shadow-scholar 0.3.0

🕶️ Meddling with papers in the dark 🎓

Shadow Scholar

Installation

To install, simply clone the repository and install with pip:

git clone [e-mail]:allenai/shadow-scholar.git
cd shadow-scholar
pip install .

Available Scripts

Each script is launched with shadow <entrypoint_name>. For a full list of all entry points, run shadow -l.

Getting Access to AWS services

To run the scripts that use AWS services, you will need to have access to the following services:

• Athena
• S3

The best way to do so is to obtain AWS credentials (access key and secret key) and set them as environment variables.

Writing your own script

To write your own script for Shadow Scholar, follow these steps:

Step 1: Choose where to add your code in Shadow Scholar. It can either be in an existing module, such as shadow_scholar.collections.athena, or in a new module.

Step 2: Understand that the entry point for your script should be a single function; think of this as the main function.

Step 3: Write your main function. For each argument you expect a user might want to provide from command line, add a corresponding argument to the function. For example:

def my_script(
    arg1: str,
    arg2: int,
    arg3: bool,
    arg4: Optional[str] = None,
):
    # Do something with the arguments
    pass

Step 4: Add the cli from shadow_scholar.cli to your script. This will allow users to run your script from the command line. For example:

from shadow_scholar.cli import cli
from typing import Optional

@cli(
    name="scripts_collection.my_script",
    arguments=...,
    requirements=...,
)
def my_script(
    arg1: str,
    arg2: int,
    arg3: bool,
    arg4: Optional[str] = None,
):
    # Do something with the arguments
    pass

The cli decorator takes three arguments: the name of the script, a list of arguments, and a list of requirements. The name of the script should be the name a user would use to run the script from the command line. In the example above, the user would run the script with shadow scripts_collection.my_script.

Step 5: Add arguments to your script. Each argument should be an instance of shadow_scholar.cli.Argument. For example:

from typing import Optional
from shadow_scholar.cli import Argument, cli

@cli(
    name="scripts_collection.my_script",
    arguments=[
        Argument(
            name="arg1",
            type=str,
            help="This is the first argument",
        ),
        Argument(
            name="arg2",
            type=int,
            help="This is the second argument",
        ),
        Argument(
            name="arg3",
            type=bool,
            help="This is the third argument",
        ),
        Argument(
            name="arg4",
            type=str,
            help="This is the fourth argument",
            default=None,
        ),
    ],
    requirements=...,
)
def my_script(
    arg1: str,
    arg2: int,
    arg3: bool,
    arg4: Optional[str] = None,
):
    # Do something with the arguments
    pass

You should have as many Arguments as you have arguments to your main function.

Step 6: Add requirements to your script. Each requirement should be an in the format used by requirements.txt. When using optional requirements, make sure to wrap them in a with safe_import() statement at the top of your script. For example:

from typing import Optional
from shadow_scholar.cli import Argument, cli, safe_import

with safe_import() as safe:
    # this will not fail if pandas is not installed
    import pandas as pd


@cli(
    name="scripts_collection.my_script",
    arguments=[
        Argument(
            name="arg1",
            type=str,
            help="This is the first argument",
        ),
        Argument(
            name="arg2",
            type=int,
            help="This is the second argument",
        ),
        Argument(
            name="arg3",
            type=bool,
            help="This is the third argument",
        ),
        Argument(
            name="arg4",
            type=str,
            help="This is the fourth argument",
            default=None,
        ),
    ],
    requirements=[
        "pandas>=1.0.0",
    ],
)
def my_script(
    arg1: str,
    arg2: int,
    arg3: bool,
    arg4: Optional[str] = None,
):
    # Do something with the arguments
    pass

Step 7: Import the function in the __init__.py file of the module. For example, if you added your script to shadow_scholar/examples.py, you would add the following to shadow_scholar/__init__.py:

from shadow_scholar.examples import my_script