Shell Programming and Scripting

Hi All,
I have many of files(.csv) of the format given below.

Code:

Date,Name,Location
04/02/2012,A,India
,B,China
,C,USA

Like this I have 1000's of rows and many columns in all my files.

I need a shell script to copy down the Date(in this example column1) to the next 2 rows below(in the same column) also.
Required output.

Code:

Date,Name,Location
04/02/2012,A,India
04/02/2012,B,China
04/02/2012,C,USA

Many thanks in advance..
Sidda

Code:

awk -F, -vOFS=, '$1=="" {$1=oldone}1 $1!="" {oldone=$1}1' file.csv

Code:

sed "s#^,#$(date "+%m/%d/%Y")&#" infile
Date,Name,Location
04/02/2012,A,India
04/02/2012,B,China
04/02/2012,C,USA

Code:

awk -F, '$1=$1?p=$1:p' OFS=, infile

Could you please explain this

Code:

'$1=$1?p=$1:p'

Thanks

Thanks Carlo and Scrutinizer

Both of your scripts worked very well. But what I found is that some of the .csv files have the following type of data in them(example).

Code:

SerialNo, Location, Country
1,Delhi,India
,Mumbai,India
,Bangalore,India
2,NewYork,USA
,SantaClara,USA
3,Frankfurt,Germany

What I need is the following output.
SerialNo, Location, Country

Code:

1,Delhi,India
1,Mumbai,India
1,Bangalore,India
2,NewYork,USA
2,SantaClara,USA
3,Frankfurt,Germany

Can you modify the scripts written by you to get my desired output.
Thanks

You mean skip the first line?

Code:

awk -F, 'NR>1{$1=$1""?p=$1:p}1' OFS=, infile

or more readable (more like CarloM's suggestion):

Code:

awk -F, 'NR>1{if($1=="")$1=p; else p=$1}1' OFS=, infile

----

This means: if $1 is unequal to "" or 0 then $1 remains $1, while we are setting p to the current $1 value ($1=p=$1) else set $1 to p.
There was a bug in this code, it should replace the value in column 1 if it is empty, not when it is 0. This could be corrected like this:

Code:

awk -F, '{$1=x$1?p=$1:p}1' OFS=, infile

So now it reads: if x$1 is unequal to "" or 0 then $1 remains $1 .. etc..
x is empty and therefore x is equal to "" . By concatenating the empty string x to $1 , $1 if forced to be interpreted as a string, not as a numerical value. Why is this necessary? If $1 contains the numerical value 0 then this would translate to false, but if $1 contains the string value "0" then this translates to "true" (only the string value "" is translates to false).

This can seem a bit complex, so a more readable version may be preferable and I think this should be considered more for learning the intricacies of awk...