accept_0conf: uncomment code to show allocations bug

[zoedberg]

To be merged once we have a fix for this. Run cargo test accept_0conf to see the issue:

thread 'accept_0conf' panicked at tests/utils/helpers.rs:947:17:
assertion `left == right` failed
  left: 2
 right: 1

i.e. the sender sees 2 allocations (the issued amount and the change amount) instead of just 1 (the change amount)

[dr-orlovsky]

It seems the issue is not related to 0 conformations, but instead the sender sees initial issuance which has no witness transaction at all (since it is in genesis). Am I right?

Also it might be that the issuance is reported since it is a public state. Are there other tests which check that the issuance is not reported if not owned by a wallet?

[zoedberg]

It seems the issue is not related to 0 conformations, but instead the sender sees initial issuance which has no witness transaction at all (since it is in genesis). Am I right?

I don't think so. The issuance doesn't need to do a new witness transaction but it requires an existing UTXO that the wallet sees as owned and should see as unowned when it spends that. The issue is related to 0 confirmations, if you add a mine like so

    let (consignment, _) = wlt_1.transfer(invoice.clone(), None, None);

    wlt_2.accept_transfer(consignment.clone());

+   mine(false);
    wlt_1.sync();
    wlt_1.check_allocations(
        contract_id,
        &iface_type_name,
        AssetSchema::Nia,
        vec![issue_supply - amt],
        false,
    );

the test passes

Also it might be that the issuance is reported since it is a public state. Are there other tests which check that the issuance is not reported if not owned by a wallet?

As just shown, if we mine, the issuance is not reported anymore as owned. Also it wouldn't make sense, we're asking for owned allocations, not for public state.

[dr-orlovsky]

i.e. the sender sees 2 allocations (the issued amount and the change amount) instead of just 1 (the change amount)

Unless the transaction is mined, in order to enable RBFs, we need to show that allocation. Thus, I think the test should be rather changed than we need to change the way allocations are presented.

The wallet receives witness txid as a part of the allocation information. With that it can use ContractState::witness_info to access the mining status and filter out mempool transactions if needed for the UX

[zoedberg]

The tests are showing the allocations in the same way the CLI does and I think it could be misleading.
I can't find any ContractState::witness_info method, could you please show me where it is?

[dr-orlovsky]

The tests are showing the allocations in the same way the CLI does and I think it could be misleading.

I think it will be good to add the information about the mining status/confirmations to the cli

I can't find any ContractState::witness_info method, could you please show me where it is?

https://github.com/RGB-WG/rgb-std/pull/269/files#diff-c9d11eb7aab562be404a98e2ffb1f9915bf3e820398303a35f1befc3c903c7dfR367

[zoedberg]

I think it will be good to add the information about the mining status/confirmations to the cli

I agree, and I think we should do the same here

https://github.com/RGB-WG/rgb-std/pull/269/files#diff-c9d11eb7aab562be404a98e2ffb1f9915bf3e820398303a35f1befc3c903c7dfR367

Ok, thanks, I thoght I should have searched in the beta 8 code. Let's leave this PR open then, when next release will be there I will use that method to add more info about mining status

[dr-orlovsky]

BTW the mining status was added to the clii in RGB-WG/rgb#253