From 62695d6145943488099d076d9f7153d39c20ae11 Mon Sep 17 00:00:00 2001
From: Yaroslav Halchenko <debian@onerussian.com>
Date: Fri, 18 Oct 2024 14:05:14 -0400
Subject: [PATCH] feat: extend and rename validation result properties

BREAKING CHANGE: we renamed .bids_version to more generic .standard + .standard_version
---
 dandi/files/bids.py     |  2 +-
 dandi/files/zarr.py     | 32 +++++++++++++++++++-------------
 dandi/validate.py       |  3 ++-
 dandi/validate_types.py | 28 ++++++++++++++++++++++------
 4 files changed, 44 insertions(+), 21 deletions(-)

diff --git a/dandi/files/bids.py b/dandi/files/bids.py
index 8541b5030..907a14c59 100644
--- a/dandi/files/bids.py
+++ b/dandi/files/bids.py
@@ -111,7 +111,7 @@ def _validate(self) -> None:
                         self._asset_metadata[bids_path] = prepare_metadata(
                             result.metadata
                         )
-                        self._bids_version = result.origin.bids_version
+                        self._bids_version = result.origin.standard_version
 
     def get_asset_errors(self, asset: BIDSAsset) -> list[ValidationResult]:
         """:meta private:"""
diff --git a/dandi/files/zarr.py b/dandi/files/zarr.py
index 1c5140c08..8dee86cd8 100644
--- a/dandi/files/zarr.py
+++ b/dandi/files/zarr.py
@@ -16,7 +16,7 @@
 import requests
 from zarr_checksum.tree import ZarrChecksumTree
 
-from dandi import get_logger
+from dandi import __version__, get_logger
 from dandi.consts import (
     MAX_ZARR_DEPTH,
     ZARR_DELETE_BATCH_SIZE,
@@ -209,6 +209,12 @@ def get_validation_errors(
         import zarr
 
         errors: list[ValidationResult] = []
+        origin: ValidationOrigin = ValidationOrigin(
+            name="zarr",
+            version=zarr.__version__,
+            standard="zarr",
+        )
+
         try:
             data = zarr.open(str(self.filepath))
         except Exception:
@@ -216,10 +222,7 @@ def get_validation_errors(
                 raise
             errors.append(
                 ValidationResult(
-                    origin=ValidationOrigin(
-                        name="zarr",
-                        version=zarr.version.version,
-                    ),
+                    origin=origin,
                     severity=Severity.ERROR,
                     id="zarr.cannot_open",
                     scope=Scope.FILE,
@@ -228,13 +231,19 @@ def get_validation_errors(
                 )
             )
             data = None
+
+        origin = ValidationOrigin(
+            name="dandi.zarr",
+            version=__version__,
+            standard="zarr",
+        )
+        # if data:
+        # TODO: figure out how to assign standard_version
+        # origin.standard_version = data.???
         if isinstance(data, zarr.Group) and not data:
             errors.append(
                 ValidationResult(
-                    origin=ValidationOrigin(
-                        name="zarr",
-                        version=zarr.version.version,
-                    ),
+                    origin=origin,
                     severity=Severity.ERROR,
                     id="zarr.empty_group",
                     scope=Scope.FILE,
@@ -248,10 +257,7 @@ def get_validation_errors(
                 raise ValueError(msg)
             errors.append(
                 ValidationResult(
-                    origin=ValidationOrigin(
-                        name="zarr",
-                        version=zarr.version.version,
-                    ),
+                    origin=origin,
                     severity=Severity.ERROR,
                     id="zarr.tree_depth_exceeded",
                     scope=Scope.FILE,
diff --git a/dandi/validate.py b/dandi/validate.py
index 3746434f8..d6ab74c00 100644
--- a/dandi/validate.py
+++ b/dandi/validate.py
@@ -49,7 +49,8 @@ def validate_bids(
     origin = ValidationOrigin(
         name="bidsschematools",
         version=bidsschematools.__version__,
-        bids_version=validation_result["bids_version"],
+        standard="bids",
+        standard_version=validation_result["bids_version"],
     )
 
     # Storing variable to not re-compute set paths for each individual file.
diff --git a/dandi/validate_types.py b/dandi/validate_types.py
index c09740d39..a379a9059 100644
--- a/dandi/validate_types.py
+++ b/dandi/validate_types.py
@@ -1,35 +1,51 @@
 from __future__ import annotations
 
 from dataclasses import dataclass
-from enum import Enum
+from enum import Enum, IntEnum
 from pathlib import Path
+from typing import Any
 
 
 @dataclass
 class ValidationOrigin:
     name: str
     version: str
-    bids_version: str | None = None
+    standard: str | None = None  # TODO: Enum for the standards??
+    standard_version: str | None = None
 
 
-class Severity(Enum):
+# TODO: decide on the naming consistency -- either prepend all with Validation or not
+class Severity(IntEnum):
     HINT = 1
-    WARNING = 2
-    ERROR = 3
+    INFO = 2  # new/unused, available in linkml
+    WARNING = 3
+    ERROR = 4
+    CRITICAL = 5  # new/unused, linkml has FATAL
 
 
 class Scope(Enum):
     FILE = "file"
     FOLDER = "folder"
+    # Isaac: make it/add "dandiset-metadata" to signal specific relation to metadata
     DANDISET = "dandiset"
     DATASET = "dataset"
 
 
+# new/unused, may be should be gone
+class ValidationObject(Enum):
+    METADATA = "metadata"
+    DATA = "data"  # e.g. actual data contained in files, not metadata (e.g. as in
+    # nwb or nifti header)
+    FILE = "file"  # e.g. file itself, e.g. truncated file or file not matching checksum
+
+
 @dataclass
 class ValidationResult:
     id: str
-    origin: ValidationOrigin
+    origin: ValidationOrigin  # metadata about underlying validator and standard
     scope: Scope
+    origin_result: Any | None = None  # original validation result from "origin"
+    object: ValidationObject | None = None
     severity: Severity | None = None
     # asset_paths, if not populated, assumes [.path], but could be smth like
     # {"path": "task-broken_bold.json",