Awk: expanding first field along the column

Question

I have a task which requires some awk scripting:

Lets say I have a file:

A,under1
,under2
,under3
,under4
,under5
B,new1
,new2
,new3
C,sm1
,sm2

How could I make it look like this using awk?

A,under1
A,under2
A,under3
A,under4
A,under5
B,new1
B,new2
B,new3
C,sm1
C,sm2

In other words, I want to expand first field along the column.

Regarding pros and cons of setting awk variables as command paramaters vs. as command ooptions, see: Are there 2 ways to set awk vars via command line?. The -v(option) syntax was introduced to deal with potential pitfalls of the paramater method shown in the two (current) answers. — Peter.O
– Peter.O, Commented Jan 17, 2013 at 6:52

Guru · Accepted Answer · 2013-01-17 02:53:50Z

3

One way:

$ awk -F, '!$1{$1=p;}$1{p=$1;}1' OFS=, file

$1{p=$1;} When $1 has a value, store it in a variable p

!$1 Mean $1 is empty. When $1 is empty, set $1 with p which contains the last $1 value.

1 to print every line.

answered Jan 17, 2013 at 2:53

Guru

5,9931 gold badge22 silver badges20 bronze badges

sparkie · Accepted Answer · 2013-01-17 04:21:12Z

2

another solution:

awk -F, '{$1=$1?P=$1:P;print}' OFS=,

sets P to $1 if $1 not empty. sets $1 to previous P otherwise

answered Jan 17, 2013 at 4:21

sparkie

1212 bronze badges

2 Answers 2