How to remove newline, tab, spaces in curly braces.. :( Pls Help?

Shell Programming and Scripting
1

Hi Everyone,

in the below "xyz (Exception e)" part... after the curly braces, there is a new line and immediately few tabs are present before closing curly brace.

xyz (Exception e) { 

}

note: there can be one or more newlines between the curly braces.

My desired output should be

xyz (Exception e) {}

Have tried a lot using awk, sed etc.. but no use. Please help asap. Thank you so much.

Reg,
NY

2

There are a lot of things that aren't specified in your requirements. Making lots of wild assumptions, the following might be something you can use:

awk '
/{/,/}/ {
	if(sub("{.*$", "{}")) print
	next
}
1' file

If you want to try this on a Solaris/SunOS system, change awk to /usr/xpg4/bin/awk , /usr/xpg6/bin/awk , or nawk .

If file contains:

xyz (Exception e) { 

}

Free text

abc (int c) {
	garbage
	garbage
	garbage
		}
free text 2

single (line test) {     	}
after single line

the above code produces the output:

xyz (Exception e) {}

Free text

abc (int c) {}
free text 2

single (line test) {}
after single line
3

Thank yo Mr. Cragun for the fast reply. Let me explain you completely.. My file contain multiple exceptions. For instance, see the below 3.

xyz (Exception) {
                                                                                                         }

xyz (Exception) {


                                                  }

xyz (Exception) { 
                                            filhfs ijsdlfh sdlfhl }

In the above three different scenarios, no change required for 3rd scenario as there is data between curly braces.

I need the 1st & 2nd scenario to be handled as there is no data between curly braces and also, if there is space/tab exists in between the braces then it should be removed..
The output for the both 1st and 2nd scenarios should be same. I.e,

xyz (Exception) {}

Thank you so much.

4

So, apparently you want something more like:

awk '
/{/,/}/ {
	if(/{/)	o = $0
	else	o = o "\n" $0
	if(/}/)	{
		sub(/{[ \t\n]*}/, "{}", o)
		print o
	}
	next
}
1' file

which with the input I showed earlier,produces the output:

xyz (Exception e) {}

Free text

abc (int c) {
	garbage
	garbage
	garbage
		}
free text 2

single (line test) {}
after single line
5

How about

awk     '/{/    {while (!(/}/)) {getline X; $0 = $0 X}
                 sub (/{[       ]*/,"{")
                 sub (/[        ]*}/, "}")
                }
         1
        ' file
xyz (Exception) {}


xyz (Exception) {}



xyz (Exception) {filhfs ijsdlfh sdlfhl}
6

Thank you again for a so quick reply Mr. Cargun.

The solution given is not working as expected. For the better clarity, I am attaching the input and the desire output.

Please go through them once. Thank you so much.

regards,
NY

7

The "solution given is not working as expected" because you posted an inappropriate, truncated sample, and the files you are using are not NIX standard text files (containing extra win <CR> line terminating chars).
However, try (based on Don Cragun's proposal):

awk     '/([Ee]xcept.*) {/      {SV = $0; L=1; next}
         !L
         L                      {SV = SV RS $0
                                 if (/}/) {sub (/{[ \t\r\n]*}/, "{}", SV)
                                           print SV
                                           L=0
                                          }
                                }
        ' input.txt