Feat/meta data native contract #3759

carpawell · 2025-12-26T08:53:59Z

Contract part of #3742.

cthulhu-rider

rebase is needed

cthulhu-rider · 2025-12-26T10:25:08Z

pkg/innerring/internal/metachain/contracts.go

+			panic(fmt.Sprintf("unexpected native contract found: %T", contract))
+		}
+	}
+	return append(newContracts, meta.NewMetadata(neoContract))


meta.NewMetadata does not seem resistant to nil arg. I'd precheck it

in general, looking at for-loop, we dont check absence of any contract. If they are required, i'd still prevent this

added required contracts checks to NewCustomNatives

cthulhu-rider · 2025-12-26T10:28:34Z

pkg/innerring/internal/metachain/contracts.go

+			newContracts = append(newContracts, contract)
+		case *native.Std, *native.Crypto, *native.Oracle, *native.Treasury:
+		default:
+			panic(fmt.Sprintf("unexpected native contract found: %T", contract))


is this done to not miss new contract?

since native.NewDefaultContracts does not state there will be no more contract, i'd say this is an error

A new contract is a significant (and very rare) event, this forces explicit handling of it. Should be ok.

is this done to not miss new contract?

yes, *native.Treasury was even added after this feature was started to be implemented. my idea was that once there is an update in neo-go, tests will start panicking, and we will immediately find out whether we need to adopt a new contract to our side chain or not

cthulhu-rider · 2025-12-29T10:05:49Z

pkg/innerring/internal/metachain/gas/gas.go

+	g := &GAS{
+		initialSupply: init,
+	}
+	defer g.BuildHFSpecificMD(g.ActiveIn())


defer seems redundant

this is the way every native contract is implemented in neo-go, so i suggest to keep it

cthulhu-rider · 2025-12-29T10:09:54Z

pkg/innerring/internal/metachain/contracts/gas/gas.go

+	return nil
+}
+
+// InitializeCache implements the Contract interface.


Suggested change

// InitializeCache implements the Contract interface.

// InitializeCache implements [native.IGAS] interface.

for all methods

cthulhu-rider · 2025-12-29T10:19:08Z

pkg/innerring/internal/metachain/contracts/gas/nep17.go

+}
+
+// makeUint160Key creates a key from the account script hash.
+func makeUint160Key(prefix byte, h util.Uint160) []byte {


unused too? Seems like some stuff is copy-pasted but not needed

cthulhu-rider · 2025-12-29T10:32:41Z

pkg/innerring/internal/metachain/meta/method_netmap.go

+			return fmt.Errorf("%d vector has incorrect REP: %w", i, err)
+		}
+		rep := repB.Int64()
+		if rep > maxREPsClauses {


i'd also check negatives

cthulhu-rider · 2025-12-29T10:36:48Z

pkg/innerring/internal/metachain/meta/method_netmap.go

+	Nodes keys.PublicKeys
+}
+
+func (p Placement) ToStackItem() (stackitem.Item, error) {


used anywhere?

for some reason lost resolution for this thread: #3742 (comment)

is used now

cthulhu-rider · 2025-12-29T10:38:34Z

pkg/innerring/internal/metachain/contracts/meta/method_objects.go

+	}
+	metaInfo, ok := metaInfoSI.Value().([]stackitem.MapElement)
+	if !ok {
+		panic(fmt.Errorf("unexpected deserialized meta information value: expected %T, %T, given", metaInfo, metaInfoSI.Value()))


Suggested change

panic(fmt.Errorf("unexpected deserialized meta information value: expected %T, %T, given", metaInfo, metaInfoSI.Value()))

panic(fmt.Errorf("unexpected deserialized meta information value: expected %T, %T given", metaInfo, metaInfoSI.Value()))

cthulhu-rider · 2025-12-29T10:46:25Z

pkg/innerring/internal/metachain/contracts/meta/method_objects.go

+		}
+
+		if foundSigs == rep {
+			continue nodesLoop


instruction seems excessive here, condition may be inverted

reimplementation of the byte-contract code from the neofs-contracts problems. fixed

cthulhu-rider · 2025-12-29T10:47:59Z

pkg/innerring/internal/metachain/contracts/meta/method_objects.go

+		panic("incorrect vub")
+	}
+	if vub.Int64() <= int64(ic.BlockHeight()) {
+		panic("incorrect vub: exceeded")


runtime values may be helpful in exception

added values everywhere a value is parsed and can be stringified

roman-khimov · 2025-12-30T10:23:34Z

pkg/innerring/internal/metachain/contracts/contracts.go

@@ -0,0 +1,45 @@
+package contracts


I'd strip this directory, can be metachain.

roman-khimov · 2025-12-30T10:27:39Z

pkg/innerring/internal/metachain/contracts.go

+			newContracts = append(newContracts, contract)
+		case *native.Std, *native.Crypto, *native.Oracle, *native.Treasury:
+		default:
+			panic(fmt.Sprintf("unexpected native contract found: %T", contract))


A new contract is a significant (and very rare) event, this forces explicit handling of it. Should be ok.

roman-khimov · 2025-12-30T10:30:18Z

pkg/innerring/internal/metachain/gas/nep17.go

+)
+
+// nep17TokenNative represents a NEP-17 token contract.
+type nep17TokenNative struct {


Wow. This should be imported from NeoGo native, really.

really sorry, but it is not now

This should be imported from NeoGo native

I don't think so. NeoGo's nep17TokenNative contains a lot of code that is specific to native Neo/Gas implementation and that is not needed for the simple meta GAS contract. I'd say that meta GAS doesn't even need nep17TokenNative, just define all methods on the GAS itself.

roman-khimov · 2025-12-30T15:29:17Z

pkg/innerring/internal/metachain/contracts/meta/meta.go

+	m.AddMethod(md, desc)
+
+	desc = native.NewDescriptor("registerMetaContainer", smartcontract.VoidType,
+		manifest.NewParameter("cID", smartcontract.Hash256Type))


roman-khimov · 2025-12-30T15:29:27Z

pkg/innerring/internal/metachain/contracts/meta/meta.go

+	m.AddMethod(md, desc)
+
+	desc = native.NewDescriptor("unregisterMetaContainer", smartcontract.VoidType,
+		manifest.NewParameter("cID", smartcontract.Hash256Type))


Here as well.

roman-khimov · 2025-12-30T16:17:17Z

pkg/innerring/internal/metachain/meta/method_objects.go

+		panic("container does not support chained metadata")
+	}
+
+	sigsVectorsRaw, ok := args[1].Value().([]stackitem.Item)


This follows current container contract, but eventually it can be removed from parameters, nodes can sign transaction directly.

this will require responding there with a signature not of an object's metadata, but with a signature of some predefined transaction. is that what you meant?

Yes. Another option is a custom verification callback.

roman-khimov · 2025-12-30T16:29:31Z

pkg/innerring/internal/metachain/meta/method_objects.go

+		}
+	}
+
+	// required


Why cid is checked separately?

it is used in other checks above (container must exist and be registered as one that supports meta)

I'd rather have similar checks at the same place.

roman-khimov · 2025-12-30T16:31:37Z

pkg/innerring/internal/metachain/meta/method_objects.go

+
+		for _, sig := range sigVectors[i] {
+			for _, node := range placement[i].Nodes {
+				if node.Verify(sig, metaHash) {


i only suggest sorting signatures by public keys. requires support on the IR side and on the SN side

It's not just about sorting. Remember, currently signatures are verified in verify() method of Proxy contract. This scales well. Checking them during transaction execution doesn't.

But sorting is beneficial anyway and easily done.

@roman-khimov, but what if number of nodes is bigger than 16? i found it occasionally when increased number of nodes in tests

do you mean that tx should be signed by nodes? or do you mean that we need Verify() method of the new native contract that should still get signatures from args and verify them (and, i guess, that new transactions should require the contract's signature to call Verify)?

Multiple signatures for a single signer.

@roman-khimov, did some multisignature magic, check, please, if i got you right

roman-khimov · 2025-12-30T16:33:32Z

pkg/innerring/internal/metachain/contracts/meta/method_objects.go

+	}
+	if v, ok := getFromMap(metaInfo, "firstPart"); ok {
+		firstPart, err := v.TryBytes()
+		if err != nil || len(firstPart) != smartcontract.Hash256Len {


sha256 extraction is pretty popular

roman-khimov · 2025-12-30T16:34:59Z

pkg/innerring/internal/metachain/meta/method_objects.go

+	if err != nil {
+		panic(err)
+	}
+


That's it, no processing?

do you mean the secondary index must be in the same chain? my first implementation only needs side chain to check metadata and then notify about it. all the storage nodes handle notification on their own

Among things current structure tries to solve is state synchronization. There is everything needed for it in blockchain, but if you're just throwing an event it all doesn't work. Primary index (based on submitted data) must be built directly on-chain. Then seconday one (header-based) is a different story.

@roman-khimov, ok, adopted the previous metaservice approach. BUT currently there is oID+cID==64 problem for storage keys. i think we will eventually find a "one chain per container" solution, and it will be easier to work with keys, but currently i can only suggest skipping a single byte from oID

Why do you care much about key size limits in native code?

codecov · 2025-12-30T17:34:03Z

Codecov Report

❌ Patch coverage is 0% with 34 lines in your changes missing coverage. Please review.
✅ Project coverage is 25.61%. Comparing base (6285ebd) to head (4844514).

Files with missing lines	Patch %	Lines
pkg/innerring/internal/metachain/contracts.go	0.00%	30 Missing ⚠️
pkg/innerring/internal/blockchain/blockchain.go	0.00%	2 Missing ⚠️
pkg/innerring/internal/metachain/chain.go	0.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #3759      +/-   ##
==========================================
- Coverage   25.63%   25.61%   -0.02%     
==========================================
  Files         657      659       +2     
  Lines       42117    42149      +32     
==========================================
+ Hits        10795    10797       +2     
- Misses      30342    30373      +31     
+ Partials      980      979       -1

☔ View full report in Codecov by Sentry.
📢 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.

Also, implement new MetaData contract and GAS contract implementation without economic. Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>

roman-khimov

Better, but still can be optimized.

roman-khimov · 2026-01-22T09:50:47Z

pkg/innerring/internal/metachain/meta/const.go

+	metaContainersPrefix = iota
+	containerPlacementPrefix
+
+	// object prefixes.


Too many of them. Per-object meta can be stored in addrIndex, then other indexes should be added only if we care about specific lookup abilities (like pick all objects of type X or find if object has any children or is it locked, etc).

so you mean a single kv addr->Serialize(metaMap) and some helper indexes, but NOT every header field we have, like for searching in metabase?

Metabase is very much different because it has generic search capabilities. You don't have this requirement here, you can add indexes needed only for specific tasks. The main feature of this DB in fact is just an OID list. Sizes are nice as well (don't forget to calculate the sum). Lock status and parent-child relations.

roman-khimov · 2026-01-22T09:53:24Z

pkg/innerring/internal/metachain/meta/method_objects.go

+		return fmt.Errorf("expected exactly 1 witness script, got %d", l)
+	}
+
+	_, pks, ok := scparser.ParseMultiSigContract(ic.Tx.Scripts[0].VerificationScript)


This bloats the transaction for no real reason. You have these keys in the contract, @AnnaShaleva can arrange a per-container verification script that would be short and have basically the same functionality (pack sigs and call some contract method to check).

roman-khimov · 2026-01-22T09:56:52Z

pkg/innerring/internal/metachain/meta/method_objects.go

+		panic(fmt.Errorf("cannot retrieve placement vector from stack item: %w", err))
+	}
+
+	err = isSignedBySNs(ic, placement)


With a correct verification script you could just check the signer account and this check better be done first.

how do i know what signer to expect? it can be any node to sign, REP out of the full placement vector. i guess i still dont get you idea. is it about a really big and complex verification script?

It's about small and easy verification script. Once you have CID you know it and you can hash it. #3759 (comment)

i think i got what i had not understood, you suggest placing all the complex logic to CALL meta_check_sigs(cid, [][]sigs) call in the verification script, i didn't have such masterful skills before, i will try

roman-khimov · 2026-01-22T10:05:50Z

pkg/innerring/internal/metachain/meta/method_objects.go

+		return fmt.Errorf("invalid verification script")
+	}
+
+	for i, vector := range placement {


And since you a single multisig you also have a problem parsing things here. Our placement policies are naturally a set of multisigs ("REP 3 + REP 2" is like "3 out of 9 + 2 out of 6"), so you should also naturally have a slice of input parameters here. Placement policies never change (even with nspcc-dev/neofs-api#352 the number of vectors is the same, only required number of signatures may change), so you can a verification script like

PUSH n_vectors PACK CALL meta_check_sigs(cid, [][]sigs)

And an invocation script of

PUSH sig1_1 PUSH sig1_2 PUSH sig1_3 PUSH 3 PACK PUSH sig2_1 PUSH sig2_2 PUSH 2 PACK // keep the order of arrays in mind

And there you have it --- two arrays of signatures in invocation script (not that convenient to sign, true, but all technically possible and not a problem once done), then an array of arrays created by verification script and passed into some contract method.

AnnaShaleva · 2026-01-23T13:52:37Z

pkg/innerring/internal/metachain/gas/gas.go

+
+// GAS represents GAS custom native contract. It always returns [DefaultBalance] as a
+// balance, has no-op `Burn`, `Mint`, `Transfer` operations.
+type GAS struct {


var _ = (native.IGAS)(&GAS{})

AnnaShaleva · 2026-01-23T13:53:12Z

pkg/innerring/internal/metachain/gas/gas.go

+	g := &GAS{
+		initialSupply: init,
+	}
+	defer g.BuildHFSpecificMD(g.ActiveIn())


AnnaShaleva · 2026-01-23T14:00:29Z

pkg/innerring/internal/metachain/gas/gas.go

+}
+
+// NewGAS returns [GAS] custom native contract.
+func NewGAS(init int64) *GAS {


You probably don't even need init argument because I suppose that initial supply won't be castomizable in meta chain, just use constant. You mock balances anyway, so the supply is not so important. I'd keep this mock GAS contract as simple as possible.

AnnaShaleva · 2026-01-23T14:03:47Z

pkg/innerring/internal/metachain/gas/gas.go

+	_, totalSupply := g.getTotalSupply(ic.DAO)
+	if totalSupply.Sign() != 0 {
+		return errors.New("already initialized")
+	}
+	h, err := getStandbyValidatorsHash(ic)
+	if err != nil {
+		return err
+	}
+	g.Mint(ic, h, big.NewInt(g.initialSupply), false)


Why do you ever need this code? You mock all the balances anyway.

@roman-khimov, is it intentionally?

Not likely, can be omitted or can be not (balance < totalSupply is strange, but there are a lot of strange things here).

AnnaShaleva · 2026-01-23T14:07:19Z

pkg/innerring/internal/metachain/gas/nep17.go

+)
+
+// nep17TokenNative represents a NEP-17 token contract.
+type nep17TokenNative struct {


This should be imported from NeoGo native

I don't think so. NeoGo's nep17TokenNative contains a lot of code that is specific to native Neo/Gas implementation and that is not needed for the simple meta GAS contract. I'd say that meta GAS doesn't even need nep17TokenNative, just define all methods on the GAS itself.

AnnaShaleva · 2026-01-23T14:23:44Z

pkg/innerring/internal/metachain/meta/method_netmap.go

+		repB, err := vectorRaw[0].TryInteger()
+		if err != nil {
+			return fmt.Errorf("%d vector has incorrect REP: %w", i, err)
+		}
+		rep := repB.Int64()
+		if rep > maxREPsClauses {


stackitem.ToUint*, that's it.

AnnaShaleva · 2026-01-23T14:24:44Z

pkg/innerring/internal/metachain/meta/method_objects.go

+	cID, err := requiredInMap(metaInfo, "cid").TryBytes()
+	if err != nil || len(cID) != smartcontract.Hash256Len {
+		panic("invalid container ID")
+	}


stackitem.ToUint256, reuse these helpers everywhere below.

AnnaShaleva · 2026-01-23T14:25:26Z

pkg/innerring/internal/metachain/meta/method_objects.go

+			panic(fmt.Sprintf("incorrect object type: %s", err.Error()))
+		}
+		switch typ.Int64() {
+		case 0, 1, 2, 3, 4: // regular, tombstone, storage group, lock, link


for some reason, i forgot we have it in SDK, will fix

AnnaShaleva · 2026-01-23T14:37:45Z

pkg/innerring/internal/metachain/meta/method_objects.go

+	metaInfoSI, err := stackitem.Deserialize(metaInfoRaw)
+	if err != nil {
+		panic(fmt.Errorf("cannot deserialize meta information from byte array: %w", err))
+	}
+	metaInfo, ok := metaInfoSI.Value().([]stackitem.MapElement)
+	if !ok {
+		panic(fmt.Errorf("unexpected deserialized meta information value: expected %T, %T given", metaInfo, metaInfoSI.Value()))
+	}


Is there any usual contract that interacts with serialized metaInfo map? If not, then we may change the serialization format to bytes or raw JSON (like it's done for native ContractManagement's NEF/Manifest arguments of deploy/update), no need to keep stackitem.Map.

AnnaShaleva · 2026-01-23T14:44:13Z

pkg/innerring/internal/metachain/contracts.go

+//   - new native metadata contract (see [meta.NewMetadata] for details).
+func NewCustomNatives(cfg neogoconfig.ProtocolConfiguration) []interop.Contract {
+	var (
+		defaultContracts = native.NewDefaultContracts(cfg)


NACK. Don't do that, don't use native.NewDefaultContracts at all, it won't work correctly because the set of default native contracts uses cross-links to default native Neo/Gas/etc implementations. Instead, follow the https://github.com/nspcc-dev/neo-go/blob/fc4a77d3ead769e0c314c65fff53613718217278/pkg/core/custom_native_test.go#L216 example. Manually initialize those default native implementations that are required, initialize your custom contracts and properly set cross-links for all contracts.

If you need some other default native implementations, then I'll just export them.

Ref. nspcc-dev/neo-go#4145.

it won't work correctly because the set of default native contracts uses cross-links

can you explain in details, please?

Manually initialize those default native implementations that are required, initialize your custom contracts and properly set cross-links for all contracts.

doesnt it look like calling NewDefaultContracts, dropping unused things, and adding new useful?

can you explain in details, please?

For example, native Management uses INEO. Default native Management implementation created by native.NewDefaultContracts will use default native Neo implementation, whereas in your list of contracts you'll replace it with custom Neo. So you need to set internal links to native interfaces properly, like this:
https://github.com/nspcc-dev/neo-go/blob/fc52479b5b7c5bddd7b4b158308c04d72947d459/pkg/core/custom_native_test.go#L234

doesnt it look like calling NewDefaultContracts, dropping unused things, and adding new useful?

No, it's different. Follow the example: https://github.com/nspcc-dev/neo-go/blob/fc52479b5b7c5bddd7b4b158308c04d72947d459/pkg/core/custom_native_test.go#L216

AnnaShaleva · 2026-01-23T14:56:12Z

pkg/innerring/internal/metachain/contracts.go

+
+// NewCustomNatives returns custom list of native contracts for metadata
+// side chain. Returned contracts:
+//   - NEO


@roman-khimov, we don't need default NeoContract for this chain. We don't need voting and rewards at all, is it true?

i think, these lines require me to use smth, i tried to drop NEO contract. changing NEO to smth no-op can be done, i went the easiest way

do you mean to fill it with smth different just to cut out voting and all?

You need to implement custom Neo contract (like you did it with custom Gas). This contract should implement minimal required functionality for meta chain: committee tracking and committee witness checking, that's it (although I'm not sure whether we need dynamic committee at meta chain; let's start with static set of keys). The rest of INEO functionality should be stubbed. Voting should be dropped.

You need to implement custom Neo contract

why i need it?

For committee management.

AnnaShaleva · 2026-01-23T14:58:24Z

pkg/innerring/internal/metachain/contracts.go

+//   - NEO
+//   - Management
+//   - Ledger
+//   - Policy
+//   - Designate
+//   - Notary
+//   - redefined GAS (see [gas.NewGAS] for details)
+//   - new native metadata contract (see [meta.NewMetadata] for details).


Reorder comments in order to match the real order of custom natives. In general, the order of natives is important, it affects OnPersist/PostPersist behaviour.

Required for nspcc-dev/neofs-node#3759. Signed-off-by: Anna Shaleva <shaleva.ann@nspcc.ru>

Required for initial version of Meta native contracts constructor, ref. nspcc-dev/neofs-node#3759. Signed-off-by: Anna Shaleva <shaleva.ann@nspcc.ru>

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>

carpawell requested a review from AnnaShaleva December 26, 2025 08:53

carpawell requested review from End-rey, cthulhu-rider and roman-khimov as code owners December 26, 2025 08:54

cthulhu-rider reviewed Dec 29, 2025

View reviewed changes

roman-khimov reviewed Dec 30, 2025

View reviewed changes

carpawell force-pushed the feat/meta-data-native-contract branch 2 times, most recently from bd68082 to c8ae0ca Compare December 30, 2025 17:31

carpawell force-pushed the feat/meta-data-native-contract branch from c8ae0ca to 1f840ee Compare December 30, 2025 17:36

carpawell force-pushed the feat/meta-data-native-contract branch 2 times, most recently from b7a798d to 13c1f32 Compare January 19, 2026 07:19

carpawell marked this pull request as draft January 19, 2026 08:09

carpawell force-pushed the feat/meta-data-native-contract branch from 13c1f32 to 978b2b0 Compare January 20, 2026 17:43

ir/internal: add possibility to deploy custom native contract

4844514

Also, implement new MetaData contract and GAS contract implementation without economic. Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>

carpawell force-pushed the feat/meta-data-native-contract branch from 978b2b0 to 4844514 Compare January 20, 2026 17:47

roman-khimov reviewed Jan 22, 2026

View reviewed changes

AnnaShaleva requested changes Jan 23, 2026

View reviewed changes

AnnaShaleva reviewed Jan 23, 2026

View reviewed changes

AnnaShaleva added a commit to nspcc-dev/neo-go that referenced this pull request Jan 23, 2026

native: export default native Policy constructor

79c015d

Required for nspcc-dev/neofs-node#3759. Signed-off-by: Anna Shaleva <shaleva.ann@nspcc.ru>

AnnaShaleva added a commit to nspcc-dev/neo-go that referenced this pull request Jan 23, 2026

native: export default native Notary constructor

6e46431

Required for nspcc-dev/neofs-node#3759. Signed-off-by: Anna Shaleva <shaleva.ann@nspcc.ru>

AnnaShaleva mentioned this pull request Jan 23, 2026

Export default native contract constructors nspcc-dev/neo-go#4145

Merged

AnnaShaleva added a commit to nspcc-dev/neo-go that referenced this pull request Jan 26, 2026

native: export default Neo implementation

b66e732

Required for initial version of Meta native contracts constructor, ref. nspcc-dev/neofs-node#3759. Signed-off-by: Anna Shaleva <shaleva.ann@nspcc.ru>

AnnaShaleva mentioned this pull request Jan 26, 2026

native: export default Neo implementation nspcc-dev/neo-go#4146

Merged

wip

00d5d10

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>

	// InitializeCache implements the Contract interface.
	// InitializeCache implements [native.IGAS] interface.

	panic(fmt.Errorf("unexpected deserialized meta information value: expected %T, %T, given", metaInfo, metaInfoSI.Value()))
	panic(fmt.Errorf("unexpected deserialized meta information value: expected %T, %T given", metaInfo, metaInfoSI.Value()))

Feat/meta data native contract #3759

Are you sure you want to change the base?

Feat/meta data native contract #3759

Uh oh!

Conversation

carpawell commented Dec 26, 2025

Uh oh!

cthulhu-rider 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!

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!

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!

carpawell Jan 20, 2026 •

edited

Loading

carpawell Jan 20, 2026 •

edited

Loading