Issue8

[kamuik16]

Description

Draft PR for #8

Additional context

I have added some initial logic to support __gap storage gaps variable.

[sonarcloud]

Quality Gate passed

Kudos, no new issues were introduced!

0 New issues
0 Security Hotspots
No data about Coverage
No data about Duplication

See analysis details on SonarCloud

[ZeroEkkusu]

Okay, I've done some research and we're going to set one rule:

__gaps must be of size 32 * n. You don't need to worry about it because it's already what we're doing.
Just a note to self that we may want to throw an error if that isn't the case.

---

1. We'll need problematicOnlyBecauseOfGap function:

IN: item from new layout, old layout
OUT: whether item from new layout overlaps only with gap from old layout

The function just needs to iterate over the items from the old layout, and using the current item's start and end, check if there's an overlap with the provided item from the new layout (by comparing to its start and end).
If there's an overlap, return false if the current item is not a __gap.
If it is a __gap, just continue.
Return true at the end, i.e., if we don't find an overlap with an item from the old layout that is not a __gap.

This should be straightforward to do.

Let's start from there - by adding this function. I'll continue to debug the code in the meantime.
Let me know if you think this is a good idea.

[kamuik16]

Okay, I've done some research and we're going to set one rule:

__gaps must be of size 32 * n. You don't need to worry about it because it's already what we're doing. Just a note to self that we may want to throw an error if that isn't the case.

1. We'll need problematicOnlyBecauseOfGap function:

IN: item from new layout, old layout OUT: whether item from new layout overlaps only with gap from old layout

The function just needs to iterate over the items from the old layout, and using the current item's start and end, check if there's an overlap with the provided item from the new layout (by comparing to its start and end). If there's an overlap, return false if the current item is not a __gap. If it is a __gap, just continue. Return true at the end, i.e., if we don't find an overlap with an item from the old layout that is not a __gap.

This should be straightforward to do.

Let's start from there - by adding this function. I'll continue to debug the code in the meantime. Let me know if you think this is a good idea.

This is the same as variableFitInGap function, isn't it?

[ZeroEkkusu]

It's another, separate function.

variableFitInGap check's if the variable fits in the gap wholly. That is to say, it starts and ends in the gap.

problematicOnlyBecauseOfGap checks for a partial overlap, and we'll use it to cover an edge case later.

Here's an example:

var starts at position 0.
gap starts at position 1.

0   var
1   gap

Let's say that we change the type of var to a type that occupies more space.

var starts at position 0, but is of a bigger type and occupies position 1 as well.
gap now starts at position 2.

0   biggerVar
2   gap

If you think about it, var now occupies both the space that it previously occupied (0) AND some space that the gap used to occupy (1).

We'll need a function to check for that, and that's why we're adding it.

Let me know if that makes sense.

[kamuik16]

It's another, separate function.

variableFitInGap check's if the variable fits in the gap wholly. That is to say, it starts and ends in the gap.

problematicOnlyBecauseOfGap checks for a partial overlap, and we'll use it to cover an edge case later.

I think it's probably good if you take it on from here, because you will have to explain each and everything and it will consume your much of time so maybe I'll close this PR and let you do your job. What do you think?

[ZeroEkkusu]

Whatever you prefer. I have something else to work on, so I won't get back to Storage Delta this week. Totally fine if you want or not want to give it a shot in the meantime.

[kamuik16]

Whatever you prefer. I have something else to work on, so I won't get back to Storage Delta this week. Totally fine if you want or not want to give it a shot in the meantime.

Okay @ZeroEkkusu! Btw it was great doing discussion with you on this and I definitely learned a lot! Thanks! Closing this PR.

[ZeroEkkusu]

Thank you,
Take care, @kamuik16!