tanmayrauth opened a new pull request, #1120:
URL: https://github.com/apache/iceberg-go/pull/1120
Adds `table.PositionDeltaWriter` for the MoR insert/reinsert split on v3
tables. Mirrors the data-file half of Java's `SparkPositionDeltaWrite`:
- `Reinsert(batch)` — survivor rows from a position-delta rewrite, with
explicit non-null `_row_id` values to preserve identity.
- `Insert(batch)` — fresh rows; writer appends a null `_row_id` column so
the reader synthesizes a new ID at scan time.
Output schema is `SchemaWithRowID` (a new helper): `_row_id` written
explicitly, `_last_updated_sequence_number` left absent so it inherits from the
manifest entry's `data_sequence_number` — exactly the value the spec requires
for rewritten rows.
Scope: data files only. The position-delete entries that pair with
reinserts (turning an UPDATE into delete-old + reinsert) are the engine
driver's responsibility — the writer doc spells this out so a reader doesn't
expect Java's full `SparkPositionDeltaWrite` semantics from the name alone.
Part 2 of #999
Closes: #999
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
