andygrove opened a new pull request, #4202: URL: https://github.com/apache/datafusion-comet/pull/4202
## Which issue does this PR close? Starts a Spark 4.1 Known Limitations section so operator-facing compatibility issues on the experimental 4.1 profile have a home. First entry seeds it with #4199. ## Rationale for this change Spark 4.1 support is experimental and we're discovering a handful of limitations as we work through the 4.1 CI failures. Some of them are real user-visible gaps (not just test-side plumbing) and deserve a visible entry in the compatibility guide so users who hit them can self-diagnose. ## What changes are included in this PR? - \`docs/source/user-guide/latest/compatibility/spark-versions.md\`: add a "Known Limitations" section under Spark 4.1, mirroring the existing section under Spark 4.0, and seed it with the \`NullType\` Parquet issue (#4199). The entry describes the on-disk encoding mismatch (\`BOOLEAN + LogicalType::Unknown\` vs parquet-rs requiring \`INT32 + Unknown\`), the exact error users will see, and concrete workarounds. Additional 4.1 limitations will be appended to the section as they are characterised. ## How are these changes tested? Docs-only change; rendered locally to confirm markdown formatting. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
