Link Search Menu Expand Document

Releases for RepoDb.EnterpriseDb.BulkOperations

View the NuGet package here or download it directly here.

RepoDb.EnterpriseDb.BulkOperations (v0.0.1-alpha) - Preview

Released: TBA

New

First release of the bulk operations extension for RepoDb.EnterpriseDb. Targets .NET 8, .NET 9, and .NET 10.

Architectural note: rows are staged through a pseudo (staging) table and loaded via RepoDb.Connector.EnterpriseDb’s EDBBulkCopy, itself built on Npgsql’s native binary COPY protocol — a genuine bulk load, not a row-by-row loop.

What’s included

  • BulkInsert, BulkMerge, BulkUpdate, BulkDelete and BulkDeleteByKey, each with an Async overload, callable against an EDBConnection with an entity list, a DataTable, or an IDataReader (BulkDeleteByKey takes a target table name and a list of primary key values directly, rather than an entity type parameter).
  • EDBBulkImportIdentityBehavior (KeepIdentity default, ReturnIdentity) — controls whether identity values are sent as-is or read back after BulkInsert/BulkMerge. ReturnIdentity for BulkMerge pre-assigns identities to unmatched rows via nextval(pg_get_serial_sequence(...)) before an ON CONFLICT DO UPDATE upsert; for BulkInsert it inserts straight from the pseudo table with a RETURNING clause.
  • EDBBulkImportPseudoTableType (Auto, Memory, Physical) — selects the staging-table strategy backing every operation; Memory creates a genuine session-scoped TEMP table, Physical an ordinary heap table under a deterministic name.
  • EDBBulkInsertMapItem — explicit source-to-destination column mapping, with an optional EDBType override. Explicit mappings are type-checked up front: an incompatible source/destination CLR type pairing (other than the allowed Guid↔string and integral-to-integral widenings) throws an InvalidTypeException immediately, rather than failing inside EnterpriseDB.
  • EDBTraceKeys — the tracing key constants for the five bulk operations, for use with ITrace.
  • BulkMerge (KeepIdentity) uses a two-statement cascade — an UPDATE ... FROM for matched rows, then INSERT ... WHERE NOT EXISTS (...) for unmatched ones. BulkMerge (ReturnIdentity) instead pre-assigns identities and runs a single INSERT ... ON CONFLICT DO UPDATE ... RETURNING.
  • BulkUpdate uses a correlated UPDATE ... FROM "PseudoTable" S WHERE (...) form. BulkDelete/BulkDeleteByKey use DELETE ... USING "PseudoTable" S WHERE (...).
  • Every pseudo table gets an identity __RepoDbBulkRowOrder__ column recording each row’s original load order, used to align generated identity values back onto the correct entity/DataRow after BulkInsert/BulkMerge with ReturnIdentity.
  • Explicit column mappings are type-checked against the target table’s DbField types before the write begins (Guid↔string and integral-to-integral widenings allowed), throwing InvalidTypeException up front on a mismatch.
  • Referenced the RepoDb package v1.16.0.
  • Referenced the RepoDb.EnterpriseDb package v0.0.1-alpha.
  • Referenced the RepoDb.Connector.EnterpriseDb package v0.0.1-alpha2.

Known limitations (v1)

  • Pseudo table names are deterministic, not per-call-unique — built from the pseudoTableType, the target table name, and the operation (e.g. PhysicalPersonMerge). Two concurrent bulk calls of the same operation against the same table, both resolving to the same pseudoTableType, can interfere with each other’s staged rows. Prefer Memory (a genuine per-session TEMP table) over Physical when running concurrent bulk operations from different connections against the same table.
  • The DbDataReader overloads of BulkInsert and BulkMerge have no identityBehavior argument — a forward-only, single-pass reader cannot be rewound to correlate generated identity values back onto a source row.
  • BulkDeleteByKey takes a tableName string directly (BulkDeleteByKey<TPrimaryKey>(connection, tableName, primaryKeys, ...)) rather than an entity type parameter.
  • This package inherits every Known limitation of RepoDb.EnterpriseDb itself (e.g. no table hints).