[ARROW-1398] [Python] No support reading columns of type decimal(19,4)

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.6.0
• Fix Version/s: 0.8.0
• Component/s: Python
• Labels: decimal python
• Environment: Ubuntu 16.04, Python 3.6
• External issue URL: https://github.com/apache/arrow/issues/17425
• Assignee: Phillip Cloud
• Reporter: Ludwik Bielczynski
• Created: 22/Aug/17 14:48
• Updated: 11/Jan/23 07:14
• Resolved: 05/Dec/17 03:40

Description

I have a localy saved parquet database created in spark from querrying an SQL database. When I run:

import pyarrow.parquet as pq

path = "path/to/parquet/dataset"
dataset = pq.ParquetDataset(path)
dataset.read()

an error indicating that there is no support for reading columns of type decimal(19,4). It's quite a common type used in SQL databases and I saw in the source code that there is an implementation for decimals. I'm stuck trying to figuring out a solution. Is there a walk around (conversion of decimals to integers during reading)?

Issue Links

• depends upon: PARQUET-1095 [C++] Read and write Arrow decimal values
• depends upon: ARROW-1839 [C++/Python] Add Decimal Parquet Read/Write Tests