This documentation is for an unreleased version of Apache Paimon. We recommend you use the latest stable version.

Data Evolution #

PyPaimon for Data Evolution mode. See Data Evolution.

Update Columns By Row ID #

You can create TableUpdate.update_by_arrow_with_row_id to update columns to data evolution tables.

The input data should include the _ROW_ID column, update operation will automatically sort and match each _ROW_ID to its corresponding first_row_id, then groups rows with the same first_row_id and writes them to a separate file.

simple_pa_schema = pa.schema([
  ('f0', pa.int8()),
  ('f1', pa.int16()),
])
schema = Schema.from_pyarrow_schema(simple_pa_schema,
                                    options={'row-tracking.enabled': 'true', 'data-evolution.enabled': 'true'})
catalog.create_table('default.test_row_tracking', schema, False)
table = catalog.get_table('default.test_row_tracking')

# write all columns
write_builder = table.new_batch_write_builder()
table_write = write_builder.new_write()
table_commit = write_builder.new_commit()
expect_data = pa.Table.from_pydict({
  'f0': [-1, 2],
  'f1': [-1001, 1002]
}, schema=simple_pa_schema)
table_write.write_arrow(expect_data)
table_commit.commit(table_write.prepare_commit())
table_write.close()
table_commit.close()

# update partial columns
write_builder = table.new_batch_write_builder()
table_update = write_builder.new_update().with_update_type(['f0'])
table_commit = write_builder.new_commit()
data2 = pa.Table.from_pydict({
  '_ROW_ID': [0, 1],
  'f0': [5, 6],
}, schema=pa.schema([
  ('_ROW_ID', pa.int64()),
  ('f0', pa.int8()),
]))
cmts = table_update.update_by_arrow_with_row_id(data2)
table_commit.commit(cmts)
table_commit.close()

# content should be:
#   'f0': [5, 6],
#   'f1': [-1001, 1002]