feat(utxo-swap): add utxo burn output for non-kmd taker fee #2112

dimxy · 2024-05-06T18:34:05Z

Adds a burn output sending 25% of the taker utxo DEX fee to a dedicated pre-burn address. Funds collected on the pre-burn address will be traded for KMD to burn them (thus additionally burning KMD supply).
This PR partially closes #2010

The requirements for this PR: #2269

In this PR:

split dex fee for non-kmd utxo coins (apart from zcoin) and added an output to the pre-burn account
old-style non-split dex fee non-kmd txns are also allowed in validation code (until all taker nodes upgraded) - this feature is removed in favour of version-based burning activation
refactored dex fee pubkey in params: instead of passing dex fee and burn pubkeys in function params new methods dex_pubkey() and burn_pubkey() were added to the SwapOps trait
add version to maker/taker negotiation message and activate burning for the new version
mocktopus was made optional dependency and activated only for development builds (as its doc suggests).
for the burn account no burn part is added (DexFee::Standard used)
sending burn part to the burn account for zcoin
sending burn part to the burn account for tendermint

NOTE: As mocktopus now is marked 'optional = true' in coins Cargo.toml and activated from the mm2_main crate by adding features = ["mocktopus"] in [dev-dependencies] section, you also need to mark your mockable code, called from other crates, this way: #[cfg_attr(feature = "mocktopus", mockable)], otherwise mocks won't work (see samples in code)

TODO:

fix burn pubkey and burn zaddr
disable non-split non-kmd dex fee (no burn output) validation when all taker nodes upgrade to new dex fee splitting
do not pay dex fee if taker is the dex pubkey (for non-privacy utxo)

* dev: docs(README): remove outdated information from the README (#2097) fix(sia): fix sia compilation after hd wallet PR merge (#2103) feat(hd_wallet): utxo and evm hd wallet and trezor (#1962) feat(sia): initial Sia integration (#2086) fix(BCH): deserialize BCH header that uses KAWPOW version correctly (#2099) fix(eth_tests): remove ETH_DEV_NODE from tests (#2101)

fix is_kmd in TestCoin

* dev: feat(tendermint): pubkey-only activation and unsigned tx (#2088) fix(tests): set txfee for some tbtc tests (#2116) fix(eth): remove my_address from sign_and_send_transaction_with_keypair (#2115) fix(utxo-swap): apply events occurred while taker down (#2114) refactor(memory): memory usage improvements (#2098) feat(app-dir): implement root application dir `.kdf` (#2102) fix tendermint fee calculation (#2106) update dockerfile (#2104)

* dev: feat(ETH): eip1559 gas fee estimator and rpcs (#2051) fix(p2pk-tests): fix p2pk tests post merge (#2119) fix(p2pk): show and spend P2PK balance (#2053) fix(swap): use tmp file for swap and order files (#2118)

fix zhtlc send and spend tests

add should_burn_dex_fee method to indicate which coin has burn output

* dev: fix(indexeddb): window usage in worker env (#2131) feat(tx-history): handle encoded transaction values (#2133) fix(core): tendermint withdraws on hd accounts (#2130) fix(core): improve validation rules for table names (#2123) fix(test): improve log wait condition to fix taker restart test (#2125)

…er test add timeout in wait for fee in qrc20 docker test

… docker test failed due to different dex fee values if trait default impl was used)

laruh

Great progress! Here is the first review iteration

mm2src/coins/lp_coins.rs

use same ovk for dex fee and burn outputs

add sanity check for dex_fee in calc_burn_amount_for_op_return

* dev: chore(RPCs): rename `get_peers_info` RPC to `get_directly_connected_peers` (#2195) chore(WASM-builds): remove `wasm-opt` overriding (#2200) fix(coins): add p2p feature to mm2_net dependency (#2210) chore(test): turn on debug assertion (#2204) feat(sia): extract sia lib to external repo (#2167) feat(eth-swap): eth tpu v2 methods, eth docker test enhancements (#2169) fix(cors): allow OPTIONS request to KDF server (#2191) docs(README): update commit badges to use dev branch (#2193) use default value for `komodo_proxy` (#2192) feat(cosmos): komodo-defi-proxy support (#2173)

* dev: fix(orders): fix cancel order race condition using time-based cache (#2232) fix(legacy-swap): taker failed spend maker payment marked as failed (#2199) chore(adex-cli): deprecate adex-cli (#2234) feat(new-RPC): connection healthcheck implementation for peers (#2194) fix(proxy-signature): add message lifetime overflows (#2233) feat(CI): handle remote files in a safer way (#2217) chore(doc): update issue address in README (#2227) fix(merge): remove duplicated db_root function (#2229) feat(wallets): add `get_wallet_names` rpc (#2202) chore(tests): don't use `.wait()` and use `block_on` instead (#2220) fix(native-rpc): remove escaped response body (#2219) fix(clippy): fix coins mod clippy warnings in wasm (#2224) feat(core): handling CTRL-C signal with graceful shutdown (#2213) docs(README): fix typos (#2212) remove the non-sense arguments (#2216) fix(db): stop creating the all-zeroes dir on KDF start (#2218)

* dev: feat(cosmos-offline-tests): prepare IBC channels inside the container (#2246) fix(cosmos): fix tx broadcasting error (#2238) chore(solana): remove solana implementation (#2239) chore(cli): remove leftover subcommands from help message (#2235)

…for tendermint

…de was refactored to enable building with mockable)

- use only dex_fee_from_taker_coin() in swap code - make burn_active param optional (to allow None when only total fee amount is needed) - add DexFeeOption enum to better indicate requested dex fee kind - more dex fee calc tests

* dev: fix(legacy-swap): check for confirmations on recover taker (#2242) fix(legacy-swap): remove the need for takers to confirm their payment (#2249) refactor(P2P): types and modules (#2256) fix(evm): correctly display eth addr in iguana v2 activation result (#2254) feat(utxo): prioritize electrum connections (#1966) refactor(SwapOps): make all methods async (#2251) refactor(SwapOps): make `send_maker_payment` async (#2250) remove old p2p implementation (#2248)

…ddress' error when syncing wallet db

* dev: fix(hd-wallet): use `CoinBalanceMap` for UTXO and QTUM (#2259) fix(tests): add more sepolia endpoints in tests (#2262)

mariocynicys

Thanks for the feat :)
first review iter. gonna need more iteration(s) to draw the full picture.
I get that we use env-vars since we run mm2 as a process in tests thus have less control over it. i think we should invest in a framework to standardize how we do this since i can imagine this will be hard to maintain.

mariocynicys · 2024-11-04T14:07:47Z

mm2src/coins/solana.rs

this is not related, no? or is it a mis-merge with dev?

mariocynicys · 2024-11-04T14:08:00Z

mm2src/coins/solana/spl.rs

... and this?

mariocynicys · 2024-11-04T17:04:01Z

mm2src/coins/z_coin/z_htlc.rs

+    let mut outputs = vec![dex_fee_out];
+    if let Some(dex_burn_amount) = dex_fee.burn_amount() {
+        let dex_burn_amount_sat =
+            sat_from_big_decimal(&Into::<BigDecimal>::into(dex_burn_amount), coin.utxo_arc.decimals)?;


We seem to allow DexFee::NoFee here. Is this intentional?

also nit: Into::<BigDecimal>::into(dex_burn_amount) -> dex_burn_amount.to_decimals()

fixed thanks

mariocynicys · 2024-11-04T17:46:26Z

mm2src/coins/Cargo.toml

@@ -72,7 +72,7 @@ mm2_number = { path = "../mm2_number"}
 mm2_p2p = { path = "../mm2_p2p" }
 mm2_rpc = { path = "../mm2_rpc" }
 mm2_state_machine = { path = "../mm2_state_machine" }
-mocktopus = "0.8.0"
+mocktopus = { version = "0.8.0", optional = true }


maybe we could add it as a for-test feature.

I think the way like it is now is convenient as we may use mocktopus both from tests inside the crate and from other crates, without extra features, and at the same time avoiding building it into releases

from tests inside the crate -> dev-deps section
from tests from outside/integration -> for-test flag to activate it, this has the downside though as u said will enable all and every extra test features we have, but it's a for-test build anyway so i wouldn't worry much about that and instead prioritize having less complex cfgs around the code base.

from tests from outside/integration -> for-test flag to activate it

I think, in this case mocktopus would be built into release what is not recommended for performance reasons.
The optional flag is a way to avoid this.

mariocynicys · 2024-11-04T18:14:18Z

mm2src/coins/lp_coins.rs

+        #[cfg(feature = "for-tests")]
+        {
+            unsafe {
+                if let Some(ref test_pk) = TEST_DEX_FEE_ADDR_RAW_PUBKEY {
+                    return test_pk.as_slice();
+                }
+            }
+        }


why have this if we can mock it

I think this would not work in this case, as this code is executed in another mm2 process (iirc I even tried this)

mariocynicys · 2024-11-05T10:59:39Z

mm2src/mm2_main/src/lp_swap/maker_swap.rs

+        #[cfg(feature = "run-docker-tests")]
+        if let Ok(env_pubkey) = std::env::var("TEST_DEX_FEE_ADDR_RAW_PUBKEY") {
+            unsafe {
+                TEST_DEX_FEE_ADDR_RAW_PUBKEY = Some(hex::decode(env_pubkey).expect("valid hex"));
+            }
+        }


can we have these only where we happen to use the variable TEST_DEX_FEE_ADDR_RAW_PUBKEY? and maybe remove the variable then.
we might also want to group these test-related env vars in some file and assign them to constants to not lose track of them or have them dangling.

I think, this var is needed because in docker swap tests we start new taker and maker mm2 processes and in couple tests I need to run taker with a test dex pubkey. I guess this is okay as there are other cases in the codebase using env vars to configure tests.
(I considered creating a feature for that but then we would need an extra cargo test call to run such tests in CI)

i meant the static variable not the environment one

Ofc I'd like to get rid of TEST_DEX_FEE_ADDR_RAW_PUBKEY static var too but for that I'd have to change dex_pubkey() return type (it's static ref now), did not like to do it just for tests sake

laruh

huge progress! I would like to clarify some moments

laruh · 2024-11-05T14:48:58Z

mm2src/mm2_main/src/lp_swap/maker_swap.rs

+/// Determine version from negotiation data if saved swap data does not store version
+/// (if the swap started before the upgrade to versioned negotiation message)
+/// In any case it is very undesirable to upgrade mm2 when any swaps are active
+fn fix_taker_version(negotiation_data: &TakerNegotiationData) -> u16 { negotiation_data.taker_version.unwrap_or(0) }


may be get_taker_version? fix_ in the naming feels like it should solve some issue. But as I can see the function returns taker version from data

Same note for

fn fix_maker_version(negotiation_data: &MakerNegotiationData) -> u16 { negotiation_data.maker_version.unwrap_or(0) }

changed both names, thanks

laruh · 2024-11-05T15:07:06Z

mm2src/mm2_main/src/lp_swap/maker_swap.rs

+        let mut msgs = vec![];
+
+        // Important to try the new versioned negotiation message first
+        if cfg!(not(feature = "for-tests")) || !env_var_as_bool("USE_NON_VERSIONED_MAKER") {
+            let maker_versioned_negotiation_msg = SwapMsgExt::NegotiationVersioned(NegotiationDataMsgVersion {
+                version: SWAP_PROTOCOL_VERSION,
+                msg: negotiation_data.clone(),
+            });
+            msgs.push((
+                swap_ext_topic(&self.uuid),
+                SwapMsgWrapper::Ext(maker_versioned_negotiation_msg),
+            ));
+        }
+
+        let maker_old_negotiation_msg = SwapMsg::Negotiation(negotiation_data);
+        msgs.push((
+            swap_topic(&self.uuid),
+            SwapMsgWrapper::Legacy(maker_old_negotiation_msg),
+        ));


could you clarify, are you pushing both versioned negotiation msg and legacy negotiation msg in production code?
could you tell why do we need this?

We need this for the upgrade period: first try if the remote node understands the old (default) version, if yes, fallback to older negotiation protocol and turn off new features

mariocynicys · 2024-11-06T20:47:12Z

mm2src/coins/lp_coins.rs

+            let burn_amount = dex_fee - min_tx_amount;
+            (min_tx_amount.clone(), burn_amount)
+        } else {
+            (dex_fee.clone(), MmNumber::from(0))


shouldn't this be (min_tx_amount.clone(), 0), if we reach this case then we know dex_fee is dust already.

dex_fee is not dust in both calc_burn_amount_for_burn_account() and calc_burn_amount_for_op_return(): it was checked in the caller fn.

mariocynicys · 2024-11-06T20:55:53Z

mm2src/coins/lp_coins.rs

+        // Default case where burn_amount is considered dust
+        (dex_fee.clone(), 0.into())


similar to the case above, we don't know for sure here if dex_fee is dust or not though.

mariocynicys · 2024-11-06T22:12:59Z

mm2src/coins/lp_coins.rs

+        let new_fee = dex_fee * &MmNumber::from(Self::DEX_FEE_SHARE);
+        let burn_amount = dex_fee - &new_fee;
+        if &new_fee >= min_tx_amount && &burn_amount >= min_tx_amount {
+            // Use the max burn value, which is 25%. Ensure burn_amount is not dust
+            return (new_fee, burn_amount);
+        }
+        // If the new dex fee is dust set it to min_tx_amount and check the updated burn_amount is not dust.
+        let burn_amount = dex_fee - min_tx_amount;
+        if &new_fee < min_tx_amount && &burn_amount >= min_tx_amount {
+            // actually currently burn_amount (25%) < new_fee (75%) so this never happens. Added for a case if 25/75 will ever change
+            return (min_tx_amount.clone(), burn_amount);
+        }


i find matches usually good for when the conditions get complicated:

match (dex_cut > min_tx_amount, burn_cut > min_tx_amount) { (true, true) => (dex_cut, burn_cut), (true, false) => (both_cuts, 0), (false, true) => (both_cuts, 0), (false, false) => (max(both_cuts, min_tx_amount), 0) }

Hmm, looks a bit unusual for me.
I guess, expressions like traditional boolean algebra logic are more comprehensible?

mariocynicys · 2024-11-07T09:56:36Z

mm2src/mm2_main/src/lp_swap/maker_swap.rs

+        #[cfg(feature = "run-docker-tests")]
+        if let Ok(env_pubkey) = std::env::var("TEST_DEX_FEE_ADDR_RAW_PUBKEY") {
+            unsafe {
+                TEST_DEX_FEE_ADDR_RAW_PUBKEY = Some(hex::decode(env_pubkey).expect("valid hex"));
+            }
+        }


i meant the static variable not the environment one

mariocynicys · 2024-11-07T10:11:01Z

mm2src/coins/lp_coins.rs

+
+        if taker_coin.is_kmd() {
+            // use a special dex fee option for kmd
+            let (fee_amount, burn_amount) = Self::calc_burn_amount_for_op_return(&dex_fee, &min_tx_amount);


if we came to find the burn_amount = 0 in this branch we should return dex fee standard (just like the non-kmd counterpart), otherwise the code down the call tree will generate a tx with a zero op_return burn and spike the miner fee (tx size) for no reason.

tbh i think we should merge Standard & WithBurn to something general that has both dex and burn fee and to always check the burn (and dex) fee aren't zero before generating their output (legacy standard & dust burn would both have 0 burn fee).

We can also use non-zero-u* for the burn field to make sure it's always set (if we don't wanna go to the route of merging both variants & checking for 0 at tx generation).

if we came to find the burn_amount = 0 in this branch we should return dex fee standard

This changes already working code. I guess we could do this only as version activated feature

tbh i think we should merge Standard & WithBurn to something general

Eventually we will eliminate the Standard option (when all nodes are upgraded), so in fact this will be done.
For now I suggest to retain it as it would be easier to eliminated the code related to Standard.

mariocynicys · 2024-11-07T10:55:58Z

mm2src/coins/swap_features.rs

+
+    pub fn is_active(feature: SwapFeature, version: u16) -> bool {
+        if let Some(found) = Self::SWAP_FEATURE_ACTIVATION.iter().find(|fv| fv.1 == feature) {
+            return version >= found.0;


this always assumes that a remote version greater than activation version will always activation such a feature (but might we not disable it?). this is not the most extensible.

e.g.: what if we activated some feature in version 3 and disabled it in version 5, a version 3/4 peer will think the feature is active since 5 >= 3.

added deactivation in 05feb72 (may be of use indeed)

mariocynicys · 2024-11-07T11:11:15Z

mm2src/mm2_main/src/lp_swap/maker_swap.rs

+        let remote_version = self
+            .r()
+            .data
+            .taker_version
+            .ok_or("No swap protocol version".to_owned())?;
+        let is_burn_active = SwapFeature::is_active(SwapFeature::SendToPreBurnAccount, remote_version);
+        let dex_fee = dex_fee_from_taker_coin(
+            self.taker_coin.deref(),
+            &self.r().data.maker_coin,
+            &taker_amount,
+            Some(self.r().other_taker_coin_htlc_pub.to_vec().as_ref()),
+            Some(is_burn_active),
+        );


i don't think this would be dev-friendly to upgrade to the next feature set. I like the versioning mechanism and think we should have one, but also think it should be abstracted away in each entities impl (here DexFee).

so how i imagine something like this would be is: taker and maker settle on a version together (deterministically, pick the min version for e.g. or have a deterministic mapping rules), and we pass that version to whoever needs it to apply some version specific rules (in here this is the DexFee, so something like DexFee::using_version(version, taker_coin, maker_coin, etc...) and inside we match against the version and apply the appropriate rule),

I started to do this but realised there is a problem here:
We support both legacy and TPU protocols, each of both may have own versioning. So it looks like it's better to deal with versions outside the lower level components like DexFee (which is called both from legacy and TPU), don't you think?

mariocynicys · 2024-11-07T11:12:55Z

mm2src/mm2_main/src/lp_swap/maker_swap.rs

+            .r()
+            .data
+            .taker_version
+            .ok_or("No swap protocol version".to_owned())?;


shouldn't this be unwrap_or(0)? or maybe just not let this be an option and default it to 0 right away.

The reason why this taker_version is optional is because it is discovered on the negotiation phase.
I think it is not correct to set it to some specific version on the swap start.
Also, if it is not defined at this stage (wait_taker_fee) - I think this is a clear logic error and probably also not very good to default it to 0 or another version (we should determine it explicitly on the negotiation phase).

mariocynicys · 2024-11-07T11:19:24Z

mm2src/mm2_main/src/lp_swap.rs

    TakerPayment(Vec<u8>),
 }

+// Extension to SwapMsg with version added to negotiation exchange
+#[derive(Clone, Debug, Eq, Deserialize, PartialEq, Serialize)]
+pub enum SwapMsgExt {
+    NegotiationVersioned(NegotiationDataMsgVersion),
+    NegotiationReplyVersioned(NegotiationDataMsgVersion),
+}


why not merge the new msgs into the old struct (and have a comment separating this is version x)?

I did this on purpose: the old struct should be intact for old nodes to be able to parse it.
(I tried to test this though, in principle it could be tweaked to allow the merged struct to be parsed but I chose not to do this cause it involves knowledge about internal lib behaviour)

mariocynicys · 2024-11-07T11:25:49Z

mm2src/coins/Cargo.toml

@@ -72,7 +72,7 @@ mm2_number = { path = "../mm2_number"}
 mm2_p2p = { path = "../mm2_p2p" }
 mm2_rpc = { path = "../mm2_rpc" }
 mm2_state_machine = { path = "../mm2_state_machine" }
-mocktopus = "0.8.0"
+mocktopus = { version = "0.8.0", optional = true }


from tests inside the crate -> dev-deps section
from tests from outside/integration -> for-test flag to activate it, this has the downside though as u said will enable all and every extra test features we have, but it's a for-test build anyway so i wouldn't worry much about that and instead prioritize having less complex cfgs around the code base.

revert preburn_account_active var (fix burning kmd coins by non-upgraded nodes)

dimxy added 9 commits April 27, 2024 14:33

make mockable optional

f3625e5

add support for utxo coin burn account to dex fee

d3f843d

remove left over dup dex_fee_rate()

7894f7d

fix coins crate mocktopus dependency for mm2_main crate tests

2984c67

fix dex fee changed due to new burn output in couple of lp_swap tests

128c4b4

fix is_kmd in TestCoin

fix swap v2 test non-kmd burn output

23f2a08

fix fmt

2c810b0

remove comment

af45b71

dimxy changed the title ~~Add utxo burn output for non-kmd taker fee~~ feat(utxo-swap): add utxo burn output for non-kmd taker fee May 6, 2024

dimxy added in progress Changes will be made from the author under review and removed in progress Changes will be made from the author labels May 6, 2024

dimxy assigned shamardy May 19, 2024

Merge branch 'dev' into burn-dex-fee-for-non-kmd

0f34d2e

* dev: feat(ETH): eip1559 gas fee estimator and rpcs (#2051) fix(p2pk-tests): fix p2pk tests post merge (#2119) fix(p2pk): show and spend P2PK balance (#2053) fix(swap): use tmp file for swap and order files (#2118)

dimxy added in progress Changes will be made from the author and removed under review labels Jun 9, 2024

dimxy added 9 commits June 14, 2024 20:54

fix zombie chain scan block num

d53dc37

fix zhtlc send and spend tests

add burn output for privacy coin,

81263e0

add should_burn_dex_fee method to indicate which coin has burn output

use custom log! instead debug! (as env_logger disabled) in qrc20 dock…

9331e40

…er test add timeout in wait for fee in qrc20 docker test

fix to make should_burn_dex_fee explicitly in each coin (to fix qrc20…

b343275

… docker test failed due to different dex fee values if trait default impl was used)

restore original test log message

587b1ae

add test for std dex fee for zcoin

58d84cf

remove unused cache dir var in z_coin init block

1243935

fix clippy

bc00ee8

laruh requested changes Jun 18, 2024

View reviewed changes

dimxy added 2 commits June 19, 2024 13:05

fix burn addr in ZOutput

4d59545

use same ovk for dex fee and burn outputs

refactor dex fee code;

dc9e710

add sanity check for dex_fee in calc_burn_amount_for_op_return

dimxy added 14 commits September 13, 2024 18:36

add dex fee burning for tendermint coins

f95b91e

fix should_burn_dex_fee to true and burn_pubkey for validate dex fee …

e19b738

…for tendermint

add mockable support for TendermintCoin (note: some TendermintCoin co…

446b3db

…de was refactored to enable building with mockable)

refactor validate_taker_fee_test with mocked txns

8aa7a06

add test to validate taker fee tx with burn for tendermint

53a553f

add test to calc DexFee with burn for nucleus ticker

2f47bdc

Refactor dex fee code:

a61d4a0

- use only dex_fee_from_taker_coin() in swap code - make burn_active param optional (to allow None when only total fee amount is needed) - add DexFeeOption enum to better indicate requested dex fee kind - more dex fee calc tests

fix await in zhtlc tests

29ad460

add TODO zhtlc feature

2ea8b9f

fix fmt

ab4f2a5

laruh mentioned this pull request Nov 4, 2024

fix(tpu-v2): fix tpu-v2 wait for payment spend and extract secret #2261

Open

dimxy added 5 commits November 4, 2024 22:11

add comments how to run zcoin native tests

8711131

add print into tests with hint how to sync wallet db

bf65293

add debug logging for updating zcoin wallet db

5f0d3d6

add scan interval in zcoin tests to prevent 'Can't assign requested a…

b3eccf8

…ddress' error when syncing wallet db

Merge branch 'dev' into burn-dex-fee-for-non-kmd

d5b9bbd

* dev: fix(hd-wallet): use `CoinBalanceMap` for UTXO and QTUM (#2259) fix(tests): add more sepolia endpoints in tests (#2262)

mariocynicys reviewed Nov 5, 2024

View reviewed changes

laruh reviewed Nov 5, 2024

View reviewed changes

dimxy added 2 commits November 7, 2024 14:56

fix review notes

b92d065

fix build err

601519d

mariocynicys reviewed Nov 7, 2024

View reviewed changes

dimxy added 5 commits November 7, 2024 20:22

add swap feature deactivation by version

05feb72

improve comments for protocol version

ce1a5e0

fix burn_pubkey as taker in DexFee::NoFee;

dcfdbbd

revert preburn_account_active var (fix burning kmd coins by non-upgraded nodes)

fix fmt

dfb9673

fix tests with burn key as taker

2559e0e

		// Default case where burn_amount is considered dust
		(dex_fee.clone(), 0.into())

feat(utxo-swap): add utxo burn output for non-kmd taker fee #2112

Are you sure you want to change the base?

feat(utxo-swap): add utxo burn output for non-kmd taker fee #2112

Conversation

dimxy commented May 6, 2024 • edited Loading

laruh left a comment

Choose a reason for hiding this comment

mariocynicys left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

dimxy Nov 7, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

laruh left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

dimxy commented May 6, 2024 •

edited

Loading

dimxy Nov 7, 2024 •

edited

Loading