Re: awk BEGIN END syntax

[Ivo Palli]

An awk program consists of 3 main parts:
 * BEGIN
This is code block that is executed BEFORE handling any data * One or more 
blocks of code per line
These get executed for every line of data * END
This code block is executed after all data has been handledCompare these 
examples:
awk 'BEGIN{print "Hello world"; exit}'
seq 1 10 | awk '{print $0}'
seq 1 10 | awk '$0%2{print "uneven: " $0} !($0%2){print "even: " $0}'
find /etc -type f -printf "%s\t%f\n" | \
  awk 'BEGIN{FS="\t"} {s=s+$1;} END{print("size: " s)}'

This should help you on your way.
On Sunday, June 23, 2024 12:03 CEST, Michael Herdman via Help-gawk 
<help-gawk@gnu.org> wrote:
 Hi,
How can I resolve the following problem? I have an infile with two
tab-separated columns the first of which contains two words separated by
a single space:

a b    g
c d    h
e f    i

When I run the following awk command, in which the quotation marks are
escaped to retain them:

awk 'BEGIN {print "<tr><td class=\"Col1_col\">"$1"</td><td
class=\"Col2_col\">"$2"</td></tr>"}' infile > outfile

I expect the outfile to contain:

<tr><td class="Col1_col"a b></td><td class="Col2_col"g></td></tr>
but $1 and $2 are not returned.

I know that the BEGIN statement has to be followed somewhere by an END
statement, but I do not see where to put it or what action to assign to it.

--
Michael Herdman
Institut Pasteur (retired)
Visit the cyanobacterial reference phylogeny site:
http://cyanophylogeny.scienceontheweb.net/