Shell Programming and Scripting

Hi,

I have a flat text file consisting of rows (each field separated by '|') from a table. e.g;

NSW|Gulliver Travels|236||5000|BW

This has to be converted to the following format

NSW "Gulliver Travels" 236 5000 BW

No data field has to be left as a blank character so that we have three blank characters in effect between the two consecutive populated fields in the above scenario.

I have actually other flat files also with the same requirement. The no. of fields are different in the other files. So I am in need of a unix script to do this which can be applied directly to all the flat files irrespective of the no. of columns or fields. I have been trying with awk and I accomplished only the following which is not desirable.

NSW Gulliver Travels 236 5000 BW

Also I used hard-coding for this.
Any pointers to solve this are greatly appreciated and would help me a lot.

Maybe something like this:

Code:

$
$ cat data.txt
NSW|Gulliver Travels|236||5000|BW
PQR|Robinson Crusoe|999|123||XY
XYZ|The Merchant of Venice|999 abc|123||XY ZZ
XYZ LMN|Alice in Wonderland|xyz 777|123|345|PQ RS ZZ
$
$ perl -ne '{$l=""; chomp; @x=split/\|/; foreach $i (@x) { if ($i =~ / /) {$l = $l."\"".$i."\" "; } elsif ($i ne "") {$l = $l.$i." ";}}print substr($l,0,-1)."\n"}' data.txt
NSW "Gulliver Travels" 236 5000 BW
PQR "Robinson Crusoe" 999 123 XY
XYZ "The Merchant of Venice" "999 abc" 123 "XY ZZ"
"XYZ LMN" "Alice in Wonderland" "xyz 777" 123 345 "PQ RS ZZ"
$
$

HTH,
tyler_durden

Here is one method when you have a known set of columns. You can use any variable names you like for the 'read' command, I just picked some as if it was a library card file.

Code:

IFS="|"
while read type title shelf row section author; do
  echo "$type \"$title\" $shelf $row $section $author" | sed 's/  */ /g'
done < pipe_delimited_file

hi tyler, thanks a lot for your quick reply. The script works well but the only problem is that it doesn't take into account the blank fields. These blank fields are ignored in the output and the immediate populated field gets displayed. For e.g;

Input

D BRUX EXCC CC||JLCSVD|1|VIC|
T BGNW BSAT SA|||7|VIC|
T PNTO SCTC E0|NGS PK B/B|JLCNSD||NSW|
P GLSE P4BR Z5||||NSW|
P GLSE P4BR Z5|PK B/B NON AUG|JLCQLD|30||
D DANB EXCC CC|OTHER|JLCSVD|1|VIC|
T PNTO SPTG E0|NGS PK B/B|JLCNSD|3|NSW|

Output

"D BRUX EXCC CC" JLCSVD 1 VIC
"T BGNW BSAT SA" 7 VIC
"T PNTO SCTC E0" "NGS PK B/B" JLCNSD NSW
"P GLSE P4BR Z5" NSW
"P GLSE P4BR Z5" "PK B/B NON AUG" JLCQLD 30
"D DANB EXCC CC" OTHER JLCSVD 1 VIC
"T PNTO SPTG E0" "NGS PK B/B" JLCNSD 3 NSW

Ideal output should have been

"D BRUX EXCC CC" JLCSVD 1 VIC (3 blank characters)
"T BGNW BSAT SA" 7 VIC (6 blank characters)
"T PNTO SCTC E0" "NGS PK B/B" JLCNSD NSW (3 blanks again)
"P GLSE P4BR Z5" NSW (9 blank characters)
"P GLSE P4BR Z5" "PK B/B NON AUG" JLCQLD 30
"D DANB EXCC CC" OTHER JLCSVD 1 VIC
"T PNTO SPTG E0" "NGS PK B/B" JLCNSD 3 NSW

Is there a way to address this also in your script. Thanks again for ur inputs.

ldapswandog- thanks to u as well for ur inputs.

Code:

awk -F"|" 'BEGIN{dq="\042";}
{ 
  for(i=1;i<=NF;i++){
   if ($i ~ / /){
    $i = dq $i dq
   }
  }
}'  file

try this in awk:

$ awk -F "|" '{
for (i=1;i<=NF;i++) {
if ( $i == "") printf (" ");
if (split($i,t," ") > 1)printf(" \"%s\"",$i)
else printf(" %s", $i)
}printf "\n"
}' filename


cheers,
Devaraj Takhellambam

thanks a many Devaraj it works. u made my evening.
cheers.