你们可以用正常的方式读前两行，其余的用 pandas.read_csv(..., sep="\s+") 我用 io.StringIO 仅用于模拟文件，但您应该使用 open() ```
data = '''14.0
7
1 SQLCHAR 0 12 "," 1 Id ""
2 SQLCHAR 0 200 "," 2 value1 SQL_Latin1_General_CP1_CI_AS
3 SQLCHAR 0 200 "," 3 value2 SQL_Latin1_General_CP1_CI_AS
4 SQLCHAR 0 200 "," 4 value3 SQL_Latin1_General_CP1_CI_AS
5 SQLCHAR 0 200 "," 5 value4 SQL_Latin1_General_CP1_CI_AS
6 SQLCHAR 0 200 "," 6 value5 SQL_Latin1_General_CP1_CI_AS
7 SQLCHAR 0 12 "\r\n" 7 Flag ""'''

import pandas as pd
import io

open file for reading

fh = io.StringIO(data)

fh = open(filename)

read first two line

two_lines = [next(fh), next(fh)]
print(two_lines)

read rest as CSV file without headers

df = pd.read_csv(fh, sep="\s+", header=None)
print(df)

skip rows which have Id or Flag in 7th column

df = df[ ~df[6].str.contains('Id|Flag') ]

assign new values in first column

df[0] = range(1, len(df)+1)

print(df)

open file for writing

fh = open('output.txt', 'w')

write first two lines

fh.write(two_lines[0])
fh.write(str(len(df)) + '\n')

if you don't need spaces then you can use this

df.to_csv(fh, sep=' ', index=False, header=False)

if you need spaces then you can use this

for row in df.itertuples():
# format line to keep spaces
line = '{:<8}{:<20}{:<8}{:<9}{:<8}{:<6}{:<28}{}'.format(*row[1:])

#print(line)
# write line
fh.write(line + '\n')

close file

fh.close()

这应该使用 re.split() 并根据要求跳过/重新编号字段：

import re
SKIP_WORDS = ['Id', 'Flag']
LAST_LINE_CRLF = r'"\r\n"'
with open('test.fmt') as input_file:
    vers = next(input_file).strip()
    _ = next(input_file)
    new_lines, new_field_no = [], 0
    for line in input_file:
        if not any(s in line for s in SKIP_WORDS):
            new_field_no += 1
            data = re.split(r'( +(?=.*))', line.strip())
            data[0] = str(new_field_no)  # renumber field
            data[10] = str(new_field_no)  # renumber field
            new_lines.append(data)
with open('test_new.fmt', 'w') as output_file:
    output_file.write(vers + '\n')
    output_file.write(str(new_field_no) + '\n')
    for idx, line in enumerate(new_lines, start=1):
        if idx == new_field_no:  # adjust last line delimiter
            line[8] = LAST_LINE_CRLF
            line[9] = '   '
        output_file.write(''.join(line) + '\n')