Lesson 2 - Array vs Mapping

[ghost]

Hi!

I am wondering if in practice, both a mapping an array would be created in order to find needed data more quickly in different situations. It seems by creating both it would be doubling data for no reason.

[mattaereal]

Without explicitly talking about lesson 2, since an array and a mapping are different structure types, unless your mapping uses a uint256 as an index mapped to the same type of the array, then you are not specifically doubling all the data. So in practice, it depends on the use case of the contract.

If by having two different data structures, you increase the storage cost, but save gas by querying them in a criteria which exploits those data structures, then great.

Talking explicitly about Lesson 2, I'm guessing you're talking about these two declarations:

    People[] public people;
    mapping(string => uint256) public nameToFavoriteNumber;

In the case you need to see what's my favorite number, you can just call nameToFavoriteNumber["matt"] and that would be it. It's complexity is O(1).

That mapping is avoiding the need to iterate through the entire people array looking for the field name to match "matt", which is O(n) complexity by the way.

In short, some data may be redundant, like favoriteNumber being twice in storage, but the tradeoff is saving gas for lookups.

[ghost]

Thank you!

[krakxn]

"both a mapping an array would be created in order to find needed data more quickly" It will not be done as that is very redundant; let alone the gas costs.

[ghost]

Thanks!