[PY-619] metadata.json parser utilities

darwin/datatypes.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from dataclasses import dataclass, field
 from enum import Enum, auto
 from pathlib import Path
@@ -20,7 +22,7 @@
 except ImportError:
     NDArray = Any  # type:ignore
 
-from darwin.path_utils import construct_full_path
+from darwin.path_utils import construct_full_path, is_properties_enabled, parse_metadata
 
 # Utility types
 
@@ -386,6 +388,96 @@ def __str__(self) -> str:
         return f"{self.major}.{self.minor}{self.suffix}"
 
 
+@dataclass
+class Property:
+    """
+    Represents a property of an annotation file.
+    """
+
+    # Name of the property
+    name: str
+
+    # Type of the property
+    type: str
+
+    # Whether the property is required or not
+    required: bool
+
+    # Property options
+    options: list[dict[str, str]]
+
+
+@dataclass
+class PropertyClass:
+    name: str
+    type: str
+    description: Optional[str]
+    color: Optional[str] = None
+    sub_types: Optional[list[str]] = None
+    properties: Optional[list[Property]] = None
+
+
+def parse_property_classes(metadata: dict[str, Any]) -> list[PropertyClass]:
+    """
+    Parses the metadata file and returns a list of PropertyClass objects.
+
+    Parameters
+    ----------
+    metadata : dict[str, Any]
+        The metadata file.
+
+    Returns
+    -------
+    list[PropertyClass]
+        A list of PropertyClass objects.
+    """
+    assert "classes" in metadata, "Metadata does not contain classes"
+
+    classes = []
+    for metadata_cls in metadata["classes"]:
+        assert (
+            "properties" in metadata_cls
+        ), "Metadata class does not contain properties"
+        classes.append(
+            PropertyClass(
+                name=metadata_cls["name"],
+                type=metadata_cls["type"],
+                description=metadata_cls.get("description"),
+                color=metadata_cls.get("color"),
+                sub_types=metadata_cls.get("sub_types"),
+                properties=[Property(**p) for p in metadata_cls["properties"]],
+            )
+        )
+
+    return classes
+
+
+def split_paths_by_metadata(
+    path, dir: str = ".v7", filename: str = "metadata.json"
+) -> tuple[Path, Optional[list[PropertyClass]]]:
+    """
+    Splits the given path into two: the path to the metadata file and the path to the properties
+
+    Parameters
+    ----------
+    path : Path
+        The path to the export directory.
+
+    Returns
+    -------
+    tuple[Path, Optional[list[PropertyClass]]]
+        A tuple containing the path to the metadata file and the list of property classes.
+    """
+    if not is_properties_enabled(path, dir, filename):
+        return path, None
+
+    metadata_path = path / dir / filename
+    metadata = parse_metadata(metadata_path)
+    property_classes = parse_property_classes(metadata)
+
+    return metadata_path, property_classes
+
+
 @dataclass
 class AnnotationFile:
     """

darwin/path_utils.py

@@ -1,4 +1,7 @@
-from pathlib import PurePosixPath
+from __future__ import annotations
+
+import json
+from pathlib import Path, PurePosixPath
 from typing import Optional, Tuple
 
 
@@ -41,3 +44,56 @@ def deconstruct_full_path(filename: str) -> Tuple[str, str]:
     """
     posix_path = PurePosixPath("/") / filename
     return str(posix_path.parent), posix_path.name
+
+
+def parse_metadata(path: Path) -> dict:
+    """
+    Returns the parsed metadata file.
+
+    Parameters
+    ----------
+    path : Path
+        The path to the metadata file.
+
+    Returns
+    -------
+    dict
+        The parsed metadata file.
+    """
+    with open(path) as f:
+        metadata = json.load(f)
+
+    return metadata
+
+
+def is_properties_enabled(
+    export_dir_path: Path,
+    dir: str = ".v7",
+    filename: str = "metadata.json",
+    annotations_dir: str = "annotations",
+) -> bool:
+    """
+    Returns whether the given export directory has properties enabled.
+
+    Parameters
+    ----------
+    export_dir_path : Path
+        The path to the export directory.
+
+    Returns
+    -------
+    bool
+        Whether the given export directory has properties enabled.
+    """
+    path = export_dir_path / dir
+    if not path.exists():
+        annotations_path = export_dir_path / annotations_dir
+        for annotation_path in annotations_path.rglob("*"):
+            with open(annotation_path) as f:
+                if '"properties"' in f.read():
+                    return True
+        return False
+
+    metadata_path = path / filename
+    metadata_classes = parse_metadata(metadata_path).get("classes", [])
+    return any(_cls.get("properties") for _cls in metadata_classes)