Commandline Tool

Command Line Tool¤

kirsche¤

Usage:

kirsche [OPTIONS] COMMAND [ARGS]...

Options:

  --help  Show this message and exit.

connections¤

Establish connections between the list of papers, either from a list of DOIs, bib file, or from download metadata file.

If no metadata_file provided, the metadata will be downloaded using parameters specified in paper_id or bib_file. If metadata_file is provided, the connections will be established using the metadata file.

About Metadata Sources¤

Download paper data from service provides (e.g., SemanticScholar).

There are two ways to provide a list of DOIs to be retrieved, provide paper DOIs directly using --paper_id or -p, or loading a bib file using --source_bib_file or -sb.

To save the downloaded data, provide a path to a file using --target or -t.

:param paper_id: Paper DOI, optional, can be multiple :param source_bib_file: Bib file path, optional :param source_metadata_path: Target data file path, optional :param connected_papers_path: path to save enhanced data file with connections calcualted :param sleep_time: Sleep time between requests, defaults to 1sec.

Usage:

kirsche connections [OPTIONS]

Options:

  -p, --paper_id TEXT             Paper ID
  -bib, --source_bib_file PATH    Bib file path
  -meta, --source_metadata_path PATH
                                  path to data file/folder with paper metadata
  -t, --connected_papers_path PATH
                                  path to save enhanced data file
  -st, --sleep_time INTEGER       Sleep time between requests
  --help                          Show this message and exit.

connections-from-metadata¤

Establish connections between the list of papers

:param source_metadata_path: path to data file with paper metadata :param connected_papers_path: path to save enhanced data file

Usage:

kirsche connections-from-metadata [OPTIONS]

Options:

  -s, --source_metadata_path PATH
                                  path to data file/folder with paper metadata
  -t, --connected_papers_path PATH
                                  path to save enhanced data file(s)
  --help                          Show this message and exit.

metadata¤

Download paper data from service provides (e.g., SemanticScholar).

There are two ways to provide a list of DOIs to be retrieved, provide paper DOIs directly using --paper_id or -p, or loading a bib file using --source_bib_file or -bib.

To save the downloaded data, provide a path to a file using --target_metadata_path or -t.

:param paper_id: Paper DOI, optional, can be multiple :param source_bib_file: Bib file path, optional :param target_metadata_path: Target data file path, optional :param sleep_time: Sleep time between requests, defaults to 1sec.

Usage:

kirsche metadata [OPTIONS]

Options:

  -p, --paper_id TEXT             Paper ID
  -bib, --source_bib_file PATH    Bib file path
  -t, --target_metadata_path PATH
                                  Target data file path
  -sleep, --sleep_time INTEGER    Sleep time between requests
  --help                          Show this message and exit.

visualization¤

Visualize the connections between the papers.

Usage:

kirsche visualization [OPTIONS]

Options:

  -p, --source_paper_id TEXT      Source: Paper ID
  -bib, --source_bib_file PATH    Source: Bib file path
  -meta, --source_metadata_path PATH
                                  Source: path to data file/folder with paper
                                  metadata
  -conn, --source_connected_papers_path PATH
                                  Source: path to save enhanced data
                                  file/folder
  --title TEXT                    title of the chart
  -t, --target_html_path PATH     Target: path to html file  [required]
  -sleep, --sleep_time INTEGER    Sleep time between requests
  --help                          Show this message and exit.

Helpers¤

`connections(paper_id, source_bib_file, source_metadata_path, connected_papers_path, sleep_time)` ¤

Establish connections between the list of papers, either from a list of DOIs, bib file, or from download metadata file.

If no metadata_file provided, the metadata will be downloaded using parameters specified in paper_id or bib_file. If metadata_file is provided, the connections will be established using the metadata file.

About Metadata Sources¤

Download paper data from service provides (e.g., SemanticScholar).

There are two ways to provide a list of DOIs to be retrieved, provide paper DOIs directly using --paper_id or -p, or loading a bib file using --source_bib_file or -sb.

To save the downloaded data, provide a path to a file using --target or -t.

Parameters:

Name	Type	Description	Default
`paper_id`	`Union[str, Iterable]`	Paper DOI, optional, can be multiple	required
`source_bib_file`	`Path`	Bib file path, optional	required
`source_metadata_path`	`Path`	Target data file path, optional	required
`connected_papers_path`	`Union[str, Path]`	path to save enhanced data file with connections calcualted	required
`sleep_time`	`int`	Sleep time between requests, defaults to 1sec.	required

Source code in kirsche/command.py

@kirsche.command()
@click.option("--paper_id", "-p", help="Paper ID", multiple=True)
@click.option("--source_bib_file", "-bib", type=click.Path(exists=True), help="Bib file path")
@click.option(
    "--source_metadata_path",
    "-meta",
    type=click.Path(exists=True),
    help="path to data file/folder with paper metadata",
)
@click.option(
    "--connected_papers_path",
    "-t",
    type=click.Path(exists=False),
    help="path to save enhanced data file",
)
@click.option("--sleep_time", "-st", default=1, help="Sleep time between requests")
def connections(
    paper_id: Union[str, Iterable],
    source_bib_file: Path,
    source_metadata_path: Path,
    connected_papers_path: Union[str, Path],
    sleep_time: int,
):
    """Establish connections between the list of papers, either from a list of DOIs, bib file, or from download metadata file.

    If no `metadata_file` provided, the metadata will be downloaded using parameters specified in `paper_id` or `bib_file`.
    If `metadata_file` is provided, the connections will be established using the metadata file.

    ## About Metadata Sources

    Download paper data from service provides (e.g., SemanticScholar).

    There are two ways to provide a list of DOIs to be retrieved, provide paper DOIs directly using `--paper_id` or `-p`, or loading a bib file using `--source_bib_file` or `-sb`.

    To save the downloaded data, provide a path to a file using `--target` or `-t`.


    :param paper_id: Paper DOI, optional, can be multiple
    :param source_bib_file: Bib file path, optional
    :param source_metadata_path: Target data file path, optional
    :param connected_papers_path: path to save enhanced data file with connections calcualted
    :param sleep_time: Sleep time between requests, defaults to 1sec.
    """
    if isinstance(connected_papers_path, str):
        connected_papers_path = Path(connected_papers_path)

    click.secho(f"Retrieving paper metadata...")
    if not source_metadata_path:
        if source_bib_file:
            logger.debug(f"Using bib file: {source_bib_file}")

        if connected_papers_path.exists():
            existing_connected_papers = load_batch_json(connected_papers_path)
        else:
            existing_connected_papers = []

        records = _metadata(
            paper_id,
            source_bib_file,
            None,
            sleep_time,
            existing_records=existing_connected_papers,
        )
    else:
        records = load_batch_json(source_metadata_path)
    click.secho(f"  Retrieved {len(records)} records.")

    click.secho(f"Connecting papers...")
    connected_papers = append_connections(records)
    click.secho(f"  Connected papers...")

    # Filter out unnecessary keys in the dictionary
    click.secho(f"Filtering and saving data...")
    connected_papers = save_connected_papers(connected_papers, target=connected_papers_path)
    click.secho(f"  Done...")

    if not connected_papers_path:
        click.secho(f"No saving path specified, printing simplified data view...")
        dv = DataViews(connected_papers)
        click.echo(dv.json_simple)

`connections_from_metadata(source_metadata_path, connected_papers_path)` ¤

Establish connections between the list of papers

Parameters:

Name	Type	Description	Default
`source_metadata_path`	`Path`	path to data file with paper metadata	required
`connected_papers_path`	`Path`	path to save enhanced data file	required

Source code in kirsche/command.py

@kirsche.command()
@click.option(
    "--source_metadata_path",
    "-s",
    type=click.Path(exists=True),
    help="path to data file/folder with paper metadata",
)
@click.option(
    "--connected_papers_path",
    "-t",
    type=click.Path(exists=False),
    help="path to save enhanced data file(s)",
)
def connections_from_metadata(source_metadata_path: Path, connected_papers_path: Path):
    """Establish connections between the list of papers

    :param source_metadata_path: path to data file with paper metadata
    :param connected_papers_path: path to save enhanced data file
    """

    connected_papers = append_connections_for_file(source_metadata_path, connected_papers_path)

    return connected_papers

`metadata(paper_id, source_bib_file, target_metadata_path, sleep_time)` ¤

Download paper data from service provides (e.g., SemanticScholar).

There are two ways to provide a list of DOIs to be retrieved, provide paper DOIs directly using --paper_id or -p, or loading a bib file using --source_bib_file or -bib.

To save the downloaded data, provide a path to a file using --target_metadata_path or -t.

Parameters:

Name	Type	Description	Default
`paper_id`	`Union[str, Iterable]`	Paper DOI, optional, can be multiple	required
`source_bib_file`	`Path`	Bib file path, optional	required
`target_metadata_path`	`Path`	Target data file path, optional	required
`sleep_time`	`Optional[int]`	Sleep time between requests, defaults to 1sec.	required

Source code in kirsche/command.py

@kirsche.command()
@click.option("--paper_id", "-p", help="Paper ID", multiple=True)
@click.option("--source_bib_file", "-bib", type=click.Path(exists=True), help="Bib file path")
@click.option(
    "--target_metadata_path",
    "-t",
    type=click.Path(exists=False),
    help="Target data file path",
)
@click.option("--sleep_time", "-sleep", default=1, help="Sleep time between requests")
def metadata(
    paper_id: Union[str, Iterable],
    source_bib_file: Path,
    target_metadata_path: Path,
    sleep_time: Optional[int],
):
    """Download paper data from service provides (e.g., SemanticScholar).

    There are two ways to provide a list of DOIs to be retrieved, provide paper DOIs directly using `--paper_id` or `-p`, or loading a bib file using `--source_bib_file` or `-bib`.

    To save the downloaded data, provide a path to a file using `--target_metadata_path` or `-t`.

    :param paper_id: Paper DOI, optional, can be multiple
    :param source_bib_file: Bib file path, optional
    :param target_metadata_path: Target data file path, optional
    :param sleep_time: Sleep time between requests, defaults to 1sec.
    """
    records = _metadata(paper_id, source_bib_file, target_metadata_path, sleep_time)

    return records

`visualization(source_paper_id, source_bib_file, source_metadata_path, source_connected_papers_path, title, target_html_path, sleep_time)` ¤

Visualize the connections between the papers.

Source code in kirsche/command.py

@kirsche.command()
@click.option("--source_paper_id", "-p", required=False, help="Source: Paper ID", multiple=True)
@click.option(
    "--source_bib_file",
    "-bib",
    required=False,
    type=click.Path(exists=True),
    help="Source: Bib file path",
)
@click.option(
    "--source_metadata_path",
    "-meta",
    required=False,
    type=click.Path(exists=True),
    help="Source: path to data file/folder with paper metadata",
)
@click.option(
    "--source_connected_papers_path",
    "-conn",
    required=False,
    type=click.Path(exists=True),
    help="Source: path to save enhanced data file/folder",
)
@click.option("--title", default="Kirsche: Paper Graph", help="title of the chart")
@click.option(
    "--target_html_path",
    "-t",
    required=True,
    type=click.Path(exists=False),
    help="Target: path to html file",
)
@click.option("--sleep_time", "-sleep", default=1, help="Sleep time between requests")
def visualization(
    source_paper_id,
    source_bib_file,
    source_metadata_path,
    source_connected_papers_path,
    title,
    target_html_path,
    sleep_time,
):
    """Visualize the connections between the papers."""
    if source_connected_papers_path:
        connected_papers = load_batch_json(source_connected_papers_path)
    else:
        click.secho(f"Retrieving paper metadata...")
        if not source_metadata_path:
            if source_bib_file:
                logger.debug(f"Using bib file: {source_bib_file}")
            records = _metadata(source_paper_id, source_bib_file, None, sleep_time)
        else:
            records = load_batch_json(source_metadata_path)
        click.secho(f"  Retrieved {len(records)} records.")

        click.secho(f"Connecting papers...")
        connected_papers = append_connections(records)
        click.secho(f"  Connected papers...")

        # Filter out unnecessary keys in the dictionary
        click.secho(f"Filtering and saving data...")
        connected_papers = save_connected_papers(connected_papers)
        click.secho(f"  Done...")

    g = PaperGraph(connected_papers, title=title)
    nodes = g.nodes
    edges = g.edges

    click.secho(f"Saving html file...")
    visualize(nodes, edges, g.title, target_html_path)

Commandline Tool