Hi,
Can some one help me with unix sort command?
I have a file like this:
01 asd 8211 abcd
02 das 8211 fsdf
03 fes 8301 fsdf
04 sda 8213 cdsc
05 lkm 8321 acsd
I want sort by first 2 char of third comlumn and then 4th conlumn
the result would be like:
01 asd 8211 abcd
04 sda 8213 cdsc
02 das 8211 fsdf
05 lkm 8321 acsd
03 fes 8301 fsdf

Thanks In advance

sort -k 3.2,3.3 -k 4,4 data.file

Thank Nails.
Could you explain how is it work?
3.2 means third conlumn first 2 char?
3.3 ????
4 means 4th conlumn but what about ",4"? by 4 char?
One more question:
If I got second conlumn blank how can I ignore it?
01 asd 8211 abcd
02 8211 fsdf
03 fes 8301 fsdf
04 sda 8213 cdsc
05 lkm 8321 acsd
It has fixed spaces. I still want sort it like this:
01 asd 8211 abcd
04 sda 8213 cdsc
02 8211 fsdf
05 lkm 8321 acsd
03 fes 8301 fsdf

Thanks again

HI:

Let's discuss your "blank" field question first: Just looking at the data, how do you know it's the second field blank? My eye says that it might be the 4th field? If it were me, I'd run a script (like an awk script?) against the data file and then do something if there's only 3 fields.

One of the options of the sort command is the key field. Within the key field you have to give it a start and an end. If you use the dot operator, that's the start and stop postions with the key field:

-k 3.2,3.3

In the above statement, start at the third field's 2nd character and end at the 3rd character.

The second key is to start by the 4th field and end by the 4th field:

-k 4,4

Since you're sorting by only 1 field, you probably can get by with:

-k 4

but I always give it an end. Do a 'man sort' for more information.

Regards,

Nails

I include

Thank you so much for your help.
For the second question I was typing in 2nd blank filed but when I submmit it just ignore the space in between. So i looks like 4th balnk. Sorry about that.
BTW, the sort command give me the result like this:
01 asd 8211 abcd
02 das 8211 fsdf
04 sda 8213 cdsc
03 fes 8301 fsdf
05 lkm 8321 acsd
seems -k3.1,3.2 not working for me properly. It doesn't stop at third's 2nd char.I'd like get result like this:
01 asd 8211 abcd
04 sda 8213 cdsc --ignore last 2 char"13"
02 das 8211 fsdf
05 lkm 8321 acsd --ignore last 2 char"21"
03 fes 8301 fsdf
I was using this command:
sort -k3.1,3.2 -k4,4 file
Did i do something wrong?

Thanks, James

James:

Take a closer look at it. Per your original question, shouldn't it be:

sort -k 3.2,3.3 -k 4 data.file

It should be the second and third characters where you are using the first and second -k 3.1,3.2

Regards,

Nails

Sorry for the confusing.
If I want sort by "first 2 character of 3rd conlumn" It should be
sort -k3.1,3.2 -k4,4 right?
In this case I only need consider 82, 83 and ignore the rest chars of 3rd column.
Thanks, James