File: gawk.info,  Node: Rewind Function,  Next: File Checking,  Prev: Filetrans Function,  Up: Data File Management

10.3.2 Rereading the Current File
---------------------------------

Another request for a new built-in function was for a function that
would make it possible to reread the current file.  The requesting user
didn't want to have to use 'getline' (*note Getline::) inside a loop.

   However, as long as you are not in the 'END' rule, it is quite easy
to arrange to immediately close the current input file and then start
over with it from the top.  For lack of a better name, we'll call the
function 'rewind()':

     # rewind.awk --- rewind the current file and start over

     function rewind(    i)
     {
         # shift remaining arguments up
         for (i = ARGC; i > ARGIND; i--)
             ARGV[i] = ARGV[i-1]

         # make sure gawk knows to keep going
         ARGC++

         # make current file next to get done
         ARGV[ARGIND+1] = FILENAME

         # do it
         nextfile
     }

   The 'rewind()' function relies on the 'ARGIND' variable (*note
Auto-set::), which is specific to 'gawk'.  It also relies on the
'nextfile' keyword (*note Nextfile Statement::).  Because of this, you
should not call it from an 'ENDFILE' rule.  (This isn't necessary
anyway, because 'gawk' goes to the next file as soon as an 'ENDFILE'
rule finishes!)

   You need to be careful calling 'rewind()'.  You can end up causing
infinite recursion if you don't pay attention.  Here is an example use:

     $ cat data
     -| a
     -| b
     -| c
     -| d
     -| e

     $ cat test.awk
     -| FNR == 3 && ! rewound {
     -|    rewound = 1
     -|    rewind()
     -| }
     -|
     -| { print FILENAME, FNR, $0 }

     $ gawk -f rewind.awk -f test.awk data
     -| data 1 a
     -| data 2 b
     -| data 1 a
     -| data 2 b
     -| data 3 c
     -| data 4 d
     -| data 5 e