Пример #1
0
def _parse_filename(filename):
    """Perform the required checks and get the final filename"""
    # Get the filename of the notebook (if not provided)
    if filename is None:
        if in_script():
            filename = get_script_filename()
        elif in_notebook():
            filename = get_notebook_name()

    # Add the right extension to the filename
    elif get_file_extension(filename) not in ['py', 'ipynb']:
        filename += '.py' if in_script() else '.ipynb'
    return filename
Пример #2
0
def test_get_notebook(mock_get_notebook_path):
    assert get_notebook_name() == 'Fake-Notebook.ipynb'

    assert get_notebook_name() == None
Пример #3
0
def commit(secret=False,
           nb_filename=None,
           files=[],
           capture_env=True,
           env_type='conda',
           notebook_id=None,
           create_new=None):
    """Save the notebook, capture environment and upload to the cloud for sharing.

    In most cases, commit works well with the default arguments. It attempts to 
    1. Save the Jupyter notebook
    2. Upload the notebook (and additional scripts, CSVs etc.) to https://jvn.io 
    3. Capture the python environment (using Anaconda or pip)
    4. Upload the python environment to cloud

    Capturing the python environment ensures that the notebook can be reproduced and 
    executed easily using the `jovian` command line tool. For more details, see 
    https://jvn.io/getting-started . 

    Issues and bugs can be reported here: https://github.com/swiftace-ai/jovian-py

    Arguments:

        secret (bool, optional): Create a secret notebook on https://jvn.io , which is only 
            accessible via the link, and is not visible on the owner's public profile. By default,
            commited notebooks are public and visible on the owner's profile.

        nb_filename (string, optional): The filename of the jupyter notebook (including 
            the .ipynb extension). This is detected automatically in most cases, but in
            certain environments like Jupyter Lab, the detection may fail and the filename
            needs to be provided using this argument.

        files (array, optional): Any additional scripts (.py files), CSVs that are required to
            run the notebook. These will be available in the files tab on https://jvn.io .

        capture_env (bool, optional): If `True`, the Python environment (python version,
            libraries etc.) are captured and uploaded along with the notebook.

        env_type (string, optional): The type of environment to be captured. Allowed options are
            'conda' and 'pip'.

        notebook_id (string, optional): If you wish to update an existing notebook owned by you,
            you can use this argument to provide the base64 ID (present in the URL) of an notebook 
            hosted on https://jvn.io . In most cases, this argument is not required, and the library
            can automatically infer whether you are looking to update an existing notebook or create
            a new one.

        create_new (bool, optional): If set to True, doesn't update the existing notebook on 
            https://jvn.io (if one is detected). Instead, it creates a new notebook when commit is called.

    """
    global _current_slug

    # Check if we're in a Jupyter environment
    if not in_notebook():
        log('Failed to detect Juptyer notebook. Skipping..', error=True)
        return

    # Save the notebook (uses Javascript, doesn't work everywhere)
    log('Saving notebook..')
    save_notebook()
    sleep(1)

    # Get the filename of the notebook (if not provided)
    if nb_filename is None:
        nb_filename = get_notebook_name()

    # Exit with help message if filename wasn't detected (or provided)
    if nb_filename is None:
        log(FILENAME_MSG)
        return

    # Check whether to create a new gist, or update an old one
    if not create_new and notebook_id is None:
        # First preference to the in-memory slug variable
        if _current_slug is not None:
            notebook_id = _current_slug
        else:
            notebook_id = get_notebook_slug(nb_filename)

    # Check if the current user can push to this slug
    if notebook_id is not None:
        gist_meta = get_gist(notebook_id)
        if not gist_meta['isOwner']:
            notebook_id = None

    # Log whether this is an update or creation
    if notebook_id is None:
        log('Creating a new notebook on https://jvn.io')
    else:
        log('Updating notebook "' + notebook_id + '" on https://jvn.io')

    # Upload the notebook & create/update the gist
    res = create_gist_simple(nb_filename, notebook_id, secret)
    if res is None:
        return

    # Extract slug and owner from created gist
    slug, owner = res['slug'], res['owner']

    # Set/update the slug information
    _current_slug = slug
    set_notebook_slug(nb_filename, slug)

    # Save & upload environment
    if capture_env:
        log('Capturing environment..')

        if env_type == 'conda':
            # Capture conda environment
            try:
                upload_conda_env(slug)
            except CondaError as e:
                log(str(e), error=True)

        elif env_type == 'pip':
            # Capture pip environment
            try:
                upload_pip_env(slug)
            except Exception as e:
                log(str(e), error=True)

    # Upload additional files
    if files and len(files) > 0:
        log('Uploading additional files..')

        # Upload each file
        for fname in files:
            if os.path.exists(fname) and not os.path.isdir(fname):
                try:
                    upload_file(slug, fname)
                except Exception as e:
                    log(str(e), error=True)
            elif os.path.isdir(fname):
                log('Ignoring directory "' + fname + '"', error=True)
            else:
                log('Ignoring "' + fname + '" (not found)', error=True)

    # Print commit URL
    log('Committed successfully! ' + WEBAPP_URL + "/" + owner['username'] +
        "/" + slug)