You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@calcite.apache.org by "Stamatis Zampetakis (Jira)" <ji...@apache.org> on 2026/05/28 12:07:00 UTC
[jira] [Updated] (CALCITE-7462) Support custom field delimiters in the CSV adapter
[ https://issues.apache.org/jira/browse/CALCITE-7462?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Stamatis Zampetakis updated CALCITE-7462:
-----------------------------------------
Fix Version/s: 1.42.0
> Support custom field delimiters in the CSV adapter
> ---------------------------------------------------
>
> Key: CALCITE-7462
> URL: https://issues.apache.org/jira/browse/CALCITE-7462
> Project: Calcite
> Issue Type: Improvement
> Reporter: Diveyam Mishra
> Priority: Minor
> Fix For: 1.42.0
>
>
> Currently, the Calcite CSV adapter (`CsvTableFactory`, etc.) hardcodes the comma (`,`) as the field separator. This prevents users from mapping schemas to other popular delimited format types, such as pipe-separated (`|`) or tab-separated (`\t`) files.
>
> This improvement introduces support for a `separator` property in the model JSON's operand map for the CSV adapter.
>
> Example Usage:
> {
> "version": "1.0",
> "defaultSchema": "TEST",
> "schemas": [
> {
> "name": "TEST",
> "type": "custom",
> "factory": "org.apache.calcite.adapter.file.CsvTableFactory",
> "operand": {
> "file": "data.csv",
> "separator": "|"
> }
> }
> ]
> }
>
> If no separator is provided in the model JSON, the adapter defaults to the standard comma (`,`) character, strictly preserving backward compatibility.
> (PS: doing this for the first time so make sure to tell me if something else is needed )
--
This message was sent by Atlassian Jira
(v8.20.10#820010)