LinuxQuestions.org - Need help parsing text file

- Programming (https://www.linuxquestions.org/questions/programming-9/)

- - Need help parsing text file (https://www.linuxquestions.org/questions/programming-9/need-help-parsing-text-file-260650/)

Need help parsing text file

Hi,

I'm writing a script that parses test files and in some cases, combines several lines into one. For example, the lines:

TTITLE1=There has go
TTITLE1=t to be a better way.

Would be output as "There has got to be a better way"

This is fine. The problem I have is if the last character is a space. For example, if I have the lines:

TTITLE2=You shook me all night
TTITLE2=long.

Even though there is a space after "night" in the file, my output comes out as:

You shook me all nightlong.

QUESTION:
How do I parse a text file such that spaces at the end of lines are recognized and retained???

Thanks in advance!

-Steve

probably just a case of wrapping your variable in quote marks, but as you've not even gievn us the code you're working on...

Putting quotes around the text should work:
title="a string ending with a space "

My apologies for not being more clear in my initial question. In my first example, imagine a text file containing:

TTITLE1=There has go
TTITLE1=t to be a better way.

I want my script to parse this as "There has got to be a better way."

In my second example, imagine my text file contains:

TTITLE2=You shook me all night
TTITLE2=long.

Here, I'd want to parse it as "You shook me all night long."

The problem is that, depending on how I parse the file, I either end up with:

There has go t to be a better way. #unwanted space in "got"
You shook me all night long.

or

There has got to be a better way.
You shook me allnight long. # No space between "all" and "night"

Regardless of whether I use cut, sed, or awk, lines that end with a space aren't beiing handled correctly. Here's the script I'm using, where $1 is the file being read:

#!/bin/bash
#
titlenum=0

while [ -n "$(grep "TTITLE$titlenum=" $1)" ]
do
tan=$(grep "TTITLE$titlenum=" $1)
title=$(echo ${tan//"TTITLE$titlenum="/} | awk '{print $0}')
echo $title >> trackfile.txt
let titlenum=titlenum+1
done

In this case, I get the following:

There has got to be a better way.
You shook me allnight long. # No space between "all" and "night"

How do I modify my script so that it properly wraps text into a single line regardless of whether or not there's a space at the end of a line?

Code:



/ibrc/users/tba17/daily17/soumen/tmp>cat abc |tr " " "*"

TTITLE1=There*has*go

TTITLE1=t*to*be*a*better*way.



TTITLE2=You*shook*me*all*night*

TTITLE2=long.



TTITLE3=hello,

TTITLE3=*

TTITLE3=how*are*you?*



/ibrc/users/tba17/daily17/soumen/tmp>cat b

#!/usr/bin/bash



fname=$1

tstr=TTITLE

ofile=trackfile.txt



> $ofile

for srl in $(cut -d"=" -f1 $fname | grep -v "^$" | sed -e 's/'$tstr'//g' |sort -u -n) ; do

        curstr=${tstr}${srl}

        echo "Processing $curstr ..."



        grep "^${curstr}=" $fname | \

        awk -F"=" '{arr[$1]=arr[$1]""$2} END{ for (i in arr) {printf("%s\n",arr[i])}}' | \

        grep -v "^$" >> $ofile

done

/ibrc/users/tba17/daily17/soumen/tmp>b abc

Processing TTITLE1 ...

Processing TTITLE2 ...

Processing TTITLE3 ...

/ibrc/users/tba17/daily17/soumen/tmp>cat trackfile.txt

There has got to be a better way.

You shook me all night long.

hello, how are you?

/ibrc/users/tba17/daily17/soumen/tmp>

The spaces in the input file [abc here], have been shown with "*". The script works even if the line is :

1. Ending/starting with space
2. Has only space(s)
3. Has nothing after "=" e.g. TTITLE1=

HTH.

Just what I needed. Thank you so much!

-Steve