Link Search Menu Expand Document

Releases for RepoDb.Firebird.BulkOperations

View the NuGet package here or download it directly here.

RepoDb.Firebird.BulkOperations (v1.0.0)

Released: September 8, 2026

First general availability (non-preview) release of the Firebird Bulk Operations package, promoting the preview below to stable v1.0.0. No functional changes since the preview — see that entry for the full feature set and known limitations.

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

Released: 2026-08-29

New

First release of the bulk operations extension for RepoDb.Firebird, built on FbBatchCommand — the FirebirdSql.Data.FirebirdClient driver’s native ADO.NET batching API — via FirebirdCommandBatcher.

Verification status: this package has been implemented and reviewed, but not yet exercised against a live Firebird instance. Verify the pseudo-table lifecycle, the EXECUTE BLOCK identity read-back paths, and the transactional behavior end-to-end before relying on this package in production.

What’s included

  • BulkInsert, BulkMerge, BulkUpdate, BulkDelete and BulkDeleteByKey, each with an Async overload, callable against an FbConnection with an entity list, a DataTable, or an IDataReader (BulkDeleteByKey takes a list of primary key values instead).
  • FirebirdCommandBatcher — an FbBatchCommand-based bulk-copy class with true asynchronous capability (FbBatchCommand.ExecuteNonQueryAsync), used internally to write into the target table (BulkInsert without ReturnIdentity) and into the pseudo (staging) table backing the other operations.
  • FirebirdBulkImportIdentityBehavior (KeepIdentity default, ReturnIdentity) — controls whether identity values are sent as-is or read back. ReturnIdentity reads generated values back via an EXECUTE BLOCK cursor loop over the staging table, correlated to the source rows by a client-assigned row-order column rather than by sorting the generated identities themselves.
  • FirebirdBulkImportPseudoTableType (Auto, Memory, Physical) — selects the staging-table strategy backing BulkMerge, BulkUpdate, BulkDelete, BulkDeleteByKey, and BulkInsert (when ReturnIdentity is used). Every staging table is created with a per-call unique name, so Memory and Physical are both genuinely functional and safe under concurrent callers writing against the same target table.
  • FirebirdCommandBatcherMapItem — explicit source-to-destination column mapping, with an optional FbDbType override.
  • FirebirdCommandBatcherColumnMappingCollection — the ColumnMappings collection backing FirebirdCommandBatcher.
  • FirebirdTraceKeys — the tracing key constants for the five bulk operations, for use with ITrace.
  • BulkMerge generates one of three statement shapes depending on whether the identity column is itself a qualifier and whether ReturnIdentity is requested: a single ANSI MERGE, an EXECUTE BLOCK loop of UPDATE OR INSERT ... RETURNING, or an EXECUTE BLOCK loop that branches per row between a plain INSERT and UPDATE OR INSERT — see BulkMerge for details.
  • When a caller-supplied FbTransaction is passed, every step of a bulk operation’s pipeline (pseudo-table DDL, the batched write, the cascading statement, and the pseudo-table drop) participates in it — and since Firebird’s DDL is itself transactional, the whole pipeline is genuinely atomic in that case.

Known limitations (v1)

  • Without an explicit caller-supplied transaction, each step of a bulk operation’s pipeline runs under its own implicit transaction, so a failure partway through (e.g. after the pseudo table is created but before it is dropped) can leave an orphaned pseudo table behind.
  • BulkCopyTimeout is accepted on every operation, but FbBatchCommand has no timeout-equivalent property to apply it to — the argument currently has no effect.
  • BulkMerge/BulkUpdate/BulkDelete/BulkDeleteByKey report Firebird’s engine-level affected-row count where available, but MERGE/EXECUTE BLOCK statements always report -1 from ExecuteNonQuery — for those, the pseudo table’s own row count is used as a proxy instead, on the assumption that every staged row is guaranteed to be either inserted, updated, or matched for deletion.
  • 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.
  • BulkUpdate returns 0 immediately, without creating a pseudo table at all, when every staged field is also a qualifier (i.e. there is nothing left to actually update).
  • Every bulk call against a table it hasn’t seen before creates its own uniquely-named staging table (CREATE TABLE/CREATE GLOBAL TEMPORARY TABLE) and drops it once the call completes, rather than reusing one across calls — trading a small amount of per-call DDL overhead for full safety under concurrent callers.

  • This package inherits every Known limitation of RepoDb.Firebird itself (e.g. no table hints, MaxParameterCount of 1500, Firebird 3.0+ required for identity detection).

  • Referenced the RepoDb package v1.16.0.
  • Referenced the RepoDb.Firebird package v0.0.1-alpha.
  • Referenced the FirebirdSql.Data.FirebirdClient package v10.3.4.