How to insert "~" at specific locations (multiple times in a single line) in .txt file

jigneshpatel87 · 08-22-2017, 07:54 AM

Hi All,

I am new on Linux.

My requirements is I have to insert "~" symbol at many specific location in single line in a .txt file.

I have to insert "~" at flowing locations.

first "~" at after 10 character
second "~" at after 4 character
third "~" at after 6 character
forth "~" at after 3 character
fifth "~" at after 7 character

Input .txt file as below:
SDBPRSNL100390ANNUALSDBPAYMENT

Output in .txt file should be as below:
SDBPRSNL10~0390~ANNUAL~SDB~PAYMENT

Thanks in advance and appreciate your help.

syg00 · 08-22-2017, 08:02 AM

Homework ?. LQ has rules about that, which you should have read.

rtmistler · 08-22-2017, 08:05 AM

Hi jigneshpatel87 and welcome to LQ.

Please review the guidelines for how to prepare your question and ask it more effectively in the LQ FAQ

Note that you have presented your question with highly specific things to do, however you have given no background about what attempts you have done to solve this problem, what language you intend to solve this problem with, or what experience you have with scripting or languages. You also have not shown any representative data from your input file.

People here will be happy to help you, however you should show some effort and also note that if this is an assignment, it is not bad that it is, however please be considerate enough to note that it is, as well as to not post any assignments verbatim, per the LQ Rules

Turbocapitalist · 08-22-2017, 08:05 AM

Welcome.

I would look at sed, awk, or perl for that. The s/// command in sed is a good way to get started with that:

Code:

sed -e 's///; s///; s///; s///; s///;' < input.txt > output.txt

For an authoritative reference see

Code:

man sed
man 7 regex

For a tutorial or guide, try these:

jigneshpatel87 · 08-22-2017, 08:20 AM

Quote:

Originally Posted by rtmistler

Hi jigneshpatel87 and welcome to LQ.

Please review the guidelines for how to prepare your question and ask it more effectively in the LQ FAQ

Note that you have presented your question with highly specific things to do, however you have given no background about what attempts you have done to solve this problem, what language you intend to solve this problem with, or what experience you have with scripting or languages. You also have not shown any representative data from your input file.

People here will be happy to help you, however you should show some effort and also note that if this is an assignment, it is not bad that it is, however please be considerate enough to note that it is, as well as to not post any assignments verbatim, per the LQ Rules

Thank you very much. I will make sure next time when I post any thread.

AwesomeMachine · 08-23-2017, 01:42 AM

I'll give a hint. You're trying to make a delimited file with "~" as the delimiting character.

MadeInGermany · 08-23-2017, 04:21 AM

Multiple s commands have the principal complication that the insertion changes the positions for the following insertions

Code:

s///; s///; s///; s///;

Of course you can solve it with some calculations and adapted positions.
But it is easier to have one big s command, with multiple \( \) groups and \n back references. A \{m\} repetition of the "any character" directly reflects the given relative positions.
Example: 4 times any character in group 1 and 3 times any character in group 2, in the substitution print group 1 plus a comma plus group 2

Code:

s/\(.\{4\}\)\(.\{3\}\)/\1,\2/;

Turbocapitalist · 08-23-2017, 06:31 AM

Quote:

Originally Posted by MadeInGermany

But it is easier to have one big s command, . . .

Quite right. Now that I think about it another time, that seems to be the way to go.

Sefyir · 08-23-2017, 10:47 AM

By adding to the string, you change the structure of it so that it needs to be offset (as noted)
However, if you make the changes in reverse order, this is no longer true.
I changed the locations from the given "relative" values to more absolute ones. Adding or changing where they go should be more trivial now.

Code:

#locations = [10, 4, 6, 3, 7]      
locations = [10, 14, 20, 23, 30]      
locations.sort(reverse=True)
dlm = '~'                 
with open('delimited_file') as _f:
    reader = _f.readlines()   
    for row in reader:            
        row = list(row)
        for location in locations:
            row.insert(location, dlm)
        print(''.join(row))

EDIT:

notebook preview of code output and refactor using a function for inserting the delimiter.
https://gist.github.com/anonymous/ca...ed379fd1211395

Code:

#!/usr/bin/env python3
def insert_delimiter(string, locations, delimiter='|'):
    string = list(string)
    locations = sorted(list(locations), reverse=True)
    for location in locations:
        string.insert(location, delimiter)
    string = ''.join(string)
    string = string.strip('\n {}'.format(delimiter))
    return string


locations = [10, 14, 20, 23, 30]
with open('delimited_file') as _f:
    for row in _f.readlines():
        delimited_row = insert_delimiter(row, locations=locations, delimiter='~')
        print(delimited_row)