pdf_parser/complaints/parse_pdf.py

from collections import defaultdict
from IPython.display import display
import os
import tika
import pandas as pd

tika.initVM()
from tika import parser
from parse_pdf_utils import (
    extract_acronyms,
    extract_case_number,
    extract_defendent,
    extract_filing_date,
    extract_firm_name,
    extract_patent_number,
    extract_plaintiff,
)


def extract_all(PATH):
    """
    Returns all the required data from the pdfs in a dataframe format.
    """
    data_dict = defaultdict(list)
    required_files = [file for file in os.listdir(PATH) if file.find(".pdf") != -1]
    for idx, file in enumerate(required_files):
        print(idx, file)
        parse_file = parser.from_file(file)["content"].strip().replace("\n", "")
        # data_dict["case_number_list"].append(extract_case_number(content=parse_file))
        # data_dict["filing_date_list"].append(extract_filing_date(content=parse_file))
        data_dict["plaintiff_list"].append(extract_plaintiff(content=parse_file))
        # data_dict["defendent_list"].append(extract_defendent(content=parse_file))
        # data_dict["acronyms_list"].append(extract_acronyms(content=parse_file))
        # data_dict["firm_name_list"].append(extract_firm_name(content=parse_file))
        # data_dict["patent_list"].append(extract_patent_number(content=parse_file))

    data = pd.DataFrame(data_dict)
    # data.to_csv("required_data.csv", index=False)
    return data


if __name__ == "__main__":
    BASE_DIR = "Code/pdf_parser/complaints"
    HOME_DIR = os.path.expanduser("~")
    PATH = os.path.join(HOME_DIR, BASE_DIR)
    display(extract_all(PATH))