[GitHub] [iceberg] rdblue commented on a change in pull request #2805: Spec: Add back distinct_counts in data_file metadata

[GitBox <gi...@apache.org>]

rdblue commented on a change in pull request #2805:
URL: https://github.com/apache/iceberg/pull/2805#discussion_r668347737



##########
File path: site/docs/spec.md
##########
@@ -375,7 +375,7 @@ The schema of a manifest file is a struct called `manifest_entry` with the follo
 | _optional_ | _optional_ | **`109  value_counts`**           | `map<119: int, 120: long>`   | Map from column id to number of values in the column (including null and NaN values) |
 | _optional_ | _optional_ | **`110  null_value_counts`**      | `map<121: int, 122: long>`   | Map from column id to number of null values in the column |
 | _optional_ | _optional_ | **`137  nan_value_counts`**       | `map<138: int, 139: long>`   | Map from column id to number of NaN values in the column |
-| _optional_ |            | ~~**`111 distinct_counts`**~~     | `map<123: int, 124: long>`   | **Deprecated. Do not write.** |
+| _optional_ | _optional_ | **`111  distinct_counts`**        | `map<123: int, 124: long>`   | Map from column id to number of distinct values in the column; distinct counts must be produced using values in the file, not on merged counts from other metadata |

Review comment:
       I think it is a bit more clear to state that this can't be a merged count. If someone is compacting data files without reading the data, like merging Parquet row groups without re-encoding then it may seem reasonable to merge the counts somehow. I'm not sure that people would consider merging an operation that doesn't reflect the distinct count in the file, that's why I mentioned it specifically.
   
   Is there anything else that you're including in cases that would not "reflect the distinct count in the file"? I'm trying to use language that allows non-exact values. I think we should be in agreement that sketch-based counts are okay.




-- 
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: issues-unsubscribe@iceberg.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@iceberg.apache.org
For additional commands, e-mail: issues-help@iceberg.apache.org