How to find text between `*~`?

Question

There are lots of special characters in these text file I get and I am trying to find and replace them with other strings. For example:

s = "BRAND~*HP*5250*RAM128*GPUAMD7990*~"

df = re.findall('BRAND~*(.+?)*~', s)

print(df)

The result yield ['*']

I want the result to be HP*5250*RAM128*GPUAMD7990

How do I achieve this?

`findall` returns a `list` not a `string`... would't be easier to use string methods `s.removeprefix("BRAND~*").removesuffix("*~")` or `re.sub`? — cards, Jan 23 '23 at 20:46

score 2 · Answer 1 · answered Jan 23 '23 at 19:54

2

Escape the * in the pattern:

s = "BRAND~*HP*5250*RAM128*GPUAMD7990*~"

out = re.findall('BRAND~\*(.+?)\*~', s)

Output:

['HP*5250*RAM128*GPUAMD7990']

answered Jan 23 '23 at 19:54

mozway

score 2 · Answer 2 · answered Jan 23 '23 at 19:55

2

You need to escape the '*'s:

df = re.findall('BRAND~\*(.+?)\*~', str)

answered Jan 23 '23 at 19:55

JustLearning

Same thing for `~` or others. – JustLearning Jan 23 '23 at 19:56
1

`~` has no special meaning in regular expressions and doesn't need escaping – kindall Jan 23 '23 at 19:58
You are right! my bad. – JustLearning Jan 23 '23 at 19:59

score 1 · Answer 3 · answered Jan 23 '23 at 20:38

1

You want to

PS: I see your question has been edited. You may discard the 'strip' part. It seems you did not need that anymore.

[match / find]

import re

## find match based on regex pattern
df = re.findall('BRAND~\*(.+?)\*~', str)
df

[out] ['HP*5250*RAM128*GPUAMD7990']

[strip / remove]

r = re.compile('(\*)')
[r.sub('', w) for w in df]

out ['HP5250RAM128GPUAMD7990']

PS: Others might assist with a more python code!

answered Jan 23 '23 at 20:38

semmyk-research

1

Hi semmyk-research. Your post is right on. Just a small warning. Forcing a match between delimiters using `.+?` could consume a delimiter if the content were empty. The adverse affect is to go to the next valid delimiter it finds. – sln Jan 23 '23 at 21:09
@sln Thanks for pointing this out. #Appreciated. Some of these things, one gets to overlook easily at ones detriment (down the line). Apparently, `.+` gets even greedy~ – semmyk-research Jan 24 '23 at 10:37
Yeah so `.*?` single character checks for the end delimiter might be better always in this cases. – sln Jan 27 '23 at 21:15

score 1 · Answer 4 · answered Jan 23 '23 at 21:01

1

A pair of lookarounds will get the center.

(?<=BRAND~\*).*?(?=\*~)

But to replace specific characters within that match requires a callback function.

answered Jan 23 '23 at 21:01

sln

4 Answers4