Extract text between two pattern in python using regex

Question

I am trying to extract out all texts including categories (i.e. A, B, C).

A     <some text1> 

B     <some text2> 

C     <some text3>

However, when I apply this regex -

ptrn='\n[A-z]*\t'     

pattern1= '(.*)'+ptrn      

f = re.findall(pattern1,test_doc)

it gives me

f[0] = A     <some text1> 

f[1] = <some text2> 

f[2] = <some text3>

But I want -

f[0] =  A     <some text1>

f[0] =  B     <some text2> 

f[2] =  C     <some text2>

this link has some raw text of many documents. each document has following pattern:

category<tab><sometext> \n

hence the whole corpus looks like this:-

category<tab><sometext1> \n 

category<tab><sometext2> \n

.

.

i want

doc[0] = category<tab><sometext1>

doc[1] = category<tab><sometext2>

.
.
and so on

Any answer/hint will be very helpful :)

Wait so you want to find all the text? Why do you need regex? Is there other text somewhere that you don't want? — pault
– pault, Commented Jan 27, 2018 at 3:44
@EvanNowak because <some text> can contain '\n' and it will split within <some text> . — IISC_Student
– IISC_Student, Commented Jan 27, 2018 at 3:51
Maybe you're looking for something like this, but it's hard to tell. Can you give us a more specific example input/output? — pault
– pault, Commented Jan 27, 2018 at 4:01
csmining.org/tl_files/Project_Datasets/r8%20r52/… this link has some raw text of many documents. each document has following pattern: category<tab><sometext1> \n category<tab><sometext2> \n . . i want doc[1] = category<tab><sometext1> doc[2] = category<tab><sometext2> and so on — IISC_Student
– IISC_Student, Commented Jan 27, 2018 at 4:10

pault · Accepted Answer · 2018-01-27 04:18:53Z

3

Try the following pattern:

import re
pattern = r"(\w+)(\t)(.*)(\b)"

Explanation

answered Jan 27, 2018 at 4:18

pault

43.7k17 gold badges121 silver badges161 bronze badges

Sign up to request clarification or add additional context in comments.

1 Answer 1