Merge pull request #2033 from MIT-LCP/alistair/action_fixes

fix pathing by switching to `$GITHUB_WORKSPACE` in the actions

Author: alistairewj

.github/actions/load-mimic-duckdb/action.yml

@@ -3,20 +3,26 @@ description: "Builds a DuckDB database from the demo hosp/icu data using import_
 
 inputs:
   mimic_data_dir:
-    description: "Directory containing the hosp/ and icu/ demo data folders."
+    description: "Directory containing the hosp/ and icu/ demo data folders. Defaults to $GITHUB_WORKSPACE."
     required: false
-    default: ${{ github.workspace }}
+    default: ""
   db_file:
-    description: "Path to the DuckDB database file to create."
+    description: "Path to the DuckDB database file to create. Defaults to $GITHUB_WORKSPACE/mimic4.db."
     required: false
-    default: ${{ github.workspace }}/mimic4.db
+    default: ""
 
 runs:
   using: "composite"
   steps:
     - name: Build DuckDB database
       shell: bash
-      # `echo n` answers the overwrite prompt defensively
+      # Resolve paths at shell time so $GITHUB_WORKSPACE points at the
+      # container-mounted path.
+      # `echo n` answers the overwrite prompt defensively.
       run: |
+        data_dir="${{ inputs.mimic_data_dir }}"
+        data_dir="${data_dir:-$GITHUB_WORKSPACE}"
+        db_file="${{ inputs.db_file }}"
+        db_file="${db_file:-$GITHUB_WORKSPACE/mimic4.db}"
         cd "$GITHUB_WORKSPACE/mimic-iv/buildmimic/duckdb"
-        echo n | ./import_duckdb.sh "${{ inputs.mimic_data_dir }}" "${{ inputs.db_file }}"
+        echo n | ./import_duckdb.sh "$data_dir" "$db_file"

.github/actions/load-mimic-psql/action.yml

@@ -3,16 +3,21 @@ description: "Creates the MIMIC-IV schemas and loads the demo hosp/icu data. Req
 
 inputs:
   mimic_data_dir:
-    description: "Directory containing the hosp/ and icu/ demo data folders."
+    description: "Directory containing the hosp/ and icu/ demo data folders. Defaults to $GITHUB_WORKSPACE."
     required: false
-    default: ${{ github.workspace }}
+    default: ""
 
 runs:
   using: "composite"
   steps:
     - name: Create schemas and load demo data
       shell: bash
+      # Resolve the data dir at shell time so $GITHUB_WORKSPACE points at the
+      # container-mounted path. The github.workspace context yields the host
+      # path, which does not exist inside a container job.
       run: |
+        data_dir="${{ inputs.mimic_data_dir }}"
+        data_dir="${data_dir:-$GITHUB_WORKSPACE}"
         psql -q -v ON_ERROR_STOP=1 -f "$GITHUB_WORKSPACE/mimic-iv/buildmimic/postgres/create.sql"
-        psql -q -v ON_ERROR_STOP=1 -v mimic_data_dir="${{ inputs.mimic_data_dir }}" \
+        psql -q -v ON_ERROR_STOP=1 -v mimic_data_dir="$data_dir" \
           -f "$GITHUB_WORKSPACE/mimic-iv/buildmimic/postgres/load_gz.sql"

.github/workflows/build-duckdb.yml

@@ -4,9 +4,6 @@ name: build duckdb
 on:
   workflow_call:
 
-env:
-  DUCKDB_FILE: ${{ github.workspace }}/mimic4.db
-
 jobs:
   build:
     runs-on: ubuntu-latest
@@ -25,13 +22,11 @@ jobs:
       - uses: ./.github/actions/download-demo
 
       - uses: ./.github/actions/load-mimic-duckdb
-        with:
-          db_file: ${{ env.DUCKDB_FILE }}
 
       - name: Validate row counts
         # validate_demo.sql is in postgres folder but also works for duckdb
         run: |
-          duckdb -csv "$DUCKDB_FILE" < mimic-iv/buildmimic/postgres/validate_demo.sql | tee results
+          duckdb -csv "$GITHUB_WORKSPACE/mimic4.db" < mimic-iv/buildmimic/postgres/validate_demo.sql | tee results
           if grep -F -q "FAILED" results; then
             echo "::error::DuckDB demo row-count validation failed:"
             grep "FAILED" results

.github/workflows/concepts-duckdb.yml

@@ -5,9 +5,6 @@ name: concepts duckdb
 on:
   workflow_call:
 
-env:
-  DUCKDB_FILE: ${{ github.workspace }}/mimic4.db
-
 jobs:
   build:
     runs-on: ubuntu-latest
@@ -27,12 +24,10 @@ jobs:
       - uses: ./.github/actions/download-demo
 
       - uses: ./.github/actions/load-mimic-duckdb
-        with:
-          db_file: ${{ env.DUCKDB_FILE }}
 
       - name: Create derived schema
         run: |
-          duckdb "$DUCKDB_FILE" -c "CREATE SCHEMA IF NOT EXISTS mimiciv_derived;"
+          duckdb "$GITHUB_WORKSPACE/mimic4.db" -c "CREATE SCHEMA IF NOT EXISTS mimiciv_derived;"
 
       - uses: ./.github/actions/transpile-concepts
         with:
@@ -41,9 +36,9 @@ jobs:
       - name: Build concepts
         working-directory: ./mimic-iv/concepts_duckdb
         run: |
-          duckdb "$DUCKDB_FILE" -c ".read duckdb.sql"
+          duckdb "$GITHUB_WORKSPACE/mimic4.db" -c ".read duckdb.sql"
 
       - uses: actions/upload-artifact@v4
         with:
           name: duckdb-db
-          path: ${{ env.DUCKDB_FILE }}
+          path: mimic4.db

.github/workflows/equivalence.yml

@@ -7,7 +7,6 @@ on:
 env:
   # Tables with known cross-engine differences
   IGNORE_TABLES: "first_day_gcs,suspicion_of_infection,oxygen_delivery,bg,first_day_bg,first_day_bg_art,kdigo_uo,kdigo_stages,urine_output_rate"
-  DUCKDB_FILE: ${{ github.workspace }}/mimic4.db
 
 jobs:
   diff:
@@ -49,10 +48,11 @@ jobs:
           name: psql-derived
 
       - name: Download DuckDB database
+        # No path: defaults to the working directory ($GITHUB_WORKSPACE inside
+        # the container), placing mimic4.db at $GITHUB_WORKSPACE/mimic4.db.
         uses: actions/download-artifact@v4
         with:
           name: duckdb-db
-          path: ${{ github.workspace }}
 
       - name: Restore PostgreSQL derived schema
         run: |
@@ -62,5 +62,5 @@ jobs:
         run: |
           mimic_utils compare_concepts \
             --pg "host=$PGHOST dbname=$PGDATABASE user=$PGUSER" \
-            --duckdb "$DUCKDB_FILE" \
+            --duckdb "$GITHUB_WORKSPACE/mimic4.db" \
             --ignore "$IGNORE_TABLES"