[SOLVED] AWK Adding a Period at the end of a line

metallica1973 · 11-21-2012, 09:43 AM

I started venturing in learning the art of using AWK/GAWK and wanted to simply added a period from line #11 to line #28 or to the end of the file if there is data. So for example:

Code:

11 	Centos.NM                         
12	dojo1                 
13	redhat.5.5.32Bit       
14	redhat.6.2.64Bit       
15	mandriva.9.2.32Bit     
16	RELEASE_WIN2003       
17	SaintBox-Ubuntu-x64     
18	SAINTexploitVM-x86      
19	SAINTVM-x64             
20	Ubuntu-12.04.x64        
21	bigbadwolf.x64          
22	WIN2003PATCHED          
23	WIN2003UNPATCH          
24	WIN-IQF3U12CJA5         
25	XPSP3PATCHED            
26	XPPROUNPATCHED          
27	WIN2012                 
28	WIN8_141

to

Code:

11 	Centos.NM.                         
12	dojo1.                 
13	redhat.5.5.32Bit.       
14	redhat.6.2.64Bit.       
15	mandriva.9.2.32Bit.     
16	RELEASE_WIN2003.       
17	SaintBox-Ubuntu-x64.     
18	SAINTexploitVM-x86.      
19	SAINTVM-x64.             
20	Ubuntu-12.04.x64.        
21	bigbadwolf.x64.          
22	WIN2003PATCHED.          
23	WIN2003UNPATCH.          
24	WIN-IQF3U12CJA5.         
25	XPSP3PATCHED.            
26	XPPROUNPATCHED.          
27	WIN2012.                 
28	WIN8_141.

This is what I had so far:

Code:

awk '{ NR==11,NR==28 print $0 "." }' < test_file 
awk: { NR==11,NR==28 print $0 "." }
awk:         ^ syntax error
awk: { NR==11,NR==28 print $0 "." }
awk:                 ^ syntax error

??

pan64 · 11-21-2012, 09:49 AM

this should do the job:
awk ' NR>=11 && NR <= 28 { print $0 "." } ' test_file

metallica1973 · 11-21-2012, 09:55 AM

Simply awesome. Many thanks

danielbmartin · 11-21-2012, 10:45 AM

Code:

awk ' NR>=11 && NR <= 28 { print $0 "." } ' test_file

Nitpick #1) This solution doesn't deal with trailing white space.
Nitpick #2) This solution discards lines 01-10 and 29-up.

Daniel B. Martin

colucix · 11-21-2012, 12:04 PM

Quote:

Originally Posted by danielbmartin

Nitpick #1) This solution doesn't deal with trailing white space.

You sure? Until the record is not split into fields, the record should be printed as it is, with trailing blank spaces (if any). Example:

Code:

$ echo "    Donald Duck" | awk '{print $0 "."}'
    Donald Duck.
$ echo "  Donald Duck" | awk '$1 = $1{print $0 "."}'
Donald Duck.

In the second example, the first field is re-assigned to itself and the record is re-built to take in account the changes. In this case the trailing blank spaces are lost.

pixellany · 11-21-2012, 12:21 PM

sed '11,28s/$/\./' filename > newfilename

danielbmartin · 11-21-2012, 12:21 PM

Quote:

Originally Posted by colucix

You sure? Until the record is not split into fields, the record should be printed as it is, with trailing blank spaces (if any).

I learn by reading posts and follow up by testing proposed solutions. In this instance I created an input file using Copy-and-Paste. This resulted in lines which had trailing blanks. OP gave an example of the desired file, and it had the added periods appended to the right-most printed character.

Daniel B. Martin

danielbmartin · 11-21-2012, 12:26 PM

Quote:

Originally Posted by pixellany

sed '11,28s/$/\./' filename > newfilename

Short and sweet! Love it!

Daniel B. Martin

colucix · 11-21-2012, 02:36 PM

Quote:

Originally Posted by danielbmartin

I learn by reading posts and follow up by testing proposed solutions. In this instance I created an input file using Copy-and-Paste. This resulted in lines which had trailing blanks. OP gave an example of the desired file, and it had the added periods appended to the right-most printed character.

Daniel B. Martin

Sorry Daniel, reading your post I realized I confused trailing blank spaces with leading spaces. You're absolutely right: the suggested solutions don't deal with trailing blank spaces. My mis-understanding due to the fact that English is not my mother language!

colucix · 11-21-2012, 02:41 PM

Here is an enhanced sed solution:

Code:

sed '11,$s/[[:blank:]]*$/./'

and a similar solution in awk:

Code:

awk 'NR >= 11 {gsub(/[[:blank:]]*$/,".")} 1' file

grail · 11-22-2012, 09:45 AM

Here's an awk with a twist:

Code:

awk '(NR >10 && NR < 29 && ORS=".\n") || ORS=RT' RS=' *\n' file

And if we are not to assume to much, ie maybe the user means at the end of text but to keep the rest of the line in tact:

Code:

awk '(NR >10 && NR < 29 && ORS="."RT) || ORS=RT' RS=' *\n' file

metallica1973 · 11-26-2012, 02:05 PM

Super!!