我有一个txt文件，看起来像这样：
Quod equidem non reprehendo;
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Quibus natura iure responderit non esse verum aliunde finem beate vivendi, a se principia rei gerendae peti; Quae enim adhuc protulisti, popularia sunt, ego autem a te elegantiora desidero. Duo Reges: constructio interrete. Tum Lucius: Mihi vero ista valde probata sunt, quod item fratri puto. Bestiarum vero nullum iudicium puto. Nihil enim iam habes, quod ad corpus referas; Deinde prima illa, quae in congressu solemus: Quid tu, inquit, huc? Et homini, qui ceteris animantibus plurimum praestat, praecipue a natura nihil datum esse dicemus?
=========================================================================
Planet Number festival animal
colour book
Mercury First firecrack phone
Venus Last kite computer
Earth Country rangoli tv
Jupiter C.COD bomb
---------------------------------------------------------------------
11 4526 diwali dog
holi bigb
12 Joe diwali 111
45 Doe sankaranti acer
65 UK diwali pan
67 22 diwali
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Planet Number festival animal
colour book
Mercury First firecrack phone
Venus Last kite computer
Earth Country rangoli tv
Jupiter C.COD bomb
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
45 5637 ganesh tiger
holi cinema
67 micael holi 222
78 john diwali xamoi
90 france diwali hp
34 34 diwali我想把这个文本文件转换成csv格式。我想展示的输出：输出：输出
我的代码：
from itertools import groupby, chain
with open("file.txt", "r") as fin,
open("file.csv", "w") as fout:
for key, group in groupby(fin, key=lambda line: bool(line.strip())):
if key:
zipped = zip(*(line.rstrip().split() for line in group))
fout.write(",".join(chain(*zipped)) + "n")
这将满足您的要求。这只是一个收集字段的问题，直到我们得到写入它们的触发器，AND忽略开头的文本，AND忽略除第一个之外的所有标题。
fin = open('file.txt')
fout = open('file.csv','w')
gather = []
skipping = True
first = True
for line in fin:
if skipping:
skipping = line.find('====') < 0
elif line.find('----') >= 0:
if gather and (first or gather[0] != 'Planet'):
print( ','.join(gather), file=fout )
gather = []
first = False
else:
gather.extend( line.strip().split() )
if gather:
print( ','.join(gather), file=fout )
文件的相关块似乎有一个大致固定宽度的列结构，因此您可以尝试在它们上使用pandas.read_fwf：
from io import StringIO
from itertools import groupby
import pandas as pd
def keep(line): return bool(line.strip()) and not line.startswith("---")
with open('file.txt', 'r') as fin,
open('file.csv','w') as fout:
while True:
if next(fin).startswith("==="): break
first = True
for key, group in groupby(fin, key=keep):
if key:
line = ",".join(
pd.read_fwf(StringIO("".join(group)), header=None)
.stack().sort_index(level=1).dropna().astype(str)
.str.replace(r"^(-?d+).0+$", r"1", regex=True)
) + "n"
if first:
header, first = line, False
fout.write(line)
elif line != header:
fout.write(line)
file.csv:中的结果
Planet,Mercury,Venus,Earth,Jupiter,Number,First,Last,Country,C.COD,festival,colour,firecrack,kite,rangoli,bomb,animal,book,phone,computer,tv
11,12,45,65,67,4526,Joe,Doe,UK,22,diwali,holi,diwali,sankaranti,diwali,diwali,dog,bigb,111,acer,pan
45,67,78,90,34,5637,micael,john,france,34,ganesh,holi,holi,diwali,diwali,diwali,tiger,cinema,222,xamoi,hp
如果您不关心数字格式，则可以删除.str.replace(r"^(-?d+).0+$", r"1", regex=True)。
但是：这真的是你文件的真实格式吗？
我相信您可以使用Pandas-lib将txt文件转换为csv
# importing panda library
import pandas as pd
# readinag given csv file
# and creating dataframe
dataframe1 = pd.read_csv("input_file.txt")
# storing this dataframe in a csv file
dataframe1.to_csv('output_file.csv',
index = None)