manirajv06 commented on issue #13855: URL: https://github.com/apache/iceberg/issues/13855#issuecomment-3281986051
> at least there seems to be a path here without changing the specification. We are not going to change the spec. As mentioned earlier, `linking schema id with file` approach has been ruled out. We are proposing "columns written" similar to `rowCount` defined in https://github.com/apache/iceberg/blob/46ab802cdc6ff39c05a3fecc608a721b26aece7f/api/src/main/java/org/apache/iceberg/Metrics.java#L34 to capture all the column id's similar to how `rowCount` being captured in https://github.com/apache/iceberg/blob/46ab802cdc6ff39c05a3fecc608a721b26aece7f/parquet/src/main/java/org/apache/iceberg/parquet/ParquetMetrics.java#L79. This should happen always, not required to be done based on any new configuration. `DataFile` depends on `Metrics` to populate its own values. Once we hook "columns written" info in `DataFile`, same can be used anywhere where `DataFile` is being used. For example, `InclusiveMetricsEvalluator`, `StrictMetricsEvalluator` evaluators does the evaluation based on file. Are we good? Does that makes sense? -- 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]
