Move text in a unix flat file

JoHarris
JoHarris used Ask the Experts™
on
I've got a file of IDs that basically looks as follows :

123456789AAAAA
123456790AAAAA
123456791AAAAA

and I want to edit that file (into another file) and swap the data elements around so that it looks like :

AAAAA123456789
AAAAA123456790
AAAAA123456791

(ie 5 char code followed by 9 char code)

How can I do that in one statement?  I managed to use cut so that that I could get the 9 chars :

cat file_in | cut -c0-9 > file_out

but that's not enough.  Do I need a sed statement?
Comment
Watch Question

Do more with

Expert Office
EXPERT OFFICE® is a registered trademark of EXPERTS EXCHANGE®
you don't even need the cut, a simple "sed" will do it...


sed 's/^\(.....\)\(.........\)/\2\1/' file_in > file_out

Note this is hardcoded to 5 leading characters, and 9 trailing characters...

That will transpose the first 14 characters, leaving anything else in the line in-place. If this is not desirable then extend the regex:

sed  's/^\(.....\)\(.........\).*$/\2\1/'


Also, you may want to search for matches that are strictly characters ([A-Z]) or strictly numerics ([0-9]) rather than any 5/9 characters.
Also, using bounds may be an easier way of argumenting the number of characters...

so...

using bounds, and strictly characters/numbers, stripping any suplurfous characters after first 14:

sed 's/^\([A-Z]\{5\}\)\([0-9]\{9\}\).*$/\2\1/' in_file > out_file


e.g.

$ echo ABCDE123456789extra | sed 's/^\([A-Z]\{5\}\)\([0-9]\{9\}\).*$/\2\1/'
123456789ABCDE

Author

Commented:
Thats great thanks - the flat file will always contain the 9 followed by the 5 so the 1st one works a treat.

Thanks for helping me out so quickly!
:-)

Do more with

Expert Office
Submit tech questions to Ask the Experts™ at any time to receive solutions, advice, and new ideas from leading industry professionals.

Start 7-Day Free Trial