feat: register a polling schedule by anxolin · Pull Request #123 · cowprotocol/composable-cow

anxolin · 2026-07-10T22:33:47Z

Part of #121. Follow uo #116

The polling requires a pre-authorization from the funding account (the registration).

This pre-authorization is done by registering a polling schedule

What

Adds the struct of the register.

The registrations are done using a scheduleId (bytes32). The PR provides a pure function that generates this id deterministically based on the funder, handler, owner and salt.

This pre-register is what will allow in a follow up PR to implement the constraints in the function that does the actual poll.

Test plan

forge test --match-contract ComposableCowPollerTest -vv

Test pass

kaze-cow

there is probably some shady security issue with simply not including the staticInput in the schedule ID, but I need to see the rest of the code put together before I will comment further.

appears there are currently merge conflicts that need fixing

kaze-cow · 2026-07-17T05:54:07Z

+    /// @dev A schedule is uniquely identified by its funder, handler, owner, and salt.
+    struct Schedule {
+        IConditionalOrderGenerator handler; // the conditional-order handler to poll (e.g. the TWAP type)
+        address funder; // source of funds; the only registrant


based solely on this comment, its unclear: in the case of polling for a TWAP order using CoW Shed and an EOA, is the funder the EOA, or the CoWShed?

IIUC, funder can be anyone that approves the Poller contract

kaze-cow · 2026-07-17T05:56:16Z

+    struct Schedule {
+        IConditionalOrderGenerator handler; // the conditional-order handler to poll (e.g. the TWAP type)
+        address funder; // source of funds; the only registrant
+        address owner; // order owner; the pull destination


from this comment I assume it is the CoWShed. I would rename this field to orderOwner and clarify in the comment that this is the CoW Settlement order, as opposed to any other kind of order that may become relevant.

I don't think this brings any value. owner can be anyone that will receive funds in the end

kaze-cow · 2026-07-17T05:58:51Z

+        bytes32 salt; // the conditional order's salt
+        bytes staticInput; // the order's staticInput


comments are kind of redundant. maybe

Suggested change

bytes32 salt; // the conditional order's salt

bytes staticInput; // the order's staticInput

/// Used to set a new poller schedule instance with the same parameters

bytes32 salt;

/// The `staticInput` parameter used for `getTradableOrder` when calling `handler`

bytes staticInput;

kaze-cow · 2026-07-17T05:59:11Z

+contract ComposableCowPoller {
+    /// @notice Parameters for a JIT funding schedule.
+    /// @dev A schedule is uniquely identified by its funder, handler, owner, and salt.
+    struct Schedule {


all of the fields in this struct comments could be docstrings (/// before the field)

kaze-cow · 2026-07-17T06:03:12Z

+    /// @notice Computes the deterministic, appData-independent schedule key.
+    /// @dev `staticInput` is excluded because its appData can depend on this key.
+    ///      Use a different salt for concurrent schedules with the same funder, handler, and owner.
+    /// @param funder The token source.
+    /// @param handler The conditional-order generator.
+    /// @param owner The conditional-order owner.
+    /// @param salt The conditional-order salt.
+    /// @return The schedule key.
+    function scheduleId(address funder, IConditionalOrderGenerator handler, address owner, bytes32 salt)
+        public
+        pure
+        returns (bytes32)
+    {
+        return keccak256(abi.encode(funder, handler, owner, salt));
+    }


Even though staticInput isnt used in the schedule ID, it still would make the most sense here to simply take in the Schedule struct to make it simpler for the front end. Also, its wierd that the order of address funder and IConditionalOrderGenerator handler have been flipped from the original structure, and this could be problematic on the frontend because the type is ultimately address in both cases.

kaze-cow · 2026-07-17T06:06:55Z

+    /// @param schedule The schedule to store.
+    /// @return id The deterministic key of the stored schedule.
+    function register(Schedule calldata schedule) external returns (bytes32 id) {
+        if (msg.sender != schedule.funder) revert OnlyFunder();


its now possible to do

Suggested change

if (msg.sender != schedule.funder) revert OnlyFunder();

require(msg.sender == schedule.funder, OnlyFunder());

While I prefer this style of reverts, this is not a standard adopted in any CoW repo IIRC.

kaze-cow · 2026-07-17T06:07:48Z

+    /// @param id The deterministic key of the schedule.
+    /// @param owner The conditional-order owner and pull destination.
+    /// @param funder The token source that registered the schedule.
+    event ScheduleRegistered(bytes32 indexed id, address indexed owner, address indexed funder);


for completeness it would be good to include all fields of Schedule in the log

The current fields are the most important ones, and adding more could just spike up gas costs at not much additional benefits (e.g. staticInput, handler etc)

kaze-cow · 2026-07-17T06:09:44Z

-///         added in follow-up PRs.
+/// @notice Exercises registering a schedule for a composable TWAP created via `createWithContext`.
 contract ComposableCowPollerTest is BaseComposableCoWTest {
+    uint256 constant PART = 100e18;


this is the amount to trade on each TWAP trade, right? The constant (or at least the naming) actually makes it harder to understand what is going on, so it seems like it would be best to rename this to something like TWAP_PART_AMOUNT, or perhaps remove it entirely.

kaze-cow · 2026-07-17T06:10:40Z

+/// @notice Exercises registering a schedule for a composable TWAP created via `createWithContext`.
 contract ComposableCowPollerTest is BaseComposableCoWTest {
+    uint256 constant PART = 100e18;
+    uint256 constant LIMIT = 1e18;


this value is only used in the _bundle function so low value in having it as a constant here.

Hardcoded magic values are really hard to understand, so I'd keep it as-is

Clarify schedule roles, accept Schedule in scheduleId, and improve TWAP test naming.

This was referenced Jul 10, 2026

Just in time polling of funds #121

Open

feat: Just in time polling of funds (PoC) #116

Closed

anxolin marked this pull request as draft July 10, 2026 22:36

anxolin changed the title ~~feat: register a polling schedule (1/3)~~ feat: register a polling schedule Jul 10, 2026

anxolin force-pushed the feat/poller-base branch from f47098d to f62db1a Compare July 10, 2026 23:12

anxolin force-pushed the feat/poller-register branch from 8fdcbe1 to 4e308b0 Compare July 10, 2026 23:12

anxolin marked this pull request as ready for review July 10, 2026 23:16

anxolin mentioned this pull request Jul 10, 2026

feat: revoke a polling schedule #124

Open

anxolin force-pushed the feat/poller-base branch from f62db1a to 6b6c223 Compare July 11, 2026 00:42

anxolin force-pushed the feat/poller-register branch from 4e308b0 to 6966303 Compare July 11, 2026 00:42

anxolin force-pushed the feat/poller-base branch from 6b6c223 to 6e29a31 Compare July 11, 2026 01:44

anxolin force-pushed the feat/poller-register branch 2 times, most recently from 04c1bde to bf350d1 Compare July 13, 2026 09:17

igorroncevic reviewed Jul 14, 2026

View reviewed changes

Comment thread src/types/ComposableCowPoller.sol Outdated

Comment thread src/types/ComposableCowPoller.sol

Comment thread src/types/ComposableCowPoller.sol

Comment thread test/ComposableCowPoller.t.sol Outdated

Comment thread test/ComposableCowPoller.t.sol Outdated

kaze-cow requested changes Jul 17, 2026

View reviewed changes

anxolin and others added 3 commits July 17, 2026 19:08

feat: register a polling schedule

181744c

chore: more tests, added natspec

039f0e8

fix: address poller registration feedback (#123)

45755b8

Clarify schedule roles, accept Schedule in scheduleId, and improve TWAP test naming.

igorroncevic force-pushed the feat/poller-register branch from 21fe323 to 45755b8 Compare July 17, 2026 17:17

igorroncevic requested a review from kaze-cow July 17, 2026 17:43

		bytes32 salt; // the conditional order's salt
		bytes staticInput; // the order's staticInput

	if (msg.sender != schedule.funder) revert OnlyFunder();
	require(msg.sender == schedule.funder, OnlyFunder());

Uh oh!

Conversation

anxolin commented Jul 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What

Test plan

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

kaze-cow left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

anxolin commented Jul 10, 2026 •

edited

Loading