thswlsqls opened a new issue, #9074: URL: https://github.com/apache/paimon/issues/9074
**Search before asking** - [x] I searched in the [issues](https://github.com/apache/paimon/issues) and found nothing similar. **Paimon version** master @ a5e87ebf6 (2.1-SNAPSHOT) **Compute Engine** Hive **Minimal reproduce step** 1. Create a Paimon table with a `CHAR(300)` (or `VARCHAR(100000)`) column via Flink/Spark — Paimon allows lengths up to `Integer.MAX_VALUE`. 2. Read that table from Hive. `PaimonSerDe.initialize()` builds a `PaimonInternalRowObjectInspector`, which calls `PaimonObjectInspectorFactory.create()` (paimon-hive/paimon-hive-connector-common, line 65-74) for every field. That method passes the Paimon length straight to the inspector constructor, which throws `RuntimeException: Char length 300 out of allowed range [1, 255]`. **What doesn't meet your expectations?** Expected: the column is read as `string`. Actual: a single long column makes the whole table unreadable from Hive. **Anything else?** The sibling `HiveTypeUtils.PaimonToHiveTypeVisitor` (paimon-hive/paimon-hive-common, line 149-164) already clamps to `string` with `HiveChar.MAX_CHAR_LENGTH` / `HiveVarchar.MAX_VARCHAR_LENGTH`. That clamp came from #1568 (fix #1565), which did not update this factory, so `varchar(2147483646)` from #1565 still fails here. **Are you willing to submit a PR?** - [x] I'm willing to submit a PR! -- 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]
