Question

I have a text file with lines like this:

  variable_name : value

I need to make a bash script to replace “value” with a fixed value I know (let say “myvalue”).

I don’t know previously which is the variable_name’s value.

I need to use bash commands that can be used in any default installation of linux box like sed, grep, awk, etc.

I know how to do it with sed when I know the “value” string, but I can’t figure out how to do it by searching in the file by the name of the variable.

EDIT: I add some complexity to this task, I have to exclude some patterns with a fixed text, for example if I have:

  variable_name : value
  variable_name : VARIABLE_NAME > lorem_ipsum

I have to exclude the replacement with the string VARIABLE_NAME. The result must be:

  variable_name : myvalue
  variable_name : VARIABLE_NAME > lorem_ipsum     

Thanks in advance,

Answer 1

are you looking for this?

echo "variable_name : value"|sed -r 's/(variable_name :)(.*)/\1myvalue/'
variable_name :myvalue

EDIT for new requirement

you said you will exclude replacement with string VARIABLE_NAME, so I assume that (space)> lorem_ipsum will be replaced by new value.

so this line may help:

awk 'BEGIN{v="variable_name";V=toupper(v); FS=OFS=" : "} $1~"^"v{$2=($2~"^"V)?V"myValue":"myValue"}1' file

take a look this:

kent$  echo "variable_name : value
variable_name : VARIABLE_NAME > lorem_ipsum"|awk 'BEGIN{v="variable_name";V=toupper(v); FS=OFS=" : "} $1~"^"v{$2=($2~"^"V)?V"myValue":"myValue"}1';                         
variable_name : myValue
variable_name : VARIABLE_NAMEmyValue

EDIT2

Neuquino, you should explain your requirement more clear….. anyway:

 kent$  echo "variable_name : value
variable_name : VARIABLE_NAME > lorem_ipsum"|awk 'BEGIN{v="variable_name";V=toupper(v); FS=OFS=" : "} $1~"^"v && $2!~"^"V{$2="myValue"}1';                                  
variable_name : myValue
variable_name : VARIABLE_NAME > lorem_ipsum