Add new Provenance combination_* attributes with combine_echodata, mi…

…rroring conversion_* attributes (#1113)

echopype/echodata/combine.py

@@ -1,4 +1,3 @@
-import datetime
 import itertools
 import re
 from pathlib import Path
@@ -13,6 +12,7 @@
 
 from ..utils.io import validate_output_path
 from ..utils.log import _init_logger
+from ..utils.prov import echopype_prov_attrs
 from .echodata import EchoData
 
 logger = _init_logger(__name__)
@@ -742,11 +742,13 @@ def _combine(
                 combined_ds.attrs.update(
                     {
                         "is_combined": True,
-                        "conversion_time": datetime.datetime.utcnow().strftime(
-                            "%Y-%m-%dT%H:%M:%SZ"
-                        ),
+                        "conversion_software_name": group_attrs["conversion_software_name"],
+                        "conversion_software_version": group_attrs["conversion_software_version"],
+                        "conversion_time": group_attrs["conversion_time"],
                     }
                 )
+                prov_dict = echopype_prov_attrs(process_type="combination")
+                combined_ds = combined_ds.assign_attrs(prov_dict)
 
             # Data holding
             tree_dict[ed_group] = combined_ds

echopype/echodata/echodata.py

@@ -3,7 +3,7 @@
 import warnings
 from html import escape
 from pathlib import Path
-from typing import TYPE_CHECKING, Any, Dict, Optional, Set, Tuple
+from typing import TYPE_CHECKING, Any, Dict, Optional, Set, Tuple, Union
 
 import fsspec
 import numpy as np
@@ -196,16 +196,28 @@ def _load_tree(self) -> None:
                 setattr(self, group, node)
 
     @property
-    def version_info(self) -> Tuple[int]:
-        if self["Provenance"].attrs.get("conversion_software_name", None) == "echopype":
-            version_str = self["Provenance"].attrs.get("conversion_software_version", None)
+    def version_info(self) -> Union[Tuple[int], None]:
+        def _get_version_tuple(provenance_type):
+            """
+            Parameters
+            ----------
+            provenance_type : str
+                Either conversion or combination
+            """
+            version_str = self["Provenance"].attrs.get(f"{provenance_type}_software_version", None)
             if version_str is not None:
                 if version_str.startswith("v"):
                     # Removes v in case of v0.4.x or less
                     version_str = version_str.strip("v")
                 version_num = version_str.split(".")[:3]
                 return tuple([int(i) for i in version_num])
-        return None
+
+        if self["Provenance"].attrs.get("combination_software_name", None) == "echopype":
+            return _get_version_tuple("combination")
+        elif self["Provenance"].attrs.get("conversion_software_name", None) == "echopype":
+            return _get_version_tuple("conversion")
+        else:
+            return None
 
     @property
     def nbytes(self) -> float:

echopype/tests/echodata/test_echodata_combine.py

@@ -1,3 +1,4 @@
+from datetime import datetime
 from textwrap import dedent
 from pathlib import Path
 import tempfile
@@ -160,6 +161,18 @@ def test_combine_echodata(raw_datasets):
 
     combined = echopype.combine_echodata(eds)
 
+    # Test Provenance conversion and combination attributes
+    for attr_token in ["software_name", "software_version", "time"]:
+        assert f"conversion_{attr_token}" in combined['Provenance'].attrs
+        assert f"combination_{attr_token}" in combined['Provenance'].attrs
+
+    def attr_time_to_dt(time_str):
+        return datetime.strptime(time_str, '%Y-%m-%dT%H:%M:%SZ')
+    assert (
+            attr_time_to_dt(combined['Provenance'].attrs['conversion_time']) <=
+            attr_time_to_dt(combined['Provenance'].attrs['combination_time'])
+    )
+
     # get all possible dimensions that should be dropped
     # these correspond to the attribute arrays created
     all_drop_dims = []
@@ -174,7 +187,6 @@ def test_combine_echodata(raw_datasets):
     all_drop_dims.append("echodata_filename")
 
     for group_name in combined.group_paths:
-
         # get all Datasets to be combined
         combined_group: xr.Dataset = combined[group_name]
         eds_groups = [

echopype/utils/prov.py

@@ -12,7 +12,7 @@
 
 from .log import _init_logger
 
-ProcessType = Literal["conversion", "processing", "mask"]
+ProcessType = Literal["conversion", "combination", "processing", "mask"]
 # Note that this PathHint is defined differently from the one in ..core
 PathHint = Union[str, Path]
 PathSequenceHint = Union[List[PathHint], Tuple[PathHint], NDArray[PathHint]]