You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@shardingsphere.apache.org by GitBox <gi...@apache.org> on 2026/09/10 03:36:11 UTC
[GH] (shardingsphere/fix-firebird-blob-error-handling): Workflow run "E2E - Operation" is working again!
The GitHub Actions job "E2E - Operation" on shardingsphere.git/fix-firebird-blob-error-handling has succeeded.
Run started by GitHub user rishi-rana (triggered by linghengqian).
Head commit for run:
9b5f8a5008ff48e3c391c0fadd1d2efb7c89d60f / rishi-rana <ri...@hotmail.com>
Address review findings on Firebird BLOB write handle validation
- Validate the BLOB handle before iterating segments in
FirebirdBatchBlobSegmentsCommandExecutor, so an empty batch on an
unknown or closed handle no longer returns a success response.
Firebird validates the handle before entering the loop in
rem_port::put_segment().
- Treat a zero BLOB id in FirebirdOpenBlobCommandExecutor as the NULL
BLOB quad and open an empty BLOB instead of raising an invalid BLOB
id, matching blb::open2(). This mirrors the existing zero-id handling
in FirebirdExecuteStatementCommandExecutor.bindBlobParameters().
- Distinguish a handle opened for reading from an unknown handle in the
new shared FirebirdBlobWriteHandleValidator, raising
isc_cannot_update_old_blob instead of isc_bad_segstr_handle, matching
BLB_put_segment().
The read-handle check uses FirebirdBlobHandleGenerator allocation state
rather than read-cache cursor presence, because
FirebirdBlobReadCache.readSegment() removes the cursor once a BLOB is
fully consumed; a drained but unclosed read handle would otherwise be
reported as unknown.
Co-Authored-By: Claude Opus 5 <no...@anthropic.com>
Claude-Session: https://claude.ai/code/session_0126yyNgnn3J8br27eT4P5as
Report URL: https://github.com/apache/shardingsphere/actions/runs/34424826314
With regards,
GitHub Actions via GitBox