Add $bucketAuto _id override and reshaped-upstream compatibility tests

docs/testing/TEST_COVERAGE.md

@@ -424,6 +424,7 @@ For each invalid_type in [string, object, array, ...]:
 - Pass-through stages (`$limit`, `$skip`, `$sort`, `$match`) must preserve all BSON types unchanged, including deprecated types
 - Reshaping stages (`$project`, `$set`, `$unset`, `$addFields`) must be tested with all BSON types as values
 - New-document stages (`$count`, `$group`, `$bucket`, `$sortByCount`) must verify output field names and types
+- `_id`-synthesizing stages (`$bucket`, `$bucketAuto`) must verify `_id` in `output`: `$bucket` rejects it; `$bucketAuto` lets an `_id` accumulator override the default `{min, max}` boundary
 
 **Stage Interactions**:
 - Multi-stage interaction tests belong in the parent `stages/` directory, not in individual stage folders. Per `FOLDER_STRUCTURE.md`, interactions between same-level features go in the parent folder (e.g., `stages/test_stages_combination_sort.py`, `stages/test_stages_position_match.py`).

documentdb_tests/compatibility/tests/core/operator/stages/bucketAuto/test_bucketAuto_output.py

@@ -225,6 +225,57 @@
     ),
 ]
 
+# Property [_id Output Override]: unlike $bucket (which reserves _id), $bucketAuto
+# permits _id in the output specification; an _id accumulator replaces the default
+# {min, max} boundary document with the accumulated value.
+BUCKET_AUTO_ID_OVERRIDE_TESTS: list[StageTestCase] = [
+    StageTestCase(
+        "id_override_replaces_boundary",
+        docs=[{"_id": 1, "x": 1, "v": 10}, {"_id": 2, "x": 5, "v": 20}],
+        pipeline=[
+            {
+                "$bucketAuto": {
+                    "groupBy": "$x",
+                    "buckets": 1,
+                    "output": {"_id": {"$sum": "$v"}},
+                }
+            }
+        ],
+        expected=[{"_id": 30}],
+        msg="$bucketAuto output _id accumulator should replace the default {min, max} boundary _id",
+    ),
+    StageTestCase(
+        "id_override_with_other_output_fields",
+        docs=[{"_id": 1, "x": 1, "v": 10}, {"_id": 2, "x": 5, "v": 20}],
+        pipeline=[
+            {
+                "$bucketAuto": {
+                    "groupBy": "$x",
+                    "buckets": 1,
+                    "output": {"_id": {"$max": "$v"}, "count": {"$sum": 1}},
+                }
+            }
+        ],
+        expected=[{"_id": 20, "count": 2}],
+        msg="$bucketAuto output _id override should coexist with other output fields",
+    ),
+    StageTestCase(
+        "id_override_applies_per_bucket",
+        docs=[{"_id": 1, "x": 1, "v": 10}, {"_id": 2, "x": 5, "v": 20}],
+        pipeline=[
+            {
+                "$bucketAuto": {
+                    "groupBy": "$x",
+                    "buckets": 2,
+                    "output": {"_id": {"$sum": "$v"}},
+                }
+            }
+        ],
+        expected=[{"_id": 10}, {"_id": 20}],
+        msg="$bucketAuto output _id override should be computed independently per bucket",
+    ),
+]
+
 # Property [Duplicate Output Field Names]: duplicate output field names resolve
 # to the last definition.
 BUCKET_AUTO_DUPLICATE_FIELD_NAME_TESTS: list[StageTestCase] = [
@@ -257,6 +308,7 @@
     + BUCKET_AUTO_NESTED_EXPR_TESTS
     + BUCKET_AUTO_PUSH_SYSTEM_VAR_TESTS
     + BUCKET_AUTO_OUTPUT_FIELD_NAME_TESTS
+    + BUCKET_AUTO_ID_OVERRIDE_TESTS
     + BUCKET_AUTO_DUPLICATE_FIELD_NAME_TESTS
 )
 

documentdb_tests/compatibility/tests/core/operator/stages/bucketAuto/test_bucketAuto_output_errors.py

@@ -198,11 +198,26 @@ def _out(value):
     ),
 ]
 
+# Property [Output Field _id Must Be Accumulator]: unlike $bucket, which reserves
+# _id and rejects it in output entirely, $bucketAuto permits _id in output but
+# still requires it to be an accumulator object; a non-accumulator _id is rejected
+# with the same error as any other non-accumulator output field.
+BUCKET_AUTO_OUTPUT_ID_NOT_ACCUMULATOR_TESTS: list[StageTestCase] = [
+    StageTestCase(
+        "output_id_non_accumulator_int",
+        docs=[{"_id": 1}],
+        pipeline=_out({"_id": 5}),
+        error_code=BUCKET_OUTPUT_NOT_ACCUMULATOR_ERROR,
+        msg="$bucketAuto should reject a non-accumulator _id value in output",
+    ),
+]
+
 BUCKET_AUTO_OUTPUT_ERROR_TESTS = (
     BUCKET_AUTO_OUTPUT_TYPE_TESTS
     + BUCKET_AUTO_OUTPUT_DOLLAR_PREFIX_TESTS
     + BUCKET_AUTO_OUTPUT_DOT_TESTS
     + BUCKET_AUTO_OUTPUT_NOT_ACCUMULATOR_TESTS
+    + BUCKET_AUTO_OUTPUT_ID_NOT_ACCUMULATOR_TESTS
 )