feat(streamer): distribute rewards immediately in the current block

[keruch]

Description

https://www.notion.so/dymension/ADR-x-Incentives-DoS-47238d8a29734ee686e25caef8b68c4f

Closes #XXX

All items are required. Please add a note to the item if the item is not applicable and
please add links to any relevant follow-up issues.

PR review checkboxes:

I have...

• Added a relevant changelog entry to the Unreleased section in CHANGELOG.md
• Targeted PR against the correct branch
• included the correct type prefix in the PR title
• Linked to the GitHub issue with discussion and accepted design
• Targets only one GitHub issue
• Wrote unit and integration tests
• Wrote relevant migration scripts if necessary
• All CI checks have passed
• Added relevant godoc comments
• Updated the scripts for local run, e.g genesis_config_commands.sh if the PR changes parameters
• Add an issue in the e2e-tests repo if necessary

SDK Checklist

• Import/Export Genesis
• Registered Invariants
• Registered Events
• Updated openapi.yaml
• No usage of go map
• No usage of time.Now()
• Used fixed point arithmetic and not float arithmetic
• Avoid panicking in Begin/End block as much as possible
• No unexpected math Overflow
• Used sendCoin and not SendCoins
• Out-of-block compute is bounded
• No serialized ID at the end of store keys
• UInt to byte conversion should use BigEndian

Full security checklist here

---

For Reviewer:

• Confirmed the correct type prefix in the PR title
• Reviewers assigned
• Confirmed all author checklist items have been addressed

---

After reviewer approval:

• In case the PR targets the main branch, PR should not be squash merge in order to keep meaningful git history.
• In case the PR targets a release branch, PR must be rebased.

[zale144]

so, I can see that the logic in both EndBlock and AfterEpochEnd roughly has some similarities:

1. GetActiveStreams
2. CalculateRewards
3. SendCoinsFromModuleToModule
4. ik.Distribute

is there a way to somehow extract that logic into a common method?
the main difference I see is that EndBlock iterates over all epochPointers while AfterEpochEnd distributes for the current epochPointer

the is that the implementation is a bit convoluted, it's not trivial to follow the logic, so maybe it would help to centralize those critical pieces of logic

[codecov]

Codecov Report

Attention: Patch coverage is 75.37994% with 81 lines in your changes missing coverage. Please review.

Project coverage is 24.95%. Comparing base (031ecad) to head (90d79e3).
Report is 24 commits behind head on main.

Files with missing lines	Patch %	Lines
x/streamer/keeper/distribute.go	64.88%	36 Missing and 10 partials ⚠️
utils/cache/ordered.go	71.64%	19 Missing ⚠️
x/incentives/keeper/distribute.go	76.92%	4 Missing and 2 partials ⚠️
x/streamer/types/stream.pb.go	66.66%	3 Missing and 1 partial ⚠️
utils/cache/linked_list.go	94.87%	1 Missing and 1 partial ⚠️
x/streamer/keeper/abci.go	71.42%	1 Missing and 1 partial ⚠️
x/streamer/keeper/hooks.go	85.71%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1173      +/-   ##
==========================================
- Coverage   25.30%   24.95%   -0.36%     
==========================================
  Files         499      513      +14     
  Lines       98751   102755    +4004     
==========================================
+ Hits        24989    25642     +653     
- Misses      70523    73786    +3263     
- Partials     3239     3327      +88     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[danwt]

I will continue the review but it's better I ask questions first, mainly I don't quite understand the intent behind the linked list and its wrapper

https://github.com/dymensionxyz/dymension/pull/1173/files

Somehow I can't see the forest through the trees when it comes to the data structures and it's usage

• please docstring all public methods on InsertionOrdered
• in general better documentation of insertion ordered data structure
• why need key func in insertion ordered?
• no need for insertion ordered upsert multiple and upsert, just have upsert take variadic
• Linked lists are not performant AT ALL afaiu, use a more hardware friendly data structure? does perf matter?
• why linked list? what property there is important?
• Use data structure library instead of implement from scratch? If implement then put in sdk-utils?, (and test with rapid?)
• if it's in common utility it should be common, but data structure seems pretty specialized to use case

Other

• possible to dry out endblock and AfterEpochEnd?
• don't interpolate data into error strings
• golang support iterators natively in 1.23 did you consider?
• general weighted pagination maybe better called weightedPagination or something
• x/streamer/keeper/abci_test.go is huge, would it have been possible to use rapid?
• confusing between weight of gauages and weight of iteration

[keruch]

@danwt thanks for the review!

regarding slices/linked lists. my point about re-allocations was that in case of multiple appends the slice can trigger re-allocations (we don't know the number in advance, but it might be 0-100.000 approx), and it might be inefficient at some scale compared to the linked list. in theory it seems reasonable, but in practice apparently it's not :)
i tried to replace a linked list in my cache with a slice and benchmark it. so you are right, slice seems to be more efficient even with a larger number of elements. i guess due to its more hardware-friendly nature and optimizations.

so i used a slice for this cache after all.

possible to dry out endblock and AfterEpochEnd?

don't think so, they have different semantics though they are very similar. i think i've done the best i could

don't interpolate data into error strings

i got your point, tried to avoid it

golang support iterators natively in 1.23 did you consider?

yes i did, they seem over-engineering for me

general weighted pagination maybe better called weightedPagination or something

kept just pagination, but added a detailed comment + renames weight to operations

x/streamer/keeper/abci_test.go is huge, would it have been possible to use rapid?

it's possible, but i'm afraid that writing them is time consuming. i created a task for it to take on in the future #1231

confusing between weight of gauages and weight of iteration

fixed! see my comments above

[danwt]

good work monsieur

[danwt]

lgtm nice work

[zale144]

beautiful

[VictorTrustyDev]

Hi @keruch, I have a question.

Is this PR moved incentives distribution (around 12k transfer as I see on Mainnet) from begin blocker to end blocker?