouldn't be changed.
So if you use pip3 install command, please ensure that it runs in an isolated virtual dev environment, such as with virtualenv (sudo apt install python3-virtualenv), or with Python3 built-in (-m venv), or at a per-user level (--user pip option, default in Ubuntu-provided pip since 2016), but not system-wide (never sudo pip3!), because pip interferes with the operation of the APT package manager and may affect Ubuntu OS components when a system-used python module is unexpectedly changed. Good luck!
P. S. All the above is for the 'ideal' solution (Debian/Ubuntu way).
If you still want to use the new pip3 v10 exclusively, there are 3 quick workarounds:
simply open a new bash session (a new terminal tab, or type bash) - and pip3 v10 becomes available (see pip3 -V). debian's pip3 v8 remains installed but is broken; or
the command $ hash -d pip3 && pip3 -V to refresh pip3 pathname in the $PATH. debian's pip3 v8 remains installed but is broken; or
the command $ sudo apt remove python3-pip && hash -d pip3 to uninstall debian's pip3 v8 completely, in favor of your new pip3 v10.
Note: You will always need to add --user flag to any non-debian-provided pip, unless you are in a virtualenv! (it deploys python packages to ~/.local/, default in debian/ubuntu-provided python3-pip and python-pip since 2016). Your use of pip 10 system-wide, outside of virtualenv, is not really supported by Ubuntu/Debian. Never sudo pip3!
Further details:
https://github.com/pypa/pip/issues/5221#issuecomment-382069604
https://github.com/pypa/pip/issues/5240#issuecomment-381673100
resolved in one step only.
I too faced this issue, But this can be resolved simply by 1 command without bothering around and wasting time and i have tried it on multiple systems it's the cleanest solution for this issue. And that's:
For python3:- sudo python3 -m pip uninstall pip && sudo apt install python3-pip --reinstall.
By this , you can simply install packages using pip3. to check use pip3 --version.
For older versions, use : sudo python -m pip uninstall pip && sudo apt install python-pip --reinstall.
By this, now you can simply install packages using pip. to check use pip --version.
Use python -m pip install instead of pip install
Example:
python -m pip install --user somepackage
python3 -m pip install --user somepackage
The pip (resp. pip3) executable is provided by your distro (python-pip package on Ubuntu 16.04) and located at /usr/bin/pip.
Therefore, it is not kept up-to date with the pip package itself as you upgrade pip, and may break.
If you just use python -m pip directly, e.g. as in:
python -m pip install --user somepackage
python3 -m pip install --user somepackage
it goes through your Python path, finds the latest version of pip and executes that file.
It relies on the fact that file is executable through import, but that is a very standard type of interface, and therefore less likely to break than the hackier Debian script.
Then I recommend adding the following aliases to your .bashrc:
pip() ( python -m pip "$#" )
pip3() ( python3 -m pip "$#" )
The Ubuntu 18.04 /usr/bin/pip3 file does:
from pip import main
and presumably main was removed from pip at some point which is what broke things.
The breaking pip commit appears to be: 95bcf8c5f6394298035a7332c441868f3b0169f4 "Move all internal APIs to pip._internal" which went into pip 18.0.
Tested in Ubuntu 16.04 after an update from pip3 9.0.1 to 18.0.
pyenv
Ultimately however, for serious Python development I would just recommend that you install your own local Python with pyenv + virtualenv, which would also get around this Ubuntu bug: https://askubuntu.com/questions/682869/how-do-i-install-a-different-python-version-using-apt-get/1195153#1195153
You can resolve this issue by reinstalling pip.
Use one of the following command line commands to reinstall pip:
Python2:
python -m pip uninstall pip && sudo apt install python-pip --reinstall
Python3:
python3 -m pip uninstall pip && sudo apt install python3-pip --reinstall
Check if pip has been cached on another path, to do so, call $ which pip and check that the path is different from the one prompted in the error, if that's the case run:
$ hash -r
When the cache is clear, pip will be working again.
reference: http://cheng.logdown.com/posts/2015/06/14/-usr-bin-pip-no-such-file-or-directory
I'm running on a system where I have sudo apt but no sudo pip. (And no su access.) I got myself into this same situation by following the advice from pip:
You are using pip version 8.1.1, however 18.0 is available. You should consider upgrading via the 'pip install --upgrade pip' command.
None of the other fixes worked for me, because I don't have enough admin privileges. However, a few things stuck with me from reading up on this:
I shouldn't have done this. Sure, pip told me to. It lied.
Using --user solves a lot of issues by focusing on the user-only directory.
So, I found this command line to work to revert me back to where I was. If you were using a different version than 8.1.1, you will obviously want to change that part of the line.
python -m pip install --force-reinstall pip==8.1.1 --user
That's the only thing that worked for me, but it worked perfectly!
I met the same problem on my Ubuntu 16.04 system. I managed to fix it by re-installing pip with the following command:
curl https://bootstrap.pypa.io/get-pip.py | sudo python3
Recover with python3 -m pip install --user pip==9.0.1 (or the version that worked)
Same thing happened to me on Pixelbook using the new LXC (strech). This solution is very similar to the accepted one, with one subtle difference, whiched fixed pip3 for me.
sudo python3 -m pip install --upgrade pip
That bumped the version, and now it works as expected.
I found it here ... Python.org: Ensure pip is up-to-date
The commands above didn't work for me but those were very helpful:
sudo apt purge python3-pip
sudo rm -rf '/usr/lib/python3/dist-packages/pip'
sudo apt install python3-pip
cd
cd .local/lib/python3/site-packages
sudo rm -rf pip*
cd
cd .local/lib/python3.5/site-packages
sudo rm -rf pip*
sudo pip3 install jupyter
In ubuntu 18.04.1 Bionic Beaver, you need to log out and log back in (restart not necessary) to get the proper environment.
$ sudo apt install python-pip
$ pip --version
pip 9.0.1 from /usr/lib/python2.7/dist-packages (python 2.7)
$ pip install --upgrade pip
$ pip --version
Traceback (most recent call last):
File "/usr/bin/pip", line 9, in <module>
from pip import main
ImportError: cannot import name main
$ exit
<login>
$ pip --version
pip 18.1 from /home/test/.local/lib/python2.7/site-packages/pip (python 2.7)
I use sudo apt remove python3-pip then pip works.
~ sudo pip install pip --upgrade
[sudo] password for sen:
Traceback (most recent call last):
File "/usr/bin/pip", line 9, in <module>
from pip import main
ImportError: cannot import name 'main'
➜ ~ sudo apt remove python3-pip
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following packages were automatically installed and are no longer required:
libexpat1-dev libpython3-dev libpython3.5-dev python-pip-whl python3-dev python3-wheel
python3.5-dev
Use 'sudo apt autoremove' to remove them.
The following packages will be REMOVED:
python3-pip
0 upgraded, 0 newly installed, 1 to remove and 0 not upgraded.
After this operation, 569 kB disk space will be freed.
Do you want to continue? [Y/n] y
(Reading database ... 215769 files and directories currently installed.)
Removing python3-pip (8.1.1-2ubuntu0.4) ...
Processing triggers for man-db (2.7.5-1) ...
➜ ~ pip
Usage:
pip <command> [options]
For Python version 2.7 #Anthony solution works perfect, by changing python3 to python as follows:
sudo python -m pip uninstall pip && sudo apt install python-pip --reinstall
What worked for me to fix the error with using pip3 was:
sudo cp -v /usr/local/bin/pip3 /usr/bin/pip3
Everything works:
demon#UbuntuHP:~$ pip -V
pip 10.0.1 from /usr/local/lib/python3.5/dist-packages/pip (python 3.5)
demon#UbuntuHP:~$ pip2 -V
pip 10.0.1 from /home/demon/.local/lib/python2.7/site-packages/pip (python 2.7)
demon#UbuntuHP:~$ pip3 -V
pip 10.0.1 from /usr/local/lib/python3.5/dist-packages/pip (python 3.5)
Maybe the new 10.0.1 version of pip doesn't update the binary in /usr/bin ? (which seems it does not)
EDIT: the same issue occurs in Ubuntu 18.04. The best solution I've found is to symlink the pip binaries from /home/<user/.local/bin to /usr/local/bin or /usr/bin (depending on your preference), as follows:
ln -sv /home/<user>/.local/bin/pip /usr/local/bin/pip
ln -sv /home/<user>/.local/bin/pip2 /usr/local/bin/pip2
ln -sv /home/<user>/.local/bin/pip2.7 /usr/local/bin/pip2.7
ln -sv /home/<user>/.local/bin/pip3 /usr/local/bin/pip3
ln -sv /home/<user>/.local/bin/pip3.6 /usr/local/bin/pip3.6
NOTE: replace <user> with your current running user
The associated versions (latest) are in:
Version 3.6:
/home/demon/.local/lib/python3.6/site-packages/pip (python 3.6)
Version 2.7:
/home/demon/.local/lib/python2.7/site-packages/pip (python 2.7)
Trick and works too
sudo -H pip install lxml
I had this same error, but python -m pip was still working, so I fixed it with the nuclear option sudo python -m pip install --upgrade pip. It did it for me.
For what it's worth, I had the problem with pip (not pip2 or pip3):
$ pip -V
Traceback (most recent call last):
File "/usr/bin/pip", line 9, in <module>
from pip import main
ImportError: cannot import name main
$ pip2 -V
pip 8.1.1 from /usr/lib/python2.7/dist-packages (python 2.7)
$ pip3 -V
pip 8.1.1 from /usr/lib/python3/dist-packages (python 3.5)
Somehow (I can't remember how) I had python stuff installed in my ~/.local directory. After I removed the pip directory from there, pip started working again.
$ rm -rf /home/precor/.local/lib/python2.7/site-packages/pip
$ pip -V
pip 8.1.1 from /usr/lib/python2.7/dist-packages (python 2.7)
Is something wrong with the packages, when it generating de file /usr/bin/pip,
you have to change the import:
from pip import main
to
from pip._internal import main
That solves the problem, I'm not sure why it generated, but it saids somthing in the following issue:
After pip 10 upgrade on pyenv "ImportError: cannot import name 'main'"
You can try this:
sudo ln -sf $( type -P pip ) /usr/bin/pip
I also run into this problem when I wanted to upgrade system pip pip3 from 9.0.1 to 19.2.3.
After running pip3 install --upgrade pip, pip version becomes 19.2.3. But main() has been moved in pip._internal in the latest version, which leaves pip3 broken.
So in file /usr/bin/pip3, replace line 9: from pip import main with from pip._internal import main. The issue will be fixed, works the same for python2-pip. (Tested on Ubuntu 18.04 distribution)
According to #Vincent H.'s answer
Please run the following commands to do the fix. After running python3 -m pip install --upgrade pip, please run the following command.
hash -r pip
Source: https://github.com/pypa/pip/issues/5221
you can simply fix the pip and pip3 paths using update-alternatives
first thing you should check is your current $PATH
run echo $PATH and see is you can find /usr/local/bin which is where pip3 and pip usually are
there is a change your system is looking here /bin/pip and /bin/pip3
so i will say fix the PATH by adding to your ~/.bash_profile file so it persists
export PATH=$PATH:/usr/local/bin
and then check is its fixed with which pip and which pip3
if not then use update-alternatives to fix it finally
update-alternatives --install /bin/pip3 pip3 /usr/local/bin/pip3 30
and if you want to point pip to pip3 then
update-alternatives --install /bin/pip pip /usr/local/bin/pip3 30
I have the same problem and solved it. Here is my solution.
First, when I run pip install something, the error came out like this:
Traceback (most recent call last):
File "/usr/bin/pip3", line 9, in <module>
from pip import main
ImportError: cannot import name 'main'
So, I cd into the file /usr/bin/ and cat pip3 to see the code in it. I see this in it:
#!/usr/bin/python3
# GENERATED BY DEBIAN
import sys
# Run the main entry point, similarly to how setuptools does it, but because
# we didn't install the actual entry point from setup.py, don't use the
# pkg_resources API.
from pip import main
if __name__ == '__main__':
sys.exit(main())
And then I think that it was not in the installation path. So I cd into the python3-pip, like this:
cd /.local/lib/python3.5/site-packages/pip
P.S.: you have to cd into the right directions in your computer
Then I cat the file to see the differences(you can use other operations to see the code):
cat __main__.py
And I saw this:
from __future__ import absolute_import
import os
import sys
# If we are running from a wheel, add the wheel to sys.path
# This allows the usage python pip-*.whl/pip install pip-*.whl
if __package__ == '':
# __file__ is pip-*.whl/pip/__main__.py
# first dirname call strips of '/__main__.py', second strips off '/pip'
# Resulting path is the name of the wheel itself
# Add that to sys.path so we can import pip
path = os.path.dirname(os.path.dirname(__file__))
sys.path.insert(0, path)
from pip._internal import main as _main # isort:skip # noqa
if __name__ == '__main__':
sys.exit(_main())
So, can you see the difference? I can figure out that I have to make the file the same as the file in /usr/bin/pip3
So, I copied the code in /.local/lib/python3.5/site-packages/pip to replace the code in /usr/bin/pip3
and the problem disappear!
P.S.: pip3 or pip have no difference in this problem.
I will be happy if my solution solve your problem!
This Worked for me !
hash -r pip # or hash -d pip
Now, uninstall the pip installed version and reinstall it using the following commands.
python -m pip uninstall pip # sudo
sudo apt install --reinstall python-pip
If pip is broken, use:
python -m pip install --force-reinstall pip
Hope it helps!
I used the following code to load a module that might need install, thus avoiding this error (which I also got) - using the latest Python and latest pip with no problem
try
from colorama import Fore, Back, Style
except:
!pip install colorama
from colorama import Fore, Back, Style
import main from pip._internal
from pip._internal import main
Edit the pip code from
sudo nano /usr/bin/pip3
As #cryptoboy said - check what pip/python version you have installed
demon#UbuntuHP:~$ pip -V
demon#UbuntuHP:~$ pip2 -V
demon#UbuntuHP:~$ pip3 -V
and then check for no-needed libraries in your .local/lib/ folder.
I did backup of settings when I was migrating to newer Kubuntu and in had .local/lib/python2.7/ folder in my home directory. Installed python 3.6. I just removed the old folder and now everything works great!
On Debian you will need to update apt first....
sudo apt-get update -qq
sudo apt-get install python-pip -qq
sudo pip install pip --upgrade --quiet
sudo pip2 install virtualenv --quiet
If you skip 'sudo apt-get update -qq' your pip will become corrupt and display the 'cannot find main' error.

Python3 virtualenv installs python2

I am not sure what is wrong but I can't seem to get python3 in a virtualenv environment. I tried upgrading my ubuntu and updating all the packages - but no luck:
python3 -m virtualenv ENV
Running virtualenv with interpreter /usr/bin/python2
New python executable in /home/ramin/projects/buybulkamerica/ENV/bin/python2
Also creating executable in /home/ramin/projects/buybulkamerica/ENV/bin/python
Installing setuptools, pkg_resources, pip, wheel...done.
What can I do to ensure that virtualenv installs python3 instead of python2?
First, uninstall existing virtualenv.
sudo apt-get remove --purge python-virtualenv if you installed it using a package manager.
pip uninstall virtualenv if you have installed it using pip.
pip3 uninstall virtualenv if you have installed it using pip3.
Any one of the above commands will work.
Now install virtualenv again. Since you want python3, you need to run the following command.
pip3 install virtualenv
That should do the trick. Now when you create a new virtualenv, it will use python3.
There may be a better way but I had the same problem and after not finding any solution I tried this and it worked.
After you have installed virtualenv using pip, it doesn't matter if you used pip or pip3 if you give the location of your python3 installation to the virtualenv command, like this.
Create new virtualenv
virtualenv --python=/usr/bin/python3.6 environmentname
Access virtualenv
source /environmentname/bin/activate
If this doesn't work, use complete path from pwd
source /complete/path/to/environmentname/bin/activate
Stop virtualenv
deactivate

pip uninstall: "No files were found to uninstall."

I have created a python module, call it 'foo_bar'.
I can install it and I can upgrade it, but I cannot uninstall it.
I build my module using bdist_wheel:
$ python3 setup.py bdist_wheel
And I install and upgrade it as follows:
$ python3 -m pip --timeout 60 install --upgrade dist/foo_bar-1.4.3-py3-none-any.whl
It is listed within Python 3.4 framework directory:
ls -al /Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/
drwxr-xr-x 12 samwise admin 408 Jun 21 02:50 foo_bar
drwxr-xr-x 9 samwise admin 306 Jun 21 02:50 foo_bar-1.4.3.dist-info
And it listed within pip freeze:
$ python3 -m pip freeze
foo-bar==1.4.3
However, if I try to perform pip uninstall, it cannot find it's files
$ python3 -m pip uninstall foo-bar
Can't uninstall 'foo-bar'. No files were found to uninstall.
Did I do something wrong within my setup.py for it not to be able to find my modules files during uninstall?
Version info is as follows:
$ python3 --version
Python 3.4.4
$ python3 -m pip --version
pip 8.1.2 from /Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages (python 3.4)
I had the same issue. Using verbose helped me to find out a bit more the reason:
$ pip3 uninstall --verbose my-homemade-package
Not sure how to uninstall: my-homemade-package e48e635 - Check: /home/olivier/my-homemade-package
Can't uninstall 'my-homemade-package'. No files were found to uninstall.
Removing everything that was 'my-homemade-package' related in /usr/local/python2.x and /usr/local/python3.x did not help.
I did a pip3 show my-homemade-package and got the location of the installed package on my computer:
$ pip3 show my-homemade-package
Name: my-homemade-package
Version: e48e635
Summary: My Home Made package
Home-page: UNKNOWN
Author: UNKNOWN
Author-email: UNKNOWN
License: Proprietary
Location: /home/olivier/my-homemade-package
Requires: pyOpenSSL, pyasn1, protobuf
Removing /home/olivier/my-homemade-package sorted it out the issue (ie: the package was not listed).
This is an old post, but it was top result in Google. The above answers are correct, however, in my case there was still line /usr/local/lib/python3.6/site-packages/easy-install.pth that I had to remove after also removing the egg files.
So I was having a similar issue to OP. I could install my package with pip install dist/mypackage.tar.gz. Installation would work fine, but at the end it would show Can't uninstall 'mypackage'. No files were found to uninstall., and indeed pip uninstall mypackage wouldn't work later on.
It sounds silly but what worked for me was to change working directory: once I left mypackage/ directory, pip uninstall mypackage worked.
I had such issue when I renamed my module in setup.py.
Old old_name.egg-info directory still existed in my_module directory. So, when I installed module with pip install -e . pip created a line in python3.8/site-packages/easy-install.pth pointing to module directory. After that module was listed by pip list with both names: new-name and old-name. And when I tried to remove old module with pip remove old-name pip showed error:
Found existing installation: old-name 0.3.0
Can't uninstall 'old-name'. No files were found to uninstall.
The solution was to remove directory old_name.egg-info from module directory. After that pip list shows only new-name.
Probably it is not direct answer to original post but one of the solutions for issue in topic-name.
Issue: User cannot uninstall a python package installed via pip:
pip uninstall youtube-dl
Found existing installation: youtube-dl 2021.12.17
Not uninstalling youtube-dl at /usr/lib/python3/dist-packages, outside environment /usr
Can't uninstall 'youtube-dl'. No files were found to uninstall.
Reason: PEBKAC.
Well, a simple
apt purge youtube-dl
did the trick. There had been a system package "youtube-dl" installed, with said version:
dpkg -l youtube-dl
ii youtube-dl 2021.12.17-1~nd110+1
At the same time users used to install packages locally using pip. Both packages of the same version (2021.12.17). And both ways (apt and pip) referred to the packages by the same name. Turned out, this tends to confuse users..
Next level: Have a package installed three ways: apt, pip --system and plain pip as user. Maybe pip as root (locally) FWIW, too.

Resources