fix(sanity): check REC006 only if either SampleAnnotation or ObjectAnn is not empty (#239)

* fix: remove REC006

Signed-off-by: ktro2828 <kotaro.uetake@tier4.jp>

* Revert "fix: remove REC006"

This reverts commit a9b763e.

Signed-off-by: ktro2828 <kotaro.uetake@tier4.jp>

* fix: update to run checking only if either sample_ann or object_ann is not empty

Signed-off-by: ktro2828 <kotaro.uetake@tier4.jp>

---------

Signed-off-by: ktro2828 <kotaro.uetake@tier4.jp>

Author: ktro2828

docs/schema/requirement.md

@@ -16,14 +16,14 @@
 
 ## Schema Record (`REC`)
 
-| ID       | Name                          | Severity | Description                             |
-| -------- | ----------------------------- | -------- | --------------------------------------- |
-| `REC001` | `scene-single`                | `ERROR`  | `Scene` record is a single.             |
-| `REC002` | `sample-not-empty`            | `ERROR`  | `Sample` record is not empty.           |
-| `REC003` | `sample-data-not-empty`       | `ERROR`  | `SampleData` record is not empty.       |
-| `REC004` | `ego-pose-not-empty`          | `ERROR`  | `EgoPose` record is not empty.          |
-| `REC005` | `calibrated-sensor-non-empty` | `ERROR`  | `CalibratedSensor` record is not empty. |
-| `REC006` | `instance-not-empty`          | `ERROR`  | `Instance` record is not empty.         |
+| ID       | Name                          | Severity | Description                                                                              |
+| -------- | ----------------------------- | -------- | ---------------------------------------------------------------------------------------- |
+| `REC001` | `scene-single`                | `ERROR`  | `Scene` record is a single.                                                              |
+| `REC002` | `sample-not-empty`            | `ERROR`  | `Sample` record is not empty.                                                            |
+| `REC003` | `sample-data-not-empty`       | `ERROR`  | `SampleData` record is not empty.                                                        |
+| `REC004` | `ego-pose-not-empty`          | `ERROR`  | `EgoPose` record is not empty.                                                           |
+| `REC005` | `calibrated-sensor-non-empty` | `ERROR`  | `CalibratedSensor` record is not empty.                                                  |
+| `REC006` | `instance-not-empty`          | `ERROR`  | `Instance` record is not empty if either 'SampleAnnotation' or 'ObjectAnn' is not empty. |
 
 ## Reference (`REF`)
 

t4_devkit/sanity/record/rec006.py

@@ -1,12 +1,20 @@
 from __future__ import annotations
 
+from typing import TYPE_CHECKING
+
+from returns.maybe import Maybe, Nothing, Some
+
 from t4_devkit.schema import SchemaName
 
 from ..checker import RuleID, RuleName, Severity
 from ..registry import CHECKERS
 from ..result import Reason
+from ..safety import load_json_safe
 from .base import RecordCountChecker
 
+if TYPE_CHECKING:
+    from ..context import SanityContext
+
 __all__ = ["REC006"]
 
 
@@ -17,9 +25,39 @@ class REC006(RecordCountChecker):
     id = RuleID("REC006")
     name = RuleName("instance-not-empty")
     severity = Severity.ERROR
-    description = "'Instance' record is not empty."
+    description = (
+        "'Instance' record is not empty if either 'SampleAnnotation' or 'ObjectAnn' is not empty."
+    )
     schema = SchemaName.INSTANCE
 
+    def can_skip(self, context: SanityContext) -> Maybe[Reason]:
+        # return skip reason if instance.json does not exist
+        match super().can_skip(context):
+            case Some(x):
+                return Maybe.from_value(x)
+
+        # instance.json should contain any records if either
+        # SampleAnnotation or ObjectAnn records exist
+        sample_ann_file = context.to_schema_file(SchemaName.SAMPLE_ANNOTATION)
+        object_ann_file = context.to_schema_file(SchemaName.OBJECT_ANN)
+
+        match (sample_ann_file.value_or(None), object_ann_file.value_or(None)):
+            case (None, _) | (_, None):
+                return Maybe.from_value(Reason("Missing 'annotation' directory"))
+            case (s, o):
+                sample_ann_count = len(load_json_safe(s).unwrap()) if s.exists() else 0
+                object_ann_count = len(load_json_safe(o).unwrap()) if o.exists() else 0
+                return (
+                    Maybe.from_value(
+                        Reason(
+                            f"Both {SchemaName.SAMPLE_ANNOTATION} "
+                            f"and {SchemaName.OBJECT_ANN} records are empty"
+                        )
+                    )
+                    if sample_ann_count == 0 and object_ann_count == 0
+                    else Nothing
+                )
+
     def check_count(self, records: list[dict]) -> list[Reason] | None:
         num_instance = len(records)
         return [Reason("'Instance' record must not be empty")] if num_instance == 0 else None

tests/sanity/test_record_checkers.py

@@ -13,6 +13,7 @@
 from t4_devkit.sanity.record.rec004 import REC004
 from t4_devkit.sanity.record.rec005 import REC005
 from t4_devkit.sanity.record.rec006 import REC006
+from t4_devkit.schema.name import SchemaName
 
 # Base sample dataset root (contains all mandatory annotation json files with non-empty records)
 SAMPLE_ROOT = Path(__file__).parent.parent.joinpath("sample", "t4dataset")
@@ -155,6 +156,18 @@ def test_rec006_pass_instance_not_empty() -> None:
     assert report.reasons is None
 
 
+def test_rec006_skip_annotation_empty(tmp_path: Path) -> None:
+    root = _make_mutated_dataset(tmp_path, {"sample_annotation.json": [], "object_ann.json": []})
+    checker = REC006()
+    report = checker(_context(root))
+    assert report.is_passed(strict=True)
+    assert (
+        report.reasons
+        and report.reasons[0]
+        == f"Both {SchemaName.SAMPLE_ANNOTATION} and {SchemaName.OBJECT_ANN} records are empty"
+    )
+
+
 def test_rec006_fail_instance_empty(tmp_path: Path) -> None:
     root = _make_mutated_dataset(tmp_path, {"instance.json": []})
     checker = REC006()