rocrate-action-recorder 0.4.0

Record calls of Python CLI commands into a Research Object Crate

rocrate-action-recorder

Python package to record calls of Python CLI commands into a Research Object Crate (RO-Crate).

Supports RO-Crate 1.1 specification. Specifically the Process Run Crate profile.

Install

pip install rocrate-action-recorder

Usage

Example of recording a CLI command (example-cli input.txt output.txt) implemented with argparse.

import argparse
from pathlib import Path
import sys

from rocrate_action_recorder import recorded_argparse

parser = argparse.ArgumentParser(prog="example-cli", description="Example CLI")
parser.add_argument("--version", action="version", version="%(prog)s 1.2.3")
parser.add_argument("--no-record", action="store_false", help="Disable RO-Crate recording.")
parser.add_argument("input", type=Path, help="Input file")
parser.add_argument("output", type=Path, help="Output file")

@recorded_argparse(
    parser=parser,
    input_files=["input"],
    output_files=["output"],
    dataset_license="CC-BY-4.0",
    enabled_argument="no_record",
)
def handler(args: argparse.Namespace):
    # For demonstration, just upper case input to output, replace with real logic
    args.output.write_text(args.input.read_text().upper())

# Prepare input
Path("input.txt").write_text("hello")
# Simulate command-line arguments
sys.argv.extend(["input.txt", "output.txt"])

args = parser.parse_args()
handler(args)

Will generate a `ro-crate-metadata.json` file in the current working directory describing the execution of the CLI command. (Click me to see crate content)

{
    "@context": [
        "https://w3id.org/ro/crate/1.1/context",
        "https://w3id.org/ro/terms/workflow-run/context"
    ],
    "@graph": [
        {
            "@id": "./",
            "@type": "Dataset",
            "conformsTo": {
                "@id": "https://w3id.org/ro/wfrun/process/0.5"
            },
            "datePublished": "2026-01-28T15:07:18.600135",
            "description": "An RO-Crate recording the files and directories that were used as input or output by example-cli.",
            "hasPart": [
                {
                    "@id": "input.txt"
                },
                {
                    "@id": "output.txt"
                }
            ],
            "license": "CC-BY-4.0",
            "name": "Files used by example-cli"
        },
        {
            "@id": "ro-crate-metadata.json",
            "@type": "CreativeWork",
            "about": {
                "@id": "./"
            },
            "conformsTo": {
                "@id": "https://w3id.org/ro/crate/1.1"
            }
        },
        {
            "@id": "https://w3id.org/ro/wfrun/process/0.5",
            "@type": "CreativeWork",
            "name": "Process Run Crate",
            "version": "0.5"
        },
        {
            "@id": "example-cli@1.2.3",
            "@type": "SoftwareApplication",
            "description": "Example CLI",
            "name": "example-cli",
            "version": "1.2.3"
        },
        {
            "@id": "input.txt",
            "@type": "File",
            "contentSize": 5,
            "description": "Input file",
            "encodingFormat": "text/plain",
            "name": "input.txt"
        },
        {
            "@id": "output.txt",
            "@type": "File",
            "contentSize": 5,
            "description": "Output file",
            "encodingFormat": "text/plain",
            "name": "output.txt"
        },
        {
            "@id": "someuser",
            "@type": "Person",
            "name": "someuser"
        },
        {
            "@id": "example-cli input.txt output.txt",
            "@type": "CreateAction",
            "agent": {
                "@id": "someuser"
            },
            "endTime": "2026-01-28T15:07:18.600135",
            "instrument": {
                "@id": "example-cli@1.2.3"
            },
            "name": "example-cli input.txt output.txt",
            "object": [
                {
                    "@id": "input.txt"
                }
            ],
            "result": [
                {
                    "@id": "output.txt"
                }
            ],
            "startTime": "2026-01-28T15:07:18.599714"
        }
    ]
}

For you the startTime, endTime, and Person name will differ.

You can also call the argument parser agnostic version of the recorder directly. (Click me to see code)

from datetime import datetime, UTC
from pathlib import Path

from rocrate_action_recorder import (
    IOArgumentPath,
    IOArgumentPaths,
    Program,
    record,
)

crate_dir = Path()
input_path = crate_dir / "input.txt"
output_path = crate_dir / "output.txt"
input_path.write_text("Hello World\n")
argv = [
    "myscript",
    "--input",
    str(input_path),
    "--output",
    str(output_path),
]
start_time = datetime(2026, 1, 16, 12, 0, 0, tzinfo=UTC)
# Simulate the script's main operation
output_path.write_text(input_path.read_text().upper())
end_time = datetime(2026, 1, 16, 12, 0, 5, tzinfo=UTC)

crate_meta = record(
    program=Program(
        name="myscript", description="My test script", version="1.2.3"
    ),
    ioargs=IOArgumentPaths(
        input_files=[
            IOArgumentPath(name="input", path=input_path, help="Input file")
        ],
        output_files=[
            IOArgumentPath(name="output", path=output_path, help="Output file")
        ],
    ),
    argv=argv,
    current_user="tester",
    start_time=start_time,
    end_time=end_time,
    crate_dir=crate_dir,
    dataset_license="CC-BY-4.0",
)
# crate_meta == Path("ro-crate-metadata.json")

Example

See the example folder for a minimal example.

Contributions

See AGENTS.md for commands and hints for contributions.

Citation

See CITATION.cff for citation information.