AoC-Day3-regexified

AoC-Day3-regexified.py

import pandas as pd
import re


def main():
    df = pd.read_csv("input.txt", index_col=False)

    re_number = re.compile(r'\d')
    re_spec_char = re.compile(r'[*&%\-/#&$@+=]')
    re_spec_char_dot = re.compile(r'[*&%\-/#&$@+=.]')

    df.columns = ['col1']
    # Split each character into a new column
    df_split = pd.DataFrame(df['col1'].apply(list).tolist())
    symbols = set()

    total_numbers = []
    for row in range(df_split.shape[0]):
        for col in range(df_split.shape[1]):
            this_cell = df_split.iloc[row, col]
            symbols.add(this_cell)
            if re.match(re_spec_char, this_cell):
                spec_char_spot = this_cell
                # see if there are any numbers around this spot
                found_numbers_already = []
                for near_row in [-1, 0, 1]:
                    for near_col in [-1, 0, 1]:
                        near_row_int = row + near_row
                        near_col_int = col + near_col
                        nearby_numbers = []
                        if re.match(re_number, df_split.iloc[near_row_int, near_col_int]):
                            # see if this number is the first spot of the number
                            if re.match(re_spec_char_dot, df_split.iloc[near_row_int, near_col_int - 1]):
                                begin_int_spot = near_col_int
                            elif re.match(re_number, df_split.iloc[near_row_int, near_col_int - 1]):
                                if  re.match(re_spec_char_dot, df_split.iloc[near_row_int, near_col_int - 2]):
                                    begin_int_spot = near_col_int - 1
                                elif re.match(re_number, df_split.iloc[near_row_int, near_col_int - 2]):
                                    begin_int_spot = near_col_int - 2
                            # find the end int spot
                            if  re.match(re_spec_char_dot, df_split.iloc[near_row_int, begin_int_spot + 1]):
                                end_int_spot = begin_int_spot
                            elif re.match(re_number,df_split.iloc[near_row_int, begin_int_spot + 1]):
                                if  re.match(re_spec_char_dot, df_split.iloc[near_row_int, begin_int_spot + 2]):
                                    end_int_spot = begin_int_spot + 1
                                elif re.match(re_number,df_split.iloc[near_row_int, begin_int_spot + 2]):
                                    end_int_spot = begin_int_spot + 2
                            spot_start = begin_int_spot
                            end_spot = end_int_spot + 1
                            nearby_number = [x for x in df_split.iloc[near_row_int, spot_start:end_spot]]
                            nearby_number = int(''.join(nearby_number))
                            if (near_row_int, spot_start) not in found_numbers_already:
                                found_numbers_already += [(near_row_int, spot_start)]
                                total_numbers += [nearby_number]
                            if row == 1:
                                print(df_split.iloc[row, col], row, col, nearby_number, found_numbers_already,
                                      total_numbers)
    print(sum(total_numbers))


if __name__ == '__main__':
    main()