CODEWITHSUNDEEP
linuxbashshellawk
MOHAMED
MOHAMED

Reputation: 43518

How I can get a field based on a prefix in the awk print?

I have a file containing the following data:

toto,1,xsd:integer
titi,true,xsd:boolean
tata,str,attr,xsd:string

infact the format of each line is:

parameter_name,value,...,xsd:type

the parameter_name is fix and it's in the position 1. The value is also fix and it's in position 2. But the type position is variable it could at any postion except the position 2 and 1. but the type field contains always the prefix xsd.

Now I m trying to execute an awk to extract only the param name, the value and the type. For the moment I m able to extract the param name and the value with

awk -F"," '{print $1"-"$2}'

But I m not able to extract the type field since its position is variable. How I can get the type field based on its prefix xsd in the awk output?

Upvotes: 1

Views: 1560

Answers (2)

Barmar
Barmar

Reputation: 780779

Loop over the rest of the fields looking for the prefix.

awk -F, '{type="";
          for (i = 3; i <= NF; i++) { if($i ~ /^xsd:/) { type = $i; break; } }
          print $1"-"$2"-"type; }'

Upvotes: 2

Avinash Raj
Avinash Raj

Reputation: 174696

You could try this GNU sed command,

$ sed -r 's/^([^,]*),([^,]*),.*(xsd:[^,]*).*/\1,\2,\3/g' file
toto,1,xsd:integer
titi,true,xsd:boolean
tata,str,xsd:string

OR

$ sed -r 's/^([^,]*),([^,]*),.*(xsd:[^,]*).*/\1-\2-\3/g' file
toto-1-xsd:integer
titi-true-xsd:boolean
tata-str-xsd:string

Upvotes: 0

Related Questions