正则表达式A“或” B

Question

I have the following lines:

  9 (1224) Starting item export: IPM.Appointment, Zomverbanden (wielen) monteren, 2,61 B, John \Calendar, E:\tmp\John Kn
  9 (1224) Starting item export: IPM.Appointment,  [JK], 7,97 KB, John Knappers\Calendar, E:
  9 (1224) Starting item export: IPM.Appointment, Niet op kantoor (Auto), 1,66 GB, John \Calendar, E:\tmp\John .
  9 (1224) Starting item export: IPM.Appointment, Bespip / Tobias , 9,13 KB, John \Calendar, E:\tmp\John K
  9 (1224) Starting item export: IPM.Appointment, Q-ware el / Mehan [JK], 8,01 MB, \Calendar, E:\tmp\J

how can I find a matching pattern for these bytes and megabytes and so on?

I have tried

res = re.findall(r'(\d*,\d* KB)|(\d*,\d* MB) | (\d*,\d* B)| (\d*,\d* GB)', i)

but it returns me dict of 4 tuples, but I need only 1 item for each line:

2,61 B
7,97 KB
1,66 GB
9,13 KB
8,01 MB

Answer 1

You can rewrite it making it shorter:

\d+,\d* (?:KB|MB|B|GB)

Live Demo

You could make it even shorter:

\d+,\d* [KMG]?B

Live Demo

This is of course assuming there's a single space in between the number and the unit. Instead of the space you could have [ \\t]+ , making it possible to be multiple spaces or even tabs.

In case 10 KB is valid (without the decimal) then you could do:

\d+(,\d+)? [KMG]?B