Neovim
Pynvim: Python client toPynvim implements support for python plugins in Nvim. It also works as a library for connecting to and scripting Nvim processes through its msgpack-rpc API.
Install
Supports python 3.10 or later.
pip3 install pynvim
You can install the package without being root by adding the --user
flag.
Anytime you upgrade Neovim, make sure to upgrade pynvim as well:
pip3 install --upgrade pynvim
Alternatively, you can install the development version by cloning this repository and executing the following at the top level:
pip3 install .
Python Plugin API
Pynvim supports python remote plugins (via the language-agnostic Nvim rplugin
interface), as well as Vim plugins (via the :python3
interface). Thus when
pynvim is installed Neovim will report support for the +python3
Vim feature.
The rplugin interface allows plugins to handle vimL function calls as well as defining commands and autocommands, and such plugins can operate asynchronously without blocking nvim. For details on the new rplugin interface, see the Remote Plugin documentation.
Pynvim defines some extensions over the vim python API:
- Builtin and plugin vimL functions are available as
nvim.funcs
- API functions are available as
vim.api
and for objects such asbuffer.api
- Lua functions can be defined using
vim.exec_lua
and called withvim.lua
- Support for thread-safety and async requests.
See the Python Plugin API documentation for usage of this new functionality.
Development
Use (and activate) a local virtualenv.
python3 -m virtualenv venv
source venv/bin/activate
If you change the code, you must reinstall for the changes to take effect:
pip install .
Use pytest
to run the tests. Invoking with python -m
prepends the current
directory to sys.path
(otherwise pytest
might find other versions!):
python -m pytest
For details about testing and troubleshooting, see the development documentation.
Usage from the Python REPL
A number of different transports are supported, but the simplest way to get
started is with the python REPL. First, start Nvim with a known address (or use
the $NVIM_LISTEN_ADDRESS
of a running instance):
$ NVIM_LISTEN_ADDRESS=/tmp/nvim nvim
In another terminal, connect a python REPL to Nvim (note that the API is similar to the one exposed by the python-vim bridge):
>>> from pynvim import attach
# Create a python API session attached to unix domain socket created above:
>>> nvim = attach('socket', path='/tmp/nvim')
# Now do some work.
>>> buffer = nvim.current.buffer # Get the current buffer
>>> buffer[0] = 'replace first line'
>>> buffer[:] = ['replace whole buffer']
>>> nvim.command('vsplit')
>>> nvim.windows[1].width = 10
>>> nvim.vars['global_var'] = [1, 2, 3]
>>> nvim.eval('g:global_var')
[1, 2, 3]
You can embed Neovim into your python application instead of connecting to a running Neovim instance.
>>> from pynvim import attach
>>> nvim = attach('child', argv=["/bin/env", "nvim", "--embed", "--headless"])
- The
--headless
argument tellsnvim
not to wait for a UI to connect. - Alternatively, use
--embed
without--headless
if your client is a UI and you wantnvim
to wait for your client tonvim_ui_attach
before continuing startup.
See the tests for more examples.
Release
- Create a release commit with title
Pynvim x.y.z
- list significant changes in the commit message
- bump the version in
pynvim/util.py
andsetup.py
(3 places in total)
- Make a release on GitHub with the same commit/version tag and copy the message.
- Run
scripts/disable_log_statements.sh
- Run
python -m build
- diff the release tarball
dist/pynvim-x.y.z.tar.gz
against the previous one.
- diff the release tarball
- Run
twine upload -r pypi dist/*
- Assumes you have a pypi account with permissions.
- Run
scripts/enable_log_statements.sh
orgit reset --hard
to restore the working dir.