Shell Programming and Scripting

I am attempting to combine sections of log that should be one line but are spaced out over 10-30 lines due to how the software is outputting the info. (If I am making a newbie mistake I apologize)

Example of log I am working with:
2009-04-14 14:51:22 access data here
info.
Info.
Info.
…….. )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here data here
info.
Info.
Info.
data here )}

I am currently trying to use the following:

cat log.txt | /scripting/gawk.exe '{d=d""$o}
/2009-04/ {
print d
d=""
}' | > new.txt

What is happening, is it is merging the lines as expected, but incorrectly merging the first line following the matched string.

I would expect the script to create two lines, one for each match of 2009-04, however what is being returned is:

2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here info. info. info. data here. )} 2009-04-14 14:51:22 get data here )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here )}


How do I stop the second matching line from being appended? Shouldn't it be on its own line since it is the next instance of 2009-04?

Thanks in advance..

What happens when you do this:

Code:

sed -e :a -e '/\.$/N; s/\.\n//; ta' filename

When I run the sed command the ouput still contains some unwanted newlines, this is the output after running the suggested command:


2009-04-14 14:51:22 access data here
info.
Info.
Info.
data here )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here
Info.
Info.
Info.
data here )}

What I need is:

2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here. info. Info. Info. data here )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here )}
2009-04-14 14:51:22 access data here. info. Info. Info. data here )}




The purpose of all this is that its easier to merge files and sort once each line has a timestamp.


Edit: The only definate is that the lines I need to merge starts with 2009 and ends with )}

sorry reposted my response

Try this:

Code:

awk '{printf("%s%c",$0,match($0,/)}$/)?RS:FS)}' log.txt