Struct RelTable 

pub struct RelTable<'a, B>
where
    B: StorageBackend,
{ /* private fields */ }

One table of a RelationalDb. Every method runs (and commits) its own transaction; use RelationalDb::write_tx or Db::write_tx to group several operations atomically.

Implementations

impl<'a, B> RelTable<'a, B>

where B: StorageBackend,

pub fn schema(&self) -> &TableSchema

The schema this handle was created with (for a handle from RelationalDb::table_named, the definition at that time — operations always use the current one).

pub fn insert( &self, values: BTreeMap<String, PropValue>, ) -> Result<PropValue, BknError>

Inserts a new row. For an auto-increment schema, values should not include the primary-key column (it’s allocated here); the generated PK is returned. For a non-auto-increment schema, values must include the primary-key column — use RelTable::insert_with_pk instead if the PK comes from elsewhere (e.g. an existing NodeId). Fails with BknError::DuplicateKey if the pk is taken.

pub fn insert_bulk( &self, rows: impl IntoIterator<Item = BTreeMap<String, PropValue>>, ) -> Result<Vec<PropValue>, BknError>

Inserts multiple rows in a single transaction, reserving PKs in one counter update if auto-increment is enabled.

pub fn insert_with_pk( &self, pk: PropValue, values: BTreeMap<String, PropValue>, ) -> Result<(), BknError>

Inserts a row under an explicit, caller-supplied PK — the mechanism for linking a relational row to an existing id from elsewhere (e.g. a graph crate::graph::NodeId). Rejected on an auto-increment schema, where PKs must come from RelTable::insert instead.

pub fn insert_with_pk_bulk( &self, rows: impl IntoIterator<Item = (PropValue, BTreeMap<String, PropValue>)>, ) -> Result<(), BknError>

Inserts multiple rows with caller-supplied PKs in a single transaction.

pub fn upsert( &self, values: BTreeMap<String, PropValue>, ) -> Result<PropValue, BknError>

Like RelTable::insert, but replaces an existing row with the same pk instead of failing. On an auto-increment schema, a row without a pk gets a fresh one and a row with a pk is upserted under it.

pub fn upsert_bulk( &self, rows: impl IntoIterator<Item = BTreeMap<String, PropValue>>, ) -> Result<Vec<PropValue>, BknError>

pub fn upsert_with_pk( &self, pk: PropValue, values: BTreeMap<String, PropValue>, ) -> Result<(), BknError>

Inserts the row, or replaces the existing row with the same PK (keeping every secondary index consistent). Works on any schema, including auto-increment ones.

pub fn upsert_with_pk_bulk( &self, rows: impl IntoIterator<Item = (PropValue, BTreeMap<String, PropValue>)>, ) -> Result<(), BknError>

pub fn get(&self, pk: &PropValue) -> Result<Option<Row>, BknError>

pub fn select(&self) -> SelectQuery<'a, B>

pub fn update(&self) -> UpdateQuery<'a, B>

pub fn delete(&self) -> DeleteQuery<'a, B>

pub fn select_prefix( &self, column: &str, prefix: &str, ) -> Result<Vec<Row>, BknError>

Rows whose column string value starts with prefix, using that column’s secondary index if available, falling back to a full scan.

Trait Implementations

impl<'a, B> Clone for RelTable<'a, B>

where B: StorageBackend,

fn clone(&self) -> RelTable<'a, B>

Returns a duplicate of the value.

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

Performs copy-assignment from source.