One liner to S & R by column

dirtdog has asked for the wisdom of the Perl Monks concerning the following question:

Hi Monks,

I'm trying to create a one liner to search and replace any occurance of the value Y in column 19 with an H excluding the header record.

Below is a sample of the file:

0TESTHLDGS201312230830201312230000001633
125459W540   CUSIPY DIREXION DAILY 20 YEAR PLUS
188554D205   CUSIPY3D SYSTEMS CORPORATION
188579Y101   CUSIPY3M CO.
1000375204   CUSIPYABB LTD
1002824100   CUSIPYABBOTT LABORATORIES
100287Y109   CUSIPYABBVIE INC
[download]

I thought there would be many examples of this on the net, but I couldn't find any and i'm struggling with the solution.

Any help is much appreciated.

Comment on One liner to S & R by column Download Code

Replies are listed 'Best First'.
Re: One liner to S & R by column by davido (Cardinal) on Dec 23, 2013 at 16:47 UTC
What have you tried? `perl -pli.bak -e '$. != 1 and substr($_,18, 1) =~ tr/Y/H/;' filename` Here's how to read it: Run `perl`. Open `filename` for reading. Rename the input file `filename.bak` Open `filename` for output and select it as the default for STDOUT. Form an implicit `while( <> ) { chomp; ... } continue { print "$_$/" }` loop. If `$.` is 1 (first line of the file), skip to the next iteration. On all subsequent iterations `$.` will no longer be one; transliterate the 18th (zero-indexed) column from 'Y' to 'H'. perlrun, perlintro, and perlop should contain 99% of what you need to learn to get a good start on the next one-liner yourself. Update: Added play by play explanation. Update2: Fixed issue where files aren't zero-indexed with respect to `$.` (Thanks Jim,Laurent_R) Dave	[reply] [d/l] [select]
Re^2: One liner to S & R by column by Jim (Curate) on Dec 23, 2013 at 17:54 UTC
If $. is 0 (first line of the file), skip to the next iteration. … On all subsequent iterations $. will no longer be zero… The number of the first line of the input file is 1, not 0. So you need to use an explicit comparison operation. `perl -pli.bak -e 'substr($_, 18, 1) =~ tr/Y/H/ if $. > 1' filename` [download] Here's the same thing using a regular expression. `perl -pli.bak -e 's/^.{18}\KY/H/ if $. > 1' filename` [download] Jim	[reply] [d/l] [select]
Re^3: One liner to S & R by column by davido (Cardinal) on Dec 23, 2013 at 18:01 UTC
Updated (and documented the change). :) Thanks. Dave	[reply]
Re^4: One liner to S & R by column by Laurent_R (Canon) on Dec 23, 2013 at 18:51 UTC
Re^2: One liner to S & R by column by dirtdog (Monk) on Dec 23, 2013 at 17:15 UTC
Thank you Dave. That worked like a charm!	[reply]
Re: One liner to S & R by column by Corion (Patriarch) on Dec 23, 2013 at 16:44 UTC
So, what code have you already written and how did it fail for you?	[reply]
Re^2: One liner to S & R by column by dirtdog (Monk) on Dec 23, 2013 at 17:07 UTC
This is what I have so far, which will work, but i would need multiple one liners to include all security types `perl -p -i -e 's/CUSIPY/CUSIPH/g' <filename>`	[reply] [d/l]
Re^3: One liner to S & R by column by Corion (Patriarch) on Dec 23, 2013 at 17:09 UTC
If you want to play it safe and list all security types, you can use alternation to list them all (see perlre and perlretut: `perl -p -i.bak -e 's/(CUSIP\|CUSIQ\|something else)Y/${1}H'` [download]	[reply] [d/l]


go ahead... be a heretic
	PerlMonks