Shell Programming and Scripting

Proposal applied to your sample in post#7 (which, btw, doesn't differ from the ones in post#1):

Code:

awk '{printf "%s%s", NF==5?RS:"", $0} END {printf RS}' FS="|" file
41|216|398555|77|provided complete NP outcome data constituted
72|192|402632|580|Completed OHTS Phase 2,through March

I can't see that (and why) it should concatenate everything into one single line...

To remove the surplus spaces as requested in post#1, I added a gsub call:

Code:

awk '{gsub (/  */, " "); printf "%s%s", NF==5?RS:"", $0} END {printf RS}' FS="|" file

The point is to replace the condition /[|]/ that means NF>=2 by the required one, NF==5 or NF>=5

Sorry to all for not using the codes.I used the code tag now.Please let me know still I need to do any changes while posting in the forum.

Thanks RudiC for your command,

I used your commands and working fine but after executing it is deleting few lines in my original file.I took the deleted lines from the original file and kept it in test file and I executed the command but it is not changing any thing I am attaching the data for your reference.

I tried below commands on test file and data is not changing.

Code:

awk '{printf "%s%s", NF==8?RS:"", $0} END {printf RS}' FS="|" test_11.txt > test_12.txt

awk '{gsub (/  */, " "); printf "%s%s", NF==8?RS:"", $0} END {printf RS}' FS="|" test_11.txt > test_12.txt

Thanks,
Narasimhatest_11.txt

That's a different story - why didn't you post representative samples in the first place? Try

Code:

awk '{while (NF < 8) {getline X; $0 = $0 " " X}}1' FS="|" /tmp/test_11.txt

.

BTW, your file contains DOS line terminators (<CR>, \r, 0x0D) which you might want to get rid of before text processing in *nix.

Thanks RudiC,

As per your update I want work on DOS line terminators before processing/using the commands which you posted.Could you please let me know how to to work on this DOS line terminators in Unix.

Thanks,
Narasimha

A word in general: we are a self-help forum! That means: we help you to help yourself, we are not doing your work for you. If you want that: hire someone.

Here is how you do it in sed (note that "^M" is a single character! You enter it i.e. in vi pressing <CTRL>-<V> and then the <ENTER> key):

Code:

sed 's/^M$//' /path/to/file

I hope this helps.

bakunin

Hi bakunin,

Thanks for your commands.

Instead of using Unix commands I have opened the file which is having pipe delimiter in Notepad++
and every line is having CRLF and only LF.
whereever we have single LF we are facing issue.To avoid the issue, I have replaced the individual value LF with CR then i exported the file into Unix server.After that It is working fine.

But I have one file which is having 850 MB (89,18,51,027 bytes) in size and i could not able to open in Notepad++.If we use the Unix command to do the same fix for this file then issue will resolve.

I used command which you have suggested and other commands to but not working as expected.

Code:

sed 's/^M$//' /path/to/file

Could you please let me know is there any way to do so.

Thanks,
Narasimha