使用Python将一个数据文本文件拆分成几个用于MySQL的文本文件

Question

我有一个数据 txt 文件，其格式设置为以以下格式（有点夸张）加载到数据库（MySQL）中：

数据.txt

name   age profession datestamp
John   23  engineer   2020-03-01
Amy    17  doctor     2020-02-27
Gordon 19  artist     2020-02-27
Kevin  25  chef       2020-03-01

以上是通过python执行以下命令生成的：

LOAD DATA LOCAL INFILE '/home/sample_data/data.txt' REPLACE INTO TABLE person_professions 
FIELDS TERMINATED BY 0x01 OPTIONALLY ENCLOSED BY '\"' LINES TERMINATED BY '\n'
 (name,age,profession,datestamp)

它创建了 data.txt； 但是，data.txt 对于一次插入所有数据库来说真的很大（设置了约 200 MB 插入限制），我想将数据分成几个块（data_1.txt、data_2.txt、data_3.txt 等） .) 并将它们一一插入以避免达到插入大小限制。 我知道你可以一行一行地寻找一个条件来切出数据，例如

with open('data.txt', 'w') as f:
    data = f.read().split('\n')
    if some condition:
       with open('data_1.txt', 'w') as f2:
            insert data 

但我不太确定如何提出条件断点以使其开始插入新的 txt 文件，除非有更好的方法。

Answer 1

我编写了一个可以完成工作的函数，具体取决于文件的大小。 代码注释中的解释。

def split_file(file_name, lines_per_file=100000):
    # Open large file to be read in UTF-8
    with open(file_name, 'r', encoding='utf-8') as rf:
        # Read all lines in file
        lines = rf.readlines()
        print ( str(len(lines)) + ' LINES READ.')
        # Set variables to count file number and count of lines written
        file_no = 0
        wlines_count = 0
        # For x from 0 to length of lines read stepping by number of lines that will be written in each file
        for x in range(0, len(lines), lines_per_file):
            # Open new "split" file for writing in UTF-8
            with open( 'data' + '-' + str(file_no) + '.txt', 'w', encoding='utf-8') as wf:
                # Write lines
                wf.writelines(lines[x:x+lines_per_file])
                # Update the written lines count
                wlines_count += (len(lines[x:x + lines_per_file]))
                # Update new "split" file count mainly for naming
                file_no+=1
        print(str(wlines_count) + " LINES WRITTEN IN " + str(file_no) + " FILES.")

# Split data.txt into files containing 100000 lines
split_file('data.txt',100000)