Append a line of text to the end of the previous line

market_garden

I need to append a line of text to the end of the previous line. My data is as follows:

dev hostname user user01 show
* deviceid:6d41fc
OK
dev hostname user user02 show
* deviceid:d8c18e
OK
dev hostname user user03 show
* deviceid:abd00d
OK

I need it in the following format and there are a few thousand lines of it:

dev hostname user user01 show * deviceid:6d41fc
OK
dev hostname user user02 show * deviceid:d8c18e
OK
dev hostname user user03 show * deviceid:abd00d
OK

or even better:

dev hostname user user01 show * deviceid:6d41fc
dev hostname user user02 show * deviceid:d8c18e
dev hostname user user03 show * deviceid:abd00d

Can anyone help with this?

ethernet

I suppose you could use this data as a source file and write a script to pull bits from it, perhaps using cut and specifying a space as the delimiter before dumping to another file (or standard output). Don't know of any fancy shortcuts for this but one of the gurus will probably have one.

Tom Dunne

I don't follow what you are trying to do.

Are you Reading from one file, outputting to another?

market_garden

Hi Tom,
I have a text file containing the following:

dev hostname user user01 show
* deviceid:6d41fc
OK
dev hostname user user02 show
* deviceid:d8c18e
OK
dev hostname user user03 show
* deviceid:abd00d
OK
.
. and so on.

I want to either rearrange the text in the file or read the file and output to another so long as I can get the information in the following format:

dev hostname user user01 show * deviceid:6d41fc
dev hostname user user02 show * deviceid:d8c18e
dev hostname user user03 show * deviceid:abd00d

Tom Dunne

I see.

The best approach would be to read in the source file, do the search, and write it out to a second file.

So, you would have a loop that does the following:

look for a string, starting with "dev" and ending with "show" -> first piece
then look for a string "*" -> second piece
then search for "deviceid" and continue on for seven characters (to catch the : and the hex number) -> third piece

I assume you don't want the "OK", so you can ignore that.

Then you combine the three pieces from above, and write them out to a file as one line.

Perl would be your man here, regular expressions are what you are looking for.

EvilMonkey

cut or awk could also be what your loking for. look at the man page

Aquafresh

The following should work:

cat data | gawk 'BEGIN{ORS=""} {if(NR % 3 == 1) print; else if(NR % 3 == 2) print " "$0; else print "\n"}'

i.e. print the first line, a space, the second line, a new line and so on for all lines mod 3. You need the ORS="" because the default output record seperator is "\n" and you don't want newlines between the first and second record etc.