fix(events): cap per-event hackathon submissions to bound storage growth

[armandocodecr]

Summary

Hackathon events have needs_application = false, so any address could call submit() and create a persistent EventSubmission entry with an arbitrarily long content_uri, with no cap on the number of distinct submissions. This let an attacker spam submit() from many fresh addresses to grow the contract's persistent state and rent burden without bound. This adds a per-event submission counter with a cap, and a length bound on content_uri.

Closes the storage bloat issue surfaced by the Almanax scan of e1f1793 (severity medium, boundless-events, event_ops.rs submit).

Closes #71

Changes

• New DataKey::EventSubmissionCount(u64), appended at the end of the enum to preserve existing key discriminants.
• New storage::submission_count and storage::append_submission, mirroring the existing append_contributor/append_applicant pattern: reserve the slot against the cap before writing, and treat an existing submission as a no-op (re-submission updates in place and must not recount).
• storage::remove_submission now decrements the counter, so withdrawing a submission frees its slot for a future submitter instead of counting against the cap forever.
• New MAX_SUBMISSIONS_PER_EVENT (5,000, matching the existing applicant/contributor caps) and MAX_CONTENT_URI_LEN (256) constants in event_ops.rs.
• submit() now rejects an oversized content_uri and reserves a submission slot before writing the entry.
• New error variant Error::TooManySubmissions (44). For the content_uri length check, I reused Error::TitleTooLong rather than adding a second new variant, since the contracterror enum was already at 49 of the 50-variant cap (one new variant was all I could add). This project already has documented precedent for exactly this tradeoff, see BACKLOG.md's L7 entry, which reused Error::InvalidPillar for an unrelated check for the same reason.

Testing

Added to contracts/events/src/tests/hackathon_pillar.rs:

• submit_beyond_cap_reverts: fast-forwards the per-event counter directly to the cap (rather than performing 5,000 real submissions) and confirms the next submit() reverts with Error::TooManySubmissions.
• submit_oversized_content_uri_reverts: a content_uri one byte over MAX_CONTENT_URI_LEN reverts.
• resubmit_by_existing_applicant_does_not_increment_submission_count: submitting twice from the same address keeps the counter at 1.
• withdraw_submission_frees_the_slot_for_future_submitters: withdrawing brings the counter back to 0.

Ran locally:

• cargo test --release (matches this repo's CI exactly): 267 passed, 0 failed (201 in boundless-events, 66 in boundless-profile).
• cargo fmt --all -- --check: clean.
• make build in contracts/events (stellar contract build): compiles, boundless_events.wasm is 60,491 bytes, under the CI's 64 KB ceiling with about 5 KB of headroom left.

I did not run cargo clippy as a gate since it isn't part of this repo's CI (verify-build.yml only builds, checks WASM size, and runs cargo test --release), and CONTRIBUTING.md's documented local check is just cargo test.

Summary by CodeRabbit

• New Features
  • Added per-event submission cap and maximum allowed content_uri length.
  • Duplicate submissions by the same applicant now update the existing entry without consuming additional capacity.
• Bug Fixes
  • Submissions beyond the cap now fail with a clear “too many” error.
  • Oversized content_uri values now fail with a “title too long” error.
  • Withdrawing (and removing nonexistent entries) no longer breaks capacity tracking; freed slots become available again.
• Tests
  • Extended coverage for submission counter behavior and error codes.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: e2ee0153-a479-4c94-b058-b677895b0467

📥 Commits

Reviewing files that changed from the base of the PR and between 12ace32 and 6c0258c.

📒 Files selected for processing (5)

• contracts/events/src/errors.rs
• contracts/events/src/event_ops.rs
• contracts/events/src/storage.rs
• contracts/events/src/tests/hackathon_pillar.rs
• contracts/events/src/types.rs

🚧 Files skipped from review as they are similar to previous changes (4)

• contracts/events/src/types.rs
• contracts/events/src/storage.rs
• contracts/events/src/event_ops.rs
• contracts/events/src/tests/hackathon_pillar.rs

---

📝 Walkthrough

Walkthrough

Hackathon submissions now enforce a 5,000-submission per-event cap and a 256-character content URI limit. Persistent submission counts support idempotent resubmissions and slot reclamation after withdrawal, with tests covering the new behaviors.

Changes

Submission limit enforcement

Layer / File(s)	Summary
Submission count storage contracts contracts/events/src/types.rs, contracts/events/src/errors.rs, contracts/events/src/storage.rs	Adds the persistent event submission count key, capped/idempotent reservation, count reads, TTL refreshes, and count cleanup on withdrawal.
Submit validation and reservation contracts/events/src/event_ops.rs	Adds submission and URI limits, rejects oversized content URIs, and reserves submission capacity before writing.
Submission behavior tests contracts/events/src/tests/hackathon_pillar.rs	Tests cap enforcement, URI length validation, withdrawal slot release, and count-preserving resubmissions.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Sequence Diagram(s)

sequenceDiagram
  participant Applicant
  participant EventOps
  participant Storage
  Applicant->>EventOps: submit content_uri
  EventOps->>EventOps: validate URI length
  EventOps->>Storage: append_submission(event_id, applicant, cap)
  Storage-->>EventOps: reserve slot or return cap error
  EventOps->>Storage: persist submission

Loading

Possibly related PRs

• boundlessfi/boundless-contract#46: Covers related Hackathon submit, withdraw, and resubmission tests.

Suggested reviewers: 0xdevcollins

Poem

I’m a rabbit guarding the gate,
Counting five thousand—then I wait.
Long links tumble into the bin,
Returning hares may hop back in.
Withdrawn slots bloom anew—
Bounded storage, carrots too!

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 75.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the main fix: capping hackathon submissions to bound storage growth.
Linked Issues check	✅ Passed	It adds per-event submission caps, URI length checks, idempotent resubmissions, and tests matching #71.
Out of Scope Changes check	✅ Passed	All changes support submission capping, storage tracking, validation, or tests for #71.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[coderabbitai]

Actionable comments posted: 2

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@contracts/events/src/event_ops.rs`:
- Around line 25-26: Replace the global MAX_SUBMISSIONS_PER_EVENT and
MAX_CONTENT_URI_LEN constants with fields on EventRecord or its event variant
payload, and update all submission-count and content-URI validation to read
those per-event values. Ensure event creation/configuration initializes the
fields so each event can set its own limits, preserving the existing defaults
where required.

In `@contracts/events/src/storage.rs`:
- Around line 413-422: Update remove_submission to first check
get_submission(env, id, applicant) and return immediately when no submission
exists; only then remove the EventSubmission key and decrement or remove the
per-event count, preserving the existing count-update behavior for present
submissions.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 05063aad-1b20-43dc-bcfc-3327a6347985

📥 Commits

Reviewing files that changed from the base of the PR and between 188bc71 and 12ace32.

📒 Files selected for processing (5)

• contracts/events/src/errors.rs
• contracts/events/src/event_ops.rs
• contracts/events/src/storage.rs
• contracts/events/src/tests/hackathon_pillar.rs
• contracts/events/src/types.rs

[0xdevcollins]

@armandocodecr Please fix conflict so i can merge

[almanax-ai]

Quota reached

Your plan allows 300 CI/CD file units per month. You've used 289 and this scan would add 28 more (total: 317).

[armandocodecr]

@0xdevcollins Done