Merge pull request #476 from tidesdb/tidesql-batch-correction

adjust tidesql reference regarding upcoming minor

Author: guycipher

src/content/docs/reference/tidesql.md

@@ -360,7 +360,7 @@ The available isolation levels are `READ_UNCOMMITTED`, `READ_COMMITTED`, `REPEAT
 
 ### Bulk DML Batching
 
-Statements that touch many rows such as `LOAD DATA INFILE`, multi row `INSERT`, `INSERT ... SELECT`, and `UPDATE` or `DELETE` over a range keep the TidesDB transaction from growing unbounded by committing mid statement in fixed size batches. The engine hooks MariaDB's `start_bulk_insert`, `start_bulk_update`, and `start_bulk_delete` callbacks, counts the row operations (data write plus secondary index maintenance) against `TIDESDB_BULK_INSERT_BATCH_OPS` (50,000 ops), and at each threshold commits the current transaction and resets it with `READ_COMMITTED` for the next batch. This keeps statement memory bounded and keeps the transaction under `TDB_MAX_TXN_OPS` regardless of statement size. Autocommit semantics are preserved so a failure rolls back only the current batch, and the statement as a whole reports the first error encountered.
+Statements that touch many rows such as `LOAD DATA INFILE`, multi row `INSERT`, `INSERT ... SELECT`, and `UPDATE` or `DELETE` over a range keep the TidesDB transaction from growing unbounded by committing mid statement in fixed size batches. The engine hooks MariaDB's `start_bulk_insert`, `start_bulk_update`, and `start_bulk_delete` callbacks, counts the row operations (data write plus secondary index maintenance) against `TIDESDB_BULK_INSERT_BATCH_OPS` (500 ops), and at each threshold commits the current transaction and resets it with `READ_COMMITTED` for the next batch. This keeps statement memory bounded regardless of statement size. Autocommit semantics are preserved so a failure rolls back only the current batch, and the statement as a whole reports the first error encountered.
 
 The mid commit logic is shared between INSERT, UPDATE, and DELETE via a single `maybe_bulk_commit` helper, so the batching threshold and the iterator plus dup cache invalidation policy are identical across the three paths.