将PHP数组操作转换为Ruby哈希

Question

I've written this script to be run in PHP, and am trying to convert it to Ruby.

Code:

$cases = array();


foreach($results as $result)
{
    if(!array_key_exists($result['id'], $cases))
        $cases[$result['id']] = array($result);
    else
        $cases[$result['id']][] = $result;
}

foreach($cases as $key => $case)
{
    foreach($case as $payment)
    {
        if(count($case) > 1)
        {

            $cases[$key]['total'] += ($payment['p1'] > $payment['p2']) ? $payment['p1'] : $payment['p2'];

        } else {

            $cases[$key]['total'] = ($payment['p1'] > $payment['p2']) ? $payment['p1'] : $payment['p2'];
        }
    }
}

In Ruby, I have the results as an array returned by MySQL (using mysql2 gem). Using the loop below, it prints every row. How can I check if the key exists when it does not provide keys?

What's the best way to implement this code in Ruby? How can I add a hash onto a hash in Ruby similar to adding an array onto an array in PHP.

results.each(:as => :array) do |row|
    puts row.inspect
end

More explanation:

PHP Structure

Array(
    0 => Array(
        Array(
            [id] => 'random id',
            [p1] => 534,
            [p2] => 105
        ),
        Array(
            [id] => 'random id',
            [p1] => 335,
            [p2] => 425
        )
    ),
    1 => Array(
        Array(
            [id] => 'random id',
            [p1] => 259,
            [p2] => 124
        )
    )
)

And i'm trying to iterate over each array inside the first, and if p1 > p2, add p1 to the total key on that array. i.e., the 0 index would look like this:

0 => Array(
    'total' => 959,
    Array(
        [id] => 'random id',
        [p1] => 534,
        [p2] => 105
    ),
    Array(
        [id] => 'random id',
        [p1] => 335,
        [p2] => 425
    )
)

Answer 1

Are you trying to group by result['id']? I believe that "translating" your PHP code to ruby will look something like this:

cases = {}
results.each do |result|
  if cases[result['id']]
    cases[result['id']] << result
  else
    cases[result['id']] = [result]
  end
end

Another way to get the same result is to use group_by:

cases = results.group_by { |result| result['id'] }

The second part "translated" to ruby will look something like this:

cases.each do |key, c|
  c.each do |payment|
    if c.count > 1 
      c['total'] += (payment['p1'] > payment['p2']) ? payment['p1'] : payment['p2']
    else
      c['total'] = (payment['p1'] > payment['p2']) ? payment['p1'] : payment['p2']
    end
  end
end

(this will actually not work in ruby, since an array cannot have an arbitrary key like 'total' - it accepts only numbers in its brackets. You will have to modify this code to hold the totals in a different structure)

Again, a more idiomatic way of summing up the max between 'p1' and 'p2' of each payment in each case might look more like this:

totals = cases.values.map do |c|
  c.inject(0) { |sum, payment| sum + [payment['p1'], payment['p2']].max }
end