This is an automated email from the ASF dual-hosted git repository.
vinishjail97 pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/incubator-xtable.git
The following commit(s) were added to refs/heads/main by this push:
new e9e05c54 [901] Document Parquet source limitations (#937)
e9e05c54 is described below
commit e9e05c54e057a2dcf708e5aa9a37d5836316450a
Author: Akash Reddy Jammula <[email protected]>
AuthorDate: Mon Sep 14 11:02:50 2026 -0700
[901] Document Parquet source limitations (#937)
Signed-off-by: Akash Reddy Jammula <[email protected]>
---
website/docs/features-and-limitations.md | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/website/docs/features-and-limitations.md
b/website/docs/features-and-limitations.md
index d3542600..806c9e2a 100644
--- a/website/docs/features-and-limitations.md
+++ b/website/docs/features-and-limitations.md
@@ -42,3 +42,7 @@ HMS and AWS Glue are the two catalogs supported right now,
support for other cat
### Delta
- When using Delta as the source for an Iceberg target, you may require field
IDs set in the parquet schema. To enable that, follow the instructions for
enabling column mapping
[here](https://docs.delta.io/latest/delta-column-mapping.html).
- When Delta is the source, Generated Columns are not synced to the target
schema. For tables that are partitioned on Generated Columns, there is limited
support. For example, we support date functions like transforming a timestamp
to `yyyy-MM-dd` format. Please file a GitHub issue or pull-request for any
cases that you think should be supported.
+
+### Parquet
+- Schema evolution across files in a Parquet source directory is not
supported. XTable derives the table schema from the footer of a file with the
most recent filesystem modification time; it does not merge or comprehensively
validate schemas across all files. If the files have different schemas,
synchronization may use a schema that does not represent every file or may fail
while processing them. Copying or restoring files can also change their
modification times and therefore change wh [...]
+- Hive-style partition columns that exist only in directory paths are not
supported. When partition extraction is configured, XTable resolves each source
field against the selected file's schema. If a partition column exists only in
a directory path, such as `region=eu/`, XTable cannot synthesize the field or
infer its type, and synchronization fails.
