Draft: Sample ID ELn base class draft

This is a draft for having a Sample ID base class. Discussions about its content have been carried out within the frame of Area A. At the moment, the Reference in children and parents' quantities does not seem to work. Additionally, results.eln.lab_ids does not seem to be populated, or at least I do not seem to be able to see it in search entries in the gui.

This is a good opportunity also to start discussing a strategy to start writing these base classes. I attach a yaml schema to test it.

sample_id_test.schema.archive.yaml

nomad/datamodel/metainfo/eln/__init__.py

@@ -18,8 +18,9 @@
 
 from nomad import utils
 from nomad.datamodel.data import EntryData, ArchiveSection
+from nomad.metainfo.metainfo import SectionProxy
 from nomad.datamodel.results import ELN, Results, Material
-from nomad.metainfo import Package, Quantity, Datetime
+from nomad.metainfo import Package, Quantity, Datetime, Reference
 
 m_package = Package(name='material_library')
 
@@ -186,4 +187,87 @@ class Measurement(ElnActivityBaseSecton):
     pass
 
 
+class SampleID(ArchiveSection):
+    ''' A ELN base section that can be used for sample IDs.
+    If the `sample_owner`, `sample_short_name`, `ìnstitute`, and `creation_datetime`
+    quantities are provided, the sample_id will be automatically created as a combination
+    of these four quantities.'''
+
+    # m_def = Section(
+    #     a_eln=dict(hide=['name', 'lab_id']))
+
+    institute = Quantity(
+        type=str,
+        description='Alias/short name of the home institute of the owner, i.e. *HZB*.',
+        a_eln=dict(component='StringEditQuantity'))
+
+    sample_owner = Quantity(
+        type=str,
+        shape=[],
+        description='Name or alias of the process operator, e.g. jmp',
+        a_eln=dict(component='StringEditQuantity'))
+
+    creation_datetime = Quantity(
+        type=Datetime,
+        description='Creation date of the sample.',
+        a_eln=dict(component='DateTimeEditQuantity'))
+
+    sample_short_name = Quantity(
+        type=str,
+        description='''A short name of the sample (the identifier scribed on the smaple,
+         or in the sample container), e.g. 4001-8, YAG-2-34.
+         This is to be managed and decided internally by the labs,
+         although we recomend to avoid the following characters on it: "_", "/", "\" and "."''',
+        a_eln=dict(component='StringEditQuantity'))
+
+    sample_id = Quantity(
+        type=str,
+        description='''Full sample id. Ideally a human readable sample id convention,
+        which is simple, understandable and still having chances of becoming unique.
+        If the `sample_owner`, `sample_short_name`, `ìnstitute`, and `creation_datetime`
+        are provided, this will be formed automatically by joining these components by an underscore (_).
+        Spaces in any of the individual components will be replaced with hyphens (-).
+        An example would be hzb_oah_20200602_4001-08''',
+        a_eln=dict(component='StringEditQuantity'))
+
+    children = Quantity(
+        type=Reference(SectionProxy('SampleID')),
+        shape=["*"],
+        descriptions='A reference to a sample which are children of this one.',
+        a_eln=dict(component='ReferenceEditQuantity'))
+
+    parents = Quantity(
+        type=Reference(SectionProxy('SampleID')),
+        descriptions='A reference to sample which are parents of this one.',
+        a_eln=dict(component='ReferenceEditQuantity'))
+
+    def normalize(self, archive, logger):
+        super().normalize(archive, logger)
+
+        if self.institute and self.sample_short_name and self.sample_owner and self.creation_datetime:
+            creation_date = self.creation_datetime.strftime('%Y%m%d')
+            sample_owner = self.sample_owner.replace(' ', '-')
+            sample_id_list = [self.institute, sample_owner, creation_date, self.sample_short_name]
+            self.sample_id = '_'.join(sample_id_list)
+
+        if isinstance(self, EntryData):
+            if archive.data == self and self.sample_id:
+                archive.metadata.entry_name = self.sample_id.replace('_', ' ')
+            EntryData.normalize(self, archive, logger)
+
+        if not archive.results:
+            archive.results = Results(eln=ELN())
+        if not archive.results.eln:
+            archive.results.eln = ELN()
+
+        if self.sample_id:
+            if archive.results.eln.lab_ids is None:
+                archive.results.eln.lab_ids = []
+            archive.results.eln.lab_ids.append(self.sample_id)
+
+        if not archive.results.eln.sections:
+            archive.results.eln.sections = []
+        archive.results.eln.sections.append(self.m_def.name)
+
+
 m_package.__init_metainfo__()

[Markus Scheidgen]

You could log a warning if no sample_id could be procuded.