Reproducing examples for the unexpected transfer history bug

[k0k0ne]

No description provided.

[zoedberg]

This is using an old RGB version. Today I plan to push updated code and merge the initiating PR, after that I'll let you know, so you can rebase this PR there to check the bug is still there.

[dr-orlovsky]

The tests from @zoedberg are merged so now you can re-base on them

[zoedberg]

@k0k0ne I've requested some changes. Also please remember to change the base branch of the PR to master and squash your 3 commits into 1

[k0k0ne]

I have modified the code as requested. Please check it.

[zoedberg]

@k0k0ne please see the requested changes

[dr-orlovsky]

But @zoedberg do you confirm that the logic of the test is correct and there is a bug in the history showcased by the test?

[k0k0ne]

I have modified the code as requested. Please check it.

[dr-orlovsky]

The compilation fails due to error: unused import: WalletError`` (see https://github.com/RGB-WG/rgb-tests/actions/runs/10936428253/job/30361992550?pr=3#step:5:54)

Can you please make sure it compiles? Also running cargo lint --all-targets should help

[zoedberg]

But @zoedberg do you confirm that the logic of the test is correct and there is a bug in the history showcased by the test?

Without investing the time to dig deeper, it's not entirely clear to me what the test should achieve and what fungible_history should return. @dr-orlovsky, since you know fungible_history way better than anyone else, can you please look into this?

[dr-orlovsky]

fungible_history should return a history of fungible asset operations. Each operation - a raw in a history, either "paid-in" or "paid-out". From what I see the test logic is correct, but I'd like your confirmation since you understand better what each call means. For instance, I do not get false in mine(false) etc

[zoedberg]

For instance, I do not get false in mine(false) etc

As answered in private:
That boolean means resume (mining), you need to put it to true in case a test stops the mining (with the stop_mining method) and then (usually after performing some checks/tasks) wants to allow mining to work again. Stopping mining prevents other tests to mine during a phase where it's important that no new blocks get mined.

@dr-orlovsky If there are other things you don't understand feel free to ask me

Anyway, the check_fungible_history is very verbose but what it does (before the panic occurs) is:

1. wlt 1 issues 100000 of a NIA asset
2. wlt 1 sends to wlt_2 200 of that asset (in witness mode)
3. wlt 2 accepts the consignment
4. a block gets mined and both wallets perform a sync

after point 1, the test checks that fungible_history returns 0 results (not sure if this should be the case)

after point 4, the first check passes (it checks that wlt 1, the sender, has a Dec(Amount(200))), while the second check fails because wlt 2, the recipient, expects a state_change of Inc(Amount(200)) but it instead contains Dec(Amount(99800)):

thread 'check_fungible_history' panicked at tests/transfers.rs:837:5:
assertion `left == right` failed
  left: Dec(Amount(99800))
 right: Inc(Amount(200))

this is the part I'd like you to tell me if it's a bug or not, since it's still unclear to me what the method should return

Speaking instead about the send_to_oneself test:

1. wlt 1 issues 100000 of a NIA asset
2. wlt 1 sends 300 of that asset (in witness mode) to itself
3. wlt 1 accepts the consignment
4. a block gets mined and the wallet performs a sync

after point 4, the test checks that the history is not empty (which succeeds)

@k0k0ne the test passes so it's unclear what bug this test should show

[dr-orlovsky]

this is the part I'd like you to tell me if it's a bug or not, since it's still unclear to me what the method should return

Wlt 2 never had 100000 assets and the only thing it ever had is 200 assets received. So this looks like a bug. But I'd like to see is this all of the history or maybe Inc(200) goes after it as a second record.

[dr-orlovsky]

I confirm the bug, I am working on a fix. @k0k0ne thank you very much for providing the demonstration of the case.

Please also let us know whether the second test demonstrate some other issue - or it is done just as a check

[k0k0ne]

The second test appears to be working after updating the environment (unlike with previous version).

[zoedberg]

@k0k0ne I have fixed and rebased the code of this PR instead of asking for a rebase (since a lot of things changed and I didn't want to bother you), thanks for your contribution!

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 37.20%. Comparing base (59dec46) to head (6353b49).
Report is 2 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master       #3      +/-   ##
==========================================
+ Coverage   36.53%   37.20%   +0.66%     
==========================================
  Files         279      279              
  Lines       42806    42806              
==========================================
+ Hits        15641    15924     +283     
+ Misses      27165    26882     -283     

Flag	Coverage Δ
rust	37.20% <ø> (+0.66%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

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

---

🚨 Try these New Features:

• Flaky Tests Detection - Detect and resolve failed and flaky tests