[ic] Order of columns for hashref returned from query?

[Ron Phipps]

> From: interchange-users-admin@interchange.redhat.com
[mailto:interchange-
> users-admin@interchange.redhat.com] On Behalf Of Kyle Cook
> 
> At 06:15 PM 4/17/02, you wrote:
> >I'm working on a piece of code that will be called by CommonAdjust to
> >only return the columns in the pricing table which are populated with
> >data for the given item.  This will force the quantity discount
> >mechanism to discount quantity orders which are equal to or exceed a
> >previous price break.  I had this code working before, but it used a
> >$Tag->data call for each column which was inefficient to do.  Below
is
> >the code as it stands right now.  It selects the data then outputs
the
> >columns.  It will be modified to only output the columns that have a
> >value once I find a solution to the problem.
> >
> >[perl tables=pricing]
> >                         my $code = $Scratch->{tmp_code};
> >                         my $available_breaks =
> >"q2,q3,q4,q5,q6,q7,q8,q9,q10,q12,q15,q25,q50,q100";
> >                         my $sql = "SELECT " . $available_breaks . "
FROM
> >pricing WHERE sku ='" . $code . "';";
> >                         my $sql_results = $Tag->query( {     sql =>
> >$sql, hashref => 'price_results' } );
> >                         my $hash = $sql_results->[0];
> >                         my $out = "";
> >                         my $key, $value;
> >                         while ( ($key, $value) = each %$hash ) {
> >                                 $out .= " $key => $value ";
> >                         }
> >                         #foreach $key (sort keys %$hash) {
> >                         #       $out .= " $key => $$hash{$key} ";
> >                         #}
> >                         return $out;
> >                 [/perl]
> >
> >The problem with this code is the ordering of the hash does not match
> >the ordering of the column list in the select statement.  It is
output
> >in this order:
> >
> >q15,q9,q25,q50,q2,q3,q10,q4,q5,q12,q6,q7,q100,q8,
> 
> Ron,
> 
> My bad, I missed the part where only some columns will be used, hmmm
> next best try would be using a sort subroutine like:
> 
> foreach $key (sort {$$hash{$a} <=> $$hash{$b}} keys %$hash) {
> 
> This will force the keys to be compared numerically, ignoring the
> 'q' at the beginning.
> 
> Kyle Cook

Thanks for the help Kyle!  I tried your suggestion and the results were
sorted on the value instead of the key.  I searched for examples of
sorting on the key and found some, but all the keys in the examples were
strings.  My attempts to modify your example to do a numerical search on
the keys did not work either.  Looks like I need to do some reading on
hashes and sorting :)

While I was relaxing it hit me, why not just split on the comma in my
available_breaks and then loop over those values and key into the hash
to see if the there is a price break.  Below is the code that I came up
with:

		[seti tmp_code][item-code][/seti]
		[perl tables=pricing]
			my $out = "pricing:price_group,";
			my $code = $Scratch->{tmp_code};
			my $available_breaks =
"q2,q3,q4,q5,q6,q7,q8,q9,q10,q12,q15,q25,q50,q100";
			my $break;
			my $sql = "SELECT " . $available_breaks . " FROM
pricing WHERE sku ='" . $code . "';";	
			my $sql_results = $Tag->query( {     sql =>
$sql, hashref => 'price_results' } );
			my $hash = $sql_results->[0];
			my @breaks = split (/,/, $available_breaks);
			foreach $break (@breaks) {
				if ($$hash{$break} ne '') {
					$out .= "$break,";
				}
			}
			return $out;
		[/perl]

Thanks again and I'll post the usertag once it's complete :)
-Ron