nmdc-runtime 2.9.0__py3-none-any.whl → 2.11.0__py3-none-any.whl

nmdc_runtime/site/changesheets/scripts/missing_neon_soils_ecosystem_data.py

@@ -0,0 +1,311 @@
+#!/usr/bin/env python
+# nmdc_runtime/site/changesheets/scripts/missing_neon_soils_ecosystem_data.py
+"""
+missing_neon_soils_ecosystem_data.py: Create a changesheet for missing ecosystem data for NEON soils samples
+"""
+
+import logging
+import os
+import time
+from pathlib import Path
+from typing import List, Tuple
+
+import click
+from dotenv import load_dotenv
+
+from nmdc_runtime.site.changesheets.base import (
+    Changesheet,
+    ChangesheetLineItem,
+    JSON_OBJECT,
+)
+from nmdc_runtime.site.normalization.gold import (
+    get_gold_biosample_name_suffix,
+    normalize_gold_id,
+)
+from nmdc_runtime.site.resources import GoldApiClient, RuntimeApiUserClient
+
+load_dotenv()
+GOLD_NEON_SOIL_STUDY_ID = "Gs0144570"
+NAME = "missing_neon_soils_ecosystem_data"
+# omics processing to biosamples file in data directory
+DATA_PATH = Path(__file__).parent.parent.joinpath("data")
+
+log_filename = f"{NAME}-{time.strftime('%Y%m%d-%H%M%S')}.log"
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s %(levelname)s %(message)s",
+    filename=log_filename,
+    encoding="utf-8",
+    filemode="w",
+)
+
+
+def read_omics_processing_to_biosample_map() -> dict:
+    """
+    Map omics processing to biosamples
+    :return: dict
+    """
+    omics_processing_to_biosamples = {}
+    with open(DATA_PATH.joinpath("OmicsProcessing-to-catted-Biosamples.tsv")) as f:
+        lines_read = 0
+        lines_skipped = 0
+        for line in f:
+            if not line.startswith("nmdc:"):
+                lines_skipped += 1
+                continue
+            lines_read += 1
+            omics_processing_id, biosample_ids = line.strip().split("\t")
+            omics_processing_to_biosamples[omics_processing_id] = biosample_ids.split(
+                "|"
+            )
+    logging.debug(
+        f"Read {lines_read} lines from OmicsProcessing-to-catted-Biosamples.tsv"
+    )
+    logging.info(
+        f"Skipped {lines_skipped} lines from OmicsProcessing-to-catted-Biosamples.tsv"
+    )
+    return omics_processing_to_biosamples
+
+
+def gold_biosample_to_nmdc_biosamples_and_omics_processing_records(
+    runtime_api_client, omprc_to_bs_map, goldbs
+) -> Tuple[List[JSON_OBJECT], List[JSON_OBJECT]]:
+    """
+    Find the corresponding NMDC biosamples and omics processing records for a
+    GOLD biosample
+    :param runtime_api_client:
+    :param omprc_to_bs_map: Dict of omics processing ID to biosample IDs
+    :param goldbs: a GOLD biosample
+    :return: (
+            List of corresponding NMDC biosamples,
+            List of corresponding NMDC omics processing records
+            )
+    """
+    goldbs_id = normalize_gold_id(goldbs["biosampleGoldId"])
+    goldbs_name_suffix = get_gold_biosample_name_suffix(goldbs["biosampleName"])
+    logging.info(f"goldbs_id: {goldbs_id}")
+    logging.info(f"goldbs_name_suffix: {goldbs_name_suffix}")
+
+    # Search for NMDC biosamples with by GOLD biosample ID
+    nmdc_biosamples = []
+    logging.info(f"Searching for NMDC biosamples with {goldbs_id}...")
+    nmdcbs = runtime_api_client.get_biosamples_by_gold_biosample_id(goldbs_id)
+    logging.info(f"Found {len(nmdcbs)} NMDC biosamples with {goldbs_id}...")
+    nmdc_biosamples.extend(nmdcbs)
+
+    # Search for NMDC biosamples via omics processing name containing GOLD biosample name suffix
+    logging.info(
+        f"Searching for NMDC omics processing name containing {goldbs_name_suffix}..."
+    )
+    omprc_records = runtime_api_client.get_omics_processing_by_name(goldbs_name_suffix)
+    for omprc in omprc_records:
+        omprc_id = omprc["id"]
+        logging.info(f"omprc_id: {omprc_id}")
+        logging.info(
+            f"Searching for NMDC biosamples with omics processing {omprc_id}..."
+        )
+        nmdcbs_ids = omprc_to_bs_map.get(omprc_id, [])
+        logging.info(f"Found {len(nmdcbs_ids)} NMDC biosamples with {omprc_id}...")
+        for nmdcbs_id in nmdcbs_ids:
+            nmdcbs_req = runtime_api_client.request("GET", f"/biosamples/{nmdcbs_id}")
+            if nmdcbs_req.status_code != 200:
+                logging.error(
+                    f"Failed to retrieve NMDC biosample {nmdcbs_id}: {nmdcbs_req.status_code}"
+                )
+                continue
+            nmdcbs = nmdcbs_req.json()
+            nmdc_biosamples.append(nmdcbs)
+
+    logging.info(f"Found {len(nmdc_biosamples)} NMDC biosamples for {goldbs_id}...")
+    return nmdc_biosamples, omprc_records
+
+
+def compare_biosamples(goldbs, nmdcbs) -> List[ChangesheetLineItem]:
+    changesheet_line_items = []
+
+    # Check for missing ecosystem metadata
+    changesheet_line_items.extend(_check_ecosystem_metadata(goldbs, nmdcbs))
+
+    # Check for missing gold biosample identifiers
+    changesheet_line_items.extend(_check_gold_biosample_identifiers(goldbs, nmdcbs))
+
+    return changesheet_line_items
+
+
+def _check_ecosystem_metadata(goldbs, nmdcbs) -> List[ChangesheetLineItem]:
+    # nmdc to gold ecosystem key map
+    ecosystem_key_map = {
+        "ecosystem": "ecosystem",
+        "ecosystem_category": "ecosystemCategory",
+        "ecosystem_type": "ecosystemType",
+        "ecosystem_subtype": "ecosystemSubtype",
+    }
+    changesheet_line_items = []
+    for nmdc_key, gold_key in ecosystem_key_map.items():
+        if not gold_key in goldbs:
+            logging.warning(f"no {gold_key} for {goldbs['biosampleGoldId']}...")
+            continue
+        if not nmdc_key in nmdcbs:
+            changesheet_line_items.append(
+                ChangesheetLineItem(
+                    id=nmdcbs["id"],
+                    action="update",
+                    attribute=nmdc_key,
+                    value=goldbs.get(gold_key),
+                )
+            )
+            continue
+        if nmdcbs[nmdc_key] != goldbs.get(gold_key):
+            changesheet_line_items.append(
+                ChangesheetLineItem(
+                    id=nmdcbs["id"],
+                    action="update",
+                    attribute=nmdc_key,
+                    value=goldbs.get(gold_key),
+                )
+            )
+            continue
+
+    return changesheet_line_items
+
+
+def _check_gold_biosample_identifiers(goldbs, nmdcbs) -> List[ChangesheetLineItem]:
+    changesheet_line_items = []
+    goldbs_id = normalize_gold_id(goldbs["biosampleGoldId"])
+    if not goldbs_id in nmdcbs["gold_biosample_identifiers"]:
+        changesheet_line_items.append(
+            ChangesheetLineItem(
+                id=nmdcbs["id"],
+                action="insert",
+                attribute="gold_biosample_identifiers",
+                value=goldbs_id + "|",
+            )
+        )
+    return changesheet_line_items
+
+
+def compare_projects(gold_project, omprc_record) -> ChangesheetLineItem:
+    gold_project_id = normalize_gold_id(gold_project["projectGoldId"])
+    if "gold_sequencing_project_identifiers" not in omprc_record:
+        return ChangesheetLineItem(
+            id=omprc_record["id"],
+            action="insert",
+            attribute="gold_sequencing_project_identifiers",
+            value=gold_project_id + "|",
+        )
+
+    if gold_project_id not in omprc_record["gold_sequencing_project_identifiers"]:
+        return ChangesheetLineItem(
+            id=omprc_record["id"],
+            action="insert",
+            attribute="gold_sequencing_project_identifiers",
+            value=gold_project_id + "|",
+        )
+
+
+@click.command()
+@click.option("--study_id", default=GOLD_NEON_SOIL_STUDY_ID, help="GOLD study ID")
+@click.option("--use_dev_api", is_flag=True, default=False, help="Use the dev API")
+def generate_changesheet(study_id, use_dev_api):
+    """
+    Generate a changesheet for missing ecosystem data for NEON soils samples by:
+    1. Retrieving GOLD biosamples for the given study
+    2. Finding the corresponding NMDC biosamples and omics processing records for each GOLD biosample
+    3. Comparing the GOLD biosample to the NMDC biosamples and omics processing records
+    4. Generating a changesheet for the differences
+
+    WARNING: This script is not idempotent. It will generate a new changesheet each time it is run.
+
+    Changesheet is written to nmdc_runtime/site/changesheets/changesheets_output
+
+    :param study_id: The GOLD study ID
+    :param use_dev_api: Use the dev API (default: False)
+    :return:
+    """
+    start_time = time.time()
+    logging.info("starting missing_neon_soils_ecosystem_data.py...")
+    logging.info(f"study_id: {study_id}")
+
+    gold_api_client = GoldApiClient(
+        base_url=os.getenv("GOLD_API_BASE_URL"),
+        username=os.getenv("GOLD_API_USERNAME"),
+        password=os.getenv("GOLD_API_PASSWORD"),
+    )
+    logging.info("connected to GOLD API...")
+
+    if use_dev_api:
+        base_url = os.getenv("API_HOST_DEV")
+        logging.info("using dev API...")
+    else:
+        base_url = os.getenv("API_HOST")
+        logging.info("using prod API...")
+
+    runtime_api_user_client = RuntimeApiUserClient(
+        base_url=base_url,
+        username=os.getenv("API_QUERY_USER"),
+        password=os.getenv("API_QUERY_PASS"),
+    )
+    logging.info("connected to NMDC API...")
+
+    # Retrieve GOLD biosamples for the given study
+    gold_biosamples = gold_api_client.fetch_biosamples_by_study(study_id)
+    logging.info(f"retrieved {len(gold_biosamples)} biosamples from GOLD API...")
+
+    # omics processing to biosamples map generated by a SPARQL query
+    omprc_to_bs_map = read_omics_processing_to_biosample_map()
+
+    changesheet = Changesheet(name=NAME)
+    # For each GOLD biosample, find the corresponding NMDC biosamples
+    nmdcbs_count = 0
+    unfindable_goldbs_ids = []
+    for goldbs in gold_biosamples:
+        (
+            nmdc_biosamples,
+            omprc_records,
+        ) = gold_biosample_to_nmdc_biosamples_and_omics_processing_records(
+            runtime_api_user_client, omprc_to_bs_map, goldbs
+        )
+        if not nmdc_biosamples:
+            logging.warning(
+                f"no corresponding NMDC biosamples found for {goldbs['biosampleGoldId']}..."
+            )
+            unfindable_goldbs_ids.append(goldbs["biosampleGoldId"])
+            continue
+        logging.info(
+            f"found {len(nmdc_biosamples)} corresponding NMDC biosamples for {goldbs['biosampleGoldId']}..."
+        )
+        nmdcbs_count += len(nmdc_biosamples)
+        for nmdcbs in nmdc_biosamples:
+            logging.info(f"nmdcbs: {nmdcbs['id']}")
+            changesheet.line_items.extend(compare_biosamples(goldbs, nmdcbs))
+
+        # Insert gold project id into omprc alternative identifiers
+        gold_projects = gold_api_client.request(
+            "/projects", params={"biosampleGoldId": goldbs["biosampleGoldId"]}
+        )
+        for gold_project in gold_projects:
+            for omprc_record in omprc_records:
+                changesheet.line_items.append(
+                    compare_projects(gold_project, omprc_record)
+                )
+
+    logging.info(f"Processed {len(gold_biosamples)} GOLD biosamples...")
+    logging.info(f"found {nmdcbs_count} corresponding NMDC biosamples...")
+    logging.info(f"unfindable_count: {len(unfindable_goldbs_ids)}...")
+    for unfindable_goldbs_ids in unfindable_goldbs_ids:
+        logging.info(f"unfindable_goldbs_id: {unfindable_goldbs_ids}...")
+    logging.info(f"changesheet has {len(changesheet.line_items)} line items...")
+
+    changesheet.write_changesheet()
+
+    logging.info("Validating changesheet...")
+    is_valid_changesheet = changesheet.validate_changesheet(base_url)
+    logging.info(f"Changesheet is valid: {is_valid_changesheet}")
+
+    logging.info(
+        f"missing_neon_soils_ecosystem_data.py completed in {time.time() - start_time} seconds..."
+    )
+
+
+if __name__ == "__main__":
+    generate_changesheet()

nmdc_runtime/site/changesheets/scripts/neon_soils_add_ncbi_ids.py

@@ -0,0 +1,210 @@
+#!/usr/bin/env python3
+# coding: utf-8
+# nmdc_runtime/site/changesheets/scripts/neon_soils_add_ncbi_ids.py
+"""
+neon_soils_add_ncbi_ids.py: Add NCBI biosample accessions to neon soils
+biosamples, NCBI bioproject accessions to omics processing, and
+NCBI Umbrella bioproject accession to neon soils study.
+"""
+import logging
+import time
+
+import click
+from dotenv import load_dotenv
+
+from nmdc_runtime.site.changesheets.base import (
+    Changesheet,
+    ChangesheetLineItem,
+    get_gold_client,
+    get_runtime_client,
+)
+
+load_dotenv()
+NAME = "neon_soils_add_ncbi_ids"
+NMDC_STUDY_ID = "nmdc:sty-11-34xj1150"
+UMBRELLA_BIOPROJECT_ACCESSION = "PRJNA1029061"
+
+log_filename = f"{NAME}-{time.strftime('%Y%m%d-%H%M%S')}.log"
+logging.basicConfig(
+    level=logging.INFO,
+    format="%(asctime)s %(levelname)s %(message)s",
+    filename=log_filename,
+    encoding="utf-8",
+    filemode="w",
+)
+
+
+def _get_change_for_biosample(biosample, ncbi_biosample_accession):
+    """
+    Get the changes for the given biosample
+    :param biosample: dict - the biosample
+    :param ncbi_biosample_accession: str - the NCBI BioSample accession
+    :return: list - the changes
+    """
+    ncbi_biosample_accessions = biosample.get("insdc_biosample_identifiers", [])
+    if ncbi_biosample_accession in ncbi_biosample_accessions:
+        return
+    biosample_id = biosample["id"]
+    logging.info(f"creating change for biosample_id: {biosample_id}")
+    return ChangesheetLineItem(
+        id=biosample["id"],
+        action="insert",
+        attribute="insdc_biosample_identifiers",
+        value="biosample:" + ncbi_biosample_accession + "|",
+    )
+
+
+def _get_change_for_omics_processing(
+    omics_processing_record, ncbi_bioproject_accession
+):
+    """
+    Get the changes for the given omics_processing_record
+    :param omics_processing_record:
+    :param ncbi_bioproject_accession:
+    :return:
+    """
+    ncbi_bioproject_accessions = omics_processing_record.get(
+        "insdc_bioproject_identifiers", []
+    )
+    if ncbi_bioproject_accession in ncbi_bioproject_accessions:
+        return
+    omics_processing_id = omics_processing_record["id"]
+    logging.info(f"creating change for omics_processing_id: {omics_processing_id}")
+    return ChangesheetLineItem(
+        id=omics_processing_id,
+        action="insert",
+        attribute="insdc_bioproject_identifiers",
+        value="bioproject:" + ncbi_bioproject_accession + "|",
+    )
+
+
+@click.command()
+@click.option("--study_id", default=NMDC_STUDY_ID, help="NMDC study ID")
+@click.option("--use_dev_api", is_flag=True, default=True, help="Use the dev API")
+def generate_changesheet(study_id, use_dev_api):
+    """
+    Generate a changesheet for neon soils study and biosamples by:
+    0. Changesheet line item: Umbrella BioProjectAccession to
+        study.insdc_project_identifiers
+    1. Retrieving all gold_study_identifiers for the neon soils study
+    2. For each gold_study_identifier, retrieve the GOLD projects
+    3. For each GOLD project,
+        A. retrieve the corresponding NMDC biosample(s). For each biosample,
+            - Changesheet line item:NCBI BioSampleAccession to
+            insdc_biosample_identifiers
+        B. Retrieve the corresponding NMDC omics_processing. For each,
+            - Changesheet line item:NCBI BioProjectAccession to
+            insdc_bioproject_identifiers
+
+    WARNING: This script is not idempotent. It will generate a new changesheet
+    each time it is run.
+    Changesheet is written to nmdc_runtime/site/changesheets/changesheets_output
+
+    :param study_id: The NMDC study ID
+    :param use_dev_api: Use the dev API (default: False)
+    :return:
+    """
+    start_time = time.time()
+    logging.info(f"Generating changesheet for {study_id}")
+    logging.info(f"Using dev API: {use_dev_api}")
+
+    # Initialize the NMDC API
+    runtime_client = get_runtime_client(use_dev_api)
+
+    # Initialize the GOLD API
+    gold_client = get_gold_client()
+
+    # Initialize the changesheet
+    changesheet = Changesheet(name=NAME)
+
+    # 1. Retrieve all gold_study_identifiers for the neon soils study
+    logging.info(f"Retrieving gold_study_identifiers for {study_id}")
+    res = runtime_client.request("GET", f"/studies/{study_id}")
+    nmdc_study = res.json()
+    changesheet.line_items.append(
+        ChangesheetLineItem(
+            id=study_id,
+            action="insert",
+            attribute="insdc_bioproject_identifiers",
+            value="bioproject:" + UMBRELLA_BIOPROJECT_ACCESSION + "|",
+        )
+    )
+
+    gold_study_identifiers = nmdc_study["gold_study_identifiers"]
+    logging.info(f"gold_study_identifiers: {gold_study_identifiers}")
+    gold_project_count = 0
+    biosample_count = 0
+    for gold_study_identifier in gold_study_identifiers:
+        # 2. For each gold_study_identifier, retrieve the GOLD projects
+        if gold_study_identifier == "gold:Gs0144570":
+            # TODO verify that this one has already been done
+            continue
+        logging.info(
+            f"Retrieving GOLD projects for gold_study_identifier: {gold_study_identifier}"
+        )
+        projects = gold_client.fetch_projects_by_study(gold_study_identifier)
+        logging.info(f"Retrieved {len(projects)} projects")
+
+        # 3. For each GOLD project,
+        for project in projects:
+            gold_project_count += 1
+            project_gold_id = project["projectGoldId"]
+            biosample_gold_id = project["biosampleGoldId"]
+            ncbi_bioproject_accession = project["ncbiBioProjectAccession"]
+            ncbi_biosample_accession = project["ncbiBioSampleAccession"]
+
+            # A. retrieve the corresponding NMDC biosample(s)
+            logging.info(
+                f"Retrieving NMDC biosamples for biosample_gold_id: {biosample_gold_id}"
+            )
+            biosamples = runtime_client.get_biosamples_by_gold_biosample_id(
+                biosample_gold_id
+            )
+            logging.info(f"Retrieved {len(biosamples)} biosamples")
+            for biosample in biosamples:
+                biosample_count += 1
+                biosample_id = biosample["id"]
+                logging.info(f"biosample_id: {biosample_id}")
+                # NcbiBioSampleAccession to insdc_biosample_identifiers
+                change = _get_change_for_biosample(biosample, ncbi_biosample_accession)
+                if change:
+                    changesheet.line_items.append(change)
+
+            # B. Retrieve the corresponding NMDC omics_processing
+            logging.info(
+                f"Retrieving NMDC omics_processing for project_gold_id: {project_gold_id}"
+            )
+            omics_processing_records = (
+                runtime_client.get_omics_processing_records_by_gold_project_id(
+                    project_gold_id
+                )
+            )
+            logging.info(f"Retrieved {len(omics_processing_records)} omics_processings")
+            for omics_processing in omics_processing_records:
+                omics_processing_id = omics_processing["id"]
+                logging.info(f"omics_processing_id: {omics_processing_id}")
+                # NcbiBioProjectAccession to insdc_experiment_identifiers
+                change = _get_change_for_omics_processing(
+                    omics_processing, ncbi_bioproject_accession
+                )
+                if change:
+                    changesheet.line_items.append(change)
+
+    logging.info(f"gold_project_count: {gold_project_count}")
+    logging.info(f"biosample_count: {biosample_count}")
+    logging.info(f"changesheet has {len(changesheet.line_items)} line items")
+
+    # Write the changesheet
+    changesheet.write_changesheet()
+
+    # Validate the changesheet
+    if changesheet.validate_changesheet(runtime_client.base_url):
+        logging.info(f"Changesheet is valid")
+    else:
+        logging.error(f"Changesheet is invalid")
+
+    logging.info(f"Completed in {time.time() - start_time} seconds")
+
+
+if __name__ == "__main__":
+    generate_changesheet()

nmdc_runtime/site/dagster.yaml

@@ -0,0 +1,53 @@
+scheduler:
+  module: dagster.core.scheduler
+  class: DagsterDaemonScheduler
+
+run_coordinator:
+  module: dagster.core.run_coordinator
+  class: QueuedRunCoordinator
+
+run_launcher:
+  module: dagster.core.launcher
+  class: DefaultRunLauncher
+
+run_storage:
+  module: dagster_postgres.run_storage
+  class: PostgresRunStorage
+  config:
+    postgres_db:
+      hostname: dagster-postgresql
+      username:
+        env: DAGSTER_POSTGRES_USER
+      password:
+        env: DAGSTER_POSTGRES_PASSWORD
+      db_name:
+        env: DAGSTER_POSTGRES_DB
+      port: 5432
+
+schedule_storage:
+  module: dagster_postgres.schedule_storage
+  class: PostgresScheduleStorage
+  config:
+    postgres_db:
+      hostname: dagster-postgresql
+      username:
+        env: DAGSTER_POSTGRES_USER
+      password:
+        env: DAGSTER_POSTGRES_PASSWORD
+      db_name:
+        env: DAGSTER_POSTGRES_DB
+      port: 5432
+
+event_log_storage:
+  module: dagster_postgres.event_log
+  class: PostgresEventLogStorage
+  config:
+    postgres_db:
+      hostname: dagster-postgresql
+      username:
+        env: DAGSTER_POSTGRES_USER
+      password:
+        env: DAGSTER_POSTGRES_PASSWORD
+      db_name:
+        env: DAGSTER_POSTGRES_DB
+      port: 5432

nmdc_runtime/site/entrypoint-daemon.sh

@@ -0,0 +1,26 @@
+#!/bin/bash
+
+set -euo pipefail
+
+file_env() {
+	local var="$1"
+	local fileVar="${var}_FILE"
+	local def="${2:-}"
+	if [ "${!var:-}" ] && [ "${!fileVar:-}" ]; then
+		echo >&2 "error: both $var and $fileVar are set (but are exclusive)"
+		exit 1
+	fi
+	local val="$def"
+	if [ "${!var:-}" ]; then
+		val="${!var}"
+	elif [ "${!fileVar:-}" ]; then
+		val="$(< "${!fileVar}")"
+	fi
+	export "$var"="$val"
+	unset "$fileVar"
+}
+
+file_env "MONGO_PASSWORD"
+file_env "DAGSTER_POSTGRES_PASSWORD"
+
+exec uv run --active dagster-daemon run

nmdc_runtime/site/entrypoint-dagit-readonly.sh

@@ -0,0 +1,26 @@
+#!/bin/bash
+
+set -euo pipefail
+
+file_env() {
+	local var="$1"
+	local fileVar="${var}_FILE"
+	local def="${2:-}"
+	if [ "${!var:-}" ] && [ "${!fileVar:-}" ]; then
+		echo >&2 "error: both $var and $fileVar are set (but are exclusive)"
+		exit 1
+	fi
+	local val="$def"
+	if [ "${!var:-}" ]; then
+		val="${!var}"
+	elif [ "${!fileVar:-}" ]; then
+		val="$(< "${!fileVar}")"
+	fi
+	export "$var"="$val"
+	unset "$fileVar"
+}
+
+file_env "MONGO_PASSWORD"
+file_env "DAGSTER_POSTGRES_PASSWORD"
+
+exec uv run --active dagit -h 0.0.0.0 -p 3000 -w workspace.yaml --read-only

nmdc_runtime/site/entrypoint-dagit.sh

@@ -0,0 +1,26 @@
+#!/bin/bash
+
+set -euo pipefail
+
+file_env() {
+	local var="$1"
+	local fileVar="${var}_FILE"
+	local def="${2:-}"
+	if [ "${!var:-}" ] && [ "${!fileVar:-}" ]; then
+		echo >&2 "error: both $var and $fileVar are set (but are exclusive)"
+		exit 1
+	fi
+	local val="$def"
+	if [ "${!var:-}" ]; then
+		val="${!var}"
+	elif [ "${!fileVar:-}" ]; then
+		val="$(< "${!fileVar}")"
+	fi
+	export "$var"="$val"
+	unset "$fileVar"
+}
+
+file_env "MONGO_PASSWORD"
+file_env "DAGSTER_POSTGRES_PASSWORD"
+
+exec uv run --active dagit -h 0.0.0.0 -p 3000 -w workspace.yaml