list 3rd column of a file with spaces only - awk

for listing 3rd column I am using
awk '{print $3}' inputfile.txt
and its output looks like
abc
xyz
lmn
pqr
But I need output like
abc xyz lmn pqr
How can I get this?


This might work for you (GNU sed):
sed -r 's/((\S*)\s){3}.*/\2/;1h;1!H;$!d;x;y/\n/ /' file
or more easily:
cut -d\ -f3 file | paste -sd\


print will always append a newline (actually, it will use ORS value). If you want more control, you can use printf:
awk '{printf "%s ", $3}'
This will also print an extra space character at the end, but for most use-cases this extra space is harmless.


Transliterate linefeeds into spaces
... | tr '\n' ' '


Use the awk Output Record Separator variable.
awk -v ORS=' ' '{print $3}' inputfile.txt


Avoiding adding a space to the beginning or end of the line:
awk '{printf "%s%s", fs, $3; fs=FS} END{print ""}' file

Related

Find the second word delimited by space or comma then insert strings before and after

I have a file containing TABLE schema.table and want to put strings around it to make a command like MARK string REJECT
the file contains many lines
TABLE SCHEMA.MYTAB, etc. etc....
or
TABLE SCHEMA.MYTAB , etc. etc....
The result is
MARK SCHEMA.MYTAB REJECT
..etc
I have
grep TABLE dirx/myfile.txt | awk -F, '{print $1}' | awk '{print $2}' | sed -e 's/^/MARK /' |sed -e 's/$/ REJECT/'
It works, but can this be tidier? I think I can combine the awk and sed into single commands but not sure how.

Maybe:
awk '/^TABLE/ {gsub(/,.*$/, ""); print "MARK " $2 " REJECT"}' dirx/myfile.txt

Replace end of line with comma and put parenthesis in sed/awk

I am trying to process the contents of a file from this format:
this1,EUR
that2,USD
other3,GBP
to this format:
this1(EUR),that2(USD),other3(GBP)
The result should be a single line.
As of now I have come up with this circuit of commands that works fine:
cat myfile | sed -e 's/,/\(/g' | sed -e 's/$/\)/g' | tr '\n' , | awk '{print substr($0, 0, length($0)- 1)}'
Is there a simpler way to do the same by just an awk command?

Another awk:
$ awk -F, '{ printf "%s%s(%s)", c, $1, $2; c = ","} END { print ""}' file
1(EUR),2(USD),3(GBP)

Following awk may help you on same.
awk -F, '{val=val?val OFS $1"("$2")":$1"("$2")"} END{print val}' OFS=, Input_file

Toying around with separators and gsub:
$ awk 'BEGIN{RS="";ORS=")\n"}{gsub(/,/,"(");gsub(/\n/,"),")}1' file
this1(EUR),that2(USD),other3(GBP)
Explained:
$ awk '
BEGIN {
RS="" # record ends in an empty line, not newline
ORS=")\n" # the last )
}
{
gsub(/,/,"(") # replace commas with (
gsub(/\n/,"),") # and newlines with ),
}1' file # output

Using paste+sed
$ # paste -s will combine all input lines to single line
$ seq 3 | paste -sd,
1,2,3
$ paste -sd, ip.txt
this1,EUR,that2,USD,other3,GBP
$ # post processing to get desired format
$ paste -sd, ip.txt | sed -E 's/,([^,]*)(,?)/(\1)\2/g'
this1(EUR),that2(USD),other3(GBP)

Print 1st word of 1st column and 2nd word of 2nd column in a single line

I would like to print first word of first column and second word of second column in to a single column
Example:
ABC 123
BCD 987
output:
ABC 987

Sweet and simple awk may help you in same.
awk 'FNR==1{val=$1;next} FNR==2{print val,$2}' Input_file

awk 'FNR==1{printf $1 OFS;next}FNR==2{print $2;exit}' infile

this will extract the diagonal for arbitrary size table
$ awk '{print $NR}' file | paste -sd' ' -

With sed
sed 'N;s/\([^ ]*\)\(.*\n[^ ]*\)\(.*\)/\1\3/' infile
Or
sed -E 'N;s/([^ ]+)(.*\n[^ ]+)(.*)/\1\3/' infile

using awk
$ awk '{printf $NR (NR==2?ORS:" ") }' file
ABC 987

This might work for you (GNU sed):
sed 'N;s/ .* / /' file
Append the second line to the first and replace everything between the first two spaces with another space.

AWK: Apply filter only if field separator is present

I surprisingly found that when you do this:
echo "hello" | awk -F'|' '{print $1;}'
you get:
hello
How to return nothing given that the field separator '|' is absent in the line ?
I do this to extract dates in beginning of log lines, but some lines don't start with a date and then give me this problem. Thanks, I am quite new in awk.

You can do this
echo "hello" | awk -F'|' 'NF>1 {print $1}'
echo "hello|1" | awk -F'|' 'NF>1 {print $1}'
hello
Only when you have more than one field, return the first field
On a file
cat testing
record1|val1
record2|val2
record3
record4|val4
awk -F'|' 'NF>1 {print $1}' testing
record1
record2
record4

Alternatively, you could use
awk -F'|' '$1==$0'
If no separator is present, then field one will contain the whole line.

awk command to change field seperator from tilde to tab

I want to replace the delimter tilde into tab space in awk command, I have mentioned below how I would have expect.
input
~1~2~3~
Output
1 2 3
this wont work for me
awk -F"~" '{ OFS ="\t"; print }' inputfile

It's really a job for tr:
tr '~' '\t'
but in awk you just need to force the record to be recompiled by assigning one of the fields to its own value:
awk -F'~' -v OFS='\t' '{$1=$1}1'

awk NF=NF FS='~' OFS='\t'
Result
1 2 3

Code for sed:
$echo ~1~2~3~|sed 'y/~/\t/'
1 2 3