How can I split a text file line by line into 2 text files by delimited column?

Question

I've tried various combinations of the following:

awk -F" ||| " '{$0=$1}1' source_file.txt > column1.txt
awk -F" ||| " '{$0=$1}2' source_file.txt > column2.txt

or

awk 'BEGIN {FS=" ||| ";}{print $1}' source_file.txt > column1.txt
awk 'BEGIN {FS=" ||| ";}{print $2}' source_file.txt > column2.txt

Instead of the desired output, I either get the entire line (ex. foo bar ||| baz) or I get only the first word (ex. foo).

If you'd like to help, here is a sample text file:

source_file.txt

foo bar ||| baz
qux ||| quux
corge grault ||| garply waldo
fred |||
xyzzy ||| thud

And here's the desired output:

column1.txt

foo bar
qux
corge grault
fred
xyzzy

column2.txt

bar
quux
garply waldo

thud

score 4 · Accepted Answer · answered Feb 22 '17 at 01:22

4

awk -F' \\|\\|\\| ?' '{print $1 > "column1"; print $2 > "column2"}' file

or more generally

awk -F' \\|\\|\\| ?' '{for(i=1;i<=NF;i++) print $i > "column"i}' file

answered Feb 22 '17 at 01:22

karakfa

Thanks. Do you also get this error? `awk: syntax error at source line 1 context is {for(i=1;i<=NF;i++) print $i > >>> "column"i <<< } awk: illegal statement at source line 1` – Ryan Feb 22 '17 at 01:26
1

No errors in my case, make sure you're using single quotes for the script. – karakfa Feb 22 '17 at 01:35

score 0 · Answer 2 · answered Feb 22 '17 at 01:32

0

You could try

cat /tmp/a | tr -s '|' | cut -d'|' -f1 #for part 1

cat /tmp/a | tr -s '|' | cut -d'|' -f2 | sed -E "s/^[[:space:]]+//g" #for part 2

The tr flag squeezes delimiters together.

answered Feb 22 '17 at 01:32

bsd

2 Answers2