nmdc-runtime 2.8.0__py3-none-any.whl → 2.10.0__py3-none-any.whl

nmdc_runtime/site/repository.py

@@ -1,6 +1,5 @@
 import json
 
-from typing import Any
 
 from dagster import (
     repository,
@@ -14,7 +13,6 @@ from dagster import (
     DagsterRunStatus,
     RunStatusSensorContext,
     DefaultSensorStatus,
-    in_process_executor,
 )
 from starlette import status
 from toolz import merge, get_in
@@ -30,8 +28,6 @@ from nmdc_runtime.site.graphs import (
     translate_metadata_submission_to_nmdc_schema_database,
     ingest_metadata_submission,
     gold_study_to_database,
-    gold_translation,
-    gold_translation_curation,
     create_objects_from_site_object_puts,
     housekeeping,
     ensure_jobs,
@@ -62,9 +58,6 @@ from nmdc_runtime.site.resources import (
 from nmdc_runtime.site.resources import (
     get_runtime_api_site_client,
 )
-from nmdc_runtime.site.translation.emsl import emsl_job, test_emsl_job
-from nmdc_runtime.site.translation.gold import gold_job, test_gold_job
-from nmdc_runtime.site.translation.jgi import jgi_job, test_jgi_job
 from nmdc_runtime.util import freeze
 from nmdc_runtime.util import unfreeze
 
@@ -249,82 +242,6 @@ def process_workflow_job_triggers(_context):
         yield SkipReason("No new jobs required")
 
 
-@asset_sensor(
-    asset_key=AssetKey(["object", "nmdc_database.json.zip"]),
-    job=ensure_jobs.to_job(name="ensure_gold_translation", **preset_normal),
-)
-def ensure_gold_translation_job(_context, asset_event):
-    mdb = get_mongo(run_config_frozen__normal_env).db
-    gold_etl_latest = mdb.objects.find_one(
-        {"name": "nmdc_database.json.zip"}, sort=[("created_time", -1)]
-    )
-    sensed_object_id = asset_materialization_metadata(asset_event, "object_id").text
-    if gold_etl_latest is None:
-        yield SkipReason("can't find sensed asset object_id in database")
-        return
-    elif gold_etl_latest["id"] != sensed_object_id:
-        yield SkipReason("later object than sensed materialization")
-        return
-
-    run_config = merge(
-        run_config_frozen__normal_env,
-        {
-            "solids": {
-                "construct_jobs": {
-                    "config": {
-                        "base_jobs": [
-                            {
-                                "workflow": {"id": "gold-translation-1.0.0"},
-                                "config": {"object_id": gold_etl_latest["id"]},
-                            }
-                        ]
-                    }
-                }
-            }
-        },
-    )
-    yield RunRequest(run_key=sensed_object_id, run_config=unfreeze(run_config))
-
-
-@asset_sensor(
-    asset_key=AssetKey(["job", "gold-translation-1.0.0"]),
-    job=gold_translation_curation.to_job(**preset_normal),
-)
-def claim_and_run_gold_translation_curation(_context, asset_event):
-    client = get_runtime_api_site_client(run_config_frozen__normal_env)
-    mdb = get_mongo(run_config_frozen__normal_env).db
-    object_id_latest = asset_materialization_metadata(
-        asset_event, "object_id_latest"
-    ).text
-    job = mdb.jobs.find_one(
-        {
-            "workflow.id": "gold-translation-1.0.0",
-            "config.object_id_latest": object_id_latest,
-        }
-    )
-    if job is not None:
-        rv = client.claim_job(job["id"])
-        if rv.status_code == status.HTTP_200_OK:
-            operation = rv.json()
-            run_config = merge(
-                run_config_frozen__normal_env,
-                {
-                    "ops": {
-                        "get_operation": {
-                            "config": {
-                                "operation_id": operation["id"],
-                            }
-                        }
-                    }
-                },
-            )
-            yield RunRequest(run_key=operation["id"], run_config=unfreeze(run_config))
-        else:
-            yield SkipReason("Job found, but already claimed by this site")
-    else:
-        yield SkipReason("No job found")
-
-
 @sensor(
     job=apply_metadata_in.to_job(name="apply_metadata_in_sensed", **preset_normal),
     default_status=DefaultSensorStatus.RUNNING,
@@ -463,11 +380,6 @@ def claim_and_run_apply_changesheet_jobs(_context):
         yield SkipReason("; ".join(skip_notes))
 
 
-# TODO ensure data_object_type values from file_type_enum
-#    see /metadata-translation/notebooks/202106_curation_updates.ipynb
-#    for details ("Create file_type_enum collection" section).
-
-
 @sensor(job=create_objects_from_site_object_puts.to_job(**preset_normal))
 def done_object_put_ops(_context):
     client = get_runtime_api_site_client(run_config_frozen__normal_env)
@@ -507,7 +419,6 @@ def on_run_fail(context: RunStatusSensorContext):
 @repository
 def repo():
     graph_jobs = [
-        gold_translation.to_job(**preset_normal),
         hello_graph.to_job(name="hello_job"),
         ensure_jobs.to_job(**preset_normal),
         apply_metadata_in.to_job(**preset_normal),
@@ -523,8 +434,6 @@ def repo():
     ]
     sensors = [
         done_object_put_ops,
-        ensure_gold_translation_job,
-        claim_and_run_gold_translation_curation,
         process_workflow_job_triggers,
         claim_and_run_apply_changesheet_jobs,
         claim_and_run_metadata_in_jobs,
@@ -534,20 +443,6 @@ def repo():
     return graph_jobs + schedules + sensors
 
 
-@repository
-def translation():
-    graph_jobs = [jgi_job, gold_job, emsl_job]
-
-    return graph_jobs
-
-
-@repository
-def test_translation():
-    graph_jobs = [test_jgi_job, test_gold_job, test_emsl_job]
-
-    return graph_jobs
-
-
 @repository
 def biosample_submission_ingest():
     normal_resources = run_config_frozen__normal_env["resources"]
@@ -574,6 +469,7 @@ def biosample_submission_ingest():
                             "study_type": "research_study",
                             "gold_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/berkeley-schema-fy24/main/assets/misc/gold_seqMethod_to_nmdc_instrument_set.tsv",
                             "include_field_site_info": False,
+                            "enable_biosample_filtering": True,
                         },
                     },
                     "export_json_to_drs": {"config": {"username": ""}},
@@ -1018,6 +914,8 @@ def database_records_stitching():
                         "config": {
                             "nmdc_study_id": "",
                             "gold_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/refs/heads/main/assets/misc/gold_seqMethod_to_nmdc_instrument_set.tsv",
+                            "include_field_site_info": False,
+                            "enable_biosample_filtering": True,
                         }
                     },
                     "export_json_to_drs": {"config": {"username": ""}},
@@ -1060,6 +958,8 @@ def database_records_stitching():
                         "config": {
                             "nmdc_study_id": "",
                             "gold_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/refs/heads/main/assets/misc/gold_seqMethod_to_nmdc_instrument_set.tsv",
+                            "include_field_site_info": False,
+                            "enable_biosample_filtering": True,
                         }
                     },
                     "export_json_to_drs": {"config": {"username": ""}},
@@ -1102,21 +1002,11 @@ def database_records_stitching():
                         "config": {
                             "nmdc_study_id": "",
                             "gold_nmdc_instrument_mapping_file_url": "https://raw.githubusercontent.com/microbiomedata/nmdc-schema/refs/heads/main/assets/misc/gold_seqMethod_to_nmdc_instrument_set.tsv",
+                            "include_field_site_info": False,
+                            "enable_biosample_filtering": True,
                         }
                     },
                 },
             },
         ),
     ]
-
-
-# @repository
-# def validation():
-#     graph_jobs = [validate_jgi_job, validate_gold_job, validate_emsl_job]
-#     return graph_jobs
-#
-#
-# @repository
-# def test_validation():
-#     graph_jobs = [test_validate_jgi_job, test_validate_gold_job, test_validate_emsl_job]
-#     return graph_jobs

nmdc_runtime/site/resources.py

@@ -109,7 +109,7 @@ class RuntimeApiUserClient(RuntimeApiClient):
             },
         )
         response.raise_for_status()
-        return response.json()["cursor"]["firstBatch"]
+        return response.json()["cursor"]["batch"]
 
     def get_omics_processing_records_by_gold_project_id(self, gold_project_id: str):
         gold_project_id = normalize_gold_id(gold_project_id)
@@ -126,7 +126,7 @@ class RuntimeApiUserClient(RuntimeApiClient):
             },
         )
         response.raise_for_status()
-        return response.json()["cursor"]["firstBatch"]
+        return response.json()["cursor"]["batch"]
 
     def get_biosamples_for_study(self, study_id: str):
         # TODO: 10000 is an arbitrarily large number that has been chosen for the max_page_size param.
@@ -170,7 +170,7 @@ class RuntimeApiUserClient(RuntimeApiClient):
             },
         )
         response.raise_for_status()
-        return response.json()["cursor"]["firstBatch"]
+        return response.json()["cursor"]["batch"]
 
     def get_study(self, study_id: str):
         response = self.request(
@@ -182,7 +182,7 @@ class RuntimeApiUserClient(RuntimeApiClient):
             },
         )
         response.raise_for_status()
-        return response.json()["cursor"]["firstBatch"]
+        return response.json()["cursor"]["batch"]
 
 
 class RuntimeApiSiteClient(RuntimeApiClient):

nmdc_runtime/site/translation/gold_translator.py

@@ -1,5 +1,4 @@
 import collections
-import csv
 import re
 from typing import List, Tuple, Union
 from nmdc_schema import nmdc
@@ -45,6 +44,7 @@ class GoldStudyTranslator(Translator):
         analysis_projects: List[JSON_OBJECT] = [],
         gold_nmdc_instrument_map_df: pd.DataFrame = pd.DataFrame(),
         include_field_site_info: bool = False,
+        enable_biosample_filtering: bool = True,
         *args,
         **kwargs,
     ) -> None:
@@ -53,15 +53,20 @@ class GoldStudyTranslator(Translator):
         self.study = study
         self.study_type = nmdc.StudyCategoryEnum(study_type)
         self.include_field_site_info = include_field_site_info
+        self.enable_biosample_filtering = enable_biosample_filtering
         # Filter biosamples to only those with `sequencingStrategy` of
-        # "Metagenome" or "Metatranscriptome"
-        self.biosamples = [
-            biosample
-            for biosample in biosamples
-            if any(
-                _is_valid_project(project) for project in biosample.get("projects", [])
-            )
-        ]
+        # "Metagenome" or "Metatranscriptome" if filtering is enabled
+        if enable_biosample_filtering:
+            self.biosamples = [
+                biosample
+                for biosample in biosamples
+                if any(
+                    _is_valid_project(project)
+                    for project in biosample.get("projects", [])
+                )
+            ]
+        else:
+            self.biosamples = biosamples
         # Fetch the valid projectGoldIds that are associated with filtered
         # biosamples on their `projects` field
         valid_project_ids = {
@@ -116,6 +121,9 @@ class GoldStudyTranslator(Translator):
         :param gold_entity: GOLD entity object
         :return: PersonValue corresponding to the first PI in the `contacts` field
         """
+        if "contacts" not in gold_entity:
+            return None
+
         pi_dict = next(
             (
                 contact
@@ -169,7 +177,7 @@ class GoldStudyTranslator(Translator):
                 project["ncbiBioSampleAccession"], default_prefix="biosample"
             )
             for project in biosample_projects
-            if project["ncbiBioSampleAccession"]
+            if project.get("ncbiBioSampleAccession")
         ]
 
     def _get_samp_taxon_id(
@@ -333,12 +341,7 @@ class GoldStudyTranslator(Translator):
         if field_value is None:
             return None
 
-        return nmdc.QuantityValue(
-            has_raw_value=field_value,
-            has_numeric_value=nmdc.Double(field_value),
-            has_unit=unit,
-            type="nmdc:QuantityValue",
-        )
+        return self._parse_quantity_value(str(field_value), unit)
 
     def _get_text_value(
         self, gold_entity: JSON_OBJECT, gold_field: str
@@ -564,13 +567,11 @@ class GoldStudyTranslator(Translator):
         gold_biosample_id = gold_biosample["biosampleGoldId"]
         return nmdc.Biosample(
             add_date=gold_biosample.get("addDate"),
-            alt=self._get_quantity_value(
-                gold_biosample, "altitudeInMeters", unit="meters"
-            ),
+            alt=self._get_quantity_value(gold_biosample, "altitudeInMeters", unit="m"),
             collected_from=nmdc_field_site_id,
             collection_date=self._get_collection_date(gold_biosample),
             depth=self._get_quantity_value(
-                gold_biosample, ("depthInMeters", "depthInMeters2"), unit="meters"
+                gold_biosample, ("depthInMeters", "depthInMeters2"), unit="m"
             ),
             description=gold_biosample.get("description"),
             diss_oxygen=self._get_quantity_value(gold_biosample, "oxygenConcentration"),
@@ -609,7 +610,7 @@ class GoldStudyTranslator(Translator):
             ),
             specific_ecosystem=gold_biosample.get("specificEcosystem"),
             subsurface_depth=self._get_quantity_value(
-                gold_biosample, "subsurfaceDepthInMeters", unit="meters"
+                gold_biosample, "subsurfaceDepthInMeters", unit="m"
             ),
             temp=self._get_quantity_value(
                 gold_biosample, "sampleCollectionTemperature"

nmdc_runtime/site/translation/neon_benthic_translator.py

@@ -11,7 +11,6 @@ from nmdc_runtime.site.util import get_basename
 from nmdc_runtime.site.translation.neon_utils import (
     _get_value_or_none,
     _create_controlled_identified_term_value,
-    _create_controlled_term_value,
     _create_geolocation_value,
     _create_quantity_value,
     _create_timestamp_value,

nmdc_runtime/site/translation/neon_soil_translator.py

@@ -10,7 +10,6 @@ from nmdc_runtime.site.util import get_basename
 from nmdc_runtime.site.translation.neon_utils import (
     _get_value_or_none,
     _create_controlled_identified_term_value,
-    _create_controlled_term_value,
     _create_geolocation_value,
     _create_quantity_value,
     _create_timestamp_value,
@@ -153,7 +152,7 @@ class NeonSoilDataTranslator(Translator):
             collection_date=_create_timestamp_value(
                 biosample_row["collectDate"].values[0]
             ),
-            temp=_create_quantity_value(biosample_row["soilTemp"].values[0], "Celsius"),
+            temp=_create_quantity_value(biosample_row["soilTemp"].values[0], "Cel"),
             depth=nmdc.QuantityValue(
                 has_minimum_numeric_value=_get_value_or_none(
                     biosample_row, "sampleTopDepth"
@@ -169,13 +168,13 @@ class NeonSoilDataTranslator(Translator):
             analysis_type=_get_value_or_none(biosample_row, "sequenceAnalysisType"),
             env_package=_create_text_value(biosample_row["sampleType"].values[0]),
             nitro=_create_quantity_value(
-                biosample_row["nitrogenPercent"].values[0], "percent"
+                biosample_row["nitrogenPercent"].values[0], "%"
             ),
             org_carb=_create_quantity_value(
-                biosample_row["organicCPercent"].values[0], "percent"
+                biosample_row["organicCPercent"].values[0], "%"
             ),
             carb_nitro_ratio=_create_quantity_value(
-                biosample_row["CNratio"].values[0], None
+                biosample_row["CNratio"].values[0], "ratio"
             ),
             ph=_create_double_value(biosample_row["soilInWaterpH"].values[0]),
             water_content=(

nmdc_runtime/site/translation/neon_surface_water_translator.py

@@ -3,7 +3,6 @@ import sqlite3
 from typing import Dict, Optional, Union
 
 import pandas as pd
-import requests
 import requests_cache
 
 from nmdc_schema import nmdc
@@ -12,7 +11,6 @@ from nmdc_runtime.site.util import get_basename
 from nmdc_runtime.site.translation.neon_utils import (
     _get_value_or_none,
     _create_controlled_identified_term_value,
-    _create_controlled_term_value,
     _create_geolocation_value,
     _create_quantity_value,
     _create_timestamp_value,

nmdc_runtime/site/translation/submission_portal_translator.py

@@ -47,6 +47,12 @@ DATA_URL_SET_AND_ANALYTE_TO_DATA_OBJECT_TYPE: dict[tuple[DataUrlSet, str], str]
     (INTERLEAVED, str(METATRANSCRIPTOME)): "Metatranscriptome Raw Reads",
 }
 
+UNIT_OVERRIDES: dict[str, dict[str, str]] = {
+    "Biosample": {
+        "depth": "m",
+    }
+}
+
 
 class EnvironmentPackage(Enum):
     r"""
@@ -272,61 +278,9 @@ class SubmissionPortalTranslator(Translator):
     def _get_quantity_value(
         self, raw_value: Optional[str], unit: Optional[str] = None
     ) -> Union[nmdc.QuantityValue, None]:
-        """Construct a nmdc:QuantityValue from a raw value string
-
-        The regex pattern minimally matches on a single numeric value (possibly
-        floating point). The pattern can also identify a range represented by
-        two numeric values separated by a hyphen. It can also identify non-numeric
-        characters at the end of the string which are interpreted as a unit. A unit
-        may also be explicitly provided as an argument to this function. If parsing
-        identifies a unit and a unit argument is provided, the unit argument is used.
-        If the pattern is not matched at all None is returned.
-
-        TODO: currently the parsed unit string is used as-is. In the future we may want
-        to be stricter about what we accept or coerce into a controlled value set
+        """Construct a nmdc:QuantityValue from a raw value string"""
 
-        :param raw_value: string to parse
-        :param unit: optional unit, defaults to None
-        :return: nmdc:QuantityValue
-        """
-        if raw_value is None:
-            return None
-
-        match = re.fullmatch(
-            "([+-]?(?=\.\d|\d)(?:\d+)?(?:\.?\d*)(?:[eE][+-]?\d+)?)(?: *- *([+-]?(?=\.\d|\d)(?:\d+)?(?:\.?\d*)(?:[eE][+-]?\d+)?))?(?: *(\S+))?",
-            raw_value,
-        )
-        if not match:
-            return None
-
-        qv = nmdc.QuantityValue(
-            has_raw_value=raw_value,
-            type="nmdc:QuantityValue",
-        )
-        if match.group(2):
-            # having group 2 means the value is a range like "0 - 1". Either
-            # group 1 or group 2 might be the minimum especially when handling
-            # negative ranges like "0 - -1"
-            num_1 = float(match.group(1))
-            num_2 = float(match.group(2))
-            qv.has_minimum_numeric_value = min(num_1, num_2)
-            qv.has_maximum_numeric_value = max(num_1, num_2)
-        else:
-            # otherwise we just have a single numeric value
-            qv.has_numeric_value = float(match.group(1))
-
-        if unit:
-            # a unit was manually specified
-            if match.group(3) and unit != match.group(3):
-                # a unit was also found in the raw string; issue a warning
-                # if they don't agree, but keep the manually specified one
-                logging.warning(f'Unit mismatch: "{unit}" and "{match.group(3)}"')
-            qv.has_unit = unit
-        elif match.group(3):
-            # a unit was found in the raw string
-            qv.has_unit = match.group(3)
-
-        return qv
+        return self._parse_quantity_value(raw_value, unit)
 
     def _get_ontology_class(
         self, raw_value: Optional[str]
@@ -475,6 +429,50 @@ class SubmissionPortalTranslator(Translator):
 
         return value
 
+    def _get_study_dois(self, metadata_submission) -> Union[List[nmdc.Doi], None]:
+        """Collect and format DOIs from submission portal schema in nmdc format DOIs
+
+        If there were no DOIs, None is returned.
+
+        :param metadata_submission: submission portal entry
+        :return: list of nmdc.DOI objects
+        """
+        data_dois = self._get_from(metadata_submission, ["studyForm", "dataDois"])
+        award_dois = self._get_from(
+            metadata_submission, ["multiOmicsForm", "awardDois"]
+        )
+        if data_dois and len(data_dois) > 0:
+            updated_data_dois = [
+                nmdc.Doi(
+                    doi_category="dataset_doi",
+                    doi_provider=doi["provider"],
+                    doi_value=self._ensure_curie(doi["value"], default_prefix="doi"),
+                    type="nmdc:Doi",
+                )
+                for doi in data_dois
+            ]
+        else:
+            updated_data_dois = []
+
+        if award_dois and len(award_dois) > 0:
+            updated_award_dois = [
+                nmdc.Doi(
+                    doi_category="award_doi",
+                    doi_provider=doi["provider"],
+                    doi_value=self._ensure_curie(doi["value"], default_prefix="doi"),
+                    type="nmdc:Doi",
+                )
+                for doi in award_dois
+            ]
+        else:
+            updated_award_dois = []
+
+        return_val = updated_data_dois + updated_award_dois
+        if len(return_val) == 0:
+            return_val = None
+
+        return return_val
+
     def _get_data_objects_from_fields(
         self,
         sample_data: JSON_OBJECT,
@@ -591,6 +589,7 @@ class SubmissionPortalTranslator(Translator):
             websites=self._get_from(
                 metadata_submission, ["studyForm", "linkOutWebpage"]
             ),
+            associated_dois=self._get_study_dois(metadata_submission),
         )
 
     def _transform_value_for_slot(
@@ -660,6 +659,17 @@ class SubmissionPortalTranslator(Translator):
                 logging.warning(f"No slot '{slot_name}' on class '{class_name}'")
                 continue
 
+            # This step handles cases where the submission portal/schema instructs a user to
+            # provide a value in a specific unit. The unit cannot be parsed out of the raw value
+            # in these cases, so we have to manually set it via UNIT_OVERRIDES. This part can
+            # go away once units are encoded in the schema itself.
+            # See: https://github.com/microbiomedata/nmdc-schema/issues/2517
+            if class_name in UNIT_OVERRIDES:
+                # If the class has unit overrides, check if the slot is in the overrides
+                unit_overrides = UNIT_OVERRIDES[class_name]
+                if slot_name in unit_overrides:
+                    unit = unit_overrides[slot_name]
+
             slot_definition = self.schema_view.induced_slot(slot_name, class_name)
             if slot_definition.multivalued:
                 value_list = value

nmdc_runtime/site/translation/translator.py

@@ -1,9 +1,13 @@
+import logging
+import re
 from abc import ABC, abstractmethod
-from typing import Any, Callable, Dict, List, Optional
+from typing import Any, Callable, Dict, List, Optional, Union
 from nmdc_schema import nmdc
 
 JSON_OBJECT = Dict[str, Any]
 
+logger = logging.getLogger(__name__)
+
 
 class Translator(ABC):
     def __init__(
@@ -27,3 +31,61 @@ class Translator(ABC):
     @abstractmethod
     def get_database(self) -> nmdc.Database:
         pass
+
+    def _parse_quantity_value(
+        self, raw_value: Optional[str], unit: Optional[str] = None
+    ) -> Union[nmdc.QuantityValue, None]:
+        """Construct a nmdc:QuantityValue from a raw value string
+
+        The regex pattern minimally matches on a single numeric value (possibly
+        floating point). The pattern can also identify a range represented by
+        two numeric values separated by a hyphen. It can also identify non-numeric
+        characters at the end of the string which are interpreted as a unit. A unit
+        may also be explicitly provided as an argument to this function. If parsing
+        identifies a unit and a unit argument is provided, the unit argument is used.
+        If the pattern is not matched at all None is returned.
+
+        :param raw_value: string to parse
+        :param unit: optional unit, defaults to None. If None, the unit is extracted from the
+            raw_value. If a unit is provided, it will override the unit extracted from the
+            raw_value.
+        :return: nmdc:QuantityValue
+        """
+        if raw_value is None:
+            return None
+
+        match = re.fullmatch(
+            "([+-]?(?=\.\d|\d)(?:\d+)?(?:\.?\d*)(?:[eE][+-]?\d+)?)(?: *- *([+-]?(?=\.\d|\d)(?:\d+)?(?:\.?\d*)(?:[eE][+-]?\d+)?))?(?: *(\S+))?",
+            raw_value,
+        )
+        if not match:
+            return None
+
+        quantity_value_kwargs = {
+            "has_raw_value": raw_value,
+            "type": "nmdc:QuantityValue",
+        }
+        if match.group(2):
+            # having group 2 means the value is a range like "0 - 1". Either
+            # group 1 or group 2 might be the minimum especially when handling
+            # negative ranges like "0 - -1"
+            num_1 = float(match.group(1))
+            num_2 = float(match.group(2))
+            quantity_value_kwargs["has_minimum_numeric_value"] = min(num_1, num_2)
+            quantity_value_kwargs["has_maximum_numeric_value"] = max(num_1, num_2)
+        else:
+            # otherwise we just have a single numeric value
+            quantity_value_kwargs["has_numeric_value"] = float(match.group(1))
+
+        if unit:
+            # a unit was manually specified
+            if match.group(3) and unit != match.group(3):
+                # a unit was also found in the raw string; issue a warning
+                # if they don't agree, but keep the manually specified one
+                logger.warning(f'Unit mismatch: "{unit}" and "{match.group(3)}"')
+            quantity_value_kwargs["has_unit"] = unit
+        elif match.group(3):
+            # a unit was found in the raw string
+            quantity_value_kwargs["has_unit"] = match.group(3)
+
+        return nmdc.QuantityValue(**quantity_value_kwargs)

nmdc_runtime/site/util.py

@@ -3,10 +3,11 @@ import os
 from functools import lru_cache
 from pymongo.database import Database as MongoDatabase
 from subprocess import Popen, PIPE, STDOUT, CalledProcessError
-from toolz import groupby
+from refscan.lib.helpers import get_collection_names_from_schema
 
-from nmdc_runtime.api.db.mongo import get_collection_names_from_schema
 from nmdc_runtime.site.resources import mongo_resource
+from nmdc_runtime.util import nmdc_schema_view
+
 
 mode_test = {
     "resource_defs": {"mongo": mongo_resource}
@@ -37,12 +38,16 @@ def run_and_log(shell_cmd, context):
 
 @lru_cache
 def schema_collection_has_index_on_id(mdb: MongoDatabase) -> dict:
+    """
+    TODO: Document this function.
+    """
+    schema_view = nmdc_schema_view()
     present_collection_names = set(mdb.list_collection_names())
     return {
         name: (
             name in present_collection_names and "id_1" in mdb[name].index_information()
         )
-        for name in get_collection_names_from_schema()
+        for name in get_collection_names_from_schema(schema_view)
     }