Clarify usage/advantage over standard PHP arrays

[patrickallaert]

I have found in README.md the "how", but not the "why".

Being a "Sparse array" implementation, I would expect that it would be much more memory efficient to do:

$array = new Judy(Judy::INT_TO_INT);
for ($i = 0; $i < 100000; ++$i) {
    $array[$i] = 0;
}
$array[42] = 42;
$array[420] = 420;

over:

$array = []; // <== this line differs
for ($i = 0; $i < 100000; ++$i) {
    $array[$i] = 0;
}
$array[42] = 42;
$array[420] = 420;

But what if we just skip the initialization part?

What is the STRING_TO_INT type useful for?

Standard PHP arrays are quite efficient in terms of dictionary, how does Judy arrays with STRING_TO_INT type behave in this regards? More memory efficient as well? Less CPU intensive?

Regarding the huge improvements made on PHP arrays in PHP 7, what are the Judy types that would still be relevant if the extension was made compatible with it?

Last question, how does PHP Judy arrays behave when accessing an element that is not set? I think this is also one important point of such type ;-)

Thanks in advance for your answers @orieg.