Re: heredoc in awk? avoid nasty quoting to print o/p with special

On 5/2/2008 12:52 PM, Rahul wrote:
> Is there something akin to a "heredoc" in awk? I need to process a set
of 
> records, extract relevant numbers from each record and then create a 
> formatted multi-line output for each record. Problem is that the output 
> would contain a lot of special characters and using a normal print or 
> printf would need me to do some nasty quoting (or not?)
> 
> The desired output snippet for each record is something like this:
> 
>   {"id":"town",
>    "label":"middleton",
>    "x_size":"10.22",
>    "y_size":"5.22",
>   },
> 
> 
> Of course, I'd be using $1 (or equivalent variables) for the town, 
> middleton, 10.22, 5.22 etc. 
> 
> Any constructs like the familiar bash
> 
> print << EOF
> static blah blah
> "${var_to_be_inserted}"
> more static blah blah
> EOF
> 

Is this what you're looking for:

$ cat file
town middleton 10.22 5.22
$ awk -v fmt='{"id":"%s",
   "label":"%s",
   "x_size":"%s",
   "y_size":"%s",
},
' '{printf fmt,$1,$2,$3,$4}' file
{"id":"town",
   "label":"middleton",
   "x_size":"10.22",
   "y_size":"5.22",
},

If not, please clarify with some small set of sample input and the
expected
output for that input.

Regards,

	Ed.