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

Hi Monks, I'm struggling with the task of capturing a date value that is populated in column 8 of a pipe delimited file and inserting it into column 11. 

row|1|CUSIP|12345678|RED|||20150125||||
row|1|CUSIP|1234A82H|RED|||20150316||||

[download]

for the 2 examples above the final result should be:

row|1|CUSIP|12345678|RED|||20150125|||20150125|
row|1|CUSIP|1234A82H|RED|||20150316|||20150316|

[download]

I know i need to use memory parens to capture the value in column 8, but can't seem to get it to work 


perl -pi -e 's/\(RED|||([0-9]{8})|||\)/\1/g'

[download]

The above is what i have so far.

Thanks for any suggestions

Replies are listed 'Best First'.
Re: one liner to capture value of field and insert in another field
by choroba (Cardinal) on Jan 28, 2015 at 15:33 UTC 
    perl -i~ -F'\|' -lane 'print join "|", @F[0..9], $F[7], ""'

    [download]

    -F specifies the separator.
    -a populates the @F array with the cell values for each row.
    -l handles the newlines.

    See perlrun for the details.

    لսႽ† ᥲᥒ⚪⟊Ⴙᘓᖇ Ꮅᘓᖇ⎱ Ⴙᥲ𝇋ƙᘓᖇ
[reply]
[d/l]
[select]
Re: one liner to capture value of field and insert in another field
by Anonymous Monk on Jan 28, 2015 at 15:35 UTC
    pipe (|) is special in regular expressions. And you shouldn't use \1 in right hand side of s///. In fact, you shouldn't use s/// at all. split is what you need. 
    $ perl -F'\|' -lane '$F[10] = $F[7];
                     print join "|", @F' file.txt

row|1|CUSIP|12345678|RED|||20150125|||20150125|
row|1|CUSIP|1234A82H|RED|||20150316|||20150316|

    [download]
    (see perlrun for details)
[reply]
[d/l]
[select]

      Works like a charm. Thanks you!

[reply]
      (choroba, great minds think alike! :)
[reply]

      I guess i jumped the gun a little bit. My apologies. It works for the sample i provided, but the real file has almost 300 fields in each record..some of those values are populated and some aren't...so simply joining a "|" at the end doesn't work for the actual file.

      is there a way to do what you did but instead of Joining one pipe at the end can we somehow join the remaining ~300 col values to each record? 

      row|1|CUSIP|35472T200|ARDIV|||20150112||||||||||||||||||||||||||||||||
+|||||||||||||||||||||20150321||||||G||||||||||||||||||||FR|||||||||||
+||||||||||||||||||||||||||||||||||||X||||||||||||||||||||||||||||||||
+|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
+|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
+||||||||||||||||||||||||||||||||||||||||LIQ||||||||||||||||||||||||||
+|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
+||||||||||||

      [download]

      Sorry for the incomplete data and thx again.

[reply]
[d/l]
        The problem with split is that by default, trailing empty fields are discarded. You can specify a negative count to prevent that, but you need to be more explicit in such a case: 
         perl -lne '@F = split /\|/, $_, -1; $F[10] = $F[7] ; print join "|", 
+@F'

        [download]
        لսႽ† ᥲᥒ⚪⟊Ⴙᘓᖇ Ꮅᘓᖇ⎱ Ⴙᥲ𝇋ƙᘓᖇ
[reply]
[d/l]
        You mean the records are spread out across several lines? And what separates them?
[reply]

Back to Seekers of Perl Wisdom