Shell Programming and Scripting

Hi everyone,

How can you tell awk to switch the input file at a given point in the program when you you have multiple file inputs? I always assumed that if you divided your code into blocks, awk would understand that the first block contains instructions for file1, the second block contains instructions for the second file...

For example:
{
//Instructions executed for file1
}
{
//Instructions executed for file2
}

Clearly, when running the program, that does not work. Can anyone please explain?

Thank you.

As you suggest, it definitely does not work that way. Instead, it read the first file, then the second file, etc. and applies all blocks as each line is read. And you can have as many blocks as you want, has nothing to do with number of files.

getline is one way to read from different files, and perhaps accomplish what you want.

Code:

FILENENAME=="file1" {
//Instructions executed for file1
next
}
{
//Instructions executed for file2
}

Thank you very much to everyone! I really appreciate it.

---------- Post updated at 06:55 PM ---------- Previous update was at 06:36 PM ----------

It works very well, but I was wondering: if I need to run the program with several pairs of files, instead of rewriting the code before each run, can I use variables?

Thank you!

If you just have two files, you can use the following:

Code:

FNR==NR{//Instructions executed for file1
        next
}
{       //Instructions executed for file2
}

Code:

awk  'FNR==1  {file++}
      file==1 {Instructions executed for file1;  next }
      file==2 {Instructions executed for file2;  next }
      file==3 {Instructions executed for file3;  next }
... etc ...