Hi guys,
brand new to this thread and very very new to UNIX...so go easy please!

Anyway I have a file that looks like this:

>>--------------------------------------------------------------------------
Date/Time/Eng.   : 2012-06-22 / 00:26 /   DS
Reported problem : (SD)  
------------------------------------------------------------------------------

|00:26| 

Time/Date : 2012-06-22  00:26
Status    : UP
-----------------------------------------------------------------------------<<

>>--------------------------------------------------------------------------
Date/Time/Eng.   : 2012-06-22 / 00:29 / mole  DS
Reported problem : (SD)  test
SO               :       7
------------------------------------------------------------------------------

|00:29| 

Time/Date : 2012-06-22  00:30
Status    : UP, up as up can be in sydney
-----------------------------------------------------------------------------<<

2012-06-22>>--------------------------------------------------------------------------
Date/Time/Eng.   : 2012-06-22 / 00:33 / mole  DS
Reported problem : (SD)  lotabort
SO               :       4
------------------------------------------------------------------------------

|00:33| 

Time/Date : 2012-06-22  00:33
Status    : UP
-----------------------------------------------------------------------------<<

sed -n '/>>/,/<</p' log.doc > outputfile.doc

Something with awk that might work for you:

awk '
    />>-+$/ {
        delete cache;
        ci = 0;
        snarf = 1;
    }
    snarf { cache[ci++] = $0 };
    /^-+<<$/ { snarf = 0; }

    END {
        for( i = 0; i < ci; i++ )
            print cache[i];
    }
' input-file >output-file

Hi agama,

Thanks for your speedy reply...

I may have left out a huge detail...hopefully not

The command I am trying to run is being run through the shell command feature in NEdit on a solaris machine...I am guessing this makes this a whole different ball game?

When I enter that code intro the command and run it I get the following:

"Unmatched'
Variable syntax
delete: Command not found
ci: Command not found
snarf: Command not found
}: Command not found 
!No file for $0
Variable syntax
END: Command not found
Badly placed 0's
No match
}: Command not found
Unmatched'"

Well that's a new one on me -- not familiar with Nedit.

It sounds like a quoting problem, but I don't have access to a Solaris box at the moment, so I cannot poke from that angle. I do know that with solaris' version of awk you'll need to change delete cache to split( "", cache, "." ) . It 'empties' the array -- that version of awk doesn't support delete. You also probalby want to use nawk or (I think) /usr/xpg4/bin/awk .

Can you save the awk script to a file, and then try running the file from Nedit? That's probably the only thing I can think to suggest without knowing the Nedit environment.

There are plenty of folks who work on Solaris more regularly than I do -- be patient I'm sure one of them will offer something that might be of more help.

Sorry.

Somehow that awk code is not protected from the shell, because it does not get run as awk code but it is run by the shell. Those are shell error messages, not awk...
It may be that every line is interpreted as separate (because the first error messages says "Unmatched' .

Does it do this also with this alternative one-liner?

/usr/xpg4/bin/awk 'f{p=p RS $0} />>--/{p=$0}{f=1} /--<</{f=0} END{print p}' infile

Thanks Scrutinizer,

I used the changes agama suggested, saved as passdown.sh, uploaded to machine, ran

chmod +x passdown.sh

./passdown.sh