How can a node verify a transaction with a script address without the script?

Question

How can a node verify a transaction with a script address without the script?

I mean when i start a node months after the transaction has been minted into a block. How can my node verify that the script conditions were met since only the hash is persisted in the chain?

Answer 1

By inspecting the current ledger CDDL specifications, you can see that each block contains a transaction_witness_sets field, where each transaction_witness_set is defined as:

transaction_witness_set =
  { ? 0: [* vkeywitness ]
  , ? 1: [* native_script ]
  , ? 2: [* bootstrap_witness ]
  }

Here, native_script, if present, contains some rules that need to be respected in order for the transaction to be valid (for instance, some minting policy):

native_script =
  [ script_pubkey
  // script_all
  // script_any
  // script_n_of_k
  // invalid_before
     ; Timelock validity intervals are half-open intervals [a, b).
     ; This field specifies the left (included) endpoint a.
  // invalid_hereafter
     ; Timelock validity intervals are half-open intervals [a, b).
     ; This field specifies the right (excluded) endpoint b.
  ]

So not only the script hash is included in the chain, but also the actual rules it contains.

Answer 2

A node initially executes a valid Plutus transaction and records it to the chain, it also adds an "isValid" tag to that transaction record.

When a new node starts syncing up and arrives at a Plutus transaction, it only looks out for the "isValid" tag and then adds that transaction to its chain if it does have the tag, this way new nodes don't have to re-execute the script but still have proof they belong on the chain.