Shell Programming and Scripting

Greetings All!!

I have a very peculiar problem where I have to parse a big text file and extract useful data out of it with starting and ending block pattern matching.

e.g. I have a input file like this:

sample data
block1
sample data
start
useful data
end
sample data
block2
sample data
start
useful data
end
sample data

My output shall be like following:

block1
useful data
block2
useful data

Any help will be appreciable.

Cheers

Arminder

Assuming that your input file is always of the same format as specified in your example, the following code will work. Save this as block.awk

Code:

#!/usr/bin/awk -f

/^block/ {
  print $0
  next
}
/^start/ {
  flag = 1
  next
}
flag == 1 {
  print $0
  flag = 0
}

Make it executable, run it over your input file, and redirect output if desired...

Code:

$ chmod +x ./block.awk
$ ./block.awk my_file > my_output
$ more my_output
block1
useful data
block2
useful data

Cheers
ZB

USe this,

Code:

 awk '{ if ($0 ~ "block") { print $0;getline;print $0 } }' testfile

hth.

No, that won't work. The OP wants "useful data" not "sample data". You'd need three "getline"s to get this to work....

Code:

awk '{ if ($0 ~ "block") { print $0;getline;getline;getline;print $0 } }' testfile

Cheers
ZB

oops.. ZB you are 100% correct.

Sorry Guys, maybe I didn't put it correctly. The amount of "useful data" isn't known i.e. it may be 2 lines long or 10/20/100 i.e. no same occurences. Same goes for the "sample data" too. Its only that "start" and "end" is that can tell me where to start from and where to end.

Later I figured a way to print the block and so currently I am using following in awk file:

/^start/./^end/

Thereafter I grep out (grep -v) the start and end text.

Cheers

Arminder