Haldayne\Fox provides algorithm strategies built using function
objects. What's this mean?
Many algorithms are generic. That is, the way the algorithm functions is
independent of what the algorithm does to the underlying data. Let's take a
specific example: array_sum and implode. How do these work? They iterate
over an array, building up a return value. That is a strategy known as
folding. Is there another PHP function that folds? Yes, array_reduce
folds, and so we can build an equilvance between these:
| Specific function | Generic equivalent |
|---|---|
array_sum($a) |
array_reduce($a, function ($t, $a) { return $t + $a; }) |
implode(',', $a) |
array_reduce(array_slice($a, 1), function ($t, $a) { return $t . ',' . $a; }, $a[0]) |
You wouldn't use the generic equivalent in production code, because the
generic version is less concise than the specific function. But what if
the algorithm you need doesn't exist as a PHP native function? That's where
Haldayne\Fox comes in.
This package implements generic strategies in a concise way. Here are some examples of strategies provided:
- Retry a piece of code until it works, failing after a certain number of attempts and exponentially waiting between calls.
- Capture PHP errors triggered when running a piece of code.