Secondary sorting keys

[deppeler]

my flat field DB fields are: [Name,Code,Category]

I have them sorted by field [3] (category=TeachersActivity or ChildrensBooks) so they are grouped together as Children's Books then Teacher Activities

open(BASE, $dbk) || do {&no_open;};
my @sorted = <BASE>;
close(BASE);
 
foreach my $line (sort cat @sorted) {
    my @show = split ',', $line;
 
 
....
 
sub cat {
    @x = split ',', $a;
    @y = split ',', $b;
    return ($x[2] cmp $y[2]);
}

the result is:

Children and Their World - Children's Books
Nature - Children's Books
Animals - Children's Books
Families - Children's Books
Learning About Myself - Children's Books
Thinking Skills - Children's Books
Multicultural Book - Children's Books
Arts and Crafts - Teacher Activities
Cooking Activities - Teacher Activities
Creative Activities - Teacher Activities
Preschool Activities - Teacher Activities
Child Guidance - Teacher Activities
 

Is there a way to further sort them so that each 'group' Children's Books or Teacher Activities would be in alphabetical order, so the result would be:

Animals - Children's Books
Children and Their World - Children's Books
Families - Children's Books
Learning About Myself - Children's Books
Multicultural Book - Children's Books
Nature - Children's Books
Thinking Skills - Children's Books
Arts and Crafts - Teacher Activities
Child Guidance - Teacher Activities
Cooking Activities - Teacher Activities
Creative Activities - Teacher Activities
Preschool Activities - Teacher Activities
 

thanks
Paul

[miller]

Yes, just add the other conditions to your sorting function:

open(BASE, $dbk) || do {&no_open;};
my @dbk = <BASE>;
close(BASE);
 
foreach my $line (sort cat @sorted) {
    my @show = split ',', $line;
    print "@show";
}
 
sub cat {
    @x = split ',', $a;
    @y = split ',', $b;
    return ($x[2] cmp $y[2]) || ($x[0] cmp $y[0]);
}
 

- Miller

[KevinADC]

use a Schwartzian Transform:

open (BASE, $dbk) || do {&no_open;};
@sorted = <BASE>;
close BASE;
@sortedNew = map  {"$_->[0] - $->}[2]"}
             sort {$a->[2] cmp $b->[2] || $a->[0] cmp $b->[0]}
             map  {chomp; [split(/,/)]} @sorted;
print "$_\n" for @sortedNew;
 

http://www.stonehenge.com/merlyn/UnixReview/col64.html

[KevinADC]

Yes, just add the other conditions to your sorting function:

open(BASE, $dbk) || do {&no_open;};
my @dbk = <BASE>;
close(BASE);
 
foreach my $line (sort cat @sorted) {
    my @show = split ',', $line;
    print "@show";
}
 
sub cat {
    @x = split ',', $a;
    @y = split ',', $b;
    return ($x[2] cmp $y[2]) || ($x[0] cmp $y[0]);
}
 

- Miller

Looks like it would be fairly slow, but if the file is not big it may not matter.

[deppeler]

thanks a lot!
many thanks
Paul

[deppeler]

One more question.
The result of the sort is:

Animals - Children's Books
Children and Their World - Children's Books
Families - Children's Books
Learning About Myself - Children's Books
Multicultural Book - Children's Books
Nature - Children's Books
Thinking Skills - Children's Books
Arts and Crafts - Teacher Activities
Child Guidance - Teacher Activities
Cooking Activities - Teacher Activities
Creative Activities - Teacher Activities
Preschool Activities - Teacher Activities

which is perfect.
I have the output going into a table, is there a way to split the output to 2 columns instead of one long column.
I would like to have Children's Books displayed in the first column and Teachers Activities in the 2nd column.

I have tried a few things and all I get is something like this:

-------------[column 1]--------------------------------------------------[column 2]
Animals - Children's Books---------------------------------Arts and Crafts - Teacher Activities
Children and Their World - Children's Books-----------(blank cell)
Families - Children's Books--------------------------------(blank cell)
Learning About Myself - Children's Books---------------Child Guidance - Teacher Activities
Multicultural Book - Children's Books---------------------Cooking Activities - Teacher Activities
Nature - Children's Books-----------------------------------(blank cell)
Thinking Skills - Children's Books-------------------------Creative Activities - Teacher Activities
--------------------------------------------------------------------Preschool Activities - Teacher Activities

I'm sure I did it once, but maybe not.
Any help would be appreciated.

thanks
Paul

[miller]

For ultra simple text formatting, use sprintf:

my @lines = (
    "hello world - what'cha doing today?\n",
    "foo bar baz biz lama lama - oh yeah\n",
    "eek - more spacing tests\n",
);
 
foreach my $line (@lines) {
    my ($col1, $col2) = split ' - ', $line;
    print sprintf("%-30s %s", $col1, $col2);
}
 

- Miller

[miller]

Looks like it would be fairly slow, but if the file is not big it may not matter.

I would call it just "slower". In truth, even for a large data file, I expect that the perl sorting and splitting functionality would be plenty fast enough for this to not be a noticable operation.

But either way, it just seemed simpler to avoid algorithmic efficiency discussions, and just focus on how one does this in a basic way. Yes, we probably wouldn't do it that way. But he didn't ask how we would, just how "one would". :)

- M

[KevinADC]

I have tried a few things and all I get is something like this:

Lets see something you have tried. "Table" could mean just about anything.

[deppeler]

Here is one I tried:

<table width='100%' border='1' cellpadding='1' cellspacing='1'>
HTML
open (BASE, $dbk) || do {&no_open;};
@sorted = <BASE>;
foreach $line (sort cat @sorted) {
    @show = split(/,/,"$line");
    $showta = "";
    $showta = "$show[0]" if $show[2] eq 'Teacher and Activity Books';
    $showcb = "";
    $showcb = "$show[0]" if $show[2] eq 'Children`s Books';
    print "<tr><td align='left' width='365'>$showta</td><td width='364'>$showcb</td></tr>";
}
close(BASE);
 
sub cat {
    @x = split(/,/,$a);
    @y = split(/,/,$b);
    return ($x[2] cmp $y[2]) || ($x[0] cmp $y[0]);
}
 
print <<"HTML";
</table>

and this gives me an output of this:

instead of this:

thanks
Paul

[KevinADC]

what you want to do would best, or most easily, be accomplished by two arrays and using a different html table structure. This is untested and not meant to be a complete solution but I think it gives you the basic concept:

my @c = ();
my @t = ();
open (BASE, $dbk) || do {&no_open;};
my @sorted = <BASE>;
close (BASE);
foreach $line (@sorted) {
   if ($line =~ /Teacher and Activity Books/) {
      push @t, $line;
   }
   else {
      push @c, $line;
   }
}
@t = sort @t;
@c = sort @c;
my ($ctable,$ttable) = (q{<table width='100%' border='1'>}, q{<table width='100%' border='1'>});
for (@c){
   $ctable .= qq{<tr><td>$_</td></tr>\n};
}
for (@t){
   $ttable .= qq{<tr><td>$_</td></tr>\n};
}
$ctable .= '</table>';
$ttable .= '</table>';
print qq{<table width='100%' border='0' cellpadding='1' cellspacing='1'>
 <tr>
  <td valign="top" width="50%">
    $ctable
  </td>
  <td valign="top" width="50%">
    $ttable
  </td>
 </tr>
</table>\n};

this is not the way I would probably do it, but unless you are ready to get into references and more complex data structres, like a hash of arrays and such, then a solution like this while a little awkward is fairly straight forward to understand.

-Kevin

[deppeler]

No that looks good except how do I stop it printing the entire line of the db?
I just want the field 1 & 2 and not the commas either.
It prints like this:
Animals,ANIML,Children's Books,

I would like it to print like this:
Animals - Children's Books

thanks a heap
Paul

[KevinADC]

Like I said, not a complete solution. Use split() to get the parts of the lines you want just like you have been doing in your previous code.

[deppeler]

Ok I tried:

my @c = ();
my @t = ();
open (BASE, $dbk) || do {&no_open;};
my @sorted = <BASE>;
close (BASE);
foreach $line (@sorted) {
 @show = split ',', $line;
   if ($line =~ /Teacher and Activity Books/) {
      push @t, $line;
   }
   else {
      push @c, $line;
   }
}
@t = sort @t;
@c = sort @c;
my ($ctable,$ttable) = (q{<table width='362' border='0' cellpadding='3' cellspacing='1' align='center' bgcolor='#F5F5F5'>}, q{<table width='361' border='0' cellpadding='3' cellspacing='1' align='center' bgcolor='#F5F5F5'>});
for (@c){
   $ctable .= qq{<tr><td>$show[0] - $show[2]</td></tr>\n};
}
for (@t){
   $ttable .= qq{<tr><td>$show[0] - $show[2]</td></tr>\n};
}
$ctable .= '</table>';
$ttable .= '</table>';
print qq{<table width='729' border='0' cellpadding='3' cellspacing='1' align='center' bgcolor='#F5F5F5'>
 <tr>
  <td valign="top" width="50%">
    $ctable
  </td>
  <td valign="top" width="50%">
    $ttable
  </td>
 </tr>
</table>\n};

But I am only get the first line repeated.
What am I doing wrong?

thanks

[KevinADC]

You have a fundamental misunderstanding of the process. See how this works:

my @c = ();
my @t = ();
open (BASE, $dbk) || do {&no_open;};
my @sorted = <BASE>;
close (BASE);
foreach $line (@sorted) {
   if ($line =~ /Teacher and Activity Books/) {
      push @t, $line;
   }
   else {
      push @c, $line;
   }
}
@t = sort @t;
@c = sort @c;
my ($ctable,$ttable) = (q{<table width='100%' border='1'>}, q{<table width='100%' border='1'>});
for (@c){
   my @temp = split(/,/);
   $ctable .= qq{<tr><td>$temp[0] - $temp[2]</td></tr>\n};
}
for (@t){
   my @temp = split(/,/);
   $ttable .= qq{<tr><td>$temp[0] - $temp[2]</td></tr>\n};
}
$ctable .= '</table>';
$ttable .= '</table>';
print qq{<table width='100%' border='0' cellpadding='1' cellspacing='1'>
 <tr>
  <td valign="top" width="50%">
    $ctable
  </td>
  <td valign="top" width="50%">
    $ttable
  </td>
 </tr>
</table>\n};