manuscript.py 文件源码

def get_manuscript_stats(text, citation_df):
    """
    Compute manuscript statistics.
    """
    stats = collections.OrderedDict()

    # Number of distinct references by type
    ref_counts = (
        citation_df
        .standard_citation
        .drop_duplicates()
        .map(lambda x: x.split(':')[0])
        .pipe(collections.Counter)
    )
    ref_counts['total'] = sum(ref_counts.values())
    stats['reference_counts'] = ref_counts
    stats['word_count'] = len(text.split())
    logging.info(f"Generated manscript stats:\n{json.dumps(stats, indent=2)}")
    return stats