import xorq as xo
import xorq.expr.datatypes as dt

con = xo.connect()
test_data = xo.memtable({
    'id': [1, 2, 3, 4, 5],
    'text_col': ['A', 'B', 'C', 'D', 'E']
})

print(test_data.to_pyarrow_batches().schema)

large_string_table = test_data.mutate(text_col=xo._.text_col.cast(dt.LargeString))


original_schema = large_string_table.to_pyarrow_batches().schema
print(original_schema)
print(f"text_col type: {original_schema.field('text_col').type}")

large_string_table.to_parquet("test_largestring.parquet")

read_back = xo.deferred_read_parquet(path="test_largestring.parquet", con=con)
final_schema = read_back.to_pyarrow_batches().schema