Static analysis to find unnecessary locations

[MaximilianAlgehed]

I don't know why this didn't work last time. Appears to be issues with CI.

@sjoerdvisscher any clue?

Pre-submit checklist:

• Branch
  • Tests are provided (if possible)
  • Commit sequence broadly makes sense
  • Key commits have useful messages
  • Relevant tickets are mentioned in commit messages
  • Formatting, materialized Nix files, PNG optimization, etc. are updated
• PR
  • Self-reviewed the diff
  • Useful pull request description
  • Reviewer requested

[MaximilianAlgehed]

@koslambrou can you please continuously re-start Hydra until it doesn't crash because it runs out of memory on PRs so we see what's actually going on?

[koslambrou]

Everything works fine when running tests with cabal test plutus-use-cases.

However, when running the tests with Nix using: nix-build -A plutus-apps.haskell.packages.plutus-use-cases.checks.plutus-use-cases-test, I'm getting error with the golden files.

multisig
    2 out of 5:                                                                    OK (0.06s)
    3 out of 5:                                                                    OK (0.07s)
    PIR:                                                                           FAIL
      Test output was different from 'test/Spec/multisig.pir'. It was:
      (program
        (let
          (nonrec)
          (datatypebind
            (datatype
              (tyvardecl Monoid (fun (type) (type)))
              (tyvardecl a (type))
              Monoid_match
              (vardecl
                CConsMonoid
                (fun [ (lam a (type) (fun a (fun a a))) a ] (fun a [ Monoid a ]))
              )
            )
          )
          (termbind
            (strict)
            (vardecl
              p1Monoid
              (all a (type) (fun [ Monoid a ] [ (lam a (type) (fun a (fun a a))) a ]))
            )
            (abs
              a
              (type)
              (lam
                v
                [ Monoid a ]
                [
                  { [ { Monoid_match a } v ] [ (lam a (type) (fun a (fun a a))) a ] }
                  (lam v [ (lam a (type) (fun a (fun a a))) a ] (lam v a v))
                ]
              )
            )
          )
          (termbind
            (strict)
            (vardecl mempty (all a (type) (fun [ Monoid a ] a)))
            (abs
              a
              (type)
              (lam
                v
                [ Monoid a ]
                [
                  { [ { Monoid_match a } v ] a }
                  (lam v [ (lam a (type) (fun a<truncated>
      Use --accept or increase --size-cutoff to see full output.
      Use -p '/multisig.PIR/' to rerun this test only.

[koslambrou]

@michaelpj Seems like something on the plutus side no?

Probably not since it only fails on Nix.

[MaximilianAlgehed]

@koslambrou I can't reproduce this. But I get:

building
use cases
  crowdfunding
    Expose 'contribute' and 'scheduleCollection' endpoints:                        OK
    make contribution:                                                             OK (0.02s)
    make contributions and collect:                                                OK (0.14s)
    cannot collect money too late:                                                 OK (0.13s)
    cannot collect unless notified:                                                OK (0.13s)
    can claim a refund:                                                            OK (0.09s)
    PIR:                                                                           OK
    script size is reasonable:                                                     OK
      Script size: 3533
    renders the log of a single contract instance sensibly:                        OK (0.14s)
    renders the emulator log sensibly:                                             OK (0.11s)
    renders an error sensibly:                                                     OK
    QuickCheck ContractModel:                                                      OK (0.10s)
      +++ OK, passed 10 tests.
      
      Actions (19 in total):
      74% CContribute
      16% CStart
      11% WaitUntil
      
      Wait interval (2 in total):
      100% <10
      
      Wait until (2 in total):
      50% 10-19
      50% <10
  vesting
    secure some funds with the vesting script:                                     OK (0.02s)
    retrieve some funds:                                                           OK (0.03s)
    cannot retrieve more than allowed:                                             OK (0.01s)
    can retrieve everything at the end:                                            OK (0.05s)
    PIR:                                                                           OK (0.04s)
    script size is reasonable:                                                     OK
      Script size: 4661
    prop_Vesting:                                                                  OK (1.43s)
      +++ OK, passed 20 tests.
      
      Actions (87 in total):
      43% WaitUntil
      34% Vest
      23% Retrieve
      
      Actions rejected by precondition (221 in total):
      54.3% Retrieve
      45.7% Vest
      
      Wait interval (37 in total):
      62% <10
      27% 10-19
       8% 20-29
       3% 30-39
      
      Wait until (37 in total):
      35% <10
      14% 10-19
      14% 20-29
      11% 30-39
       8% 100-199
       5% 50-59
       3% 40-49
       3% 60-69
       3% 70-79
       3% 80-89
       3% 90-99
    prop_CheckNoLockedFundsProof:                                                  OK (7.71s)
      +++ OK, passed 20 tests.
      
      Actions (452 in total):
      49.6% WaitUntil
      32.7% Retrieve
      13.7% Vest
       4.0% Unilateral
      
      Wait interval (224 in total):
      48.2% <10
      34.8% 10-19
      15.2% 20-29
       1.8% 30-39
      
      Wait until (224 in total):
      21.0% 100-199
      17.4% 10-19
      15.6% 20-29
      10.3% <10
       8.9% 30-39
       6.2% 60-69
       4.9% 80-89
       4.5% 40-49
       4.0% 90-99
       3.1% 70-79
       2.2% 50-59
       1.8% 200-299
  error handling
    throw an error:                                                                OK
    catch an error:                                                                OK
  futures
    setup tokens:                                                                  OK (0.05s)
    can initialise and obtain tokens:                                              OK (0.21s)
    can increase margin:                                                           OK (0.30s)
    can settle early:                                                              OK (0.20s)
    can pay out:                                                                   OK (0.24s)
    PIR:                                                                           OK (0.02s)
    script size is reasonable:                                                     OK
      Script size: 9313
  multisig
    2 out of 5:                                                                    OK (0.03s)
    3 out of 5:                                                                    OK (0.04s)
    PIR:                                                                           OK
  multi sig state machine tests
    lock, propose, sign 3x, pay - SUCCESS:                                         OK (0.23s)
    lock, propose, sign 2x, pay - FAILURE:                                         OK (0.17s)
    lock, propose, sign 3x, pay x2 - SUCCESS:                                      OK (0.44s)
    lock, propose, sign 3x, pay x3 - FAILURE:                                      OK (0.48s)
    PIR:                                                                           OK (0.06s)
    script size is reasonable:                                                     OK
      Script size: 8364
  currency
    can create a new currency:                                                     OK (0.05s)
    script size is reasonable:                                                     OK (0.01s)
  pubkey
    works like a public key output:                                                OK (0.03s)
  escrow
    can pay:                                                                       OK (0.02s)
    can redeem:                                                                    OK (0.12s)
    can redeem even if more money than required has been paid in:                  OK (0.11s)
    can refund:                                                                    OK (0.08s)
    script size is reasonable:                                                     OK (0.04s)
      Script size: 5166 (0.04s)
    QuickCheck ContractModel:                                                      OK (0.77s)
      +++ OK, passed 10 tests:
      50% Redeemable
      30% Contains Redeem
      
      Actions (42 in total):
      50% Pay
      21% BadRefund
      19% WaitUntil
       10% Redeem
      
      Actions rejected by precondition (22 in total):
      77% Redeem
       9% BadRefund
       9% Refund
       5% Pay
      
      Bad refund attempts (9 in total):
      89% steal refund
      11% early refund
      
      Wait interval (8 in total):
      62% <10
      25% 10-19
      12% 20-29
      
      Wait until (8 in total):
      50% 10-19
      12% 20-29
      12% 30-39
      12% 50-59
      12% <10
    QuickCheck NoLockedFunds:                                                      OK (2.86s)
      +++ OK, passed 10 tests:
      80% Redeemable
      60% Contains Redeem
      
      Actions (162 in total):
      43.8% Pay
      20.4% WaitUntil
      13.6% Refund
       9.9% Redeem
       6.8% Unilateral
       5.6% BadRefund
      
      Bad refund attempts (9 in total):
      78% steal refund
      22% early refund
      
      Wait interval (33 in total):
      45% 30-39
      36% <10
      18% 20-29
      
      Wait until (33 in total):
      64% 40-49
      24% 10-19
      12% <10
  simple-escrow
    can lock some value in the contract:                                           OK
    can lock and redeem:                                                           OK (0.02s)
    can lock and refund:                                                           OK (0.06s)
    only locking wallet can request refund:                                        OK (0.05s)
    can't redeem if you can't pay:                                                 OK
  game with secret arguments tests
    run a successful game trace:                                                   OK (0.03s)
    run a failed trace:                                                            OK (0.03s)
    PIR:                                                                           OK
    script size is reasonable:                                                     OK
      Script size: 2301
  game state machine with secret arguments tests
    run a successful game trace:                                                   OK (0.21s)
    run a 2nd successful game trace:                                               OK (0.25s)
    run a successful game trace where we try to leave 1 Ada in the script address: OK (0.14s)
    run a failed trace:                                                            OK (0.09s)
    PIR:                                                                           OK (0.09s)
    script size is reasonable:                                                     OK
      Script size: 9750
    can always get the funds out:                                                  OK (1.45s)
      +++ OK, passed 10 tests (70% Unlocking funds).
      
      Actions (50 in total):
      52% Guess
      32% GiveToken
      14% Lock
       2% WaitUntil
      
      Wait interval (1 in total):
      100% <10
      
      Wait until (1 in total):
      100% <10
    sanity check the contract model:                                               OK
      +++ OK, failed as expected. Falsified (after 2 tests and 14 shrinks):
      Actions 
       [Lock (Wallet 2) "*******" 2000000]
    game state machine crash tolerance:                                            OK (3.31s)
      +++ OK, passed 20 tests.
      
      Actions (170 in total):
      41.8% GiveToken
      28.2% Guess
      11.2% WaitUntil
       8.8% Lock
       6.5% Crash
       3.5% Restart
      
      Actions rejected by precondition (28 in total):
      64% Restart
      36% Guess
      
      Wait interval (19 in total):
      84% <10
      16% 10-19
      
      Wait until (19 in total):
      37% 10-19
      37% <10
      21% 20-29
       5% 40-49
  showBlockchain
    renders a crowdfunding scenario sensibly:                                      OK (0.11s)
    renders a game guess scenario sensibly:                                        FAIL (0.23s)
      Test output was different from 'test/Spec/renderGuess.txt'. It was:
      ==== Slot #0, Tx #0 ====
      TxId:       ef0ca0fb043642529818003be5a6cac88aac499e4f8f1cbc3bdb35db2b7f6958
      Fee:        -
      Mint:       Ada:      Lovelace:  1000000000
      Signatures  -
      Inputs:
        
      
      
      Outputs:
        ---- Output 0 ----
        Destination:  PaymentPubKeyHash: 2e0ad60c3207248cecd47dbde3d752e0aad141d6... (Wallet c30efb78b4e272685c1f9f0c93
787fd4b6743154)
        Value:
          Ada:      Lovelace:  10000000
      
        ---- Output 1 ----
        Destination:  PaymentPubKeyHash: 2e0ad60c3207248cecd47dbde3d752e0aad141d6... (Wallet c30efb78b4e272685c1f9f0c93
787fd4b6743154)
        Value:
          Ada:      Lovelace:  10000000
      
        ---- Output 2 ----
        Destination:  PaymentPubKeyHash: 2e0ad60c3207248cecd47dbde3d752e0aad141d6... (Wallet c30efb78b4e272685c1f9f0c93
787fd4b6743154)
        Value:
          Ada:      Lovelace:  10000000
      
        ---- Output 3 ----
        Destination:  PaymentPubKeyHash: 2e0ad60c3207248cecd47dbde3d752e0aad141d6... (Wallet c30efb78b4e272685c1f9f0c93
787fd4b6743154)
        Value:
          Ada:      Lovelace:  10000000
      
        ---- Output 4 ----
        Desti<truncated>
      Use --accept or increase --size-cutoff to see full output.
      Use -p '/renders a game guess scenario sensibly/' to rerun this test only.
    renders a vesting scenario sensibly:                                           OK (0.06s)
  token account
    Create a token account:                                                        OK (0.10s)
    Pay into the account:                                                          OK (0.07s)
    Transfer & redeem all funds:                                                   OK (0.10s)
  pingpong
    activate endpoints:                                                            OK (0.12s)
    Stop the contract:                                                             OK (0.08s)
  PRISM
    withdraw:                                                                      OK (0.21s)
    QuickCheck property:                                                           OK (1.39s)
      +++ OK, passed 15 tests.
      
      Actions (45 in total):
      36% Call
      36% Revoke
      24% Issue
       4% WaitUntil
      
      Wait interval (2 in total):
      100% <10
      
      Wait until (2 in total):
      50% 10-19
      50% 30-39
  Stablecoin
    mint reservecoins:                                                             OK (0.13s)
    mint reservecoins and stablecoins:                                             OK (0.21s)
    mint reservecoins, stablecoins and redeem stablecoin at a different price:     OK (0.29s)
    Cannot exceed the maximum reserve ratio:                                       OK (0.20s)
  auction
    run an auction:                                                                OK (0.38s)
    run an auction with multiple bids:                                             OK (0.61s)
    QuickCheck property:                                                           OK (2.56s)
      +++ OK, passed 10 tests.
      
      Actions (31 in total):
      39% Bid
      39% WaitUntil
      23% Init
      
      Wait interval (12 in total):
      75% 90-99
       8% 10-19
       8% 70-79
       8% <10
      
      Wait until (12 in total):
      83% 100-199
       8% 20-29
       8% <10
    NLFP fails:                                                                    Expected funds of W[2] to change by
  Value (Map [(,Map [("",-87240078)]),(363d3944282b3d16b239235a112c0f6e2f1195de5067f61c0dfc0f5f,Map [("token",1)])])
but they changed by
  Value (Map [(,Map [("",-89240078)])])
a discrepancy of
  Value (Map [(,Map [("",-2000000)]),(363d3944282b3d16b239235a112c0f6e2f1195de5067f61c0dfc0f5f,Map [("token",-1)])])
Expected funds of W[1] to change by
  Value (Map [(,Map [("",87240078)]),(363d3944282b3d16b239235a112c0f6e2f1195de5067f61c0dfc0f5f,Map [("token",-1)])])
but they changed by
  Value (Map [(,Map [("",-2000000)]),(363d3944282b3d16b239235a112c0f6e2f1195de5067f61c0dfc0f5f,Map [("token",-1)])])
a discrepancy of
  Value (Map [(,Map [("",-89240078)])])
Test failed.
Emulator log:
OK (0.96s)
      +++ OK, failed as expected. Assertion failed (after 2 tests):
      DLScript
        [Do $ Init, 
         Do $ Bid (Wallet 2) 89240078]
      
      The ContractModel's Unilateral behaviour for Wallet 2 does not match the actual behaviour for actions:
      Actions 
       [Var 0 := Init,
        Var 1 := Bid (Wallet 2) 89240078,
        Var 2 := Unilateral (Wallet 2),
        Var 3 := WaitUntil (Slot {getSlot = 100})]
    prop_Reactive:                                                                 OK (0.11s)
      +++ OK, passed 1000 tests.
  sealed bid auction
    packInteger is injective:                                                      OK
      +++ OK, passed 100 tests; 12 discarded.
    prop_AuctionModelCorrect:                                                      OK (2.52s)
      +++ OK, passed 20 tests.
      
      Actions (94 in total):
      40% WaitUntil
      18% Init
      17% Bid
      15% Reveal
       10% Payout
      
      Actions rejected by precondition (183 in total):
      40.4% Bid
      39.3% Reveal
      20.2% Payout
      
      Reveals (14 in total):
      71% no bid
      14% dishonest
      14% honest
      
      Wait interval (38 in total):
      71% <10
       8% 30-39
       8% 50-59
       5% 20-29
       3% 10-19
       3% 40-49
       3% 60-69
      
      Wait until (38 in total):
      29% 70-79
      21% 60-69
      16% 40-49
      11% 80-89
       8% 30-39
       8% <10
       3% 20-29
       3% 50-59
       3% 90-99
  governance tests
    vote all in favor, 2 rounds - SUCCESS:                                         OK (1.51s)
    vote 60/40, accepted - SUCCESS:                                                OK (0.83s)
    vote 50/50, rejected - SUCCESS:                                                OK (0.81s)
    PIR:                                                                           OK (0.06s)
    script size is reasonable:                                                     Script size: 9246
OK
  uniswap
    can create a liquidity pool and add liquidity:                                 OK (0.52s)
    prop_UniswapAssertions:                                                        OK (0.56s)
      +++ OK, passed 1000 tests.
      
      Actions (25269 in total):
      19.724% Bad
      15.822% AddLiquidity
      15.810% PerformSwap
      14.718% RemoveLiquidity
      13.515% WaitUntil
       7.978% SetupTokens
       7.745% CreatePool
       3.376% Start
       1.314% ClosePool
      
      Actions rejected by precondition (8299 in total):
      53.26% ClosePool
      17.04% CreatePool
      15.57% Bad
      14.13% Start
      
      Bad actions (4984 in total):
      34.95% RemoveLiquidity
      28.71% AddLiquidity
      28.19% PerformSwap
       8.15% BadRemoveLiquidity
      
      Wait interval (3415 in total):
      31.07% <10
      29.19% 10-19
      25.27% 20-29
      11.95% 30-39
       2.43% 40-49
       0.09% 50-59
      
      Wait until (3415 in total):
      22.78% 100-199
      14.64% 200-299
       9.75% 300-399
       6.06% 400-499
       4.57% 60-69
       3.95% 500-599
       3.66% 50-59
       3.63% 30-39
       3.57% 40-49
       3.51% 90-99
       3.28% 70-79
       3.28% 80-89
       2.87% 1000-1999
       2.75% 600-699
       2.72% 10-19
       2.69% <10
       2.02% 700-799
       1.55% 20-29
       1.46% 800-899
       1.02% 900-999
       0.20% 2000-2999
    prop_NLFP:                                                                     OK (0.67s)
      +++ OK, passed 250 tests.

[michaelpj]

Argh. Maybe this is the same issue that we had in plutus: somehow what you did has introduced instability in the compiler output between the Nix and non-Nix builds. We let it slide there but if it's popping up more here that's a problem.

[MaximilianAlgehed]

We should probably bump the core dep even further here as we just merged a relevant bugfix there today. I'll get to that next week.

[MaximilianAlgehed]

This is now stuck on cached OOM failure.

[koslambrou]

I'm still getting failing test cases :( .

I sure don't understand why you're not getting them.. I'm on the 94b881ddd8ff0c4e8dd6e0cd970080083b3493bd commit of your branch.

use cases
  crowdfunding
    Expose 'contribute' and 'scheduleCollection' endpoints:                        OK (0.02s)
    make contribution:                                                             OK (0.04s)
    make contributions and collect:                                                OK (0.26s)
    cannot collect money too late:                                                 OK (0.30s)
    cannot collect unless notified:                                                OK (0.27s)
    can claim a refund:                                                            OK (0.20s)
    PIR:                                                                           OK (0.01s)
    script size is reasonable:                                                     OK
      Script size: 3533
    renders the log of a single contract instance sensibly:                        OK (0.26s)
    renders the emulator log sensibly:                                             OK (0.25s)
    renders an error sensibly:                                                     OK
    QuickCheck ContractModel:                                                      OK (0.36s)
      +++ OK, passed 10 tests.
      
      Actions (31 in total):
      68% CContribute
      16% CStart
      16% WaitUntil
      
      Wait interval (5 in total):
      80% <10
      20% 10-19
      
      Wait until (5 in total):
      60% <10
      40% 10-19
  vesting
    secure some funds with the vesting script:                                     OK (0.03s)
    retrieve some funds:                                                           OK (0.12s)
    cannot retrieve more than allowed:                                             OK (0.03s)
    can retrieve everything at the end:                                            OK (0.14s)
    PIR:                                                                           OK (0.03s)
    script size is reasonable:                                                     OK (0.01s)
      Script size: 4661 (0.01s)
    prop_Vesting:                                                                  OK (1.77s)
      +++ OK, passed 20 tests.
      
      Actions (51 in total):
      43% Vest
      31% WaitUntil
      25% Retrieve
      
      Actions rejected by precondition (128 in total):
      57.8% Retrieve
      42.2% Vest
      
      Wait interval (16 in total):
      69% <10
      31% 10-19
      
      Wait until (16 in total):
      38% <10
      31% 10-19
      12% 20-29
       6% 30-39
       6% 50-59
       6% 60-69
    prop_CheckNoLockedFundsProof:                                                  OK (15.37s)
      +++ OK, passed 20 tests.
      
      Actions (465 in total):
      54.2% WaitUntil
      28.4% Retrieve
      13.3% Vest
       4.1% Unilateral
      
      Wait interval (252 in total):
      42.5% <10
      40.1% 10-19
      16.7% 20-29
       0.8% 30-39
      
      Wait until (252 in total):
      18.3% 100-199
      15.5% 20-29
      15.1% 10-19
      10.3% <10
       9.1% 30-39
       8.3% 60-69
       5.6% 40-49
       5.2% 90-99
       4.0% 70-79
       3.6% 50-59
       2.8% 80-89
       2.0% 200-299
       0.4% 300-399
  error handling
    throw an error:                                                                OK
    catch an error:                                                                OK
  futures
    setup tokens:                                                                  OK (0.14s)
    can initialise and obtain tokens:                                              OK (0.49s)
    can increase margin:                                                           OK (0.71s)
    can settle early:                                                              OK (0.51s)
    can pay out:                                                                   OK (0.61s)
    PIR:                                                                           OK (0.04s)
    script size is reasonable:                                                     OK (0.02s)
      Script size: 9313 (0.02s)
  multisig
    2 out of 5:                                                                    OK (0.07s)
    3 out of 5:                                                                    OK (0.08s)
    PIR:                                                                           FAIL
      Test output was different from 'test/Spec/multisig.pir'. It was:
      (program
        (let
          (nonrec)
          (datatypebind
            (datatype
              (tyvardecl Monoid (fun (type) (type)))
              (tyvardecl a (type))
              Monoid_match
              (vardecl
                CConsMonoid
                (fun [ (lam a (type) (fun a (fun a a))) a ] (fun a [ Monoid a ]))
              )
            )
          )
          (termbind
            (strict)
            (vardecl
              p1Monoid
              (all a (type) (fun [ Monoid a ] [ (lam a (type) (fun a (fun a a))) a ]))
            )
            (abs
              a
              (type)
              (lam
                v
                [ Monoid a ]
                [
                  { [ { Monoid_match a } v ] [ (lam a (type) (fun a (fun a a))) a ] }
                  (lam v [ (lam a (type) (fun a (fun a a))) a ] (lam v a v))
                ]
              )
            )
          )
          (termbind
            (strict)
            (vardecl mempty (all a (type) (fun [ Monoid a ] a)))
            (abs
              a
              (type)
              (lam
                v
                [ Monoid a ]
                [
                  { [ { Monoid_match a } v ] a }
                  (lam v [ (lam a (type) (fun a<truncated>
      Use --accept or increase --size-cutoff to see full output.
      Use -p '/multisig.PIR/' to rerun this test only.
  multi sig state machine tests
    lock, propose, sign 3x, pay - SUCCESS:                                         OK (0.56s)
    lock, propose, sign 2x, pay - FAILURE:                                         OK (0.38s)
    lock, propose, sign 3x, pay x2 - SUCCESS:                                      OK (1.10s)
    lock, propose, sign 3x, pay x3 - FAILURE:                                      OK (1.19s)
    PIR:                                                                           FAIL (0.09s)
      Test output was different from 'test/Spec/multisigStateMachine.pir'. It was:
      (program
        (let
          (nonrec)
          (termbind (strict) (vardecl unitval (con unit)) (con unit ()))
          (let
            (rec)
            (datatypebind
              (datatype
                (tyvardecl List (fun (type) (type)))
                (tyvardecl a (type))
                Nil_match
                (vardecl Nil [ List a ])
                (vardecl Cons (fun a (fun [ List a ] [ List a ])))
              )
            )
            (let
              (rec)
              (termbind
                (strict)
                (vardecl go (fun [ List (con bytestring) ] [ (con list) (con data) ]))
                (lam
                  ds
                  [ List (con bytestring) ]
                  {
                    [
                      [
                        {
                          [ { Nil_match (con bytestring) } ds ]
                          (all dead (type) [ (con list) (con data) ])
                        }
                        (abs dead (type) [ (builtin mkNilData) unitval ])
                      ]
                      (lam
                        x
                        (con bytestring)
                        (lam
                          xs
          <truncated>
      Use --accept or increase --size-cutoff to see full output.
      Use -p '/multi sig state machine tests.PIR/' to rerun this test only.
    script size is reasonable:                                                     OK
      Script size: 8370
  currency
    can create a new currency:                                                     OK (0.13s)
    script size is reasonable:                                                     OK (0.03s)
  pubkey
    works like a public key output:                                                OK (0.08s)
  escrow
    can pay:                                                                       OK (0.04s)
    can redeem:                                                                    OK (0.25s)
    can redeem even if more money than required has been paid in:                  OK (0.28s)
    can refund:                                                                    OK (0.29s)
    script size is reasonable:                                                     OK
      Script size: 5166 (0.01s)
    QuickCheck ContractModel:                                                      OK (1.74s)
      +++ OK, passed 10 tests:
      50% Redeemable
      30% Contains Redeem
      
      Actions (39 in total):
      72% Pay
      15% Redeem
      10% WaitUntil
       3% BadRefund
      
      Actions rejected by precondition (17 in total):
      100% Redeem
      
      Bad refund attempts (1 in total):
      100% steal refund
      
      Wait interval (4 in total):
      100% <10
      
      Wait until (4 in total):
      75% 10-19
      25% <10
    QuickCheck NoLockedFunds:                                                      OK (8.51s)
      +++ OK, passed 10 tests:
      70% Redeemable
      60% Contains Redeem
      
      Actions (204 in total):
      39.7% Pay
      24.0% WaitUntil
      13.7% Refund
       8.3% Redeem
       7.4% BadRefund
       6.9% Unilateral
      
      Bad refund attempts (15 in total):
      100% steal refund
      
      Wait interval (49 in total):
      51% <10
      27% 30-39
      16% 20-29
       6% 10-19
      
      Wait until (49 in total):
      49% 40-49
      20% 10-19
      18% <10
       6% 20-29
       6% 30-39
  simple-escrow
    can lock some value in the contract:                                           OK (0.02s)
    can lock and redeem:                                                           OK (0.09s)
    can lock and refund:                                                           OK (0.20s)
    only locking wallet can request refund:                                        OK (0.18s)
    can't redeem if you can't pay:                                                 OK (0.02s)
  game with secret arguments tests
    run a successful game trace:                                                   OK (0.09s)
    run a failed trace:                                                            OK (0.07s)
    PIR:                                                                           OK
    script size is reasonable:                                                     OK
      Script size: 2301
  game state machine with secret arguments tests
    run a successful game trace:                                                   OK (0.44s)
    run a 2nd successful game trace:                                               OK (0.64s)
    run a successful game trace where we try to leave 1 Ada in the script address: OK (0.40s)
    run a failed trace:                                                            OK (0.26s)
    PIR:                                                                           FAIL (0.08s)
      Test output was different from 'test/Spec/gameStateMachine.pir'. It was:
      (program
        (let
          (nonrec)
          (termbind (strict) (vardecl w (con integer)) (con integer 0))
          (termbind (strict) (vardecl w (con integer)) (con integer 1))
          (datatypebind
            (datatype (tyvardecl Unit (type))  Unit_match (vardecl Unit Unit))
          )
          (termbind (strict) (vardecl unitval (con unit)) (con unit ()))
          (datatypebind
            (datatype
              (tyvardecl Tuple2 (fun (type) (fun (type) (type))))
              (tyvardecl a (type)) (tyvardecl b (type))
              Tuple2_match
              (vardecl Tuple2 (fun a (fun b [ [ Tuple2 a ] b ])))
            )
          )
          (termbind
            (strict)
            (vardecl
              fail (fun (con unit) [ [ Tuple2 (con bytestring) ] (con bytestring) ])
            )
            (lam
              ds
              (con unit)
              (let
                (nonrec)
                (termbind
                  (strict)
                  (vardecl thunk (con unit))
                  [
                    {
                      [
                        Unit_match
                        [ [ { (builtin trace) Unit } (con string "Lg") ] Unit ]
       <truncated>
      Use --accept or increase --size-cutoff to see full output.
      Use -p '/game state machine with secret arguments tests.PIR/' to rerun this test only.
    script size is reasonable:                                                     OK
      Script size: 9750
    can always get the funds out:                                                  OK (3.45s)
      +++ OK, passed 10 tests (70% Unlocking funds).
      
      Actions (40 in total):
      42% Guess
      32% GiveToken
      18% Lock
       8% WaitUntil
      
      Wait interval (3 in total):
      100% <10
      
      Wait until (3 in total):
      67% 10-19
      33% <10
    sanity check the contract model:                                               OK
      +++ OK, failed as expected. Falsified (after 1 test and 16 shrinks):
      Actions 
       [Lock (Wallet 2) "secret" 2000000]
    game state machine crash tolerance:                                            OK (6.61s)
      +++ OK, passed 20 tests.
      
      Actions (159 in total):
      45.3% GiveToken
      22.6% Guess
      17.0% WaitUntil
       9.4% Lock
       5.0% Crash
       0.6% Restart
      
      Actions rejected by precondition (20 in total):
      70% Restart
      30% Guess
      
      Wait interval (27 in total):
      63% <10
      22% 10-19
      15% 20-29
      
      Wait until (27 in total):
      37% 10-19
      19% 100-199
      19% <10
      11% 30-39
       4% 20-29
       4% 50-59
       4% 70-79
       4% 90-99
  showBlockchain
    renders a crowdfunding scenario sensibly:                                      OK (0.25s)
    renders a game guess scenario sensibly:                                        OK (0.41s)
    renders a vesting scenario sensibly:                                           OK (0.14s)
  token account
    Create a token account:                                                        OK (0.23s)
    Pay into the account:                                                          OK (0.14s)
    Transfer & redeem all funds:                                                   OK (0.21s)
  pingpong
    activate endpoints:                                                            OK (0.27s)
    Stop the contract:                                                             OK (0.14s)
  PRISM
    withdraw:                                                                      OK (0.39s)
            QuickCheck property:                                                           OK (5.99s)
      +++ OK, passed 15 tests.
      
      Actions (56 in total):
      43% Call
      27% Revoke
      23% Issue
       7% WaitUntil
      
      Actions rejected by precondition (2 in total):
      100% Issue
      
      Wait interval (4 in total):
      50% 10-19
      50% <10
      
      Wait until (4 in total):
      50% 20-29
      25% 40-49
      25% 50-59
  Stablecoin
<    mint reservecoins:                                                             OK (0.28s)
    mint reservecoins and stablecoins:                                             OK (0.48s)
    mint reservecoins, stablecoins and redeem stablecoin at a different price:     OK (0.71s)
    Cannot exceed the maximum reserve ratio:                                       OK (0.46s)
  auction
    run an auction:                                                                OK (0.80s)
    run an auction with multiple bids:                                             OK (1.42s)
    QuickCheck property:                                                           OK (6.14s)
      +++ OK, passed 10 tests.
      
      Actions (31 in total):
      42% Bid
      35% WaitUntil
      23% Init
      
      Wait interval (11 in total):
      64% 90-99
      27% 80-89
       9% <10
      
      Wait until (11 in total):
      91% 100-199
       9% 10-19
    NLFP fails:                                                                    Expected funds of W[4] to change by
  Value (Map [(,Map [("",-44316956)]),(363d3944282b3d16b239235a112c0f6e2f1195de5067f61c0dfc0f5f,Map [("token",1)])])
but they changed by
  Value (Map [(,Map [("",-46316956)])])
a discrepancy of
  Value (Map [(,Map [("",-2000000)]),(363d3944282b3d16b239235a112c0f6e2f1195de5067f61c0dfc0f5f,Map [("token",-1)])])
Expected funds of W[1] to change by
  Value (Map [(,Map [("",44316956)]),(363d3944282b3d16b239235a112c0f6e2f1195de5067f61c0dfc0f5f,Map [("token",-1)])])
but they changed by
  Value (Map [(,Map [("",-2000000)]),(363d3944282b3d16b239235a112c0f6e2f1195de5067f61c0dfc0f5f,Map [("token",-1)])])
a discrepancy of
  Value (Map [(,Map [("",-46316956)])])
Test failed.
Emulator log:
OK (2.12s)
      +++ OK, failed as expected. Assertion failed (after 2 tests):
      DLScript
        [Do $ Init, 
         Do $ Bid (Wallet 4) 46316956]
      
      The ContractModel's Unilateral behaviour for Wallet 4 does not match the actual behaviour for actions:
      Actions 
       [Var 0 := Init,
        Var 1 := Bid (Wallet 4) 46316956,
        Var 2 := Unilateral (Wallet 4),
        Var 3 := WaitUntil (Slot {getSlot = 100})]
    prop_Reactive:                                                                 OK (0.24s)
      +++ OK, passed 1000 tests.
  sealed bid auction
    packInteger is injective:                                                      OK
      +++ OK, passed 100 tests; 15 discarded.
    prop_AuctionModelCorrect:                                                      OK (5.57s)
      +++ OK, passed 20 tests.
      
      Actions (75 in total):
      36% Bid
      28% WaitUntil
      24% Init
       7% Reveal
       5% Payout
      
      Actions rejected by precondition (108 in total):
      43.5% Reveal
      29.6% Bid
      26.9% Payout
      
      Reveals (5 in total):
      100% no bid
      
      Wait interval (21 in total):
      52% <10
      14% 30-39
       10% 20-29
       10% 40-49
       5% 10-19
       5% 50-59
       5% 60-69
      
      Wait until (21 in total):
      33% 60-69
      24% 50-59
      19% <10
      14% 40-49
       5% 20-29
       5% 30-39
  governance tests
    vote all in favor, 2 rounds - SUCCESS:                                         OK (3.47s)
    vote 60/40, accepted - SUCCESS:                                                OK (1.81s)
    vote 50/50, rejected - SUCCESS:                                                OK (1.83s)
    PIR:                                                                           Script size: 9246
FAIL (0.10s)
      Test output was different from 'test/Spec/governance.pir'. It was:
      (program
        (let
          (nonrec)
          (termbind (strict) (vardecl unitval (con unit)) (con unit ()))
          (datatypebind
            (datatype
              (tyvardecl Tuple2 (fun (type) (fun (type) (type))))
              (tyvardecl a (type)) (tyvardecl b (type))
              Tuple2_match
              (vardecl Tuple2 (fun a (fun b [ [ Tuple2 a ] b ])))
            )
          )
          (let
            (rec)
            (datatypebind
              (datatype
                (tyvardecl List (fun (type) (type)))
                (tyvardecl a (type))
                Nil_match
                (vardecl Nil [ List a ])
                (vardecl Cons (fun a (fun [ List a ] [ List a ])))
              )
            )
            (let
              (nonrec)
              (datatypebind
                (datatype
                  (tyvardecl Bool (type))
      
                  Bool_match
                  (vardecl True Bool) (vardecl False Bool)
                )
              )
              (let
                (rec)
                (termbind
                  (strict)
                  (vardecl
                    go
                    (fun
                      [ List [ [ Tuple2 (con bytestring) ] Bool ] ]
      <truncated>
      Use --accept or increase --size-cutoff to see full output.
      Use -p '/governance tests.PIR/' to rerun this test only.
    script size is reasonable:                                                     OK
  uniswap
    can create a liquidity pool and add liquidity:                                 OK (1.09s)
    prop_UniswapAssertions:                                                        OK (1.43s)
      +++ OK, passed 1000 tests.
      
      Actions (25982 in total):
      18.417% Bad
      16.715% AddLiquidity
      16.007% PerformSwap
      15.699% RemoveLiquidity
      13.225% WaitUntil
       7.805% SetupTokens
       7.705% CreatePool
       3.295% Start
       1.132% ClosePool
      
      Actions rejected by precondition (8255 in total):
      55.83% ClosePool
      16.35% CreatePool
      14.08% Start
      13.74% Bad
      
      Bad actions (4785 in total):
      32.94% RemoveLiquidity
      29.36% PerformSwap
      29.07% AddLiquidity
       8.63% BadRemoveLiquidity
      
      Wait interval (3436 in total):
      31.78% <10
      28.58% 10-19
      23.43% 20-29
      13.18% 30-39
       2.97% 40-49
       0.06% 50-59
      
      Wait until (3436 in total):
      22.82% 100-199
      13.80% 200-299
       9.84% 300-399
       7.33% 400-499
       4.31% 500-599
       3.78% 60-69
       3.55% 50-59
       3.26% 600-699
       3.08% 40-49
       3.06% 30-39
       3.03% 1000-1999
       3.00% 10-19
       2.97% 90-99
       2.88% 80-89
       2.79% <10
       2.74% 70-79
       2.39% 700-799
       2.07% 800-899
       1.57% 20-29
       1.34% 900-999
       0.41% 2000-2999
    prop_NLFP:                                                                     OK (1.60s)
      +++ OK, passed 250 tests.

4 out of 95 tests failed (87.30s)
error: builder for '/nix/store/2m411l786d09h3vnnhnmkm42kimf9g6j-plutus-use-cases-test-plutus-use-cases-test-0.1.0.0-check.drv' failed with exit code 1;
       last 10 log lines:
       >        2.74% 70-79
       >        2.39% 700-799
       >        2.07% 800-899
       >        1.57% 20-29
       >        1.34% 900-999
       >        0.41% 2000-2999
       >     prop_NLFP:                                                                     OK (1.60s)
       >       +++ OK, passed 250 tests.
       >
       > 4 out of 95 tests failed (87.30s)
       For full logs, run 'nix log /nix/store/2m411l786d09h3vnnhnmkm42kimf9g6j-plutus-use-cases-test-plutus-use-cases-test-0.1.0.0-check.drv'.

[MaximilianAlgehed]

@koslambrou I can not reproduce these issues on my machine. Does this happen both when you run it through nix and when you just do cabal test plutus-use-cases?

[MaximilianAlgehed]

One concerning thing is that I just had a golden test fail on that branch - even though the last thing I did on that branch was run cabal test plutus-use-cases --test-options="--accept". Something is not right here.

[koslambrou]

@koslambrou I can not reproduce these issues on my machine. Does this happen both when you run it through nix and when you just do cabal test plutus-use-cases?

Thes issues happen when I run the tests through Nix. Surprisingly, everything works fine when doing cabal test plutus-use-cases.

One concerning thing is that I just had a golden test fail on that branch - even though the last thing I did on that branch was run cabal test plutus-use-cases --test-options="--accept". Something is not right here.

:( I don't understand either..

[MaximilianAlgehed]

@michaelpj this seems like a nix issue to me. Any clue what's going on?

[MaximilianAlgehed]

Also, can @michaelpj maybe do something about all these cached failures?

[michaelpj]

@michaelpj this seems like a nix issue to me. Any clue what's going on?

No. I presume it's the same thing that caused this issue on your PR to plutus. I have asked @zliu41 to try and figure out what's going on there, so maybe we'll know something in a bit.

Also, can @michaelpj maybe do something about all these cached failures?

I restarted all the failing jobs, let's see what happens.

[MaximilianAlgehed]

Ok, no idea what's up with "wrong ELF type"...

[michaelpj]

Remaining failures appear to be actual test failures.

[MaximilianAlgehed]

@michaelpj see above about these failures. They don't reproduce with cabal test so we can't update the golden tests.

[MaximilianAlgehed]

I've bumped the core dependency here to:

1. Include the latest bugfix
2. Try to magically fix this weird non-reproducability problem

[MaximilianAlgehed]

@michaelpj restart build plx?

[sjoerdvisscher]

Maybe pulling in the latest changes from main to get GHC 8.10.7 might help? My latest PR using that went smoothly, but that could have been a fluke.

[MaximilianAlgehed]

@sjoerdvisscher I rebased this morning - has it changed since?

Also, does that solve the OOM issues in hydra?

[sjoerdvisscher]

Also, does that solve the OOM issues in hydra?

Apparently not 😞

[MaximilianAlgehed]

@michaelpj could you please restart the failing CI jobs here?

[MaximilianAlgehed]

I bumped the dep on core to include some bugfixes, maybe that could magically solve the flaky-inconsistent-PIR-golden-tests issue?

[koslambrou]

@MaximilianAlgehed I'm getting the same PIR errors as before when running the tests in Nix. Latest commit is d144191

[MaximilianAlgehed]

@koslambrou Ok. I don't know what to make of this issue. Probably it's a good idea to file this as an issue in some issue tracker somewhere and try to get some eyes on it because as far as I can tell this is an issue with the compiler / nix / reproducibility. I've spoken a bit to @michaelpj about it but I don't know what the right bug report to file is, or even where to file it.

[michaelpj]

We have a ticket for it already and someone is going to look at it.

[MaximilianAlgehed]

@michaelpj any progress on this reproducibility issue?

[michaelpj]

Some progress. Try updating to a later version of the plutus release branch, see what happens.

[MaximilianAlgehed]

@michaelpj restart hydra plz...

[michaelpj]

done

[michaelpj]

Weird, it is continuing to very reliably die in plutus-contract...

[MaximilianAlgehed]

@michaelpj PIR reproducibility problem unfortunately wasn't fixed :(

[michaelpj]

cc @zliu41, maybe we can repeat the investigation effort. Might be something similar...

[MaximilianAlgehed]

Though this one fails when I run CI locally too! Improvement! But cabal test doesn't give us the failure 🤕

[MaximilianAlgehed]

@michaelpj and @zliu41 any progress on this issue?

[MaximilianAlgehed]

@koslambrou because progress on this reproducibility issue is a bit slow I think the right thing to do is to turn off the tests here, get this merged, and fix things in a separate push later. If this turns green, could you merge this?

[koslambrou]

@MaximilianAlgehed Yes I agree.

[zliu41]

Sorry I just saw this. It's probably a similar issue. I'll open a ticket which hopefully I (or someone else) can work on in the next few days.