How can I extract a substring from multiple lines from a file in python

Question

I have a file that contains thousands of lines, and in the file, there are some lines like:

Line 115463: 08:59:25.106   08:59:24.992877 ASDF_IIS_CFGDB  GenMod  Ta-SNS__GENERATED_EVENTS (Event2f, DIR = 13)    rrc_UlUtranMsg.c (../../../HEDGE/UL3/ASDF/UtranMsg/Uplink/Code/Src) 987 
Line 236362: 08:59:28.647   08:59:28.597827 ASDF_IIS_CFGDB  GenMod  Ta-SNS__GENERATED_EVENTS (Eventab, DIR = 1) rrc_UlUtranMsg.c (../../../HEDGE/UL3/ASDF/UtranMsg/Uplink/Code/Src) 934 
Line 324964: 08:59:40.456   08:59:40.403644 ASDF_IIS_CFGDB  GenMod  Ta-SNS__GENERATED_EVENTS (Eventac, DIR = 1) rrc_UlUtranMsg.c (../../../HEDGE/UL3/ASDF/UtranMsg/Uplink/Code/Src) 934 
Line 341172: 08:59:40.659   08:59:40.616565 ASDF_IIS_CFGDB  GenMod  Ta-SNS__GENERATED_EVENTS (Eventfb, DIR = 13)    rrc_UlUtranMsg.c (../../../HEDGE/UL3/ASDF/UtranMsg/Uplink/Code/Src) 987 
Line 373186: 08:59:41.174   08:59:41.104755 ASDF_IIS_CFGDB  GenMod  Ta-SNS__GENERATED_EVENTS (Event2f, DIR = 1) rrc_UlUtranMsg.c (../../../HEDGE/UL3/ASDF/UtranMsg/Uplink/Code/Src) 934 
Line 480217: 08:59:44.481   08:59:44.389453 ASDF_IIS_CFGDB  GenMod  Ta-SNS__GENERATED_EVENTS (Eventx1, DIR = 1) rrc_UlUtranMsg.c (../../../HEDGE/UL3/ASDF/UtranMsg/Uplink/Code/Src) 934 
Line 505424: 08:59:44.777   08:59:44.701709 ASDF_IIS_CFGDB  GenMod  Ta-SNS__GENERATED_EVENTS (Event1a, DIR = 1) rrc_UlUtranMsg.c (../../../HEDGE/UL3/ASDF/UtranMsg/Uplink/Code/Src) 934 

I only need to extract the substring

'1a'

from

'SNS__GENERATED_EVENTS (Event1a, DIR = 1)'

and so on. So, basically, the two characters after '(Event'

And I need to store these in a list or somewhere else where I can use them.

How can I do this?

So far, I have tried the following code but it gives me some values mixed in:

events = []

for line in input_txt_file:
        if "Ta-SNS__GENERATED_EVENTS " not in line: continue
        parts = line.split('Event')
        event_temp = [0]
        for i,part in enumerate(parts):
            if part.endswith("Ta-SNS__GENERATED_EVENTS ("): event_temp[0] = parts[i+1].split(None,1)[0].split(',',2)[0]
            events.append(event_temp)

print events

The output I am getting is:

[[0], [0], ['2f'], ['2f'], ['ab'], ['ab'], [0], [0], ['ac'], ['ac'], ['fb'], .......]

Answer 1

If the line position is always fixed, Wayne's answer is the most efficient. If the position can vary a bit, this is a decent situation in which to use regex:

import re
events = []

for line in input_txt_file:
    match = re.search(r'SNS__GENERATED_EVENTS.*?Event(..)', line)
    if match:
        events.append(match.group(1))

This searches each line for SNS__GENERATED_EVENTS, followed by possibly some characters, followed by Event and then two more characters, and grabs those two characters.

Answer 2

Short solution using re.findall() function:

# change to your actual file path
with open('./text_files/events.txt', 'r') as fh:
    l = re.findall(r'(?<=Ta-SNS__GENERATED_EVENTS \(Event)\w+', fh.read(), re.M)

print(l)

The output:

['2f', 'ab', 'ac', 'fb', '2f', 'x1', '1a']

Answer 3

If you know that it's always going to be in that position you can simply do:

hexes = [line[99:101] for line in file]

If there are lines that don't contain that text you can do:

hexes = [line[99:101] for line in file if 'Ta-SNS__GENERATED_EVENTS' in line]

Answer 4

I would do this as a substring search using the re module, personally.

import re

for line in input_txt_file:
    val = ''
    val = re.search('SNS__GENERATED_EVENTS \(Event(.+?), DIR\)', line).group(1)
    print(val)

Answer 5

No need for regex here: just split according to Ta-SNS__GENERATED_EVENTS (Event and take 2 first letters of the 2nd field if there's one:

events=[]
for line in input_txt_file:
    toks = line.split("Ta-SNS__GENERATED_EVENTS (Event")
    if len(toks)>1:
        events.append(toks[1][:2])

EDIT: found a cool one-liner equivalent:

events=[tok[:2] for line in input_txt_file for i,tok in enumerate(line.split("Ta-SNS__GENERATED_EVENTS (Event")) if i==1]

Uses enumerate and tests if the index of the splitted item is 1: means there are at least 2 items. In that case, take 2 first chars from the token.

EDIT2: Amber has even better using partition to avoid the enumerate hack:

events=[t for t in (l.partition("Ta-SNS__GENERATED_EVENTS (Event")[2] for l in input_txt_file) if t]