Struct Dataset 

pub struct Dataset {
    pub id: String,
    pub name: String,
    pub version: Option<String>,
    pub schema: Vec<(String, String)>,
    pub row_count: Option<u64>,
    pub content_hash: String,
    pub url: Option<String>,
    pub license: Option<String>,
    pub provenance: Provenance,
    pub created: String,
}

v0.33: Dataset as a first-class kernel object.

A Dataset is a versioned, content-addressed reference to data that anchors empirical claims. Before v0.33, datasets were strings in Provenance.title or entity-typed mentions in assertions — a claim could say “we used ADNI” without anchoring which release of ADNI the analysis ran against, and re-running the same code on a refreshed cohort silently produced a “different” claim.

vd_<id> is content-addressed over `name + version + content_hash

• url`. Two dataset records with the same name but different versions get distinct ids; two records pointing at the same snapshot collapse to the same id. Claims can reference the exact bytes they rest on, not only a dataset name in prose.

Fields

id: String

vd_<16hex>, content-addressed; see Dataset::content_address.

name: String

Human-readable name (e.g. “ADNI”, “TRAILBLAZER-ALZ”, “MIMIC-IV”).

version: Option<String>

Semantic version or release tag (e.g. “ADNI-3”, “v2.2”, “SR0”). Two entries differing only in version are distinct kernel objects.

schema: Vec<(String, String)>

Optional column-level schema as (name, type) pairs. For non-tabular datasets, leave empty.

row_count: Option<u64>

Number of rows / observations / records, when known.

content_hash: String

SHA-256 of the canonical contents, when computable. For large datasets stored remotely, this is the publisher’s declared content hash; integrity verification is the puller’s job (same pattern as vfr_* snapshots).

url: Option<String>

Where the dataset is reachable (https URL, file://, s3://, etc.).

license: Option<String>

License identifier or URL (e.g. “CC-BY-4.0”, a Crossref license).

provenance: Provenance

Provenance of the dataset itself — typically the paper or release that publishes it. Reuses Provenance for shape parity with findings.

created: String

RFC 3339 creation timestamp.

Implementations

impl Dataset

pub fn content_address( name: &str, version: Option<&str>, content_hash: &str, url: Option<&str>, ) -> String

Compute the content-addressed ID per v0.33 spec: SHA-256(name | version | content_hash | url). Returns first 16 hex chars prefixed with “vd_”.

pub fn new( name: impl Into<String>, version: Option<String>, content_hash: impl Into<String>, url: Option<String>, license: Option<String>, provenance: Provenance, ) -> Self

Construct a new Dataset with a freshly-derived id and created timestamp set to now.

Trait Implementations

impl Clone for Dataset

fn clone(&self) -> Dataset

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for Dataset

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for Dataset

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for Dataset

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.