FORMAT_CONVERSION_MAXCOMPUTE/format_convert/test_walk.py

import copy
import os
import random
import re
import sys
import time
from bs4 import BeautifulSoup
from datetime import datetime
from multiprocessing import Process
import datetime as dt
sys.path.append(os.path.dirname(os.path.abspath(__file__)) + "/../")
from format_convert.utils import file_lock


def run():

    f = file_lock(os.path.abspath(os.path.dirname(__file__)) + '/19022.lock')
    print("acquire file_lock! process " + str(os.getpid()))
    for i in range(10):
        print("process " + str(os.getpid()) + " " + str(i))
        time.sleep(random.randint(0, 1))
    f.close()


def merge_table():
    with open(r'C:\Users\Administrator\Desktop\2.html', 'r') as f:
        html_str = f.read()
    html_str_origin = copy.deepcopy(html_str)

    try:
        match1 = re.finditer('<table', html_str)
        match2 = re.finditer('</table>', html_str)
        table_index_list = []
        for m1, m2 in zip(match1, match2):
            table_index_list.append([m1.span()[0], m1.span()[1], m2.span()[0], m2.span()[1]])
        print(table_index_list)

        soup = BeautifulSoup(html_str)
        tables = soup.find_all('table')
        table_td_cnt_list = []
        for table in tables:
            tds = table.tr.find_all('td')
            table_td_cnt_list.append(len(list(tds)))
        print(table_td_cnt_list)

        if len(table_index_list) == len(table_td_cnt_list):
            merge_index_list = []
            temp_index = []
            for i in range(1, len(table_index_list)):
                last_index = table_index_list[i-1]
                index = table_index_list[i]
                last_tds = table_td_cnt_list[i-1]
                tds = table_td_cnt_list[i]
                if index[0] - last_index[-1] == 0 and last_tds == tds:
                    temp_index += [i-1, i]
                    temp_index = list(set(temp_index))
                else:
                    if temp_index:
                        merge_index_list.append(temp_index)
                    temp_index = []
            if temp_index:
                merge_index_list.append(temp_index)
            print(merge_index_list)

            print('before len(html_str)', len(html_str))
            for merge in merge_index_list:
                start_index = table_index_list[merge[0]][0]
                end_index = table_index_list[merge[-1]][-1]
                table_replace = re.sub('<table[^>]*>|</table>', '', html_str[start_index:end_index])
                table_replace = '<table border="1">' + table_replace + '</table>'
                table_replace += ' '*(end_index-start_index-len(table_replace))
                html_str = html_str[:start_index] + table_replace + html_str[end_index:]
            print('after len(html_str)', len(html_str))

            if len(html_str_origin) == len(html_str):
                with open(r'C:\Users\Administrator\Desktop\3.html', 'w') as f:
                    f.write(html_str)
                return html_str
            else:
                return html_str_origin
        else:
            return html_str_origin
    except:
        return html_str_origin


if __name__ == '__main__':
    # process_list = []
    # for j in range(10):
    #     p1 = Process(target=run,)
    #     p1.start()
    #     process_list.append(p1)
    #
    # for p in process_list:
    #     p.join()

    print('|'.join(['a', 'n']))
    _t = datetime.strptime('2023-04-26', '%Y-%m-%d')
    _t2 = datetime.strptime('2023-04-02', '%Y-%m-%d')
    print(abs((_t2-_t).days))
    print(datetime.strftime(_t + dt.timedelta(days=10), '%Y-%m-%d'))

    # merge_table()

    print(datetime.now())