Python解析然后放入数据框架中

huangapple
go评论104阅读模式
英文:

Python parse then put in a dataframe

问题

我想创建一个类似的数据框：

DATA1 ERROR1
123456 500
56789 505
英文:

I have a file with a data like this:

  1. ------------------------------
  2. ------------------------------
  3. <TIME:2020-01-01 01:25:10> 
  4. <TIME:2020-01-01 01:25:10> 
  5. <TIME:2020-01-01 01:25:10> 
  6. <TIME:2020-01-01 01:25:10>
  8. ------
  9. ++++++
  10. %%RequestHandler
  11. 	DATA1 = 123456
  12. 	ERROR1 = 500
  13. 	DATA2 = 56789
  14. 	ERROR2 = 505
  16. Count = 4
  17. ---

I would like to create a dataframe like

DATA1 ERROR1
123456 500
56789 505

答案1

得分: 2

以下是您要的代码翻译：

  1. import re
  2. import pandas as pd
  4. # 读取文件
  5. with open("file.txt", "r") as file:
  6.     content = file.read()
  8. # 使用正则表达式从原始结构化文本文件中提取值
  9. data = re.findall(r"DATA\d+\s*=\s*(\d+)", content)
  10. error = re.findall(r"ERROR\d+\s*=\s*(\d+)", content)
  12. # 创建一个数据框
  13. df = pd.DataFrame({"DATA": data, "ERROR": error})
  14. print(df)

示例:

  1. import re
  2. import pandas as pd
  4. content = '''
  5. ------------------------------
  6. ------------------------------
  7. <TIME:2020-01-01 01:25:10> 
  8. <TIME:2020-01-01 01:25:10> 
  9. <TIME:2020-01-01 01:25:10> 
  10. <TIME:2020-01-01 01:25:10>
  12. ------
  13. ++++++
  14. %%RequestHandler
  15.     DATA1 = 123456
  16.     ERROR1 = 500
  17.     DATA2 = 56789
  18.     ERROR2 = 505
  20. Count = 4
  21. ---
  22. '''
  24. data = re.findall(r"DATA\d+\s*=\s*(\d+)", content)
  25. error = re.findall(r"ERROR\d+\s*=\s*(\d+)", content)
  27. df = pd.DataFrame({"DATA": data, "ERROR": error})
  28. print(df)

输出:

  1.      DATA ERROR
  2. 0  123456   500
  3. 1   56789   505

（注意：代码中的 &quot; 在中文翻译中并没有特殊意义，因此我将其保留为英文引号 "。）

英文:

Here is the code that you want, you can regular expressions to extract desired data from raw structured text file:

  1. import re
  2. import pandas as pd
  4. # Read the file
  5. with open(&quot;file.txt&quot;, &quot;r&quot;) as file:
  6.     content = file.read()
  8. # Use regular expressions to extract the values
  9. data = re.findall(r&quot;DATA\d+\s*=\s*(\d+)&quot;, content)
  10. error = re.findall(r&quot;ERROR\d+\s*=\s*(\d+)&quot;, content)
  12. # Create a dataframe
  13. df = pd.DataFrame({&quot;DATA&quot;: data, &quot;ERROR&quot;: error})
  14. print(df)

Example:

  1. import re
  2. import pandas as pd
  4. content = &#39;&#39;&#39;
  5. ------------------------------
  6. ------------------------------
  7. &lt;TIME:2020-01-01 01:25:10&gt; 
  8. &lt;TIME:2020-01-01 01:25:10&gt; 
  9. &lt;TIME:2020-01-01 01:25:10&gt; 
  10. &lt;TIME:2020-01-01 01:25:10&gt;
  12. ------
  13. ++++++
  14. %%RequestHandler
  15.     DATA1 = 123456
  16.     ERROR1 = 500
  17.     DATA2 = 56789
  18.     ERROR2 = 505
  20. Count = 4
  21. ---
  22. &#39;&#39;&#39;
  24. data = re.findall(r&quot;DATA\d+\s*=\s*(\d+)&quot;, content)
  25. error = re.findall(r&quot;ERROR\d+\s*=\s*(\d+)&quot;, content)
  27. df = pd.DataFrame({&quot;DATA&quot;: data, &quot;ERROR&quot;: error})
  28. print(df)

Output:

  1.      DATA ERROR
  2. 0  123456   500
  3. 1   56789   505

答案2

得分: 2

另一种使用 pivot 的正则表达式方法：

  1. import re
  3. # 或者使用 file.read()
  4. out = (pd.DataFrame(re.findall(r'^\s+(\w+)(\d+) = (\d+)', text, flags=re.M))
  5.          .pivot(index=1, columns=0, values=2)
  6.          .rename_axis(index=None, columns=None)
  7.       )
  9. print(out)

输出结果：

  1.      DATA ERROR
  2. 1  123456   500
  3. 2   56789   505

使用的输入：

  1. text = '''------------------------------
  2. ------------------------------
  3. <TIME:2020-01-01 01:25:10> 
  4. <TIME:2020-01-01 01:25:10> 
  5. <TIME:2020-01-01 01:25:10> 
  6. <TIME:2020-01-01 01:25:10>;
  8. ------
  9. ++++++
  10. %%RequestHandler
  11.     DATA1 = 123456
  12.     ERROR1 = 500
  13.     DATA2 = 56789
  14.     ERROR2 = 505
  16. Count = 4'''

正则表达式演示

英文:

Another regex approach with pivot:

  1. import re
  3.                                                          # or file.read()
  4. out = (pd.DataFrame(re.findall(r&#39;^\s+(\w+)(\d+) = (\d+)&#39;, text, flags=re.M))
  5.          .pivot(index=1, columns=0, values=2)
  6.          .rename_axis(index=None, columns=None)
  7.       )
  9. print(out)

Output:

  1.      DATA ERROR
  2. 1  123456   500
  3. 2   56789   505

Used input:

  1. text = &#39;&#39;&#39;------------------------------
  2. ------------------------------
  3. &lt;TIME:2020-01-01 01:25:10&gt; 
  4. &lt;TIME:2020-01-01 01:25:10&gt; 
  5. &lt;TIME:2020-01-01 01:25:10&gt; 
  6. &lt;TIME:2020-01-01 01:25:10&gt;
  8. ------
  9. ++++++
  10. %%RequestHandler
  11.     DATA1 = 123456
  12.     ERROR1 = 500
  13.     DATA2 = 56789
  14.     ERROR2 = 505
  16. Count = 4&#39;&#39;&#39;

regex demo

huangapple
  • 本文由 发表于 2023年7月12日 20:31:06
  • 转载请务必保留本文链接：https://go.coder-hub.com/76670562.html
  • dataframe
  • pandas
  • python
匿名

发表评论

匿名网友

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen:

确定