Support date-only version folders and increase scan limit in list_import_summaries

agents/common/recipes/gcp/gcs/list-import-summaries.md

@@ -45,7 +45,7 @@ the version URI only when the exact summary is needed.
 
 ## Required bounds
 
-Scan up to 100 matching summary object names plus one overflow sentinel (101
+Scan up to 1000 matching summary object names plus one overflow sentinel (1001
 names maximum). Return at most five timestamp-named versions and download at
 most those five summaries.
 

agents/common/scripts/list_import_summaries.py

@@ -33,9 +33,11 @@
 _VERSION_PATTERN = re.compile(
     r'^(?P<year>\d{4})_(?P<month>\d{2})_(?P<day>\d{2})T'
     r'\d{2}_\d{2}_\d{2}(?:_\d{1,6})?_\d{2}_\d{2}$')
+_DATE_VERSION_PATTERN = re.compile(
+    r'^(?P<year>\d{4})-(?P<month>\d{2})-(?P<day>\d{2})$')
 _SUMMARY_FILENAME = 'import_summary.json'
 _MAX_RESULT_LIMIT = 5
-_SCAN_LIMIT = 100
+_SCAN_LIMIT = 1000
 
 
 def _define_flags() -> None:
@@ -79,6 +81,8 @@ def normalize_import_name(absolute_import_name: str) -> dict[str, str]:
 
 def _version_date(version: str) -> str | None:
     match = _VERSION_PATTERN.fullmatch(version)
+    if not match:
+        match = _DATE_VERSION_PATTERN.fullmatch(version)
     if not match:
         return None
     try:
@@ -181,7 +185,7 @@ def list_import_summaries(absolute_import_name: str,
             continue
         candidates.append((version, version_date, blob))
 
-    candidates.sort(key=lambda item: item[0], reverse=True)
+    candidates.sort(key=lambda item: item[0].replace('_', '-'), reverse=True)
     for version, version_date, blob in candidates[:limit]:
         batch_job_id, issue = _read_batch_job_id(blob, version,
                                                  identity['simple_import_name'])

agents/common/scripts/list_import_summaries_test.py

@@ -78,15 +78,15 @@ def test_derives_exact_prefix_and_bounded_glob(self):
                                        client=client)
 
         self.assertEqual('scripts/a:Import', result['absolute_import_name'])
-        self.assertEqual(100, result['scan_limit'])
+        self.assertEqual(1000, result['scan_limit'])
         self.assertEqual(1, len(client.calls))
         bucket, kwargs = client.calls[0]
         self.assertEqual('bucket', bucket)
         self.assertEqual('scripts/a/Import/', kwargs['prefix'])
         self.assertEqual('scripts/a/Import/*/import_summary.json',
                          kwargs['match_glob'])
-        self.assertEqual(101, kwargs['max_results'])
-        self.assertEqual(101, kwargs['page_size'])
+        self.assertEqual(1001, kwargs['max_results'])
+        self.assertEqual(1001, kwargs['page_size'])
         self.assertEqual('items(name),nextPageToken', kwargs['fields'])
 
     def test_returns_newest_five_with_date_and_batch_job_id(self):
@@ -184,10 +184,45 @@ def test_reports_invalid_or_mismatched_selected_summaries(self):
             'summary_job_id_missing', 'summary_missing'
         ], [issue['code'] for issue in result['issues']])
 
+    def test_returns_date_only_versions(self):
+        versions = [f'2026-08-0{day}' for day in (3, 1, 7, 2, 6, 4, 5)]
+        blobs = [_blob(version) for version in versions]
+
+        result = list_import_summaries('scripts/a:Import',
+                                       'project',
+                                       'bucket',
+                                       client=_StorageClient(blobs))
+
+        self.assertEqual([
+            '2026-08-07',
+            '2026-08-06',
+            '2026-08-05',
+            '2026-08-04',
+            '2026-08-03',
+        ], [item['version'] for item in result['results']])
+        self.assertEqual('2026-08-07', result['results'][0]['date'])
+
+    def test_sorts_mixed_version_formats_by_date(self):
+        versions = [
+            '2026_08_03T01_02_03_123456_07_00',
+            '2026-08-04',
+        ]
+
+        result = list_import_summaries(
+            'scripts/a:Import',
+            'project',
+            'bucket',
+            client=_StorageClient([_blob(version) for version in versions]))
+
+        self.assertEqual([
+            '2026-08-04',
+            '2026_08_03T01_02_03_123456_07_00',
+        ], [item['version'] for item in result['results']])
+
     def test_returns_no_history_when_scan_limit_is_exceeded(self):
         blobs = [
             _blob(f'2026_07_{(index % 28) + 1:02d}T01_02_03_{index:06d}_07_00')
-            for index in range(101)
+            for index in range(1001)
         ]
 
         result = list_import_summaries('scripts/a:Import',
@@ -196,7 +231,7 @@ def test_returns_no_history_when_scan_limit_is_exceeded(self):
                                        client=_StorageClient(blobs))
 
         self.assertTrue(result['scan_truncated'])
-        self.assertEqual(101, result['scanned_summary_count'])
+        self.assertEqual(1001, result['scanned_summary_count'])
         self.assertEqual([], result['results'])
         self.assertEqual(0, sum(blob.download_count for blob in blobs))
         self.assertEqual('summary_scan_limit_exceeded',