Refactor ConfirmedSnapshot

[pgrange]

An InitialSnapshot always has number 0 and an empty list of confirmed transactions.
So we make it explicit in the type system to avoid issues with that.

[github-actions]

Transactions Costs

Sizes and execution budgets for Hydra protocol transactions. Note that unlisted parameters are currently using arbitrary values and results are not fully deterministic and comparable to previous runs.

Metadata
Generated at	2022-10-10 12:32:57.52295167 UTC
Max. memory units	14000000
Max. CPU units	10000000000
Max. tx size (kB)	16384

Cost of Init Transaction

Parties	Tx size	% max Mem	% max CPU	Min fee ₳
1	4832	11.06	4.39	0.48
2	5037	12.95	5.12	0.51
3	5243	13.91	5.46	0.53
5	5652	20.15	7.93	0.62
10	6677	28.55	11.12	0.75
45	13855	99.77	38.62	1.84

Cost of Commit Transaction

Currently only one UTxO per commit allowed (this is about to change soon)

UTxO	Tx size	% max Mem	% max CPU	Min fee ₳
1	5768	19.92	8.05	0.62

Cost of CollectCom Transaction

Parties	Tx size	% max Mem	% max CPU	Min fee ₳
1	13058	20.60	8.25	0.95
2	13522	38.08	15.44	1.16
3	13780	56.09	22.89	1.37
4	14135	77.87	31.97	1.63

Cost of Close Transaction

Parties	Tx size	% max Mem	% max CPU	Min fee ₳
1	9327	8.19	3.36	0.65
2	9363	8.05	3.16	0.65
3	9694	10.14	4.42	0.69
5	10055	12.14	5.49	0.73
10	10845	15.65	7.60	0.81
30	14219	32.63	17.16	1.16
71	16305	42.00	15.90	1.32

Cost of Contest Transaction

Parties	Tx size	% max Mem	% max CPU	Min fee ₳
1	9365	8.56	3.49	0.66
2	9485	8.94	3.79	0.67
3	9692	10.12	4.40	0.69
5	10022	11.72	5.32	0.72
10	10882	16.06	7.75	0.81
30	14213	32.32	17.03	1.16
43	16358	42.60	22.95	1.38

Cost of Abort Transaction

Some variation because of random mixture of still initial and already committed outputs.

Parties	Tx size	% max Mem	% max CPU	Min fee ₳
1	13704	25.73	10.91	1.04
2	13816	38.20	16.25	1.18
3	13927	52.63	22.47	1.35
4	14037	69.21	29.65	1.54
5	14433	97.27	42.48	1.88

Cost of FanOut Transaction

Involves spending head output and burning head tokens. Uses ada-only UTxO for better comparability.

UTxO	Tx size	% max Mem	% max CPU	Min fee ₳
1	13415	9.88	4.31	0.85
2	13452	11.40	5.19	0.87
3	13492	13.22	6.19	0.90
5	13564	15.96	7.83	0.93
10	13743	23.82	12.33	1.04
50	15182	84.80	47.57	1.85
60	15540	99.99	56.37	2.05

[ch1bo]

This is already a good improvement!

Maybe we can make some access onto this structure simpler in the HeadLogic module (number . getSnapshot) confirmedSnapshot + 1 == sn is a bit verbose), but that can be story of another PR.

[ch1bo]

The tests are failing because you need to update the api.yaml with the new JSON schema for ConfirmedSnapshot, here https://github.com/input-output-hk/hydra-poc/blob/5681f52646fc17224453be6b1624a61794cbbcd7/hydra-node/json-schemas/api.yaml#L610

[github-actions]

Unit Test Results

257 tests  ±0   251 ✔️ ±0   14m 59s ⏱️ + 1m 0s
  91 suites ±0       6 💤 ±0 
    5 files   ±0       0 ❌ ±0 

Results for commit 3bd7fac. ± Comparison against base commit 4027ca0.

♻️ This comment has been updated with latest results.

[pgrange]

I convert it back to a draft as I'd like to go through some checklist proposal:

Pre-submit checklist

• No errors in the CI (don't state the obvious)
• CHANGELOG is up to date

[ghost]

I would argue that No Errors in the CI should not be part of the checklist as it's automatically tested in the PR. I mean you won't merge a PR when CI is red would you (except in exceptional circumstnaces like the ones we faced with this failing ci/hydra job...)

[ch1bo]

Still approve :)