我有一个凌乱的文本文件，在15分钟后重复了相同的数据 我想使用python
从该txt文件创建一个数据框，其中open（fname）为f： 内容= f.read（）。splitlines（）
2019-09-26 14:15:44 discount=1e
019-09-26 14:16:44 discount=4e
019-09-26 14:17:44 discount=2e
019-09-26 14:18:44 discount=3e
019-09-26 14:19:44 discount=2e
some text
some text
some text
Products: sold = 5, bought = 5, left = 0 (20% profit),
New data and new data in the same format
期望的数据帧
date discount profit
2019-09-26 14:15:44 1 20%
2019-09-26 14:16:44 4 20%
2019-09-26 14:17:44 2 20%
2019-09-26 14:18:44 3 20%
2019-09-26 14:19:44 2 20%
答案 0 :(得分：0)
这适用于您的示例数据，但您可能需要对真实数据进行一些调整。
数据：
2019-09-26 14:15:44 discount=1e
2019-09-26 14:16:44 discount=4e
2019-09-26 14:17:44 discount=2e
2019-09-26 14:18:44 discount=3e
2019-09-26 14:19:44 discount=2e
some text
some text
some text
Products: sold = 5, bought = 5, left = 0 (20% profit),
2019-09-26 14:20:44 discount=1e
2019-09-26 14:21:44 discount=4e
2019-09-26 14:22:44 discount=2e
2019-09-26 14:23:44 discount=3e
2019-09-26 14:24:44 discount=2e
some text
some text
some text
Products: sold = 5, bought = 5, left = 0 (15% profit)
在：
# range(12) because the expected input was 12 fields. May need to change this to the number of expected fields
df = pd.read_clipboard(names=[x for x in range(12)])
# 10 is the column name with the profit. May need to change this.
df[10] = df[10].bfill()
df['date'] = pd.to_datetime(df[0] +' '+ df[1], errors='coerce')
df = df[df['date'].notnull()]
df['discount'] = df[2].str.strip('discount=e')
df['profit'] = df[10].str.strip('()')
df[['date', 'discount', 'profit']]
出局：
| date | discount | profit |
|:-------------------:|----------|--------|
| 2019-09-26 14:15:44 | 1 | 20% |
| 2019-09-26 14:16:44 | 4 | 20% |
| 2019-09-26 14:17:44 | 2 | 20% |
| 2019-09-26 14:18:44 | 3 | 20% |
| 2019-09-26 14:19:44 | 2 | 20% |
| 2019-09-26 14:20:44 | 1 | 15% |
| 2019-09-26 14:21:44 | 4 | 15% |
| 2019-09-26 14:22:44 | 2 | 15% |
| 2019-09-26 14:23:44 | 3 | 15% |
| 2019-09-26 14:24:44 | 2 | 15% |