bash, (g)awk, local variables

[MR ZenWiz]

On Sat, Jun 30, 2012 at 6:13 AM, Colin Law <clanlaw at googlemail.com> wrote:
> On 30 June 2012 14:03, franz.reitinger <franz.reitinger at htl-wels.at> wrote:
>> Hi,
>> I'm using awk for string manipulations within a (backup)bash script. Among
>> others I have to do some kind of substring operations using the built-in
>> index-function:
>>
>> Docu:
>> index(in, find)
>> This searches the string in for the first occurrence of the string find, and
>> returns the position in characters where that occurrence begins in the
>> string in. For example:
>>
>> awk 'BEGIN { print index("peanut", "an") }'
>>
>> Of course this example works; however I want to replace both literals with
>> the content of local bash / environment variables like:
>>
>> a= $(cat ANY_FILE);
>> b="anyString"
>>
>> awk 'BEGIN { print index($a, $b) }'
>>
>> Where I'm wrong & how can I use bash variables within built-in functions of
>> awk?
>
> I am not an expert but if in a terminal I do
> a = "a string"
> echo $a
> then I see
> a string
> as expected.  However if I do
> echo '$a'
> then I see
> $a
>
> but echo "$a" shows a string.  So I suggest trying double quotes in
> your awk command.
>

Not quite.

All '$' references inside awk are to its own variables, usually
(always?) just the positional variables for the elements in the input
line (i.e., $0 = the whole line, $1 is field #1, etc).  If you put a $
in front of a variable name, it will attempt to access that element in
the input line (i.e., if the variable 'field' == 1, then $field is
equivalent to $1).

To send in variable values from outside awk, you need to define them
on the command line:

awk -v mybash_a=$a -v mybash_b=$b '<awk script>'

and inside the awk script, you do NOT use $ to access the variables,
just their names.  In this case, use 'mybash_a' to access the value of
your bash $a variable, etc.

Cheers.

MR