I know it's pretty late, but I hope it helps you or others who might read this article.
If we see it as a recursion, it becomes easier to interpret & understand; in that manner the base case is to reach an identical field to sort upon(_id in this example).
let’s start with the first condition:
$or: [
{ price: { $gt: 99 } },
]
for the next step, we should assume the price is the same, and think of only 2 fields to sort upon and exactly do as explained in the article:
$or: [
{ price: { $gt: 99 } },
{
price: 99,
$or: [
{ rating: { $gt: 4 } },
{
rating: 4,
_id: { $gt: ObjectId("5f7f1f7d67f67ddaa622b796") },
},
],
},
],
and as the final step, we should assume rate is also the same, and add the base case where we only sort upon the identical field:
$or: [
{ price: { $gt: 99 } },
{
price: 99,
$or: [
{ rating: { $gt: 4 } },
{
rating: 4,
_id: { $gt: ObjectId("5f7f1f7d67f67ddaa622b796") },
},
],
},
{
price: 99,
rating: 4,
_id: { $gt: ObjectId("5f7f1f7d67f67ddaa622b796") },
},
],
It is a bit complex, but its time complexity is still O(1). When there is more than 2 fields to sort upon, and there is not a large amount of data, I encourage you to use the skip-limit method. Also, you need to consider applying proper indexing in both cases (as I mentioned in Worse Performance and Fix section).
