awk and change file content

Ayman.mashal · 05-30-2005, 04:17 AM

Hi all

i have text files with the below format :

VERSION--->41
AAA-203 04150 060 120 3000 99 99
0010350000 00000054 0303 0843 011454 F033 0844 011454 000000
0100000000 00000000 0000 0000 000000 0000 0000 000000 000000
3148210035 13830131 4821 0035 138302 7000 0000 084317 084408
0002160000 00000385 0303 0855 011454 F114 0913 011465 000549

i am trying to use AWK to do the following :

1- delete the VERSION--->41 Line
2- move the AAA-* line (the seconed line) to the end of each line in the file
and delete it .

i want the result to be :
0010350000 00000054 0303 0843 011454 F033 0844 011454 000000 AAA-203 04150 060 120 3000 99 99
0100000000 00000000 0000 0000 000000 0000 0000 000000 000000 AAA-203 04150 060 120 3000 99 99
3148210035 13830131 4821 0035 138302 7000 0000 084317 084408 AAA-203 04150 060 120 3000 99 99
0002160000 00000385 0303 0855 011454 F114 0913 011465 000549 AAA-203 04150 060 120 3000 99 99

thanks

slackie1000 · 05-30-2005, 04:32 AM

hi there,

Quote:

Originally posted by Ayman.mashal

i am trying to use AWK to do the following : ...

please post what you tried so far..
"sed" will be a best option for this problem..
regards
slackie1000

eddiebaby1023 · 06-01-2005, 08:59 AM

I'd use a script that read in each line. Ignore the first and save the AAA line in a variable, then echo each line with the variable appended. Too simple to warrant the complications of sed, really. If the requirement gets more complicated then sed would definitely be the way to go, though; awk can be a bit clunky (IMO).

Ayman.mashal · 06-02-2005, 01:25 AM

thanks ppl

i tried with sed and i wrote this command :

sed -e '/VERSION/d;/AAA-/{h;d};{G;s/\n//}' *.* > all_files.txt

but i think that the replacement of the \n dosnt work , its work when i open the file in windows

but in linux i see the concatenated line in a new line !

slackie1000 · 06-02-2005, 06:03 AM

hi there,

Quote:

Originally posted by eddiebaby1023
I'd use a script that read in each line. Ignore the first and save the AAA line in a variable, then echo each line with the variable appended. Too simple to warrant the complications of sed, really. If the requirement gets more complicated then sed would definitely be the way to go, though; awk can be a bit clunky (IMO).

ok. agreed.
the task is too simple to use sed...
i will give some tips about sed anyway....
personally, i prefer to write more statements than trying complex regexp...
something in these lines... let's say that your file is called dummy...

Code:

...
anything=`sed -n "/AAA/p" dummy`
anything2=`sed -n "/VERSION/d;/AAA/d" dummy`
...

you just need to append $anything to each line of the output from $anything2...
attention if you need this thing readable in *nix and Windows env.. you have to handel EOF ...
regards
slackie1000