fix #221 (#225)

dascore/io/prodml/utils.py

@@ -5,7 +5,7 @@
 import dascore as dc
 from dascore.core.coordmanager import get_coord_manager
 from dascore.core.coords import get_coord
-from dascore.utils.misc import maybe_get_attrs, unbyte
+from dascore.utils.misc import maybe_get_attrs, unbyte, iterate
 
 # --- Getting format/version
 
@@ -119,9 +119,24 @@ def _get_prodml_attrs(fi, extras=None) -> list[dict]:
     return out
 
 
+def _get_dims(node):
+    """Get the dimension names in the form of a tuple."""
+    # we use distance rather than locus, setup mapping to relect this.
+    map_ = {"locus": "distance", "Locus": "distance", "Time": "time"}
+    dims = unbyte(getattr(node.RawData._v_attrs, "Dimensions", "time, distance"))
+    if isinstance(dims, str):
+        dims = dims.replace(",", " ")
+        dims = tuple(map_.get(x, x) for x in dims.split())
+    else:
+        unbytes = [unbyte(x) for x in iterate(dims)]
+        dims = tuple(map_.get(x, x) for x in unbytes)
+    return dims
+
+
 def _get_data_attr(attrs, node, time, distance):
     """Get a new attributes with adjusted time/distance and data array."""
-    cm = get_coord_manager(attrs["coords"], dims=("time", "distance"))
+    dims = _get_dims(node)
+    cm = get_coord_manager(attrs["coords"], dims=dims)
     new_cm, data = cm.select(array=node.RawData, time=time, distance=distance)
     return data, new_cm
 
@@ -133,8 +148,7 @@ def _read_prodml(fi, distance=None, time=None, attr_cls=dc.PatchAttrs):
     out = []
     for attrs, node in zip(attr_list, nodes):
         data, coords = _get_data_attr(attrs, node, time, distance)
-        dims = ("time", "distance")  # dims are fixed for this file format
         if data.size:
             pattrs = attr_cls(**attrs)
-            out.append(dc.Patch(data=data, attrs=pattrs, dims=dims, coords=coords))
+            out.append(dc.Patch(data=data, attrs=pattrs, coords=coords))
     return out

tests/test_io/test_prodml/test_prod_ml.py

@@ -1,7 +1,10 @@
 """Generic tests for prodml support."""
 from __future__ import annotations
 
+import shutil
+
 import pytest
+import tables
 
 import dascore as dc
 from dascore.io.core import read
@@ -34,6 +37,20 @@ class TestProdMLFile:
     the Silixa file is technical just ProdML v2.1.
     """
 
+    @pytest.fixture(scope="class")
+    def issue_221_patch_path(self, tmp_path_factory):
+        """Ensure dims are correctly ascertained."""
+        tmp_path = tmp_path_factory.mktemp("issue_221")
+        path = dc.utils.downloader.fetch("prodml_2.0.h5")
+        new_path = shutil.copy2(path, tmp_path / "prod_2_monkey_patched.h5")
+        with tables.open_file(new_path, "a") as fi:
+            # monkey patch dimensions to simulate issue.
+            new_dims = "time, locus"
+            parent_node = fi.root.Acquisition["Raw[0]"]
+            node = parent_node["RawData"]
+            node._v_attrs.Dimensions = new_dims
+        return new_path
+
     @pytest.fixture(scope="class")
     def silixa_h5_patch(self, idas_h5_example_path):
         """Get the silixa file, return Patch."""
@@ -49,6 +66,11 @@ def test_has_gauge_length(self, silixa_h5_patch):
         patch = silixa_h5_patch
         assert hasattr(patch.attrs, "gauge_length")
 
+    def test_issue_221(self, issue_221_patch_path):
+        """Ensure dims are correctly ascertained."""
+        patch = dc.read(issue_221_patch_path)[0]
+        assert isinstance(patch, dc.Patch)
+
 
 class TestReadQuantXV2:
     """Tests for reading the QuantXV2 format."""