change webbpsf data caching in github CI

download webbpsf data once per week to update
a cache ``webbpsf-<hash>`` that can be used as
a ``cache-restore-keys`` entry for the cache in
other ci jobs. This can also be triggered when
a PR is specifically labeled to update webbpsf
data.

Update other ``roman_ci.yml`` to use most recent
``webbpsf-<hash>`` to look up hash to then construct
a combined cache key (for the combined crds/webbpsf data
cache) using the webbpsf data as a restore key to initialize
the cache with the pre-fetched webbpsf data.

``roman_ci_cron.yml`` will need similar updates.

.github/workflows/roman_ci.yml

@@ -26,8 +26,11 @@ jobs:
     env:
       OBSERVATORY: roman
       CRDS_SERVER_URL: https://roman-crds.stsci.edu
+      # CRDS_PATH here should match DATA_PATH in webbpsf_data.yml
       CRDS_PATH: /tmp/data
+      GH_TOKEN: ${{ github.token }}
     outputs:
+      # TODO simplify data_path and crds_path as they are the same
       data_path: ${{ steps.data.outputs.path }}
       webbpsf_path: ${{ steps.webbpsf_path.outputs.path }}
       data_hash: ${{ steps.data_hash.outputs.hash }}
@@ -36,6 +39,8 @@ jobs:
       crds_server: ${{ steps.crds_server.outputs.url }}
     steps:
       # crds:
+      - id: data
+        run: echo "path=/tmp/data" >> $GITHUB_OUTPUT
       - id: crds_context
         run: >
           echo "pmap=$(
@@ -52,22 +57,18 @@ jobs:
       - uses: actions/checkout@v3
         with:
           fetch-depth: 0
-      - id: data
-        run: |
-          echo "path=/tmp/data" >> $GITHUB_OUTPUT
-          echo "webbpsf_url=https://stsci.box.com/shared/static/qxpiaxsjwo15ml6m4pkhtk36c9jgj70k.gz" >> $GITHUB_OUTPUT
-      - run: |
-          mkdir -p tmp/data/
-          mkdir -p ${{ steps.data.outputs.path }}
-      - run: wget ${{ steps.data.outputs.webbpsf_url }} -O tmp/minimal-webbpsf-data.tar.gz
-      - run: tar -xzvf tmp/minimal-webbpsf-data.tar.gz -C tmp/data/
-      - id: data_hash
-        run: echo "hash=${{ hashFiles( 'tmp/data' ) }}" >> $GITHUB_OUTPUT
-      - run: mv tmp/data/* ${{ steps.data.outputs.path }}
-      - uses: actions/cache@v3
+      # webbpsf:
+      # checkout repo (required for gh below, might be possible to do this with arguments to gh)
+      - uses: actions/checkout@v3
         with:
-          path: ${{ steps.data.outputs.path }}
-          key: data-${{ steps.data_hash.outputs.hash }}-${{ steps.crds_context.outputs.pmap }}
+          fetch-depth: 0
+      - id: data_hash
+        run: |
+          # use actions/gh-actions-cache to allow filtering by key
+          gh extension install actions/gh-actions-cache
+
+          # get most recent webbpsf hash (from the cache key)
+          echo "hash=$(gh actions-cache list --key webbpsf- --sort created-at | cut -f 1 | head -n 1 | cut -d '-' -f 2)" >> $GITHUB_OUTPUT
       - id: webbpsf_path
         run: echo "path=${{ steps.data.outputs.path }}/webbpsf-data" >> $GITHUB_OUTPUT
   check:
@@ -93,6 +94,7 @@ jobs:
         DD_GIT_BRANCH: ${{ github.ref_name }}
       cache-path: ${{ needs.data.outputs.data_path }}
       cache-key: data-${{ needs.data.outputs.data_hash }}-${{ needs.data.outputs.crds_context }}
+      cache-restore-keys: webbpsf=${{ needs.data.outputs.data_hash }}
       envs: |
         - linux: py39-oldestdeps-cov
           pytest-results-summary: true

.github/workflows/webbpsf_data.yml

@@ -0,0 +1,51 @@
+name: check/update webbpsf cache
+
+on:
+  schedule:
+    - cron: "42 4 * * 3"
+  pull_request:
+    # We also want this workflow triggered if the `update webbpsf data` label is
+    # added or present when PR is updated
+    types:
+      - opened
+      - reopened
+      - labeled
+      - unlabeled
+      - synchronize
+  workflow_dispatch:
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  webbpsf-data:
+    if: (github.repository == 'spacetelescope/romancal' && (github.event_name == 'schedule' || github.event_name == 'push' || github.event_name == 'workflow_dispatch' || contains(github.event.pull_request.labels.*.name, 'update webbpsf data')))
+    name: fetch, check, and possibly update WEBBPSF data cache
+    runs-on: ubuntu-latest
+    env:
+      # DATA_PATH here should match CRDS_PATH in roman_ci.yml and roman_ci_cron.yml
+      DATA_PATH: /tmp/data
+      WEBBPSF_DATA_URL: https://stsci.box.com/shared/static/qxpiaxsjwo15ml6m4pkhtk36c9jgj70k.gz
+    outputs:
+      path: ${{ steps.cache_path.outputs.path }}
+      hash: ${{ steps.data_hash.outputs.hash }}
+    steps:
+      - id: cache_path
+        run: |
+          echo "path=${{ env.DATA_PATH }}" >> $GITHUB_OUTPUT
+      - id: data_hash
+        run: |
+          mkdir -p tmp/data
+          wget ${{ env.WEBBPSF_DATA_URL }} -O tmp/webbpsf-data.tar.gz
+          echo "hash=$( shasum tmp/webbpsf-data.tar.gz | cut -d ' ' -f 1 )" >> $GITHUB_OUTPUT
+      - id: cache_check
+        uses: actions/cache@v3
+        with:
+          path: ${{ steps.cache_path.outputs.path }}
+          key: webbpsf-${{ steps.data_hash.outputs.hash }}
+      - if: ${{ steps.cache_check.outputs.cache-hit != 'true' }}
+        name: Initialize cache
+        run: |
+          mkdir -p ${{ steps.cache_path.outputs.path }}
+          tar -xzvf tmp/webbpsf-data.tar.gz -C ${{ steps.cache_path.outputs.path }}