18

I have been using "ipython --script" to automatically save a .py file for each ipython notebook so I can use it to import classes into other notebooks. But this recenty stopped working, and I get the following error message:

`--script` is deprecated. You can trigger nbconvert via pre- or post-save hooks:
ContentsManager.pre_save_hook
FileContentsManager.post_save_hook
A post-save hook has been registered that calls:
ipython nbconvert --to script [notebook]
which behaves similarly to `--script`.

As I understand this I need to set up a post-save hook, but I do not understand how to do this. Can someone explain?

Improve this question

3 Answers 3

Reset to default
22

[UPDATED per comment by @mobius dumpling]

Find your config files:

Jupyter / ipython >= 4.0

jupyter --config-dir

ipython <4.0

ipython locate profile default

If you need a new config:

Jupyter / ipython >= 4.0

jupyter notebook --generate-config

ipython <4.0

ipython profile create

Within this directory, there will be a file called [jupyter | ipython]_notebook_config.py, put the following code from ipython's GitHub issues page in that file:

import os
from subprocess import check_call

c = get_config()

def post_save(model, os_path, contents_manager):
    """post-save hook for converting notebooks to .py scripts"""
    if model['type'] != 'notebook':
        return # only do this for notebooks
    d, fname = os.path.split(os_path)
    check_call(['ipython', 'nbconvert', '--to', 'script', fname], cwd=d)

c.FileContentsManager.post_save_hook = post_save

For Jupyter, replace ipython with jupyter in check_call.

Note that there's a corresponding 'pre-save' hook, and also that you can call any subprocess or run any arbitrary code there...if you want to do any thing fancy like checking some condition first, notifying API consumers, or adding a git commit for the saved script.

Cheers,

-t.

Improve this answer
Sign up to request clarification or add additional context in comments.

9 Comments

This works great, thanks! I had to remove my old profile_default and create a new one to get it working.
Cool, glad it worked! For future reference, if you have a default profile that doesn't have the config files in it, you can still run create profile on top of the existing profile.
Update: This solution is broken in iPython version 4, because of "The Big Split" of Jupyter from iPython. To adjust this solution to version 4, use the command jupyter notebook --generate-config to create a config file. The command jupyter --config-dir finds out which directory contains the config files. And the code snippet given by @TristanReid should be added to the file named jupyter_notebook_config.py. The rest works as before.
I had an issue with configparser (AttributeError: ConfigParser instance has no attribute 'read_file'), had to upgrade to the last version (pip install --upgrade configparser)
To check your version use conda list jupyter and check the number next to jupyter-core
|
2

Here is another approach that doesn't invoke a new thread (with check_call). Add the following to jupyter_notebook_config.py as in Tristan's answer:

import io
import os
from notebook.utils import to_api_path

_script_exporter = None

def script_post_save(model, os_path, contents_manager, **kwargs):
    """convert notebooks to Python script after save with nbconvert

    replaces `ipython notebook --script`
    """
    from nbconvert.exporters.script import ScriptExporter

    if model['type'] != 'notebook':
        return

    global _script_exporter
    if _script_exporter is None:
        _script_exporter = ScriptExporter(parent=contents_manager)
    log = contents_manager.log

    base, ext = os.path.splitext(os_path)
    py_fname = base + '.py'
    script, resources = _script_exporter.from_filename(os_path)
    script_fname = base + resources.get('output_extension', '.txt')
    log.info("Saving script /%s", to_api_path(script_fname, contents_manager.root_dir))
    with io.open(script_fname, 'w', encoding='utf-8') as f:
        f.write(script)

c.FileContentsManager.post_save_hook = script_post_save

Disclaimer: I'm pretty sure I got this from SO somwhere, but can't find it now. Putting it here so it's easier to find in future (:

Improve this answer

1 Comment

Is the var py_fname needed? And in any case, this answer is very very very similar to an entry in Jupyter docs: jupyter-notebook.readthedocs.io/en/latest/extending/…
1

I just encountered a problem where I didn't have rights to restart my Jupyter instance, and so the post-save hook I wanted couldn't be applied.

So, I extracted the key parts and could run this with python manual_post_save_hook.py:

from io import open
from re import sub
from os.path import splitext
from nbconvert.exporters.script import ScriptExporter

for nb_path in ['notebook1.ipynb', 'notebook2.ipynb']:
    base, ext = splitext(nb_path)
    script, resources = ScriptExporter().from_filename(nb_path)
    # mine happen to all be in Python so I needn't bother with the full flexibility
    script_fname = base + '.py'
    with open(script_fname, 'w', encoding='utf-8') as f:
        # remove 'In [ ]' commented lines peppered about
        f.write(sub(r'[\n]{2}# In\[[0-9 ]+\]:\s+[\n]{2}', '\n', script))

You can add your own bells and whistles as you would with the standard post save hook, and the config is the correct way to proceed; sharing this for others who might end up in a similar pinch where they can't get the config edits to go into action.

Improve this answer

Comments

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.