Module Tezos_raw_protocol_022_PsRiotum.Constants_repr

val version_value : string
val version : string
val fitness_version_number : string
val proof_of_work_nonce_size : int
val nonce_length : int
val max_anon_ops_per_block : int
val max_proposals_per_delegate : int
val max_operation_data_length : int
val max_micheline_node_count : int

A global size limit on the size of Micheline expressions after expansion.

We want to prevent constants from being used to create huge values that could potentially do damage if ever printed or sent over the network. We arrived at this number by finding the largest possible contract in terms of number of nodes. The number of nodes is constrained by the current "max_operation_data_length" (32768) to be ~10,000 ( see "large_flat_contract.tz" in the tezt suite for the largest contract with constants that can be originated). As a first approximation, we set the node size limit to 5 times this amount.

val max_micheline_bytes_limit : int

Same as max_micheline_node_count but for limiting the combined bytes of the strings, ints and bytes in a expanded Micheline expression.

val max_allowed_global_constant_depth : int

Represents the maximum depth of an expression stored in the table after all references to other constants have (recursively) been expanded, where depth refers to the nesting of Prim and/or Seq nodes.

The size was chosen arbitrarily to match the typechecker in Script_ir_translator.

val michelson_maximum_type_size : int

A global size limit on the size of Michelson types.

The size of a type is the number of nodes in its AST representation. See Script_typed_ir.TYPE_SIZE.

val denunciation_period : int

The number of **full cycles** after a misbehaviour during which it can still be denounced.

More precisely, the window for denouncing a misbehaviour that happens during cycle n spans from the misbehaviour itself until the end of cycle n + denunciation_period. So it indeed lasts denunciation_period full cycles plus part of cycle n.

(Technically, denunciations of the misbehaviour are actually allowed from the beginning of cycle n, even on earlier levels than the misbehaviour's. But this is only relevant if someone crafts double blocks or consensus operations in the future for some reason. In practice, we expect misbehaviours to only happen around the time when the chain reaches their level, which is why we don't count cycle n itself as a full cycle for denouncing.)

Note that we must have denunciation_period >= 1, otherwise if a misbehaviour happens at the very last block of a cycle, there is no time frame to denounce it.

Currently denunciation_period = 1 which is considered enough time for all misbehaviours to be denounced.

/!\ Several parts of the codebase may assume that denunciation_period = 1 **without being parametrized using this constant**. So if it is ever modified, the codebase needs to be examined extensively; searching for places that use this constant is not enough.

val slashing_delay : int

The number of **full cycles** between a misbehaviour and its slashing (assuming that it has been denounced).

That is, a misbehaviour that happens during cycle n is always slashed at the end of cycle n + slashing_delay, regardless of when it has been denounced (and if it has not been denounced by then, then it is too late and it will never be slashed).

Note that we must have slashing_delay >= denunciation_period, otherwise the chain would accept denunciations for which the slashing can no longer happen because it is too late.

Also note that to make the slashing possible, information on baking rights for cycle n needs to be kept in the context until the end of cycle n + slashing_delay.

/!\ Several parts of the codebase may assume that slashing_delay = 1 **without being parametrized using this constant**. So if it is ever modified, the codebase needs to be examined extensively; searching for places that use this constant is not enough.

val sc_max_wrapped_proof_binary_size : int

A size limit for Sc_rollups.wrapped_proof binary encoding.

val sc_rollup_message_size_limit : int

A limit on the size of the binary encoding for sc rollup messages: Sc_rollup_inbox_message_repr.t and Sc_rollup_outbox_message_repr.t

val sc_rollup_max_number_of_messages_per_level : Tezos_protocol_environment_022_PsRiotum.Z.t

A limit on the number of messages in a inbox level enforced in Sc_rollup_inbox_repr.t.

type fixed
type Tezos_protocol_environment_022_PsRiotum.Error_monad.error +=
  1. | Invalid_protocol_constants of string

performs some consistency checks on the protocol parameters

module Generated : sig ... end
module Derived : sig ... end

Pseudo-constants that are neither fixed as the ones defined above, nor protocol parameters from Constants_parametric_repr. Instead, they are derived from the protocol parameters.

type t = private {
  1. fixed : fixed;
  2. parametric : Constants_parametric_repr.t;
  3. derived : Derived.t;
}
val all_of_parametric : Constants_parametric_repr.t -> t
val cache_layout_size : int

For each subcache, a size limit needs to be declared once. However, depending how the protocol will be instantiated (sandboxed mode, test network, ...) we may want to change this limit. For each subcache, a parametric constant can be used to change the limit (see parametric).

The number of subcaches and the limits for all those subcaches form together what is called the cache_layout.

val cache_layout : Constants_parametric_repr.t -> int list

The cache_layout depends on parametric constants.