Import variables in to awk.

[Chris Torek]

>In article <10531 at thorin.cs.unc.edu> warner at unc.cs.unc.edu (Byron Warner)
>writes:
[file foo]
>>{ print import,$0 }
[command]
>>awk -F: -f foo /etc/passwd import='hello
>>why do I get just a list of logins?

In article <15919 at bloom-beacon.MIT.EDU> jik at athena.mit.edu
(Jonathan I. Kamens) writes:
>  First of all, I have never known the C-shell to allow the syntax
>"foo=bar" on a command-line to import a variable into a program.

It does not.  However, awk does.  That is, you are looking at the wrong
program.

>  Second, the only way to do what you want is to actually make the
>creation of this variable part of the awk script.  Like this:

Not so:  Within some limits, you can set awk variables from its
invocation.  For instance:

	% cat t
	BEGIN { print "BEGIN: " this; }
	{ print "INPUT: " this " " $0; }
	END { print "END: " this; }
	% cat u
	first line
	second line
	% awk -f t u this=that
	BEGIN: 
	INPUT:  first line
	INPUT:  second line
	END: that
	% awk -f t this=that u
	BEGIN:
	INPUT: that first line
	INPUT: that second line
	END: that
	% rm t u

The `BEGIN' statement is done before any `files' are opened; the `END'
statement is done after all `files' have been read.  Any `files' of
the form `a=b' set variable `a' to value `b'.

All of the above is with respect to the 4.3BSD flavour of `awk'.  The
new awk (as described in the awk book) appears to open the first `file'
before executing the BEGIN statement, so that any assignments that
appear before the first real file happen before the BEGIN.  What GNU
awk does, I do not know (but the above technique will tell you).
-- 
In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7163)
Domain:	chris at cs.umd.edu	Path:	uunet!mimsy!chris