Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 58.92%. Comparing base (f09f8af) to head (ee7102e).
⚠️ Report is 567 commits behind head on main.

@@             Coverage Diff              @@
##               main    #2492      +/-   ##
============================================
+ Coverage     56.12%   58.92%   +2.80%     
- Complexity      976     1457     +481     
============================================
  Files           119      146      +27     
  Lines         11743    13550    +1807     
  Branches       2251     2356     +105     
============================================
+ Hits           6591     7985    +1394     
- Misses         4012     4343     +331     
- Partials       1140     1222      +82     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

I'm seeing a difference in the explain plan (but not the simplified plan) for tpc-ds q9 with Spark 3.5

expected:<...m#35, count#36, sum#[37, count#38]
 but was:<...m#35, count#36, sum#[15, count#37]

I'm seeing a difference in the explain plan (but not the simplified plan) for tpc-ds q9 with Spark 3.5

expected:<...m#35, count#36, sum#[37, count#38]
 but was:<...m#35, count#36, sum#[15, count#37]

I cannot reproduce this issue locally 😞

The expected plan has:

(18) CometHashAggregate
Input [5]: [count#34, sum#35, count#36, sum#37, count#38]
Keys: []
Functions [3]: [count(1), avg(UnscaledValue(ss_ext_discount_amt#31)), avg(UnscaledValue(ss_net_paid#32))]

The actual plan has:

(18) CometHashAggregate
Input [5]: [count#34, sum#35, count#36, sum#15, count#37]
Keys: []
Functions [3]: [count(1), avg(UnscaledValue(ss_ext_discount_amt#31)), avg(UnscaledValue(ss_net_paid#32))]

I'm seeing a difference in the explain plan (but not the simplified plan) for tpc-ds q9 with Spark 3.5

expected:<...m#35, count#36, sum#[37, count#38]
 but was:<...m#35, count#36, sum#[15, count#37]

I cannot reproduce this issue locally 😞

The expected plan has:

(18) CometHashAggregate
Input [5]: [count#34, sum#35, count#36, sum#37, count#38]
Keys: []
Functions [3]: [count(1), avg(UnscaledValue(ss_ext_discount_amt#31)), avg(UnscaledValue(ss_net_paid#32))]

The actual plan has:

(18) CometHashAggregate
Input [5]: [count#34, sum#35, count#36, sum#15, count#37]
Keys: []
Functions [3]: [count(1), avg(UnscaledValue(ss_ext_discount_amt#31)), avg(UnscaledValue(ss_net_paid#32))]

I can reproduce this now

This PR is too large to review. I am going to create smaller PRs to build up to this.