使用sed / awk编辑文件

Question

I have a file which contains the following data:

wednesday
Weekday

thursday
Weekday

friday
Weekday

saturday
MaybeNot

sunday
MaybeNot

monday
Weekday

tuesday
Weekday

I would like to insert a string $:$ in front of each alternate line such that the file is rewritten as follows:

wednesday
$:$Weekday

thursday
$:$Weekday

friday
$:$Weekday

saturday
$:$MaybeNot

sunday
$:$MaybeNot

monday
$:$Weekday

tuesday
$:$Weekday

How would I achieve this using awk/sed?

Answer 1

ok. here.

cat << EOF | awk -vRS="" '{print $1 "\n$:$" $2 "\n" }'       
wednesday
Weekday

thursday
Weekday

friday
Weekday

saturday
MaybeNot

sunday
MaybeNot

monday
Weekday

tuesday
Weekday
EOF

gives...

wednesday
$:$Weekday

thursday
$:$Weekday

friday
$:$Weekday

saturday
$:$MaybeNot

sunday
$:$MaybeNot

monday
$:$Weekday

tuesday
$:$Weekday

Answer 2

If your file doesn't start with a newline and has newlines as shown in your question then use:

sed -i.bak '2s/^/$.$/;3,${n;n;s/^/$.$/;}' file

If file has no new lines and you want to put $.$ before every alternate line as you state in question:

sed -i.bak 'n;s/^/$.$/' file

Answer 3

To prepend $:$ to every 3rd line starting with line 2 do this with GNU sed:

sed -i '2~3s/^/$:$/' infile

Note that this will overwrite infile so be careful.

Answer 4

Using awk - the variable n is initialised in the BEGIN pattern, then the lines are printed as required by testing the odd/eveness of n .

awk 'BEGIN{n=1}
{
     if (n++ % 2)
       print $0
     else
       print "$:$"$0
}' datafile

Answer 5

ssed use perlre, so if you know perlre, it simple. use -i if you want modify target file.

cat t|ssed 's/^\([A-Z]\)/$:$&/'

wednesday

$:$Weekday

thursday

$:$Weekday


friday

$:$Weekday

saturday

$:$MaybeNot

sunday

$:$MaybeNot

monday

$:$Weekday

tuesday
$:$Weekday

Answer 6

awk '/^[[:upper:]]/ {$0="$:$"$0}1' file

to replace in the file

awk '/^[[:upper:]]/ {$0="$:$"$0}1' file > tmp ; mv tmp file

EDIT Another version (this needs the formatting to be the same all time):

awk 'NR%3==2 {$0="$:$"$0} 1' file