1

I have a file delimited with Cedilla, but the records are in a single line. I have to convert this to a multiline file.

Sample record:

P002365Ç1200ÇMastercardÇcarolinaÇBasildonÇEnglandÇUnited kingdomÇP002368Ç2100ÇVisaÇGouyaÇEchucaÇVictoriaÇAustraliaÇP002373Ç3600ÇMastercardÇRenee ElisabethÇTel AvivÇTel AvivÇIsraelÇP002382Ç6300ÇDinersÇbarbaraÇHyderabadÇAndhra PradeshÇIndia

Need to be converted to:

P002365Ç1200ÇMastercardÇcarolinaÇBasildonÇEnglandÇUnited Kingdom    
P002368Ç2100ÇVisaÇGouyaÇEchucaÇVictoriaÇAustralia    
P002373Ç3600ÇMastercardÇRenee ElisabethÇTel AvivÇTel AvivÇIsrael    
P002382Ç6300ÇDinersÇbarbaraÇHyderabadÇAndhra PradeshÇIndia

Can we achieve this using awk command?

Tom Fenech
  • 72,334
  • 12
  • 107
  • 141
code_baby
  • 103
  • 1
  • 1
  • 4
  • Would this be what you are looking for? http://stackoverflow.com/questions/15559979/split-file-on-nth-occurrence-of-delimiter HTHs – YFP Oct 12 '14 at 11:47
  • 1
    If you can use not only awk then this should work for you: perl -pe 's|Ç|\n|g' file.txt | paste -dÇ - - - - - - - – Slava Semushin Oct 12 '14 at 11:55

2 Answers2

4

You could use something like this:

awk -FÇ '{for (i=1;i<=NF;++i) printf "%s%s", $i, (i%7==0?RS:FS)}' file
P002365Ç1200ÇMastercardÇcarolinaÇBasildonÇEnglandÇUnited kingdom
P002368Ç2100ÇVisaÇGouyaÇEchucaÇVictoriaÇAustralia
P002373Ç3600ÇMastercardÇRenee ElisabethÇTel AvivÇTel AvivÇIsrael
P002382Ç6300ÇDinersÇbarbaraÇHyderabadÇAndhra PradeshÇIndia

A breakdown of what's going on here:

  • -FÇ - This command line argument sets the FS variable (Field Separator) to the Ç character
  • The for loop (executed on each line) loops from 1 to NF (Number [of] Fields)
  • In each iteration, a printf is executed that prints two strings (%s%s), the first being the content of the actual field ($i) and the second being one of two options:
    1. If the Field Number is a multiple of 7*, a newline character is inserted (RS Record Separator),
    2. Otherwise the FS is printed. (Defined as the Ç character).

* The number 7 is used "arbitrarily" because it was your definition for splitting the records according to the example output you supplied.

Tom Fenech
  • 72,334
  • 12
  • 107
  • 141
  • `RS` (Record Separator) is the newline character in this case? – Lix Oct 12 '14 at 11:58
  • @Lix by default, yes it is. – Tom Fenech Oct 12 '14 at 12:01
  • 1
    As I suspected :P Cool.. I'm going to add a breakdown of what's going on here (mostly to make sure I understand it myself). Please let me know if I get something wrong :) – Lix Oct 12 '14 at 12:02
  • @Lix I just edited, hopefully that explains things a little more. Feel free to edit if you have any more detail to add. – Tom Fenech Oct 12 '14 at 12:05
  • Thanks, @tom. I've broken the explanation out into bullet points. Can you please confirm that I'm not misleading anyone or over simplifying :) – Lix Oct 12 '14 at 12:10
  • Looks good to me, I've removed my explanation as it was basically a less detailed version of yours! – Tom Fenech Oct 12 '14 at 12:12
  • 1
    Fantastic :) I should really follow the awk tag more... Really helps me to brush up on one of my favorite tools... – Lix Oct 12 '14 at 12:14
0

This might work for you (GNU sed):

sed 's/Ç/\n/7;P;D' file

This replaces every 7th Ç with a newline.

potong
  • 55,640
  • 6
  • 51
  • 83