Rachelint commented on code in PR #15591:
URL: https://github.com/apache/datafusion/pull/15591#discussion_r2072859880
##########
datafusion/functions-aggregate-common/src/aggregate/groups_accumulator.rs:
##########
@@ -507,3 +509,157 @@ pub(crate) fn slice_and_maybe_filter(
Ok(sliced_arrays)
}
}
+
+// ===============================================
+// Useful tools for group index
+// ===============================================
+
+/// Operations about group index parsing
+///
+/// There are mainly 2 `group index` needing parsing: `flat` and `blocked`.
+///
+/// # Flat group index
+/// `flat group index` format is like:
+///
+/// ```text
+/// | block_offset(64bit) |
+/// ```
+///
+/// It is used in `flat GroupValues/GroupAccumulator`, only a single block
+/// exists, so its `block_id` is always 0, and use all 64 bits to store the
+/// `block offset`.
+///
+/// # Blocked group index
+/// `blocked group index` format is like:
+///
+/// ```text
+/// | block_id(32bit) | block_offset(32bit)
+/// ```
+///
+/// It is used in `blocked GroupValues/GroupAccumulator`, multiple blocks
+/// exist, and we use high 32 bits to store `block_id`, and low 32 bit to
+/// store `block_offset`.
+///
+/// The `get_block_offset` method requires to return `block_offset` as u64,
+/// that is for compatible for `flat group index`'s parsing.
+///
+pub trait GroupIndexOperations: Debug {
Review Comment:
Addressed.
--
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: github-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
