Skip to content

fix: stop crediting the fulfiller with offer items sent to someone else - #978

Closed
Xowiek wants to merge 1 commit into
ProjectOpenSea:mainfrom
Xowiek:fix/forwarded-offer-balance-check
Closed

Xowiek wants to merge 1 commit into
ProjectOpenSea:mainfrom
Xowiek:fix/forwarded-offer-balance-check

Conversation

@Xowiek

@Xowiek Xowiek commented Aug 22, 2026

Copy link
Copy Markdown
Contributor

Symptom

Accepting a bid and forwarding the proceeds with recipientAddress produces a fulfillment the SDK is happy with and the chain is not:

reverted with panic code 0x11 (Arithmetic operation overflowed ...)

Nothing in that points at the cause.

Why

validateStandardFulfillBalancesAndApprovals adds the offered items to the fulfiller's balances before it checks consideration. Requirement 2 in its own doc comment spells out the reasoning — the fulfiller "does not need to own the ERC20 item as it will first be received from the offerer".

True, so long as the offer reaches the fulfiller. recipientAddress sends it elsewhere, and the check never learns that: fulfillStandardOrder and fulfillAvailableOrders both accept recipientAddress, neither passes it down. The credit gets applied regardless.

A bid makes it concrete. Offerer bids 10 ERC20 with a 2.5% fee, so the fulfiller takes in 10 and pays 0.25 back out of it — fine while they keep the proceeds. Send those to a third party and the 0.25 has to come from the fulfiller's own balance. Holding none, they get the revert above instead of a balance error.

Change

Apply the credit only when the offer lands with the fulfiller, and thread fulfillerAddress down so the check can tell.

The zero address means the offer stays with the fulfiller. Naming the fulfiller explicitly amounts to the same thing — I checked on chain and that case really does settle — so it keeps the credit too, rather than treating any non-zero recipient as a forward.

What still works

Two of the three new tests in gifting.spec.ts pass on main as well, deliberately — they pin the behaviour this shouldn't disturb:

  • fee paid out of the proceeds when the fulfiller keeps them → still fulfills, balance lands at 9.75
  • recipientAddress set to the fulfiller's own address → same

The third, forwarding to a third party with an empty balance, fails on main (the call is fulfilled rather than rejected) and now raises "The fulfiller does not have the balances needed".

Suite 212.

validateStandardFulfillBalancesAndApprovals adds the offered items to the
fulfiller's balances before checking consideration, per requirement 2 in its
own doc comment: the fulfiller "does not need to own the ERC20 item as it
will first be received from the offerer".

That only holds when the offer actually reaches the fulfiller.
recipientAddress forwards it somewhere else, and the check never sees that
value -- fulfillStandardOrder and fulfillAvailableOrders both take
recipientAddress but never pass it down. The credit is applied either way.

Accepting a bid is where it shows. The offerer bids 10 ERC20 with a 2.5%
fee, so the fulfiller receives 10 and settles 0.25 out of it. Hand the
proceeds to a third party with recipientAddress and the fee has to come from
the fulfiller's own balance instead. With a zero balance the SDK reports no
problem and hands back a fulfillment that reverts on transfer with panic
0x11, which says nothing about what went wrong.

Only apply the credit when the offer lands with the fulfiller. The zero
address means it stays with them, and naming the fulfiller explicitly is the
same thing, so both keep the existing behaviour -- checked on chain, that
second case really does settle.
@codecov-commenter

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.97%. Comparing base (f618d19) to head (bb76394).
⚠️ Report is 368 commits behind head on main.

Additional details and impacted files
@@            Coverage Diff             @@
##             main     #978      +/-   ##
==========================================
+ Coverage   98.27%   98.97%   +0.69%     
==========================================
  Files          35       50      +15     
  Lines       14526    18565    +4039     
  Branches      660      988     +328     
==========================================
+ Hits        14276    18375    +4099     
+ Misses        245      183      -62     
- Partials        5        7       +2     

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

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

ryanio added a commit that referenced this pull request Aug 22, 2026
#978 stopped the per-order check crediting the fulfiller with offer items
that recipientAddress sends elsewhere. #984 added a batch-wide check in the
same release, and it credits the batch's whole offer unconditionally, so the
same class of shortfall is still accepted on the batch path.

A fulfiller holding 0.3 ERC20 who accepts two bids with a 250bp fee each
owes 0.5 with the proceeds forwarded away. Each order passes the per-order
check alone, and the batch check clears the total against a credit for
20 ERC20 the fulfiller never receives. It reverts on transfer.
@ryanio

ryanio commented Aug 22, 2026

Copy link
Copy Markdown
Collaborator

Good find, and thank you for tracing it back to requirement 2 in the doc comment rather than just patching the symptom. A validation pass that clears a fulfillment which then reverts with panic 0x11 is about the least useful failure mode available.

Keeping the credit when recipientAddress names the fulfiller explicitly is right, and I appreciate that you checked it onchain rather than assuming. Worth noting for later readers that this makes offerItemsLandWithFulfiller slightly more permissive than the existing isRecipientSelf check used to pick the basic fulfill route, which only treats the zero address as self. That is fine, they are answering different questions, but they now differ.

The two control tests that pass on main as well are the reason I trust this one. Reverting the source leaves those two green and fails only the third, on "expected promise to be rejected ... but it was fulfilled".

This has landed in #985 with seven other contributions from the same day. Closing in favour of that one.

One follow-up your change surfaced, for visibility: #984 landed in the same release and added a batch-wide balance check that credits each order's offer unconditionally, with no knowledge of recipientAddress. So the batch path still had the hole you closed on the single-order path. #986 fixes it by reusing your offerItemsLandWithFulfiller helper.

@ryanio ryanio closed this Aug 22, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants