Struct Corpus 

pub struct Corpus {
    pub id: i32,
    pub path: String,
    pub name: String,
    pub complex: bool,
    pub description: String,
    pub parent_corpus_id: Option<i32>,
    pub selection: Option<Value>,
    pub public_id: Uuid,
}

A minimal description of a document collection. Defined by a name, path and simple/complex file system setup.

Fields

id: i32

auto-incremented postgres id

path: String

file system path to corpus root (a corpus is held in a single top-level directory)

name: String

a human-readable name for this corpus

complex: bool

are we using multiple files to represent a document entry? (if unsure, always use “true”)

description: String

a human-readable description of the corpus, maybe allow markdown here?

parent_corpus_id: Option<i32>

For a sandbox corpus: the parent corpus it was carved from (else None). See [crate::backend::sandbox].

selection: Option<Value>

For a sandbox corpus: the filter predicate it was built from — the JSON {service, severity, category, what} selection over the parent. This IS the sandbox’s provenance (“why these entries”), so no per-task origin link is kept.

public_id: Uuid

Stable external handle (UUIDv7, DB-generated), independent of the mutable name — for public/API references that must survive a rename. Immutable once assigned (Arm 3 / D8).

Implementations

impl Corpus

pub fn find_by_name( name_query: &str, connection: &mut PgConnection, ) -> Result<Self, Error>

ORM-like until diesel.rs introduces finders for more fields

pub fn find_by_id( id_query: i32, connection: &mut PgConnection, ) -> Result<Self, Error>

Find a corpus by its primary key (used to resolve a sandbox’s parent).

pub fn sandbox_id(&self) -> Option<i32>

The id under which this corpus’s result archives are name-scoped, or None for an ordinary corpus. A sandbox (parent_corpus_id.is_some()) scopes its outputs by its own id so a rerun can’t clobber the parent’s archives — see crate::helpers::result_archive_path (F-6).

pub fn task_count(&self, connection: &mut PgConnection) -> Result<i64, Error>

Total number of tasks registered under this corpus (across all services) — the blast radius a Corpus::destroy would remove. Used to preview a destructive delete before committing.

pub fn find_by_path( path_query: &str, connection: &mut PgConnection, ) -> Result<Self, Error>

ORM-like until diesel.rs introduces finders for more fields

pub fn all(connection: &mut PgConnection) -> Result<Vec<Self>, Error>

Returns all registered corpora, ordered by name.

pub fn document_counts(connection: &mut PgConnection) -> HashMap<i32, i64>

Document count per corpus id, for every corpus in one query (no N+1) — the number of import-service (id 2) tasks, which is one per ingested document. Used to show each corpus’s scale on the overview/landing without a per-corpus count. A corpus with no import tasks (none ingested yet) is simply absent from the map (treat as 0).

pub fn to_hash(&self) -> HashMap<String, String>

Return a hash representation of the corpus, usually for frontend reports

pub fn select_services( &self, connection: &mut PgConnection, ) -> Result<Vec<Service>, Error>

Return a vector of services currently activated on this corpus

pub fn destroy(self, connection: &mut PgConnection) -> Result<usize, Error>

Deletes a corpus and all its dependent rows — the log_* messages, the tasks, and the corpus registration — consuming the object. Runs in a single transaction so a crash mid-delete can’t leave a half-deleted corpus (crash-consistency, docs/DESIGN_PRINCIPLES.md).

The log_* tables have no foreign key to tasks (the only FK is historical_tasks.task_id → tasks ON DELETE CASCADE), so their rows must be deleted explicitly before the tasks or they orphan — this is why deletion lives in one complete primitive rather than a bare DELETE FROM corpora (the CLAUDE.md “deleting a corpus orphans log_* rows” hazard, now closed at the source so every caller is safe).

Trait Implementations

impl AsChangeset for Corpus

type Target = table

The table which Self::Changeset will be updating

type Changeset = <(Grouped<Eq<path, <String as AsExpression<<path as Expression>::SqlType>>::Expression>>, Grouped<Eq<name, <String as AsExpression<<name as Expression>::SqlType>>::Expression>>, Grouped<Eq<complex, <bool as AsExpression<<complex as Expression>::SqlType>>::Expression>>, Grouped<Eq<description, <String as AsExpression<<description as Expression>::SqlType>>::Expression>>, Option<Grouped<Eq<parent_corpus_id, <i32 as AsExpression<<parent_corpus_id as Expression>::SqlType>>::Expression>>>, Option<Grouped<Eq<selection, <Value as AsExpression<<selection as Expression>::SqlType>>::Expression>>>, Grouped<Eq<public_id, <Uuid as AsExpression<<public_id as Expression>::SqlType>>::Expression>>) as AsChangeset>::Changeset

The update statement this type represents

fn as_changeset(self) -> <Self as AsChangeset>::Changeset

Convert self into the actual update statement being executed

impl<'update> AsChangeset for &'update Corpus

where (Eq<path, &'update String>, Eq<name, &'update String>, Eq<complex, &'update bool>, Eq<description, &'update String>, Option<Eq<parent_corpus_id, &'update i32>>, Option<Eq<selection, &'update Value>>, Eq<public_id, &'update Uuid>): AsChangeset,

type Target = table

The table which Self::Changeset will be updating

type Changeset = <(Grouped<Eq<path, <&'update String as AsExpression<<path as Expression>::SqlType>>::Expression>>, Grouped<Eq<name, <&'update String as AsExpression<<name as Expression>::SqlType>>::Expression>>, Grouped<Eq<complex, <&'update bool as AsExpression<<complex as Expression>::SqlType>>::Expression>>, Grouped<Eq<description, <&'update String as AsExpression<<description as Expression>::SqlType>>::Expression>>, Option<Grouped<Eq<parent_corpus_id, <&'update i32 as AsExpression<<parent_corpus_id as Expression>::SqlType>>::Expression>>>, Option<Grouped<Eq<selection, <&'update Value as AsExpression<<selection as Expression>::SqlType>>::Expression>>>, Grouped<Eq<public_id, <&'update Uuid as AsExpression<<public_id as Expression>::SqlType>>::Expression>>) as AsChangeset>::Changeset

The update statement this type represents

fn as_changeset(self) -> <Self as AsChangeset>::Changeset

Convert self into the actual update statement being executed

impl<__FK> BelongsTo<&Corpus> for Task

where __FK: Hash + Eq, for<'__a> &'__a i32: Into<Option<&'__a __FK>>, for<'__a> &'__a Corpus: Identifiable<Id = &'__a __FK>,

type ForeignKey = __FK

The foreign key of this struct

type ForeignKeyColumn = corpus_id

The database column representing the foreign key of the table this struct represents

fn foreign_key(&self) -> Option<&Self::ForeignKey>

Returns the foreign key for self

fn foreign_key_column() -> Self::ForeignKeyColumn

Returns the foreign key column of this struct’s table

impl<__FK> BelongsTo<&Corpus> for HistoricalRun

where __FK: Hash + Eq, for<'__a> &'__a i32: Into<Option<&'__a __FK>>, for<'__a> &'__a Corpus: Identifiable<Id = &'__a __FK>,

type ForeignKey = __FK

The foreign key of this struct

type ForeignKeyColumn = corpus_id

The database column representing the foreign key of the table this struct represents

fn foreign_key(&self) -> Option<&Self::ForeignKey>

Returns the foreign key for self

fn foreign_key_column() -> Self::ForeignKeyColumn

Returns the foreign key column of this struct’s table

impl<__FK> BelongsTo<Corpus> for Task

where __FK: Hash + Eq, for<'__a> &'__a i32: Into<Option<&'__a __FK>>, for<'__a> &'__a Corpus: Identifiable<Id = &'__a __FK>,

type ForeignKey = __FK

The foreign key of this struct

type ForeignKeyColumn = corpus_id

The database column representing the foreign key of the table this struct represents

fn foreign_key(&self) -> Option<&Self::ForeignKey>

Returns the foreign key for self

fn foreign_key_column() -> Self::ForeignKeyColumn

Returns the foreign key column of this struct’s table

impl<__FK> BelongsTo<Corpus> for HistoricalRun

where __FK: Hash + Eq, for<'__a> &'__a i32: Into<Option<&'__a __FK>>, for<'__a> &'__a Corpus: Identifiable<Id = &'__a __FK>,

type ForeignKey = __FK

The foreign key of this struct

type ForeignKeyColumn = corpus_id

The database column representing the foreign key of the table this struct represents

fn foreign_key(&self) -> Option<&Self::ForeignKey>

Returns the foreign key for self

fn foreign_key_column() -> Self::ForeignKeyColumn

Returns the foreign key column of this struct’s table

impl Clone for Corpus

fn clone(&self) -> Corpus

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Corpus

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

Formats the value using the given formatter.

impl HasTable for Corpus

type Table = table

The table this type is associated with.

fn table() -> <Self as HasTable>::Table

Returns the table this type is associated with.

impl<'ident> Identifiable for &'ident Corpus

type Id = &'ident i32

The type of this struct’s identifier.

fn id(self) -> <Self as Identifiable>::Id

Returns the identifier for this record.

impl<__DB: Backend, __ST0, __ST1, __ST2, __ST3, __ST4, __ST5, __ST6, __ST7> Queryable<(__ST0, __ST1, __ST2, __ST3, __ST4, __ST5, __ST6, __ST7), __DB> for Corpus

where (i32, String, String, bool, String, Option<i32>, Option<Value>, Uuid): FromStaticSqlRow<(__ST0, __ST1, __ST2, __ST3, __ST4, __ST5, __ST6, __ST7), __DB>,

type Row = (i32, String, String, bool, String, Option<i32>, Option<Value>, Uuid)

The Rust type you’d like to map from.

fn build( row: (i32, String, String, bool, String, Option<i32>, Option<Value>, Uuid), ) -> Result<Self>

Construct an instance of this type

impl Serialize for Corpus

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

where __S: Serializer,

Serialize this value into the given Serde serializer.