Normalize CIF datablock names during import

src/easydiffraction/io/cif/parse.py

@@ -4,11 +4,18 @@
 
 from __future__ import annotations
 
+import re
+
 import gemmi
 
 # Minimum raw-string length for CIF surrounding-quote detection
 _MIN_QUOTED_LEN = 2
 
+# Model datablock names use a backend-safe subset of CIF block-name
+# characters. Hyphens and underscores are retained because they are used
+# throughout EasyDiffraction's public naming conventions.
+_UNSUPPORTED_DATABLOCK_NAME_CHARS = re.compile(r'[^a-z0-9_-]')
+
 
 def document_from_path(path: str) -> gemmi.cif.Document:
     """Read a CIF document from a file path."""
@@ -26,9 +33,9 @@ def pick_sole_block(doc: gemmi.cif.Document) -> gemmi.cif.Block:
 
 
 def name_from_block(block: gemmi.cif.Block) -> str:
-    """Extract a model name from the CIF block name."""
-    # TODO: Need validator or normalization?
-    return block.name
+    """Extract and normalize a model name from the CIF block name."""
+    lowercase_name = block.name.lower()
+    return _UNSUPPORTED_DATABLOCK_NAME_CHARS.sub('', lowercase_name)
 
 
 def read_cif_str(block: gemmi.cif.Block, tag: str) -> str | None:

tests/unit/easydiffraction/datablocks/experiment/item/test_factory.py

@@ -29,6 +29,15 @@ def test_experiment_factory_from_scratch():
     assert ex.experiment_type.sample_form.value == SampleFormEnum.POWDER.value
 
 
+def test_from_cif_str_normalizes_datablock_name():
+    from easydiffraction.datablocks.experiment.item.factory import ExperimentFactory
+
+    experiment = ExperimentFactory.from_cif_str('data_83267-ICSD\n')
+
+    assert experiment.name == '83267-icsd'
+    assert experiment.as_cif.startswith('data_83267-icsd\n')
+
+
 def test_from_cif_str_restores_non_default_peak_profile_type():
     """
     Loading a CIF with a non-default peak profile type must reconstruct

tests/unit/easydiffraction/datablocks/structure/item/test_factory.py

@@ -9,6 +9,13 @@ def test_from_scratch():
     assert m.name == 'abc'
 
 
+def test_from_cif_str_normalizes_datablock_name():
+    structure = StructureFactory.from_cif_str('data_83267-ICSD\n')
+
+    assert structure.name == '83267-icsd'
+    assert structure.as_cif.startswith('data_83267-icsd\n')
+
+
 def test_from_cif_str_accepts_underscore_style_structure_tags():
     cif = """\
 data_legacy

tests/unit/easydiffraction/io/cif/test_parse.py

@@ -31,6 +31,17 @@ def test_name_from_block(self):
         name = name_from_block(block)
         assert name == 'silicon'
 
+    def test_name_from_block_normalizes_unsupported_name(self):
+        from easydiffraction.io.cif.parse import document_from_string
+        from easydiffraction.io.cif.parse import name_from_block
+        from easydiffraction.io.cif.parse import pick_sole_block
+
+        cif = 'data_My+83267-ICSD.example\n_cell.length_a 5.43\n'
+        doc = document_from_string(cif)
+        block = pick_sole_block(doc)
+        name = name_from_block(block)
+        assert name == 'my83267-icsdexample'
+
 
 class TestDocumentFromPath:
     def test_valid_file(self, tmp_path):