From 1ec93f5234976b5d55a0d9dde6205d35b6fb9e21 Mon Sep 17 00:00:00 2001
From: brimoor <brimoor@umich.edu>
Date: Thu, 26 Dec 2024 20:18:03 -0600
Subject: [PATCH] add accesses to index stats

---
 fiftyone/core/collections.py     | 18 ++++++++++++++++--
 fiftyone/core/dataset.py         | 29 ++++++++++-------------------
 tests/unittests/dataset_tests.py |  6 ++++--
 3 files changed, 30 insertions(+), 23 deletions(-)

diff --git a/fiftyone/core/collections.py b/fiftyone/core/collections.py
index 2aafc32ee2..1cbc54d127 100644
--- a/fiftyone/core/collections.py
+++ b/fiftyone/core/collections.py
@@ -9506,8 +9506,8 @@ def get_index_information(self, include_stats=False):
         details on the structure of this dictionary.
 
         Args:
-            include_stats (False): whether to include the size and build status
-                of each index
+            include_stats (False): whether to include the size, usage, and
+                build status of each index
 
         Returns:
             a dict mapping index names to info dicts
@@ -9528,6 +9528,13 @@ def get_index_information(self, include_stats=False):
                 if key in sample_info:
                     sample_info[key]["in_progress"] = True
 
+            for d in self._dataset._sample_collection.aggregate(
+                [{"$indexStats": {}}]
+            ):
+                key = d["name"]
+                if key in sample_info:
+                    sample_info[key]["accesses"] = d["accesses"]
+
         for key, info in sample_info.items():
             if len(info["key"]) == 1:
                 field = info["key"][0][0]
@@ -9550,6 +9557,13 @@ def get_index_information(self, include_stats=False):
                     if key in frame_info:
                         frame_info[key]["in_progress"] = True
 
+                for d in self._dataset._frame_collection.aggregate(
+                    [{"$indexStats": {}}]
+                ):
+                    key = d["name"]
+                    if key in frame_info:
+                        frame_info[key]["accesses"] = d["accesses"]
+
             for key, info in frame_info.items():
                 if len(info["key"]) == 1:
                     field = info["key"][0][0]
diff --git a/fiftyone/core/dataset.py b/fiftyone/core/dataset.py
index 65e75cd910..e69d17ef3a 100644
--- a/fiftyone/core/dataset.py
+++ b/fiftyone/core/dataset.py
@@ -1202,21 +1202,13 @@ def stats(
         return stats
 
     def _sample_collstats(self):
-        conn = foo.get_db_conn()
-        return conn.command(
-            "collstats",
-            self._sample_collection_name,
-        )
+        return _get_collstats(self._sample_collection)
 
     def _frame_collstats(self):
         if self._frame_collection_name is None:
             return None
 
-        conn = foo.get_db_conn()
-        return conn.command(
-            "collstats",
-            self._frame_collection_name,
-        )
+        return _get_collstats(self._frame_collection)
 
     def first(self):
         """Returns the first sample in the dataset.
@@ -7770,15 +7762,6 @@ def _get_frame_collection(self, write_concern=None):
             self._frame_collection_name, write_concern=write_concern
         )
 
-    @property
-    def _frame_indexes(self):
-        frame_collection = self._frame_collection
-        if frame_collection is None:
-            return None
-
-        index_info = frame_collection.index_information()
-        return [k["key"][0][0] for k in index_info.values()]
-
     def _apply_sample_field_schema(self, schema):
         for field_name, field_or_str in schema.items():
             kwargs = foo.get_field_kwargs(field_or_str)
@@ -9104,6 +9087,14 @@ def _get_single_index_map(coll):
     }
 
 
+def _get_collstats(coll):
+    pipeline = [
+        {"$collStats": {"storageStats": {}}},
+        {"$replaceRoot": {"newRoot": "$storageStats"}},
+    ]
+    return next(coll.aggregate(pipeline))
+
+
 def _add_collection_with_new_ids(
     dataset,
     sample_collection,
diff --git a/tests/unittests/dataset_tests.py b/tests/unittests/dataset_tests.py
index 9ef306e471..af3164e695 100644
--- a/tests/unittests/dataset_tests.py
+++ b/tests/unittests/dataset_tests.py
@@ -671,7 +671,7 @@ def test_indexes(self):
             dataset.create_index("non_existent_field")
 
     @drop_datasets
-    def test_index_sizes(self):
+    def test_index_stats(self):
         gt = fo.Detections(detections=[fo.Detection(label="foo")])
         sample = fo.Sample(filepath="video.mp4", gt=gt)
         sample.frames[1] = fo.Frame(gt=gt)
@@ -700,7 +700,9 @@ def test_index_sizes(self):
         self.assertSetEqual(set(dataset.list_indexes()), indexes)
         self.assertSetEqual(set(info.keys()), indexes)
         for d in info.values():
-            self.assertTrue(d.get("size") is not None)
+            self.assertTrue(d["size"] is not None)
+            self.assertTrue("ops" in d["accesses"])
+            self.assertTrue("since" in d["accesses"])
 
     @drop_datasets
     def test_index_in_progress(self):