fix(testcontainers): also monkeypatch oceanbase_py's get_columns

Fixing has_table() alone wasn't enough: get_columns() -- the actual
column-introspection call OceanBaseEngineSpec.get_columns() needs, and
what this test's second half exercises -- has the identical
connection.execute(raw string) bug, confirmed on real CI as the same
ObjectNotExecutableError. Patched the same way, reusing has_table()
correctly since self.has_table already resolves to the earlier patch.

.codecov.yml

@@ -3,9 +3,7 @@ codecov:
         after_n_builds: 4
 ignore:
   - "superset/migrations/versions/*.py"
-  - "superset-frontend/packages/superset-ui-demo/**/*"
   - "**/*.stories.tsx"
-  - "**/*.stories.jsx"
 coverage:
   status:
     project:

.github/ISSUE_TEMPLATE/bug-report.yml

@@ -1,6 +1,6 @@
 name: Bug report
 description: Report a bug to improve Superset's stability
-labels: ["bug"]
+labels: ["#bug"]
 body:
   - type: markdown
     attributes:

.github/ISSUE_TEMPLATE/cosmetic.md

@@ -1,7 +1,7 @@
 ---
 name: Cosmetic Issue
 about: Describe a cosmetic issue with CSS, positioning, layout, labeling, or similar
-labels: "cosmetic-issue"
+labels: "#bug:cosmetic"
 ---
 
 ## Screenshot

.github/dependabot.yml

@@ -5,6 +5,10 @@ updates:
     directory: "/"
     schedule:
       interval: "daily"
+    groups:
+      security:
+        applies-to: "security-updates"
+        patterns: ["*"]
     cooldown:
       default-days: 7
 
@@ -61,6 +65,9 @@ updates:
       - npm
       - dependabot
     groups:
+      security:
+        applies-to: "security-updates"
+        patterns: ["*"]
       rjsf:
         patterns:
           - "@rjsf/*"
@@ -98,6 +105,10 @@ updates:
     labels:
       - pip
       - dependabot
+    groups:
+      security:
+        applies-to: "security-updates"
+        patterns: ["*"]
     cooldown:
       default-days: 7
 
@@ -105,6 +116,10 @@ updates:
     directory: ".github/actions"
     schedule:
       interval: "daily"
+    groups:
+      security:
+        applies-to: "security-updates"
+        patterns: ["*"]
     open-pull-requests-limit: 10
     versioning-strategy: increase
     cooldown:
@@ -115,6 +130,9 @@ updates:
     schedule:
       interval: "daily"
     groups:
+      security:
+        applies-to: "security-updates"
+        patterns: ["*"]
       storybook:
         patterns:
           - "@storybook/*"
@@ -142,6 +160,10 @@ updates:
     labels:
       - npm
       - dependabot
+    groups:
+      security:
+        applies-to: "security-updates"
+        patterns: ["*"]
     versioning-strategy: increase
     cooldown:
       default-days: 7
@@ -153,6 +175,10 @@ updates:
     labels:
       - npm
       - dependabot
+    groups:
+      security:
+        applies-to: "security-updates"
+        patterns: ["*"]
     open-pull-requests-limit: 10
     versioning-strategy: increase
     cooldown:

.github/labeler.yml

@@ -29,7 +29,7 @@
 "dependencies:python":
 - changed-files:
   - any-glob-to-any-file:
-    - 'superset/requirements/**'
+    - 'requirements/**'
     - 'superset/translations/requirements.txt'
     - 'RELEASING/requirements.txt'
 

.github/workflows/bump-python-package.yml

@@ -21,6 +21,9 @@ on:
   #schedule:
   #  - cron: '0 0 * * *'  # Runs daily at midnight UTC
 
+permissions:
+  contents: read
+
 jobs:
   bump-python-package:
     runs-on: ubuntu-slim
@@ -45,7 +48,7 @@ jobs:
           python-version: "3.11"
 
       - name: Install uv
-        uses: astral-sh/setup-uv@c771a70e6277c0a99b617c7a806ffedaca235ff9 # v9.0.0
+        uses: astral-sh/setup-uv@20cfd1bf945f4377ade1205e4dbc17946fc9a30d # v10.0.1
         with:
           python-version: "3.11"
           enable-cache: true

.github/workflows/check-db-migration-confict.yml

@@ -16,6 +16,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.run_id }}
   cancel-in-progress: true
 
+permissions:
+  contents: read
+
 jobs:
   check_db_migration_conflict:
     name: Check DB migration conflict

.github/workflows/codeql-analysis.yml

@@ -14,6 +14,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.run_id }}
   cancel-in-progress: true
 
+permissions:
+  contents: read
+
 jobs:
   changes:
     runs-on: ubuntu-26.04
@@ -64,7 +67,7 @@ jobs:
 
       # Initializes the CodeQL tools for scanning.
       - name: Initialize CodeQL
-        uses: github/codeql-action/init@5595ccaf912efad79be6eef63a5619ff05969be3 # v4.37.6
+        uses: github/codeql-action/init@ff2f1c621b7f889edc0d3c761ac2e6a3f8cdb0dd # v4.37.7
         with:
           languages: ${{ matrix.language }}
           # If you wish to specify custom queries, you can do so here or in a config file.
@@ -75,6 +78,6 @@ jobs:
           # queries: security-extended,security-and-quality
 
       - name: Perform CodeQL Analysis
-        uses: github/codeql-action/analyze@5595ccaf912efad79be6eef63a5619ff05969be3 # v4.37.6
+        uses: github/codeql-action/analyze@ff2f1c621b7f889edc0d3c761ac2e6a3f8cdb0dd # v4.37.7
         with:
           category: "/language:${{matrix.language}}"

.github/workflows/docker.yml

@@ -51,6 +51,53 @@ jobs:
           echo "matrix_config=${MATRIX_CONFIG}" >> $GITHUB_OUTPUT
           echo $GITHUB_OUTPUT
 
+  # Runs unconditionally (no dependency on `changes`, and no build-preset
+  # matrix restriction) so a regression in the PY_VER override logic is
+  # always caught on PRs. Without this, the real docker-build job only runs
+  # when the change detector flags docker/python/frontend changes (a
+  # workflow-only edit like this one does not), and even then the PR build
+  # matrix never includes the "py311"/"py312" presets that logic protects -
+  # so a break here would otherwise first surface on a push to master.
+  pyver-override-check:
+    name: verify docker build PY_VER override
+    runs-on: ubuntu-26.04
+    timeout-minutes: 5
+    steps:
+      - name: Checkout
+        uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
+        with:
+          persist-credentials: false
+      - name: Setup supersetbot
+        uses: ./.github/actions/setup-supersetbot/
+      - name: Assert PY_VER override applies to every preset except py311/py312
+        shell: bash
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          set -euo pipefail
+          # Asserts against the actual buildx command line `supersetbot docker
+          # --dry-run` would run, not just this repo's own extra-flags helper,
+          # so a regression in supersetbot itself (dropping the py311/py312
+          # PY_VER pin, or reordering args so our override no longer lands
+          # last) is caught here too, instead of only surfacing on master.
+          assert_effective_py_ver() {
+            local preset="$1" expected="$2" extra_flags command actual
+            extra_flags="$(scripts/docker-build-extra-flags.sh "$preset" dummy-tag)"
+            command="$(supersetbot docker --preset "$preset" --platform linux/amd64 --extra-flags "$extra_flags" --dry-run)"
+            # docker buildx keeps the LAST value of a repeated --build-arg key.
+            actual="$(grep -oE -- '--build-arg PY_VER=[^[:space:]]+' <<<"$command" | tail -1)"
+            if [ "$actual" != "--build-arg PY_VER=$expected" ]; then
+              echo "::error::preset '$preset' expected effective --build-arg PY_VER=$expected, got: ${actual:-<none>} (full command: $command)"
+              exit 1
+            fi
+          }
+          for preset in dev lean websocket dockerize; do
+            assert_effective_py_ver "$preset" "3.11.14-slim-trixie"
+          done
+          assert_effective_py_ver py311 "3.11-slim-bookworm"
+          assert_effective_py_ver py312 "3.12-slim-bookworm"
+          echo "PY_VER override logic verified against the assembled buildx command for all build presets"
+
   docker-build:
     name: docker-build
     needs: [setup_matrix, changes]
@@ -124,19 +171,21 @@ jobs:
           # the whole job. buildx reuses the buildkit layer cache from the
           # failed attempt, so a retry mostly re-does just the failed push.
           #
-          # supersetbot's "dev"/"lean" presets pin their own --build-arg
-          # PY_VER, which lands ahead of --extra-flags on the assembled
-          # buildx command line; docker/buildx keeps the last value for a
-          # repeated --build-arg key, so appending PY_VER here overrides
-          # supersetbot's pin and keeps the build on the Dockerfile's own
-          # supported Python version.
+          # See scripts/docker-build-extra-flags.sh for why "py311"/"py312"
+          # are excluded from the PY_VER override applied to every other
+          # preset; that logic is also exercised on every PR by the
+          # always-on pyver-override-check job below, since this job itself
+          # only runs when the change detector flags docker/python/frontend
+          # changes and the PR build matrix never includes py311/py312.
+          EXTRA_FLAGS="$(scripts/docker-build-extra-flags.sh "$BUILD_PRESET" "$IMAGE_TAG")"
+
           for attempt in 1 2 3; do
             if supersetbot docker \
               $PUSH_OR_LOAD \
               --preset "$BUILD_PRESET" \
               --context "$EVENT" \
               --context-ref "$RELEASE" $FORCE_LATEST \
-              --extra-flags "--build-arg PY_VER=3.11.14-slim-trixie --build-arg INCLUDE_CHROMIUM=false --tag $IMAGE_TAG" \
+              --extra-flags "$EXTRA_FLAGS" \
               $PLATFORM_ARG; then
               break
             fi

.github/workflows/enforce-single-migration-head.yml

@@ -33,6 +33,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.run_id }}
   cancel-in-progress: true
 
+permissions:
+  contents: read
+
 jobs:
   enforce-single-migration-head:
     runs-on: ubuntu-26.04

.github/workflows/frontend-bundle-size-nightly.yml

@@ -0,0 +1,136 @@
+name: Frontend bundle size (nightly baseline + analyzer)
+
+# Refreshes the bundle-size baseline that superset-frontend.yml's `bundle-size`
+# job compares PRs against, and publishes a browsable bundle-analyzer treemap
+# report of the same build. Deliberately NOT triggered on every push to
+# master: a day-old baseline/report is fine for catching relative
+# regressions on PRs and for browsing what's actually in the bundle, and
+# building the production bundle on every one of the many pushes master
+# gets per day would burn CI time for no benefit a nightly refresh doesn't
+# already cover.
+on:
+  schedule:
+    - cron: "0 6 * * *"
+  workflow_dispatch: {}
+
+concurrency:
+  group: ${{ github.workflow }}
+  cancel-in-progress: true
+
+env:
+  TAG: apache/superset:bundle-size-nightly-${{ github.run_id }}
+
+permissions:
+  contents: read
+
+jobs:
+  refresh-baseline:
+    runs-on: ubuntu-26.04
+    timeout-minutes: 30
+    env:
+      NETLIFY_SITE_ID: ${{ secrets.NETLIFY_BUNDLE_ANALYZER_SITE_ID }}
+    steps:
+      - name: "Checkout master"
+        uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
+        with:
+          persist-credentials: false
+          ref: master
+
+      - name: Build Docker Image
+        run: |
+          docker buildx build \
+            -t $TAG \
+            --cache-from=type=registry,ref=apache/superset-cache:3.11-slim-trixie \
+            --target superset-node-ci \
+            .
+
+      # Same cache the PR-time bundle-size job restores/writes -- webpack's
+      # persistent filesystem cache turns a warm production build into ~20s
+      # instead of several minutes. See superset-frontend.yml for the
+      # matching restore step and why it's keyed this way.
+      - name: Restore webpack build cache
+        uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
+        with:
+          path: superset-frontend/.temp_cache
+          key: >-
+            webpack-prod-cache-${{ hashFiles('superset-frontend/package-lock.json',
+            'superset-frontend/babel.config.js', 'superset-frontend/tsconfig.json',
+            'superset-frontend/webpack.config.js') }}
+
+      # Only ever pull the last recorded data point off the cache, keyed by
+      # run ID -- `restore-keys` prefix-matches the most recently created
+      # entry. Absent on the very first run ever; benchmark-action starts a
+      # fresh history in that case.
+      - name: Restore bundle size history
+        uses: actions/cache/restore@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
+        with:
+          path: bundle-size-history.json
+          key: bundle-size-history-${{ github.run_id }}
+          restore-keys: |
+            bundle-size-history-
+
+      # BUNDLE_ANALYZER rides along in the same build as BUNDLE_SIZE_STATS --
+      # they're independent env-gated additions in webpack.config.js (one
+      # sets `config.stats`, the other pushes plugins), so one production
+      # build produces both the numeric stats.json and the analyzer's
+      # report.html. Only report.html is mounted out, not
+      # BUNDLE_ANALYZER's sibling `statistics.html` sunburst -- that file is
+      # documented in webpack.config.js as routinely exceeding 100MB for
+      # this app (it's .gitignore'd for exactly that reason), too large to
+      # publish as a static site page.
+      - name: Build production bundle with stats and analyzer report
+        run: |
+          mkdir -p ${{ github.workspace }}/superset-frontend/bundle-stats
+          mkdir -p ${{ github.workspace }}/superset-frontend/.temp_cache
+          mkdir -p ${{ github.workspace }}/superset/static/assets
+          docker run \
+          -v ${{ github.workspace }}/superset-frontend/bundle-stats:/app/superset-frontend/bundle-stats \
+          -v ${{ github.workspace }}/superset-frontend/.temp_cache:/app/superset-frontend/.temp_cache \
+          -v ${{ github.workspace }}/superset/static/assets:/app/superset/static/assets \
+          --rm $TAG \
+          bash -c \
+          "npm i && BUNDLE_SIZE_STATS=true BUNDLE_ANALYZER=true npm run build -- --json=bundle-stats/stats.json"
+
+      - name: Summarize bundle size
+        run: |
+          node superset-frontend/scripts/bundle-size-summary.js \
+            superset-frontend/bundle-stats/stats.json > bundle-size-summary.json
+          rm -rf superset-frontend/bundle-stats
+
+      # No PR to comment on here, so comment-on-alert is off -- the job
+      # summary (summary-always) is the only surface for this run.
+      - name: Update bundle size baseline
+        uses: benchmark-action/github-action-benchmark@52576c92bccf6ac60c8223ec7eb2565637cae9ba # v1.22.1
+        with:
+          tool: customSmallerIsBetter
+          output-file-path: bundle-size-summary.json
+          external-data-json-path: bundle-size-history.json
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          fail-on-alert: false
+          summary-always: true
+
+      - name: Save bundle size history
+        uses: actions/cache/save@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
+        with:
+          path: bundle-size-history.json
+          key: bundle-size-history-${{ github.run_id }}
+
+      # Publishes the treemap to Netlify (the same host already used for
+      # superset-storybook.netlify.app and docs previews, reusing the
+      # existing NETLIFY_AUTH_TOKEN). Skipped until
+      # NETLIFY_BUNDLE_ANALYZER_SITE_ID exists -- create a new (free)
+      # Netlify site named superset-bundle-analyzer and add its site ID as
+      # that secret to turn this on; nothing else in this workflow depends
+      # on it.
+      - name: Publish bundle analyzer report to Netlify
+        if: ${{ env.NETLIFY_SITE_ID != '' }}
+        env:
+          NETLIFY_AUTH_TOKEN: ${{ secrets.NETLIFY_AUTH_TOKEN }}
+        run: |
+          mkdir -p netlify-publish
+          cp superset/static/assets/report.html netlify-publish/index.html
+          # zizmor: ignore[adhoc-packages] - netlify-cli is a one-shot CI deploy
+          # tool, not an application dependency; a global/npx install has no
+          # lockfile context. Version pinned above the floor set by other
+          # ad-hoc installs in this repo (bump deliberately when upgrading).
+          npx --yes netlify-cli@27.0.1 deploy --prod --dir=netlify-publish

.github/workflows/github-action-validator.yml

@@ -45,5 +45,8 @@ jobs:
       - name: Run Script
         run: bash .github/workflows/github-action-validator.sh
 
+      - name: Test docs-deploy freshness gate
+        run: bash .github/workflows/scripts/check-docs-deploy-freshness.test.sh
+
       - name: Check for security issues on GHA workflows
         uses: zizmorcore/zizmor-action@3dc1ecc9bcb9e94e9b2c709687979e1298497054 # v0.6.2

.github/workflows/issue-creation.yml

@@ -7,6 +7,9 @@ on:
   pull_request:
     types: [opened, edited]
 
+permissions:
+  contents: read
+
 jobs:
   superbot-orglabel:
     runs-on: ubuntu-slim

.github/workflows/labeler.yml

@@ -7,6 +7,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.run_id }}
   cancel-in-progress: true
 
+permissions:
+  contents: read
+
 jobs:
   labeler:
     permissions:

.github/workflows/latest-release-tag.yml

@@ -3,6 +3,9 @@ on:
   release:
     types: [published] # This makes it run only when a new released is published
 
+permissions:
+  contents: read
+
 jobs:
   latest-release:
     name: Add/update tag to new release

.github/workflows/mirror-service-images.yml

@@ -93,7 +93,7 @@ jobs:
           password: ${{ secrets.GITHUB_TOKEN }}
 
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@bb05f3f5519dd87d3ba754cc423b652a5edd6d2c # v4.2.0
+        uses: docker/setup-buildx-action@37fe631027851001ddb9b187196cc803df7f5f0e # v4.3.0
 
       - name: Copy image to GHCR
         env:

.github/workflows/pr-lint.yml

@@ -19,6 +19,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.run_id }}
   cancel-in-progress: false
 
+permissions:
+  contents: read
+
 jobs:
   lint-check:
     runs-on: ubuntu-slim

.github/workflows/scripts/check-docs-deploy-freshness.sh

@@ -0,0 +1,49 @@
+#!/bin/bash
+#
+# Licensed to the Apache Software Foundation (ASF) under one or more
+# contributor license agreements.  See the NOTICE file distributed with
+# this work for additional information regarding copyright ownership.
+# The ASF licenses this file to You under the Apache License, Version 2.0
+# (the "License"); you may not use this file except in compliance with
+# the License.  You may obtain a copy of the License at
+#
+#    http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# Shared freshness gate used by the Docs Deployment workflow
+# (superset-docs-deploy.yml) both up front (check-freshness) and again right
+# before the deploy step (recheck-freshness). Writes an output declaring
+# whether BUILD_SHA is still master's current tip, so a superseded run can
+# skip cleanly instead of racing (and clobbering, or being force-cancelled
+# by) a fresher run.
+#
+# Required env vars:
+#   BUILD_SHA     - the commit SHA this run is building
+#   REPO          - "owner/repo" to query, e.g. github.repository
+#   OUTPUT_NAME   - the GITHUB_OUTPUT key to write, e.g. "is-current"
+#   GITHUB_OUTPUT - path to append outputs to (set by the Actions runner)
+# Optional env vars:
+#   EVENT_NAME    - if "workflow_dispatch", bypasses the check and always
+#                   reports current, since a manual dispatch is a deliberate,
+#                   one-off action rather than something racing other triggers
+#   GH_TOKEN      - passed through to `gh`, needed to call the GitHub API
+
+set -euo pipefail
+
+if [ "${EVENT_NAME:-}" = "workflow_dispatch" ]; then
+  echo "${OUTPUT_NAME}=true" >>"$GITHUB_OUTPUT"
+  exit 0
+fi
+
+latest_sha="$(gh api "repos/${REPO}/commits/master" --jq .sha)"
+if [ "${latest_sha}" = "${BUILD_SHA}" ]; then
+  echo "${OUTPUT_NAME}=true" >>"$GITHUB_OUTPUT"
+else
+  echo "${OUTPUT_NAME}=false" >>"$GITHUB_OUTPUT"
+  echo "::notice::master has moved on to ${latest_sha} since ${BUILD_SHA} was triggered — skipping this stale run."
+fi

.github/workflows/scripts/check-docs-deploy-freshness.test.sh

@@ -0,0 +1,100 @@
+#!/bin/bash
+#
+# Licensed to the Apache Software Foundation (ASF) under one or more
+# contributor license agreements.  See the NOTICE file distributed with
+# this work for additional information regarding copyright ownership.
+# The ASF licenses this file to You under the Apache License, Version 2.0
+# (the "License"); you may not use this file except in compliance with
+# the License.  You may obtain a copy of the License at
+#
+#    http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# Exercises check-docs-deploy-freshness.sh against a stubbed `gh`, covering
+# the dispatch-bypass, current-tip and stale-tip branches so the output
+# contract (is-current / still-current) can't silently regress. Run
+# directly, no extra tooling required:
+#   bash .github/workflows/scripts/check-docs-deploy-freshness.test.sh
+
+set -euo pipefail
+
+script_dir="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+script_under_test="${script_dir}/check-docs-deploy-freshness.sh"
+
+failures=0
+
+# Runs the script under test with a stubbed `gh` reporting $1 as master's
+# latest sha, asserting that GITHUB_OUTPUT ends up containing exactly $4.
+run_case() {
+  local case_name="$1"
+  local latest_sha="$2"
+  local build_sha="$3"
+  local event_name="$4"
+  local expected_line="$5"
+
+  local workdir
+  workdir="$(mktemp -d)"
+  trap 'rm -rf "${workdir}"' RETURN
+
+  # Fake `gh` that just echoes back the requested "latest" sha regardless of
+  # arguments, so the script under test never touches the network.
+  cat >"${workdir}/gh" <<EOF
+#!/bin/bash
+echo '${latest_sha}'
+EOF
+  chmod +x "${workdir}/gh"
+
+  local output_file="${workdir}/github_output"
+  : >"${output_file}"
+
+  if PATH="${workdir}:${PATH}" \
+    GITHUB_OUTPUT="${output_file}" \
+    OUTPUT_NAME="is-current" \
+    REPO="apache/superset" \
+    BUILD_SHA="${build_sha}" \
+    EVENT_NAME="${event_name}" \
+    GH_TOKEN="fake-token" \
+    bash "${script_under_test}"; then
+    :
+  else
+    echo "FAIL (${case_name}): script exited non-zero"
+    failures=$((failures + 1))
+    return
+  fi
+
+  local actual
+  actual="$(cat "${output_file}")"
+  if [ "${actual}" = "${expected_line}" ]; then
+    echo "PASS (${case_name})"
+  else
+    echo "FAIL (${case_name}): expected '${expected_line}', got '${actual}'"
+    failures=$((failures + 1))
+  fi
+}
+
+# `gh` prints "should-not-be-called" for the dispatch case above the trick:
+# it's never actually invoked since the bypass short-circuits before the
+# `gh api` call, but the fake still needs a body.
+run_case "workflow_dispatch bypasses the check" \
+  "unused" "abc123" "workflow_dispatch" \
+  "is-current=true"
+
+run_case "build sha matches master's tip" \
+  "abc123" "abc123" "push" \
+  "is-current=true"
+
+run_case "build sha is stale" \
+  "def456" "abc123" "push" \
+  "is-current=false"
+
+if [ "${failures}" -gt 0 ]; then
+  echo "${failures} case(s) failed"
+  exit 1
+fi
+
+echo "All cases passed"

.github/workflows/showtime-cleanup.yml

@@ -17,6 +17,9 @@ env:
   GITHUB_ORG: ${{ github.repository_owner }}
   GITHUB_REPO: ${{ github.event.repository.name }}
 
+permissions:
+  contents: read
+
 jobs:
   cleanup-expired:
     name: Clean up expired showtime environments

.github/workflows/showtime-trigger.yml

@@ -26,6 +26,9 @@ env:
   GITHUB_REPO: ${{ github.event.repository.name }}
   GITHUB_ACTOR: ${{ github.actor }}
 
+permissions:
+  contents: read
+
 jobs:
   sync:
     name: 🎪 Sync PR to desired state

.github/workflows/superset-docs-deploy.yml

@@ -18,16 +18,6 @@ on:
 
   workflow_dispatch: {}
 
-# Serialize deploys: the action pushes to apache/superset-site without
-# rebasing, so concurrent runs race on the final push and the loser fails
-# with `! [rejected] asf-site -> asf-site (fetch first)`. Cancel any
-# in-progress run as soon as a newer one starts — the destination repo
-# isn't touched until the final push step, so canceling mid-build is safe,
-# and the freshest content always wins.
-concurrency:
-  group: docs-deploy-asf-site
-  cancel-in-progress: true
-
 permissions:
   contents: read
   actions: read
@@ -48,17 +38,69 @@ jobs:
 
         env:
           SUPERSET_SITE_BUILD: ${{ (secrets.SUPERSET_SITE_BUILD != '' && secrets.SUPERSET_SITE_BUILD != '') || '' }}
+
+  # Master gets frequent, sometimes bursty pushes, and each one can trigger a
+  # deploy attempt. Rather than let every superseded attempt get force-killed
+  # by the build-deploy concurrency group below (which shows up as a
+  # `cancelled` — i.e. red/failing-looking — check on that commit), have each
+  # run check up front whether it's still building master's current tip and,
+  # if not, skip cleanly. Deliberately outside the docs-deploy-asf-site
+  # concurrency group so it runs immediately for every trigger without
+  # blocking or being blocked by anything.
+  check-freshness:
+    runs-on: ubuntu-26.04
+    outputs:
+      is-current: ${{ steps.check.outputs.is-current }}
+    steps:
+      # Sparse checkout: this job's only job is to be fast, so it fetches
+      # nothing but the freshness-check script itself.
+      - name: Checkout freshness-check script
+        uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
+        with:
+          persist-credentials: false
+          sparse-checkout: |
+            .github/workflows/scripts
+          sparse-checkout-cone-mode: false
+      - name: "Check whether this is still master's current commit"
+        id: check
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          BUILD_SHA: ${{ github.event.workflow_run.head_sha || github.sha }}
+          EVENT_NAME: ${{ github.event_name }}
+          REPO: ${{ github.repository }}
+          OUTPUT_NAME: is-current
+        run: .github/workflows/scripts/check-docs-deploy-freshness.sh
+
   build-deploy:
-    needs: config
+    needs: [config, check-freshness]
+    # Only the run for master's current tip proceeds; anything superseded
+    # already skipped at check-freshness above instead of landing here.
     # For workflow_run triggers, only deploy when the triggering run originated
     # from this repository (not a fork), ensuring the checked-out code and any
     # local actions executed with deploy credentials are trusted.
     if: >-
       needs.config.outputs.has-secrets &&
+      needs.check-freshness.outputs.is-current == 'true' &&
       (github.event_name != 'workflow_run' ||
        github.event.workflow_run.head_repository.full_name == github.repository)
     name: Build & Deploy
     runs-on: ubuntu-26.04
+    # Serialize deploys: the action pushes to apache/superset-site without
+    # rebasing, so concurrent runs race on the final push and the loser fails
+    # with `! [rejected] asf-site -> asf-site (fetch first)`. Queue instead of
+    # canceling: a run that already passed check-freshness can still be
+    # sitting in the queue for a runner when a newer run starts and finishes
+    # first. cancel-in-progress would let that stale, queued run kill the
+    # newer run's in-progress deploy the moment it's finally scheduled, and
+    # then skip itself at the re-check below — losing the deploy entirely.
+    # Queuing means the stale run just waits its turn and then no-ops at the
+    # re-check, so the fresher content that already deployed is never
+    # clobbered or lost. The check-freshness gate above means it should be
+    # rare for more than one run to reach this point, so the queue stays
+    # short in practice.
+    concurrency:
+      group: docs-deploy-asf-site
+      cancel-in-progress: false
     steps:
       - uses: Kesin11/actions-timeline@57fc93f20c6da7fbc14063c6d24a2a5627c799ad # v3.2.0
         with:
@@ -99,7 +141,7 @@ jobs:
           yarn install --check-cache
       - name: Download database diagnostics (if triggered by integration tests)
         if: github.event_name == 'workflow_run' && github.event.workflow_run.conclusion == 'success'
-        uses: dawidd6/action-download-artifact@b6e2e70617bc3265edd6dab6c906732b2f1ae151 # v21
+        uses: dawidd6/action-download-artifact@d63b86af1b34672e53c440b1b83979861906bad7 # v24
         continue-on-error: true
         with:
           workflow: superset-python-integrationtest.yml
@@ -108,7 +150,7 @@ jobs:
           path: docs/src/data/
       - name: Try to download latest diagnostics (for push/dispatch triggers)
         if: github.event_name != 'workflow_run'
-        uses: dawidd6/action-download-artifact@b6e2e70617bc3265edd6dab6c906732b2f1ae151 # v21
+        uses: dawidd6/action-download-artifact@d63b86af1b34672e53c440b1b83979861906bad7 # v24
         continue-on-error: true
         with:
           workflow: superset-python-integrationtest.yml
@@ -130,7 +172,24 @@ jobs:
         working-directory: docs
         run: |
           yarn build
+      # The check-freshness job above narrows the window but doesn't close it: an
+      # older run can observe is-current=true, then sit through this build while a
+      # newer run's own freshness check also passes and it deploys and finishes
+      # first. If this (stale) run then wins entry into the concurrency group, it
+      # would overwrite the newer content that already deployed. Re-check right
+      # before the one step that actually mutates superset-site, so a stale run
+      # skips deploying instead of clobbering a fresher one that already ran.
+      - name: "Re-check freshness immediately before deploying"
+        id: recheck-freshness
+        if: github.event_name != 'workflow_dispatch'
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          BUILD_SHA: ${{ github.event.workflow_run.head_sha || github.sha }}
+          REPO: ${{ github.repository }}
+          OUTPUT_NAME: still-current
+        run: .github/workflows/scripts/check-docs-deploy-freshness.sh
       - name: deploy docs
+        if: github.event_name == 'workflow_dispatch' || steps.recheck-freshness.outputs.still-current == 'true'
         uses: ./.github/actions/github-action-push-to-another-repository
         env:
           API_TOKEN_GITHUB: ${{ secrets.SUPERSET_SITE_BUILD }}

.github/workflows/superset-docs-verify.yml

@@ -140,7 +140,7 @@ jobs:
         run: |
           yarn install --check-cache
       - name: Download database diagnostics from integration tests
-        uses: dawidd6/action-download-artifact@b6e2e70617bc3265edd6dab6c906732b2f1ae151 # v21
+        uses: dawidd6/action-download-artifact@d63b86af1b34672e53c440b1b83979861906bad7 # v24
         with:
           workflow: superset-python-integrationtest.yml
           run_id: ${{ github.event.workflow_run.id }}

.github/workflows/superset-e2e.yml

@@ -26,6 +26,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.run_id }}
   cancel-in-progress: true
 
+permissions:
+  contents: read
+
 jobs:
   changes:
     runs-on: ubuntu-26.04
@@ -75,14 +78,14 @@ jobs:
       USE_DASHBOARD: ${{ github.event.inputs.use_dashboard == 'true' || 'false' }}
     services:
       postgres:
-        image: postgres:17-alpine
+        image: ghcr.io/apache/superset/ci/postgres:17-alpine
         env:
           POSTGRES_USER: superset
           POSTGRES_PASSWORD: superset
         ports:
           - 15432:5432
       redis:
-        image: redis:7-alpine
+        image: ghcr.io/apache/superset/ci/redis:7-alpine
         ports:
           - 16379:6379
     steps:
@@ -183,14 +186,14 @@ jobs:
       GITHUB_TOKEN: ${{ github.token }}
     services:
       postgres:
-        image: postgres:17-alpine
+        image: ghcr.io/apache/superset/ci/postgres:17-alpine
         env:
           POSTGRES_USER: superset
           POSTGRES_PASSWORD: superset
         ports:
           - 15432:5432
       redis:
-        image: redis:7-alpine
+        image: ghcr.io/apache/superset/ci/redis:7-alpine
         ports:
           - 16379:6379
     steps:

.github/workflows/superset-frontend.yml

@@ -212,3 +212,100 @@ jobs:
       - uses: Kesin11/actions-timeline@57fc93f20c6da7fbc14063c6d24a2a5627c799ad # v3.2.0
         with:
           expand-composite-actions: true
+
+  # Compares a PR's own bundle size against the last nightly-recorded
+  # baseline (see frontend-bundle-size-nightly.yml, which owns actually
+  # persisting new baselines). PR-only: a push to master doesn't need this
+  # check re-run against itself, and re-persisting the baseline on every
+  # push to master -- which happens many times a day -- would burn a full
+  # production build for no benefit nightly refresh doesn't already cover.
+  bundle-size:
+    needs: frontend-build
+    if: needs.frontend-build.outputs.should-run == 'true' && github.event_name == 'pull_request'
+    runs-on: ubuntu-26.04
+    timeout-minutes: 15
+    permissions:
+      contents: read
+      pull-requests: write
+    steps:
+      - name: Checkout Code
+        uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
+        with:
+          persist-credentials: false
+          ref: ${{ github.event_name == 'pull_request' && github.event.pull_request.head.sha || github.sha }}
+
+      - name: Download Docker Image Artifact
+        uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8
+        with:
+          name: docker-image
+
+      - name: Load Docker Image
+        run: |
+          zstd -d < docker-image.tar.zst | docker load
+
+      # webpack's persistent filesystem cache (superset-frontend/webpack.config.js)
+      # turns a warm production build into ~20s instead of several minutes,
+      # but GH-hosted runners are fresh VMs with nothing carried over between
+      # jobs -- without restoring it explicitly, every single PR would pay
+      # the full cold-build cost. Keyed on the same files webpack's own
+      # `buildDependencies` invalidates on, so a stale cache is never used.
+      - name: Restore webpack build cache
+        uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
+        with:
+          path: superset-frontend/.temp_cache
+          key: >-
+            webpack-prod-cache-${{ hashFiles('superset-frontend/package-lock.json',
+            'superset-frontend/babel.config.js', 'superset-frontend/tsconfig.json',
+            'superset-frontend/webpack.config.js') }}
+
+      # Only ever pull the last recorded data point off the cache, keyed by
+      # run ID -- `restore-keys` prefix-matches the most recently created
+      # entry, which is always the latest nightly run. Absent before the
+      # first nightly run ever happens; benchmark-action starts a fresh
+      # history in that case.
+      - name: Restore bundle size history
+        uses: actions/cache/restore@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
+        with:
+          path: bundle-size-history.json
+          key: bundle-size-history-${{ github.run_id }}
+          restore-keys: |
+            bundle-size-history-
+
+      - name: Build production bundle with stats
+        run: |
+          mkdir -p ${{ github.workspace }}/superset-frontend/bundle-stats
+          mkdir -p ${{ github.workspace }}/superset-frontend/.temp_cache
+          docker run \
+          -v ${{ github.workspace }}/superset-frontend/bundle-stats:/app/superset-frontend/bundle-stats \
+          -v ${{ github.workspace }}/superset-frontend/.temp_cache:/app/superset-frontend/.temp_cache \
+          --rm $TAG \
+          bash -c \
+          "npm i && BUNDLE_SIZE_STATS=true npm run build -- --json=bundle-stats/stats.json"
+
+      - name: Summarize bundle size
+        run: |
+          node superset-frontend/scripts/bundle-size-summary.js \
+            superset-frontend/bundle-stats/stats.json > bundle-size-summary.json
+          rm -rf superset-frontend/bundle-stats
+
+      # Comparison + alert only -- this job never persists. See
+      # frontend-bundle-size-nightly.yml for why.
+      #
+      # comment-on-alert is gated to same-repo PRs: on a fork PR,
+      # GITHUB_TOKEN is forced read-only regardless of the `permissions`
+      # block above, so once the alert threshold is crossed the action's
+      # `pulls.createReview` call 403s. That error isn't gated by
+      # fail-on-alert (which only governs the deliberate alert-threshold
+      # failure) -- it propagates and fails the job outright. Fork PRs
+      # still get the comparison via the job summary (summary-always).
+      - name: Compare bundle size against nightly baseline
+        uses: benchmark-action/github-action-benchmark@52576c92bccf6ac60c8223ec7eb2565637cae9ba # v1.22.1
+        with:
+          tool: customSmallerIsBetter
+          output-file-path: bundle-size-summary.json
+          external-data-json-path: bundle-size-history.json
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          comment-on-alert: ${{ github.event.pull_request.head.repo.full_name == github.repository }}
+          alert-threshold: "110%"
+          fail-on-alert: false
+          summary-always: true

.github/workflows/superset-helm-release.yml

@@ -27,6 +27,9 @@ concurrency:
   group: helm-release
   cancel-in-progress: false
 
+permissions:
+  contents: read
+
 jobs:
   release:
     runs-on: ubuntu-26.04

.github/workflows/superset-playwright.yml

@@ -22,6 +22,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.run_id }}
   cancel-in-progress: true
 
+permissions:
+  contents: read
+
 jobs:
   changes:
     runs-on: ubuntu-26.04

.github/workflows/superset-python-integrationtest.yml

@@ -53,9 +53,7 @@ jobs:
         mysql+mysqldb://superset:superset@127.0.0.1:13306/superset?charset=utf8mb4&binary_prefix=true
     services:
       mysql:
-        image: mysql:8.0
-        # Authenticated pulls use our higher Docker Hub rate limit. Empty on
-        # fork PRs (secrets unavailable) -> runner falls back to anonymous.
+        image: ghcr.io/apache/superset/ci/mysql:8.0
         env:
           MYSQL_ROOT_PASSWORD: root
         ports:
@@ -66,7 +64,7 @@ jobs:
           --health-timeout=5s
           --health-retries=5
       redis:
-        image: redis:7-alpine
+        image: ghcr.io/apache/superset/ci/redis:7-alpine
         options: --entrypoint redis-server
         ports:
           - 16379:6379
@@ -143,7 +141,7 @@ jobs:
       SUPERSET__SQLALCHEMY_DATABASE_URI: postgresql+psycopg2://superset:superset@127.0.0.1:15432/superset
     services:
       postgres:
-        image: postgres:17-alpine
+        image: ghcr.io/apache/superset/ci/postgres:17-alpine
         env:
           POSTGRES_USER: superset
           POSTGRES_PASSWORD: superset
@@ -152,7 +150,7 @@ jobs:
           # GitHub action runner's default installations
           - 15432:5432
       redis:
-        image: redis:7-alpine
+        image: ghcr.io/apache/superset/ci/redis:7-alpine
         ports:
           - 16379:6379
     steps:
@@ -202,7 +200,7 @@ jobs:
         sqlite:///${{ github.workspace }}/.temp/examples.db?check_same_thread=true
     services:
       redis:
-        image: redis:7-alpine
+        image: ghcr.io/apache/superset/ci/redis:7-alpine
         ports:
           - 16379:6379
     steps:

.github/workflows/superset-python-presto-hive.yml

@@ -14,6 +14,9 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.run_id }}
   cancel-in-progress: true
 
+permissions:
+  contents: read
+
 jobs:
   changes:
     runs-on: ubuntu-26.04
@@ -49,7 +52,7 @@ jobs:
       SUPERSET__SQLALCHEMY_EXAMPLES_URI: presto://localhost:15433/memory/default
     services:
       postgres:
-        image: postgres:17-alpine
+        image: ghcr.io/apache/superset/ci/postgres:17-alpine
         env:
           POSTGRES_USER: superset
           POSTGRES_PASSWORD: superset
@@ -58,7 +61,7 @@ jobs:
           # GitHub action runner's default installations
           - 15432:5432
       presto:
-        image: starburstdata/presto:350-e.6
+        image: ghcr.io/apache/superset/ci/presto:350-e.6
         env:
           POSTGRES_USER: superset
           POSTGRES_PASSWORD: superset
@@ -67,7 +70,7 @@ jobs:
           # GitHub action runner's default installations
           - 15433:8080
       redis:
-        image: redis:7-alpine
+        image: ghcr.io/apache/superset/ci/redis:7-alpine
         ports:
           - 16379:6379
     steps:
@@ -113,7 +116,7 @@ jobs:
       UPLOAD_FOLDER: /tmp/.superset/uploads/
     services:
       postgres:
-        image: postgres:17-alpine
+        image: ghcr.io/apache/superset/ci/postgres:17-alpine
         env:
           POSTGRES_USER: superset
           POSTGRES_PASSWORD: superset
@@ -122,7 +125,7 @@ jobs:
           # GitHub action runner's default installations
           - 15432:5432
       redis:
-        image: redis:7-alpine
+        image: ghcr.io/apache/superset/ci/redis:7-alpine
         ports:
           - 16379:6379
     steps:

.github/workflows/superset-python-unittest.yml

@@ -53,12 +53,6 @@ jobs:
         python-version: ${{ github.event_name == 'pull_request' && fromJSON('["current"]') || fromJSON('["current", "next"]') }}
     env:
       PYTHONPATH: ${{ github.workspace }}
-      # Promotes the SQLAlchemy 2.0 deprecation warnings already locked in as
-      # errors via pytest.ini's `filterwarnings` to actually run in CI, so a
-      # regression on those fails the build instead of relying on a
-      # contributor remembering to set this locally. See the migration
-      # battleplan: https://github.com/apache/superset/discussions/40273
-      SQLALCHEMY_WARN_20: "1"
     steps:
       - name: "Checkout ${{ github.ref }} ( ${{ github.sha }} )"
         uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
@@ -75,13 +69,18 @@ jobs:
           SUPERSET_SECRET_KEY: not-a-secret
         run: |
           pytest --durations-min=0.5 --cov-report= --cov=superset ./tests/common ./tests/unit_tests --cache-clear --maxfail=50 --junit-xml=test-results/junit-unit.xml
+      # COVERAGE_FILE keeps these scoped gates off the default .coverage that
+      # the step above wrote. pytest-cov starts a fresh data file per run, so
+      # without it the last gate replaces the full-suite data and the report
+      # uploaded below contains only that gate's subtree -- every other file
+      # then reads as uncovered, and patch coverage on an unrelated PR is 0%.
       - name: Python 100% coverage unit tests
         env:
           SUPERSET_TESTENV: true
           SUPERSET_SECRET_KEY: not-a-secret
         run: |
-          pytest --durations-min=0.5 --cov=superset/sql/ ./tests/unit_tests/sql/ --cache-clear --cov-fail-under=100 --junit-xml=test-results/junit-sql-coverage.xml
-          pytest --durations-min=0.5 --cov=superset/semantic_layers/ ./tests/unit_tests/semantic_layers/ --cache-clear --cov-fail-under=100 --junit-xml=test-results/junit-semantic-layers-coverage.xml
+          COVERAGE_FILE="${RUNNER_TEMP}/.coverage.sql" pytest --durations-min=0.5 --cov=superset/sql/ ./tests/unit_tests/sql/ --cache-clear --cov-fail-under=100 --junit-xml=test-results/junit-sql-coverage.xml
+          COVERAGE_FILE="${RUNNER_TEMP}/.coverage.semantic_layers" pytest --durations-min=0.5 --cov=superset/semantic_layers/ ./tests/unit_tests/semantic_layers/ --cache-clear --cov-fail-under=100 --junit-xml=test-results/junit-semantic-layers-coverage.xml
       - name: Upload code coverage
         uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f # v7.0.0
         with:

.github/workflows/supersetbot.yml

@@ -13,6 +13,9 @@ on:
         required: true
         type: string
 
+permissions:
+  contents: read
+
 jobs:
   supersetbot:
     runs-on: ubuntu-26.04

.github/workflows/testcontainers.yml

@@ -0,0 +1,188 @@
+# db_engine_specs tests against real databases (testcontainers)
+name: Testcontainers
+
+# Spins up real Docker containers (see tests/testcontainers/ for the current
+# dialect list) via testcontainers-python, which catches real dialect/driver
+# regressions -- the kind mocked db_engine_specs unit tests structurally
+# cannot, e.g. apache/superset#42899 (Trino emitting OFFSET before LIMIT).
+# Runs on a nightly cron (catches drift from a driver's own releases, not
+# just from Superset's changes) and on pull_request, scoped via `paths` to
+# only PRs that actually touch this test suite or the workflow itself, so
+# unrelated PRs across the repo are never affected.
+#
+# A matrix entry can set `nightly_only: true` to run only on the cron (or a
+# manual workflow_dispatch), never on pull_request -- for a dialect whose
+# image is too heavy (a multi-service cluster, a many-GB image, a slow
+# licensed installer) to justify adding its wall-clock/resource cost to
+# every PR that merely touches this suite. Omit the field entirely for a
+# normal dialect; it isn't nightly-only by default.
+permissions:
+  contents: read
+
+on:
+  schedule:
+    - cron: "0 5 * * *"
+  workflow_dispatch: {}
+  pull_request:
+    paths:
+      - ".github/workflows/testcontainers.yml"
+      - "tests/testcontainers/**"
+      - "superset/db_engine_specs/**"
+      - "pyproject.toml"
+      - "requirements/development.in"
+      - "requirements/development.txt"
+
+concurrency:
+  # Scoped by ref, not just workflow name -- otherwise every PR run and the
+  # nightly cron share one group, and starting the workflow on another PR
+  # (or the nightly firing mid-PR-run) cancels an unrelated in-progress run.
+  group: ${{ github.workflow }}-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  testcontainers:
+    runs-on: ubuntu-26.04
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          # One job per dialect rather than one job for the whole suite: a
+          # single slow container would otherwise inflate the wall-clock
+          # time for every dialect, not just its own. Running in parallel
+          # means the suite's total time is bounded by the slowest dialect,
+          # not the sum of all of them. Db2's first-boot init is documented
+          # upstream as notably slow (a real instance bring-up, not just a
+          # process start) and untested locally here (no arm64 image), so
+          # it gets a wider timeout margin than the rest until real CI data
+          # says otherwise.
+          - dialect: cockroachdb
+            timeout: 10
+          - dialect: crate
+            timeout: 10
+          - dialect: trino
+            timeout: 10
+          - dialect: mssql
+            timeout: 10
+          - dialect: elasticsearch
+            timeout: 10
+          - dialect: oracle
+            timeout: 15
+          - dialect: db2
+            timeout: 25
+          - dialect: mariadb
+            timeout: 10
+          - dialect: timescaledb
+            timeout: 10
+          - dialect: yugabytedb
+            timeout: 10
+          - dialect: monetdb
+            timeout: 10
+          - dialect: mongodb
+            timeout: 10
+          - dialect: postgres
+            timeout: 10
+          - dialect: mysql
+            timeout: 10
+          - dialect: clickhouse
+            timeout: 10
+          # StarRocks' allin1-ubuntu image brings up both FE and BE in one
+          # container, which is a heavier bring-up than a single-process
+          # database -- wider margin until real CI data says otherwise.
+          - dialect: starrocks
+            timeout: 15
+          - dialect: databend
+            timeout: 10
+          - dialect: risingwave
+            timeout: 10
+          - dialect: firebird
+            timeout: 10
+          - dialect: ydb
+            timeout: 10
+          # OceanBase bootstraps a distributed-style cluster even in
+          # single-node MODE=MINI -- too heavy for every PR's CI budget, so
+          # it runs on the nightly cron / manual dispatch only.
+          - dialect: oceanbase
+            timeout: 20
+            nightly_only: true
+    timeout-minutes: ${{ matrix.timeout }}
+    env:
+      PYTHONPATH: ${{ github.workspace }}
+      SUPERSET_TESTENV: true
+      SUPERSET_SECRET_KEY: not-a-secret
+      # This job's matrix installs exactly one dialect's testcontainers
+      # driver for exactly this job, so treat that driver as required: a
+      # broken/missing import should fail the job, not silently skip to a
+      # misleadingly green, zero-tests-run result. See _driver.py.
+      SUPERSET_TESTCONTAINERS_STRICT: true
+    steps:
+      - name: Checkout
+        uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7.0.1
+        with:
+          persist-credentials: false
+      - name: Setup Python
+        uses: ./.github/actions/setup-backend/
+        with:
+          python-version: current
+      - name: Install db2 driver (ibm-db-sa)
+        # ibm-db (the db2 DBAPI) ships no Linux arm64 wheel, so it's kept out
+        # of the baseline dev install (requirements/development.in) to avoid
+        # breaking the multi-platform dev Docker image build. Install it here
+        # instead, only for this leg of the matrix.
+        if: matrix.dialect == 'db2'
+        run: uv pip install --system -e .[db2]
+      - name: Install oceanbase driver (oceanbase_py)
+        # oceanbase_py pins sqlalchemy-utils>=0.38.3,<0.39, which conflicts
+        # outright with Superset's own sqlalchemy-utils==0.42.1 pin -- kept
+        # out of the baseline dev install for the same reason as db2 above.
+        # Installed as its own standalone package (not via `-e .[oceanbase]`)
+        # so --no-deps only skips *oceanbase_py's* dependencies -- applied
+        # to `-e .[oceanbase]` instead, --no-deps blocks pip from installing
+        # anything the extras marker pulls in, including oceanbase_py
+        # itself, which "succeeds" without actually installing it
+        # (confirmed on real CI: the install step reported success, but the
+        # module was still missing). This job only needs oceanbase_py's
+        # dialect module importable, not its sqlalchemy-utils dependency
+        # satisfied, since nothing here calls into it.
+        if: matrix.dialect == 'oceanbase'
+        run: uv pip install --system --no-deps "oceanbase_py>=0.0.1.2"
+      - name: Install Firebird client library (libfbclient2)
+        # sqlalchemy-firebird's driver (firebird-driver) is a pure-Python
+        # ctypes wrapper (its wheel is py3-none-any) that dynamically loads
+        # the native Firebird client library from the host at import time
+        # -- it doesn't bundle that library itself, so it has to come from
+        # the system package manager, only for this leg of the matrix.
+        if: matrix.dialect == 'firebird'
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y libfbclient2
+      - name: Run testcontainers db_engine_specs tests (${{ matrix.dialect }})
+        # A job-level `if:` can't reference `matrix` (only github/inputs/
+        # needs/vars are available there), so the nightly_only skip has to
+        # live on the step instead. A dialect without `nightly_only` set
+        # evaluates the left side true (unset is null, and `null != true`
+        # is true) and always runs; one WITH it set only runs on the cron
+        # or a manual dispatch, never on pull_request.
+        if: >-
+          matrix.nightly_only != true ||
+          github.event_name == 'schedule' ||
+          github.event_name == 'workflow_dispatch'
+        run: |
+          pytest --durations-min=2 -v -m testcontainers \
+            ./tests/testcontainers/db_engine_specs/test_${{ matrix.dialect }}.py \
+            --junit-xml=test-results/junit-testcontainers-${{ matrix.dialect }}.xml
+      - name: Upload JUnit test results
+        if: always()
+        uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7
+        with:
+          name: junit-results-testcontainers-${{ matrix.dialect }}
+          path: test-results/
+          retention-days: 7
+
+  actions-timeline:
+    needs: [testcontainers]
+    if: always()
+    runs-on: ubuntu-26.04
+    permissions:
+      actions: read
+    steps:
+      - uses: Kesin11/actions-timeline@57fc93f20c6da7fbc14063c6d24a2a5627c799ad # v3.2.0

.github/workflows/welcome-new-users.yml

@@ -5,6 +5,9 @@ on:
   pull_request_target:
     types: [opened]
 
+permissions:
+  contents: read
+
 jobs:
   welcome:
     runs-on: ubuntu-slim

README.md

@@ -31,6 +31,8 @@ under the License.
 [![Open PRs](https://img.shields.io/github/issues-pr/apache/superset)](https://github.com/apache/superset/pulls)
 [![Get on Slack](https://img.shields.io/badge/slack-join-orange.svg)](https://bit.ly/join-superset-slack)
 [![Documentation](https://img.shields.io/badge/docs-apache.org-blue.svg)](https://superset.apache.org)
+[![Storybook](https://img.shields.io/badge/storybook-live-ff4785.svg)](https://superset-storybook.netlify.app)
+[![Bundle Analyzer](https://img.shields.io/badge/bundle%20analyzer-nightly-8dd6f9.svg)](https://superset-bundle-analyzer.netlify.app)
 
 <picture width="500">
   <source

UPDATING.md

@@ -24,6 +24,32 @@ assists people when migrating to a new version.
 
 ## Next
 
+- `SAMPLES_ROW_LIMIT` is now the default for `/datasource/samples` requests without a valid explicit `per_page`, rather than a hard per-request ceiling; explicit limits are honored up to the existing global row-limit ceiling, matching `/chart/data` SAMPLES requests.
+- The `cockroachdb` extra (`pip install apache-superset[cockroachdb]`) now installs `sqlalchemy-cockroachdb` instead of the abandoned `cockroachdb` package, whose SQLAlchemy dialect could not be imported under SQLAlchemy 2.0. Existing environments with the old package installed must `pip uninstall cockroachdb` before reinstalling the extra -- both packages register the same `cockroachdb` SQLAlchemy dialect entry point, so leaving the old one in place can still load the abandoned implementation.
+
+### MCP tool results preserve stored string values
+
+Structured MCP tool results no longer add `<UNTRUSTED-CONTENT>` wrappers or
+rewrite delimiter-looking text inside string fields. Tool-result content remains
+user-controlled data, but clients must convey that trust boundary outside domain
+values instead of recognizing or removing marker strings.
+
+Clients that handled the former delimiter convention should stop stripping marker
+text: the same text can be legitimate stored content. Response models and content
+types are unchanged, and no metadata-database migration is required. Automated
+read-modify-write workflows should be paused or pinned away from older instances
+until every serving instance is upgraded; a mixed-version response has no reliable
+signal that tells a client whether its text is decorated. Redis-backed MCP response
+caches use a new internal namespace after the upgrade, so upgraded instances do not
+reuse older cached results.
+
+Values that a client already wrote back with presentation wrappers cannot be
+distinguished safely from intentional content. Operators should review possible
+`<UNTRUSTED-CONTENT>` / `</UNTRUSTED-CONTENT>` wrappers and
+`[ESCAPED-UNTRUSTED-CONTENT-OPEN]` /
+`[ESCAPED-UNTRUSTED-CONTENT-CLOSE]` substitutions rather than applying an automatic
+marker-removal migration.
+
 ### OAuth2 database callback metrics include their outcome
 
 The unqualified `DatabaseRestApi.oauth2` StatsD counter has been replaced with
@@ -31,7 +57,10 @@ The unqualified `DatabaseRestApi.oauth2` StatsD counter has been replaced with
 `DatabaseRestApi.oauth2.error`. Update monitoring rules and dashboards that consume
 the old counter to use the outcome-specific replacements.
 
+- [42930](https://github.com/apache/superset/pull/42930): Dataset import data-URI fetches no longer honor an HTTP(S) proxy when `DATASET_IMPORT_ALLOW_INTERNAL_DATA_URLS` is `False` (the default): the connection is now made directly to the destination so the peer-address check validates the real target instead of a proxy's. Deployments that require an egress proxy to reach legitimate external data URLs for dataset import should set `DATASET_IMPORT_ALLOW_INTERNAL_DATA_URLS = True` or otherwise ensure those URLs resolve without one.
 - [42935](https://github.com/apache/superset/pull/42935): The MCP service now refuses to start (`MCPAuthConfigError`) when `MCP_JWT_ISSUER` trusts more than one issuer and no `MCP_USER_RESOLVER` is configured, instead of only logging a warning. This was already a documented misconfiguration (the default resolver isn't issuer-scoped, so distinct trusted issuers minting the same username/email would resolve to the same Superset user); deployments trusting multiple issuers must configure an `MCP_USER_RESOLVER` that derives its identity from the token's `iss` claim before upgrading. Single-issuer deployments are unaffected.
+- [42429](https://github.com/apache/superset/pull/42429): The Country Map chart's Iran GeoJSON now gives Alborz province its own ISO 3166-2 code, `IR-32`, instead of `IR-30`. `ISO` is the join key used to color/filter provinces on this chart, so any existing dataset keyed on `IR-30` for Alborz will silently stop matching after upgrading; re-key that data to `IR-32`.
+- [43388](https://github.com/apache/superset/pull/43388): The MCP service now refuses to start (`MCPAuthConfigError`) if `MCP_DEV_USERNAME` and `MCP_AUTH_ENABLED = True` are both set, and separately if `MCP_AUTH_ENABLED = True` but no usable JWT key material is configured (RSA key/JWKS, or an explicit `MCP_JWT_SECRET` for HMAC) — both previously started with authentication silently weaker than configured. Deployments combining a dev-mode username with JWT auth enabled, or enabling JWT auth without key material, must pick one before upgrading: unset `MCP_DEV_USERNAME` for a real auth deployment, or unset `MCP_AUTH_ENABLED` (or configure the key material) for a dev-mode one. Response caching (`MCP_CACHE_CONFIG["enabled"] = True`) now also excludes every tool with a side effect by default, not only a partial list, so a previously-cached mutating tool call is no longer served from cache; no config change is needed to pick this up.
 - [42393](https://github.com/apache/superset/pull/42393): Exported dataset YAML now carries a `uuid` for each metric and column so that custom folder assignments (which reference metrics/columns by UUID) survive an import into another workspace. This affects any export bundle that contains datasets, not just a dataset export: chart, dashboard, database and full-asset exports all embed the same dataset YAML, so a dashboard exported from this release also fails to import into an older one even though no dataset was exported directly. As with `folders` and `currency_code_column`, the affected `datasets/` files fail schema validation (`Unknown field: uuid`) when imported into Superset releases that predate this change; regenerate or hand-edit exports for older targets in mixed-version fleets.
 - [42300](https://github.com/apache/superset/pull/42300): Timeseries charts (line/area/bar) with a Y-axis bound in effect — either an explicit `yAxisBounds` or one derived from `truncateYAxis` — now clamp out-of-range data points to that bound instead of letting ECharts drop the point (and the line segments around it) entirely. Any existing chart with a configured Y-axis bound and data outside it will look different after upgrading: a gap becomes a point pinned to the boundary. The clamp also rewrites the value ECharts reads for that point's tooltip and data label, so the displayed value is the bound rather than the true observation.
 - [42087](https://github.com/apache/superset/pull/42087): Stored calculated-column and metric expressions are validated when a query is built, under the same sub-query policy already applied to adhoc expressions. Previously only the dataset update path checked them on save, so expressions written by v1 import, by dataset duplication, or before that check existed were never validated. Since `ALLOW_ADHOC_SUBQUERY` defaults to `False` (see [19242](https://github.com/apache/superset/pull/19242)), a dataset whose stored expression contains a sub-query works before upgrading and afterwards fails at chart render with `Custom SQL fields cannot contain sub-queries.` There is no migration step, and the error does not name the offending dataset column, so audit stored expressions before upgrading: either rewrite them without the sub-query, or set `ALLOW_ADHOC_SUBQUERY = True` to keep the previous behaviour for both stored and adhoc expressions.
@@ -99,6 +128,23 @@ dialect; each package's constraint in `pyproject.toml` documents why.
 No application-level configuration changes are required for deployments
 that don't touch SQLAlchemy directly.
 
+### New metric aggregates: MEDIAN, Sample Standard Deviation, Sample Variance
+
+`MEDIAN`, `STDDEV_SAMP`, and `VAR_SAMP` are now available anywhere a metric
+aggregate is chosen (every chart type, SQL Lab, MCP), not only in Pivot
+Table's controls. Support is opt-in per database engine *spec class*,
+verified against a live instance before being enabled: Postgres, MySQL
+(`STDDEV_SAMP`/`VAR_SAMP` only, no `MEDIAN`), DuckDB, and Redshift (inherits
+Postgres's support, not yet separately verified) ship enabled in this
+release. Engine specs that subclass one of those (e.g. MariaDB, Aurora
+MySQL/Postgres, TimescaleDB) inherit the same support, on the same
+not-yet-independently-verified basis. Picking one of these aggregates on a
+database that has not opted in returns a clear "not supported on this
+database" error rather than a failed query. See
+`docs/sip/median-stddev-variance-aggregates.md` for the full design
+rationale, including why this is safe to add without reintroducing the
+totals/subtotals correctness bug fixed by #41184 (SIP-216).
+
 ### Soft delete is on by default, and purging is live
 
 `SOFT_DELETE` now ships **on** (`DEFAULT_FEATURE_FLAGS`), so deleting a
@@ -140,10 +186,12 @@ misrepresents the entity as unchanged.
 - **Storage growth.** Capture writes shadow rows per save, so the metadata
   database grows with edit volume. The `version_history.prune_old_versions`
   beat task removes rows whose transaction is older than
-  `SUPERSET_VERSION_HISTORY_RETENTION_DAYS` (default 30). A deployment that
-  replaces `CELERY_CONFIG` rather than inheriting it must carry both the
-  `superset.tasks.version_history_retention` import and the beat entry; a
-  startup warning names whichever is absent.
+  `SUPERSET_VERSION_HISTORY_RETENTION_DAYS` (default 30).
+- **Check a replaced `CELERY_CONFIG`.** Carry both the
+  `superset.tasks.version_history_retention` import and the
+  `version_history.prune_old_versions` beat entry; see
+  [Version-history retention (pruning)](#version-history-retention-pruning) for
+  the startup-warning behavior.
 - **`PUT` responses change shape.** Entity updates now return populated
   `old_version_uuid` / `new_version_uuid` fields and an `ETag` header, which
   were null or absent while capture was off.
@@ -152,7 +200,10 @@ misrepresents the entity as unchanged.
 kill-switch — not removed with the rollout toggles. Setting it to a falsy value
 stops capture within a restart, without a revert-and-redeploy. Unlike the
 soft-delete toggle, turning it off is a clean stop: existing version rows remain
-readable and no entity state is altered.
+readable and no entity state is altered. Restore is unavailable (404) while
+capture is off. A full rollback also sets
+`FEATURE_FLAGS = {"VERSION_HISTORY": False}` to hide the panel — capture off
+with the panel left on shows an empty or stale history.
 
 ### Scheduled report execution now enforces one application deadline
 
@@ -614,9 +665,9 @@ ALTER TABLE tagged_object DROP CONSTRAINT <constraint_name>;
 ALTER TABLE tagged_object DROP FOREIGN KEY <constraint_name>;
 ```
 
-### Entity version-history infrastructure (gated off by default)
+### Entity version-history infrastructure
 
-Introduces the schema and SQLAlchemy-Continuum wiring that captures version history for charts, dashboards, and datasets, plus read-only `GET /api/v1/{chart,dashboard,dataset}/<uuid>/versions/` endpoints. This ships **inert**: a new config flag `ENABLE_VERSIONING_CAPTURE` defaults to `False`, so no save writes any version rows and the endpoints return empty. It is an operational kill-switch (a release toggle that becomes a permanent ops switch), not a feature flag — set it to `True` to enable capture once validated. The migration is additive; existing entity `PUT` responses gain `old_version_uuid` / `new_version_uuid` body fields and an `ETag` header (both null/absent when capture is off).
+Introduces the schema and SQLAlchemy-Continuum wiring that captures version history for charts, dashboards, and datasets, plus read-only `GET /api/v1/{chart,dashboard,dataset}/<uuid>/versions/` endpoints. Capture is governed by the `ENABLE_VERSIONING_CAPTURE` config value — an operational kill-switch (a release toggle that became a permanent ops switch), not a feature flag; see "Version history is on by default" above for the shipped default. With capture off, no save writes version rows; the endpoints continue to serve already-captured rows read-only. The migration is additive; existing entity `PUT` responses gain `old_version_uuid` / `new_version_uuid` body fields and an `ETag` header (both null/absent when capture is off).
 
 A few save- and import-path internals change **unconditionally** (independent of the flag), because the versioned mappers must behave correctly whether or not capture is enabled:
 
@@ -637,7 +688,7 @@ A read-only companion to the version-history endpoints: each entity type gains a
 | `q`                  | string                       | —          | Case-insensitive search over the full history, applied before pagination (so `count` reflects matches) |
 | `page` / `page_size` | integer                      | `0` / `25` | Pagination (`page_size` clamped to 200)                                                                |
 
-Authorization reuses the resource's `can_read` permission and per-object `raise_for_access`; related-entity rows are visibility-filtered to what the caller may see. The stream is empty unless version capture is on (`ENABLE_VERSIONING_CAPTURE`).
+Authorization reuses the resource's `can_read` permission and per-object `raise_for_access`; related-entity rows are visibility-filtered to what the caller may see. The stream reflects captured history; with capture off it remains readable but stops accruing new entries.
 
 ### Version-history retention (pruning)
 
@@ -657,7 +708,7 @@ Purging is **live by default** (`SOFT_DELETE_PURGE_DRY_RUN=False`), so the reten
 
 Deployments that replace the default `CELERY_CONFIG` must ensure workers register `superset.tasks.deletion_retention` and schedule the `deletion_retention.purge_soft_deleted` task themselves. The shipped Docker development config uses `imports` and includes both entries. While `SOFT_DELETE` is statically enabled, a missing beat entry logs a startup warning; when the override explicitly defines `imports`, a missing purge module is also reported.
 
-Operators can immediately erase a specific entity for compliance (GDPR) via `superset deletion-retention force-purge --uuid <uuid>`; this applies legacy hard-delete semantics — a live chart referencing a force-purged dataset is left without a datasource until re-pointed (the chart is not modified), and it purges the named entity even when it was never soft-deleted. Every scheduled evaluation writes a provisional, content-free record to the new `purge_audit_log` table before the cascade starts. Meaningful retained outcomes survive the entity they name. Consecutive scheduled evaluations with the same blocked outcome suppress only the redundant current provisional record; completed outcomes, outcome transitions, and every force-purge attempt remain independent and immutable. The **scheduled** purge fails closed when its provisional record cannot be written, while **force-purge** proceeds even if the audit write fails — the operator is present and deletion outranks audit for a compliance erasure. Operators can monitor `deletion_retention.blocked_audit_suppressed` and `deletion_retention.blocked_audit_dedupe_fallback` to verify suppression and fail-safe fallback behavior without changing the existing blocked-workload gauge.
+Operators can immediately erase a specific entity for compliance (GDPR) via `superset deletion-retention force-purge --uuid <uuid>`; this applies legacy hard-delete semantics — a live chart referencing a force-purged dataset is left without a datasource until re-pointed (the chart is not modified), and it purges the named entity even when it was never soft-deleted. Every scheduled evaluation writes a provisional, content-free record to the new `purge_audit_log` table before the cascade starts. Meaningful retained outcomes survive the entity they name. Blocked audit records carry a stable machine-readable `reason` code (`report_schedule`, `user_attribute`, or `cascade_integrity_failure` for an unexpected cascade failure caused by a database integrity constraint) so the audit table alone answers why an entity was not purged; records finalized before the column existed keep a NULL reason. Apply the migration before rolling out the new code: the audit model declares the column, so a worker on the new code with an un-migrated table fails its write-ahead write and the scheduled purge fails closed until the migration lands. During a rolling deploy, workers still on the old code write reason-less blocked rows and suppress on status alone; both effects are self-healing, since a NULL-reason record never matches a reason code and the next all-new-code run re-anchors the entity. Consecutive scheduled evaluations blocked with the same status **and reason** suppress only the redundant current provisional record — a reason change writes one new blocked record carrying the new code; completed outcomes, outcome transitions, and every force-purge attempt remain independent and immutable. Retained transition records are not automatically expired, so entities whose block reason changes repeatedly can accumulate multiple audit rows. The **scheduled** purge fails closed when its provisional record cannot be written, while **force-purge** proceeds even if the audit write fails — the operator is present and deletion outranks audit for a compliance erasure. Operators can monitor `deletion_retention.blocked_audit_suppressed` and `deletion_retention.blocked_audit_dedupe_fallback` to verify suppression and fail-safe fallback behavior without changing the existing blocked-workload gauge.
 
 ### Recently Archived view and permanent delete (purge) endpoints
 
@@ -853,7 +904,7 @@ The migration is transactional (all-or-nothing) and idempotent — it can be saf
 
 ### Soft delete and restore for datasets
 
-**The soft-delete behavior in this section applies only when the `SOFT_DELETE` feature flag is enabled. The flag defaults to `False`** (`@lifecycle: development`), so on a default deployment `DELETE /api/v1/dataset/<id>` continues to **hard-delete permanently** — nothing is recoverable. Enable `SOFT_DELETE` to get the behavior described below.
+**The soft-delete behavior in this section applies only when the `SOFT_DELETE` feature flag is enabled. The flag defaults to `True`** (`@lifecycle: testing`), so on a default deployment `DELETE /api/v1/dataset/<id>` uses the recoverable soft-delete behavior described below. Setting `SOFT_DELETE` to `False` restores legacy permanent hard-delete behavior for subsequent deletes.
 
 **Flag-toggle caveat:** the soft-delete visibility filter is evaluated per query while the flag is on. If datasets are soft-deleted during a flag-on window and the flag is later turned **off**, those rows reappear as live datasets in all lists, lookups, and relationship loads (including charts that reference them). The `POST /<uuid>/restore` endpoint and the `dataset_deleted_state` list filter remain functional regardless of the flag, deliberately, so rows soft-deleted during a flag-on window stay discoverable and restorable after a rollback of the flag.
 
@@ -883,7 +934,7 @@ With the flag enabled: `DELETE /api/v1/dataset/<id>` no longer hard-deletes the
 
 ### Soft delete and restore for charts
 
-**Everything in this section applies only when the `SOFT_DELETE` feature flag is enabled. The flag defaults to `False`** (`@lifecycle: development`), so on a default deployment `DELETE /api/v1/chart/<id>` continues to **hard-delete permanently** — nothing is recoverable. Enable `SOFT_DELETE` to get the behavior described below.
+**Everything in this section applies only when the `SOFT_DELETE` feature flag is enabled. The flag defaults to `True`** (`@lifecycle: testing`), so on a default deployment `DELETE /api/v1/chart/<id>` uses the recoverable soft-delete behavior described below. Setting `SOFT_DELETE` to `False` restores legacy permanent hard-delete behavior for subsequent deletes.
 
 **Flag-toggle caveat:** the soft-delete visibility filter is evaluated per query while the flag is on. If charts are soft-deleted during a flag-on window and the flag is later turned **off**, those rows reappear as live charts in all lists, lookups, and relationship loads (including dashboards that contained them). The `POST /<uuid>/restore` endpoint and the `chart_deleted_state` list filter remain functional regardless of the flag, deliberately, so rows soft-deleted during a flag-on window stay discoverable and restorable after a rollback of the flag.
 
@@ -907,7 +958,7 @@ With the flag enabled: `DELETE /api/v1/chart/<id>` no longer hard-deletes the ch
 
 ### Soft delete and restore for dashboards
 
-**Everything in this section applies only when the `SOFT_DELETE` feature flag is enabled. The flag defaults to `False`** (`@lifecycle: development`), so on a default deployment `DELETE /api/v1/dashboard/<id>` continues to **hard-delete permanently** — nothing is recoverable. Enable `SOFT_DELETE` to get the behavior described below.
+**Everything in this section applies only when the `SOFT_DELETE` feature flag is enabled. The flag defaults to `True`** (`@lifecycle: testing`), so on a default deployment `DELETE /api/v1/dashboard/<id>` uses the recoverable soft-delete behavior described below. Setting `SOFT_DELETE` to `False` restores legacy permanent hard-delete behavior for subsequent deletes.
 
 **Flag-toggle caveat:** the soft-delete visibility filter is evaluated per query while the flag is on. If dashboards are soft-deleted during a flag-on window and the flag is later turned **off**, those rows reappear as live dashboards in all lists and lookups (including slug lookups — if a soft-deleted dashboard's slug was reused while the flag was on, both rows become visible with the same slug). The `POST /<uuid>/restore` endpoint and the `dashboard_deleted_state` list filter remain functional regardless of the flag, deliberately, so rows soft-deleted during a flag-on window stay discoverable and restorable after a rollback of the flag.
 

docs/admin_docs/configuration/alerts-reports.mdx

@@ -14,6 +14,20 @@ Users can configure automated alerts and reports to send dashboards or charts to
 
 Alerts and reports are disabled by default. To turn them on, you'll need to change configuration settings and install a suitable headless browser in your environment.
 
+## Content Format Options
+
+When scheduling an alert or report, you can choose the format used to deliver the dashboard or chart:
+
+- **PDF** – a full-page screenshot rendered as a PDF attachment. Available for both dashboards and charts.
+- **PNG** – a screenshot delivered as an attachment. Emails embed the image inline in the message body; Slack and webhook recipients receive it as an uploaded file. Available for both dashboards and charts.
+- **CSV** – chart data attached as a `.csv` file. Available for charts only.
+- **XLSX (Excel)** – chart data attached as a `.xlsx` file. Available for charts only. If the chart's data spans multiple server-paginated files, email delivery detects the bundle and renames the attachment to `.zip`; Slack and webhook deliveries always name the file with an `.xlsx` extension even when the contents are a multi-file ZIP archive.
+- **Text** – chart data embedded directly in the email or Slack message body. Available only for charts using a text-based visualization type (e.g. Table, Pivot Table, Paired t-test).
+
+Dashboard reports and alerts are limited to the PDF and PNG formats; the CSV, XLSX, and Text options are only available when scheduling a report or alert for an individual chart.
+
+For alerts (not scheduled reports), PNG/PDF screenshots and chart CSV/XLSX data are only generated when the `ALERTS_ATTACH_REPORTS` [feature flag](/admin-docs/configuration/configuring-superset#feature-flags) is enabled (the default); with it disabled, an alert notification is still sent, but without the attachment.
+
 ## Requirements
 
 ### Commons

docs/admin_docs/configuration/cache.mdx

@@ -76,6 +76,27 @@ value defined in `DATA_CACHE_CONFIG`.
 Note, that by setting the cache timeout to `-1`, caching for charting data can be disabled, either
 per chart, dataset or database, or by default if set in `DATA_CACHE_CONFIG`.
 
+Native filter option queries (the dropdown values for native filters) go through this same
+chart-data cache, but their freshness needs often differ from regular chart queries, especially for
+datasets whose visible values change frequently, including RLS-constrained datasets. Set
+`NATIVE_FILTER_OPTIONS_CACHE_TIMEOUT` in `superset_config.py` to give these queries a dedicated
+timeout, checked before the chart/dataset/database chain and the `DATA_CACHE_CONFIG` default above:
+
+```python
+NATIVE_FILTER_OPTIONS_CACHE_TIMEOUT = 60  # seconds
+```
+
+- `None` (default): native filter option queries fall through to the normal
+  chart/dataset/database/`DATA_CACHE_CONFIG` resolution chain.
+- `-1`: disables caching for native filter option queries entirely.
+- `0`: passed directly to the cache backend; behavior is backend-specific, so use `-1` if the intent
+  is to disable caching.
+- A positive integer: cache native filter option queries for that many seconds.
+
+This setting only applies to requests detected as native filter option queries. It takes precedence
+over the per-chart/dataset/database timeouts, but not over an explicit per-request
+`custom_cache_timeout` override (e.g. "Force refresh").
+
 ## Limiting Cached Result Size
 
 Very large chart or SQL query results can flood the cache backend (Redis/Memcached), evicting many

docs/admin_docs/configuration/configuring-superset.mdx

@@ -97,6 +97,54 @@ for more information on how to configure it.
 
 At the very least, you'll want to change `SECRET_KEY` and `SQLALCHEMY_DATABASE_URI`. Continue reading for more about each of these.
 
+## Localizing D3 date and time labels
+
+`BABEL_DEFAULT_LOCALE` controls Superset's application translations, while
+`D3_TIME_FORMAT` provides localized date and time names to visualizations that
+use the D3 formatter registry, including Calendar Heatmap. Configure both when
+you want the application and chart labels to use the same locale.
+
+`D3_TIME_FORMAT` accepts partial overrides. For example, Russian month names
+can be configured in `superset_config.py` as follows:
+
+```python
+BABEL_DEFAULT_LOCALE = "ru"
+
+D3_TIME_FORMAT = {
+    "months": [
+        "Январь",
+        "Февраль",
+        "Март",
+        "Апрель",
+        "Май",
+        "Июнь",
+        "Июль",
+        "Август",
+        "Сентябрь",
+        "Октябрь",
+        "Ноябрь",
+        "Декабрь",
+    ],
+    "shortMonths": [
+        "Янв",
+        "Фев",
+        "Мар",
+        "Апр",
+        "Май",
+        "Июн",
+        "Июл",
+        "Авг",
+        "Сен",
+        "Окт",
+        "Ноя",
+        "Дек",
+    ],
+}
+```
+
+Restart Superset after changing `superset_config.py` so the frontend receives
+the updated formatter configuration.
+
 ## Chart-data query timing
 
 Set `CHART_DATA_INCLUDE_TIMING = True` to add an optional versioned timing object

docs/admin_docs/configuration/importing-exporting-datasources.mdx

@@ -83,6 +83,26 @@ The optional username flag **-u** sets the user used for the datasource import.
 superset import_datasources -p <path / filename> -u 'admin'
 ```
 
+## Importing a Directory of Assets
+
+The `import_directory` command imports a directory of exported assets (databases, datasets,
+dashboards, charts) in the same layout produced by the ZIP-based export. Saved queries and tags
+included in a full export are not imported by this command:
+
+```bash
+superset import_directory <path / directory>
+```
+
+As with `import_datasources`, the optional username flag **-u** sets the user assigned as the
+owner of the imported assets. The default is 'admin'. Example:
+
+```bash
+superset import_directory <path / directory> -u 'admin'
+```
+
+If the specified user does not exist, the command fails immediately with an error rather than
+importing the assets without an owner.
+
 ## Dashboard Import Overwrite Behavior
 
 When importing a dashboard ZIP with the **overwrite** option enabled, any existing charts that are part of the dashboard are **replaced** rather than duplicated. This applies to:

docs/admin_docs/configuration/mcp-server.mdx

@@ -505,6 +505,8 @@ All MCP settings go in `superset_config.py`. Defaults are defined in `superset/m
 | `MCP_DEV_USERNAME`   | --            | Superset username for development mode (no auth)                                                                                                                                                                                                                             |
 | `MCP_RBAC_ENABLED`   | `True`        | Enforce Superset's role-based access control on MCP tool calls. When `True`, each tool checks that the authenticated user has the required FAB permission before executing. Disable only for testing or trusted-network deployments.                                         |
 | `MCP_DISABLED_TOOLS` | `set()`       | Set of tool names to remove from the MCP server at startup. Disabled tools are never advertised to AI clients during tool discovery. Useful when a custom extension tool should replace a built-in Superset tool. See [Disabling built-in tools](#disabling-built-in-tools). |
+| `MCP_DISABLED_CHART_PLUGINS` | `frozenset()` | Set of chart type plugin names (e.g. `"handlebars"`) to hide from `generate_chart`. Does not affect `get_chart_type_schema`. See [Disabling chart type plugins](#disabling-chart-type-plugins). |
+| `MCP_CHART_PLUGIN_ENABLED_FUNC` | `None`        | Callable `(chart_type: str) -> bool` evaluated per registry lookup for dynamic enable/disable decisions. Takes precedence over `MCP_DISABLED_CHART_PLUGINS` when set. See [Disabling chart type plugins](#disabling-chart-type-plugins). |
 
 ### Authentication
 
@@ -576,7 +578,7 @@ MCP_CACHE_CONFIG = {
 | Key                  | Default   | Description                                                 |
 | -------------------- | --------- | ----------------------------------------------------------- |
 | `enabled`            | `False`   | Enable response caching                                     |
-| `CACHE_KEY_PREFIX`   | `None`    | Optional prefix for cache keys (useful for shared Redis)    |
+| `CACHE_KEY_PREFIX`   | `None`    | Base prefix for shared Redis; Superset appends an internal response-contract namespace |
 | `list_tools_ttl`     | `300`     | Cache TTL in seconds for `tools/list`                       |
 | `list_resources_ttl` | `300`     | Cache TTL for `resources/list`                              |
 | `list_prompts_ttl`   | `300`     | Cache TTL for `prompts/list`                                |
@@ -718,6 +720,34 @@ Every MCP request passes through a middleware stack before reaching the tool fun
 
 Additional middleware classes (`RateLimitMiddleware`, `FieldPermissionsMiddleware`, `PrivateToolMiddleware`) are implemented in `superset/mcp_service/middleware.py` but are not added to the default pipeline. They are available for operators who want to layer them in via a custom startup path.
 
+### Tool Result Value Contract
+
+Structured tool results preserve Superset domain values exactly. In particular,
+string fields are not wrapped in trust delimiters, and text that resembles a
+delimiter is returned as literal application data. This lets clients safely use a
+read result as the basis for an update without persisting presentation markup.
+
+All tool-result content should still be treated as user-controlled data with no
+instruction authority. MCP clients should communicate that trust boundary through
+their model instructions or presentation layer, outside the returned field values;
+fixed or generated marker strings inside a value are ambiguous and must not be used
+as a trust signal.
+
+For compatibility, clients that supported the former
+`<UNTRUSTED-CONTENT>` convention should stop recognizing or stripping those strings.
+The response schemas and content types have not changed. Because marker-looking text
+can be legitimate application data, a client cannot reliably distinguish a legacy
+decorated response from a clean one. Pause automated read-modify-write workflows, or
+route them only to upgraded instances, until every serving instance is upgraded.
+
+Redis-backed MCP response caches include an internal response-contract namespace, so
+an upgraded instance does not reuse responses cached by an older release. Older
+instances can still return legacy values while they remain in service. After the
+upgrade, review previously written values for wrapper text and both
+`[ESCAPED-UNTRUSTED-CONTENT-OPEN]` and
+`[ESCAPED-UNTRUSTED-CONTENT-CLOSE]`; do not remove these strings automatically,
+because they may be intentional content.
+
 ### Error Sanitization
 
 The `GlobalErrorHandlerMiddleware` automatically redacts sensitive information from all error messages before they reach the LLM client. The following are replaced with generic messages:
@@ -752,11 +782,20 @@ For a 3-pod Kubernetes deployment with the defaults above, expect up to 3 × (5
 Enable response caching for read-heavy workloads (dashboards/datasets that don't change frequently). With the in-memory backend (default when `MCP_STORE_CONFIG` is disabled), caching is per-process. Use Redis-backed caching for consistent cache hits across multiple pods:
 
 ```python
-MCP_CACHE_CONFIG = {"enabled": True, "call_tool_ttl": 3600}
+MCP_CACHE_CONFIG = {
+    "enabled": True,
+    # Cache keys don't include the requesting principal and hits are served
+    # ahead of auth/RBAC, so a shared cache can return one caller's response
+    # to another. Required for caching to actually start -- only appropriate
+    # when every request is guaranteed to come from the same principal.
+    "dangerously_share_cache_across_principals": True,
+    "CACHE_KEY_PREFIX": "mcp_cache_",
+    "call_tool_ttl": 3600,
+}
 MCP_STORE_CONFIG = {"enabled": True, "CACHE_REDIS_URL": "redis://redis:6379/0"}
 ```
 
-Mutating tools (`generate_chart`, `update_chart`, `execute_sql`, `generate_dashboard`) are always excluded from caching regardless of this setting.
+Every tool with a side effect (create/update/delete/execute) is always excluded from caching regardless of this setting -- see the `excluded_tools` default in `superset/mcp_service/mcp_config.py` for the current list.
 
 ---
 
@@ -856,6 +895,39 @@ MCP_DISABLED_TOOLS = {"extensions.myorg.myextension.some_tool"}
 Specifying a tool name that does not exist logs a warning at startup and is otherwise ignored — it will not prevent the server from starting.
 :::
 
+## Disabling chart type plugins
+
+The `generate_chart` tool dispatches per chart type (`xy`, `table`, `pie`, `pivot_table`, `mixed_timeseries`, `handlebars`, `big_number`, `histogram`, `box_plot`, `waterfall`) to a registered chart type plugin. Two settings let operators enable or disable individual chart type plugins at runtime, without a code deploy.
+
+### Static deny-list
+
+Set `MCP_DISABLED_CHART_PLUGINS` in your `superset_config.py` to a set of chart type names:
+
+```python
+# superset_config.py
+
+# Emergency kill switch: hide "handlebars" from all callers
+MCP_DISABLED_CHART_PLUGINS = {"handlebars"}
+```
+
+Disabled chart types stay registered but are filtered out at lookup time: they're never listed in `generate_chart`'s supported chart types, and `generate_chart` calls for them are rejected. `get_chart_type_schema` consults its own static schema/example map rather than the registry filter, so a disabled chart type's schema remains queryable through that tool even though `generate_chart` will reject it.
+
+### Dynamic predicate
+
+For per-request control (A/B tests, gradual rollout, entitlement checks), set `MCP_CHART_PLUGIN_ENABLED_FUNC` to a callable. It's evaluated as `enabled_func(chart_type: str) -> bool` on every registry lookup, and it takes precedence over `MCP_DISABLED_CHART_PLUGINS` when set:
+
+```python
+# superset_config.py
+from flask import g
+
+
+def MCP_CHART_PLUGIN_ENABLED_FUNC(chart_type: str) -> bool:
+    flags = getattr(g, "feature_flags", {})
+    return flags.get(f"mcp_chart_{chart_type}", True)
+```
+
+The callable must be cheap and in-process (consult already-loaded feature flags or request-local context) -- do not perform network I/O per call. If it raises, the registry fails closed (the plugin is hidden) and logs a warning.
+
 ## Security Best Practices
 
 - **Use TLS** for all production MCP endpoints -- place the server behind a reverse proxy with HTTPS

docs/admin_docs/configuration/theming.mdx

@@ -88,6 +88,27 @@ THEME_DARK = {
 # - OS preference detection is automatically enabled
 ```
 
+### Default Theme Mode
+
+By default, Superset mimics the visitor's OS/browser preference (light or dark) for
+sessions that don't have a saved user preference. Use `THEME_DEFAULT_MODE` to override
+that starting point instance-wide for the standard application:
+
+```python
+# Default theme mode for sessions without a saved user preference.
+# One of "default" (always light), "dark" (always dark), or "system" (mimic OS preference).
+THEME_DEFAULT_MODE = "dark"
+```
+
+- `"system"` (the default) preserves the existing behavior of following the OS/browser
+  preference, provided a dark theme is configured (`THEME_DARK` is not `None`). If no dark
+  theme is available, Superset always starts in light mode regardless of this setting.
+- `"default"` or `"dark"` forces that starting mode for first-time visitors; users can still switch themes manually afterward if both `THEME_DEFAULT` and `THEME_DARK` are available.
+- A user's own saved preference, once they toggle the theme switcher, always takes precedence over `THEME_DEFAULT_MODE`.
+- `THEME_DEFAULT_MODE` has no effect on embedded dashboards: the embed SDK sets the
+  starting mode via its own `themeMode` URL parameter, which takes precedence and falls
+  back to light mode when the parameter is absent.
+
 ### App Branding
 
 The application name shown in the browser title bar and navigation can be

docs/admin_docs/security/security.mdx

@@ -243,6 +243,42 @@ Each subject in the response includes flat scalar ids (`user_id`, `role_id`, `gr
 than a nested object, so callers can match directly on whichever id they already have — only the
 id field matching the subject's `type` is populated; the others are `null`.
 
+#### Extending Editorship with EXTRA_EDITORS_RESOLVER
+
+Deployments that grant edit access to a dashboard or chart through a mechanism outside
+Superset's own Subject-based `editors` list — for example, a folder-permission system or an
+internal directory service — can plug that logic in with `EXTRA_EDITORS_RESOLVER`:
+
+```python
+def extra_editors_resolver(resource):
+    # `resource` is the Dashboard or Slice instance being checked.
+    # Return Subject instances, raw subject ids, or dicts with an `id` key.
+    return [...]
+
+
+EXTRA_EDITORS_RESOLVER = extra_editors_resolver
+```
+
+The resolver's result is unioned with the resource's own `editors` for editorship checks: it
+feeds `is_editor`, `raise_for_editorship`, save-as, and soft-delete restore. When
+`EXTRA_EDITORS_RESOLVER` is configured, the usual lockout-prevention behavior — automatically
+re-adding a non-admin who removes themselves from a resource's editors — is skipped, since the
+deployment has its own way of keeping the resource editable. This skip is global to the setting,
+not per-resource: it still applies on a resource where the resolver currently returns no
+subjects, so a resolver that can't guarantee an alternate editor for every resource can let the
+last editor remove themselves and leave it uneditable by non-admins.
+
+Resolved subject ids are also surfaced as `extra_editors` in the chart and dashboard `GET`
+responses, so API clients can distinguish resolver-granted editorship from the resource's own
+`editors` list. This field is attached after serialization and isn't part of the OpenAPI response
+schema, so generated API clients won't see it as a typed field.
+
+Because the resolver is arbitrary per-deployment Python rather than a SQL-expressible condition,
+editorship it grants cannot be reflected in list-view filtering (for example, the soft-deleted
+archive is scoped to editors via a SQL query). It does still run once per row on chart and
+dashboard list responses to populate `extra_editors`, so a slow or unavailable external resolver
+affects ordinary list requests, not just direct per-object checks.
+
 ### Dashboard Access Control
 
 Access to dashboards is managed via editors (subjects that have edit permissions to the dashboard).
@@ -400,7 +436,7 @@ Once enabled, each user manages their own keys from their profile page:
 1. Open the user menu (top-right) and click **Info** to navigate to the User Info page
 2. Expand the **API Keys** section
 3. Click **+ API Key**
-4. Enter a name and (optionally) an expiration date
+4. Enter a name and optionally select resource scopes
 5. Copy the generated token — it is shown only once
 
 Only users with the `can_read` and `can_write` permissions on `ApiKey` (granted by default to Admins) can manage API keys.
@@ -415,6 +451,18 @@ Authorization: Bearer <your-api-key>
 
 This works for all REST API endpoints and the MCP server. The request is executed with the permissions of the user who created the key.
 
+#### API Key Scopes
+
+The creation dialog can restrict an API key to MCP resource actions such as
+`superset:dashboard:read` or `superset:chart:write`. A scope is an additional
+restriction: it never grants a permission that the creating user does not
+already have through Superset RBAC. Write scopes also cover update and delete
+operations for that resource; `superset:sqllab:write` covers SQL execution.
+
+Keys created without scopes retain legacy RBAC-only behavior. The scoped-key
+restrictions described here are enforced by the MCP server; regular REST API
+routes continue to apply their existing Superset RBAC checks.
+
 #### Use Cases
 
 - **CI/CD pipelines** — automated chart/dashboard exports and imports
@@ -571,6 +619,28 @@ SELECT * FROM (
   queries run against tables that have associated datasets with RLS filters will then have
   the appropriate predicates injected automatically.
 
+#### RLS Indicator in the Dataset List and Explore
+
+When a dataset has one or more RLS filters that apply to it, Superset shows a lock
+icon badge next to the dataset name in the **Datasets** list and next to the dataset
+selector in **Explore**. Hovering over the badge shows a tooltip listing each
+applicable filter's name, filter type (Regular or Base), group key (if any), assigned
+subjects (labeled "Roles" in the tooltip, but may include users and groups too), and
+clause.
+
+This badge also surfaces filters that are inherited from the physical tables
+referenced by a virtual (SQL-based) dataset, as described above. Inherited filters
+are marked "from underlying table" in the tooltip, and a summary note is shown
+whenever any of the listed filters are inherited rather than assigned directly to
+the dataset. Inherited-filter detection depends on Superset's SQL parser being able
+to identify the referenced tables and match them to a physical dataset by name,
+schema, and database, so it's best-effort: unparseable or unmatched references won't
+surface a filter on the badge even if one would apply at query time.
+
+The badge is a visibility aid only — it does not change which filters are applied to
+a query. Use the RLS REST API described below if you need to confirm exactly which
+filters affect a dataset.
+
 #### Checking RLS Filters via the API
 
 You can use the RLS REST API to audit which filters are configured and which datasets

docs/data/countries.json

@@ -86,6 +86,7 @@
     "Israel",
     "Italy",
     "Italy (regions)",
+    "Italy (regions and autonomous provinces)",
     "Ivory Coast",
     "Japan",
     "Jordan",

docs/docs/faq.mdx

@@ -46,13 +46,43 @@ import FAQSchema from '@site/src/components/FAQSchema';
       answer:
         'You need to register a free account at Mapbox.com, obtain an API key, and add it to your .env file at the key MAPBOX_API_KEY.',
     },
+    {
+      question: 'How to limit the timed refresh on a dashboard?',
+      answer:
+        'To exclude specific slices from the timed refresh process, add the timed_refresh_immune_slices key to the dashboard JSON Metadata field with the slice IDs to exclude.',
+    },
+    {
+      question: "Why does 'flask fab' or Superset freeze, hang, or not respond when started (my home directory is NFS mounted)?",
+      answer:
+        'By default, Superset creates and uses an SQLite database at ~/.superset/superset.db. SQLite is known to not work well if used on NFS due to broken file locking. Override the path with the SUPERSET_HOME environment variable or change SQLALCHEMY_DATABASE_URI in superset_config.py.',
+    },
+    {
+      question: 'What if the table schema changed?',
+      answer:
+        'Go to Data -> Datasets, click the edit icon next to the dataset whose schema has changed, and hit Sync columns from source from the Columns tab. The new columns will get merged.',
+    },
     {
       question: 'What database engine can I use as a backend for Superset?',
       answer:
         'Superset is tested using MySQL, PostgreSQL, and SQLite backends for storing its internal metadata. While Superset supports many databases as data sources, only these are recommended for the metadata store in production.',
     },
     {
-      question: 'Does Superset work with my database?',
+      question: 'How can I configure OAuth authentication and authorization?',
+      answer:
+        'Take a look at the Flask-AppBuilder OAuth configuration example, which shows how to configure OAuth authentication and authorization in Superset.',
+    },
+    {
+      question: 'Is there a way to force the dashboard to use specific colors?',
+      answer:
+        'It is possible on a per-dashboard basis by providing a mapping of labels to colors in the JSON Metadata attribute using the label_colors key. You can use a full hex color, a named color, or the index in the current color palette.',
+    },
+    {
+      question: 'How do I expand all chart descriptions on a dashboard by default?',
+      answer:
+        'Add the expand_all_slices key to the dashboard JSON Metadata field with a value of true. Charts that have already been manually expanded or collapsed keep that per-chart override regardless of the setting.',
+    },
+    {
+      question: 'Does Superset work with [insert database engine here]?',
       answer:
         'Superset supports any database with a Python SQLAlchemy dialect and DBAPI driver. Check the Connecting to Databases documentation for the full list of supported databases.',
     },
@@ -61,15 +91,30 @@ import FAQSchema from '@site/src/components/FAQSchema';
       answer:
         'Yes, Superset has a public REST API documented using Swagger. Enable FAB_API_SWAGGER_UI in superset_config.py to access interactive API documentation at /swagger/v1.',
     },
+    {
+      question: 'How can I see usage statistics (e.g., monthly active users)?',
+      answer:
+        'This functionality is not included with Superset, but you can extract and analyze the logs table in Superset\u2019s metadata database to see what actions have occurred.',
+    },
+    {
+      question: 'What does Hours Offset in the Edit Dataset view do?',
+      answer:
+        'In the Edit Dataset view, the hours offset lets you configure the number of hours to be added or subtracted from the time column. This can be used, for example, to convert UTC time to local time.',
+    },
     {
       question: 'Does Superset collect any telemetry data?',
       answer:
         'Superset uses Scarf by default to collect basic telemetry data to help maintainers understand version usage. Users can opt out by setting the SCARF_ANALYTICS environment variable to false.',
     },
     {
-      question: 'Does Superset have a trash bin to recover deleted assets?',
+      question: 'Does Superset have an archive panel or trash bin from which a user can recover deleted assets?',
       answer:
-        'No, there is no built-in way to recover deleted dashboards, charts, or datasets. It is recommended to take periodic backups of the metadata database and use export functionality for recovery.',
+        'No. Currently, there is no way to recover a deleted Superset dashboard, chart, dataset, or database from the UI. It is recommended to take periodic backups of the metadata database and use export functionality for recovery.',
+    },
+    {
+      question: 'I ran a security scan of the Superset container image and it showed dozens of "high" and "critical" vulnerabilities! Can you release a version of Superset without these?',
+      answer:
+        'These are dependency CVEs in software that Superset uses, mostly in the Linux kernel or Python. Superset addresses them by regularly updating dependencies and welcomes pull requests that fix dependency CVEs. The Superset security team focuses primarily on vulnerabilities in Superset itself.',
     },
   ]}
 />
@@ -214,8 +259,7 @@ frontend falls back to a built-in default of `5000` milliseconds.
 SUPERSET_DASHBOARD_MANUAL_REFRESH_STAGGER_MS = 0
 ```
 
-**Why does ‘flask fab’ or superset freeze/hang/not responding when started (my home directory is
-NFS mounted)?**
+## Why does `flask fab` or Superset freeze, hang, or not respond when started (my home directory is NFS mounted)?
 
 By default, Superset creates and uses an SQLite database at `~/.superset/superset.db`. SQLite is
 known to [not work well if used on NFS](https://www.sqlite.org/lockingv3.html) due to broken file

docs/docs/using-superset/creating-your-first-dashboard.mdx

@@ -406,6 +406,18 @@ ECharts option overrides bypass Superset's validation layer. Invalid option keys
 
 When the **Search Box** is visible in a Table chart, the **Download** action exports only the rows currently visible after the search filter is applied — not the full underlying dataset. This matches the visual output and is intentional. To export the full dataset regardless of search state, use the **Download as CSV** option from the chart's three-dot menu in the dashboard or from the Explore chart toolbar before applying a search filter.
 
+### Exporting a Chart as an Image or PDF
+
+Alongside the raw-data export options (CSV, JSON, Excel), a chart's three-dot menu — in a dashboard or from the Explore chart toolbar — offers a few ways to export a visual snapshot of the chart:
+
+- **Export screenshot (jpeg)** — a single-click JPEG screenshot of the chart.
+- **Export screenshot (png)** — opens a submenu with **Transparent background** and **Solid background** options. The solid option uses the current theme's background color. PNG produces a higher-quality image than the JPEG export.
+- **Export as PDF** — downloads the chart as a PDF file.
+
+The dropdown menu is briefly hidden while the screenshot or PDF is being captured so it doesn't appear in the exported file. In Explore, these image and PDF options are available from the **Export All Data** submenu, and also from the **Export current view** submenu when the chart type supports current-view export.
+
+These menu items respect your permissions: the dashboard export menu only appears if you can download, and the image/PDF options are disabled if you lack image-export permission.
+
 ### Sharing a Specific Tab
 
 When a dashboard has tabs, each tab gets its own shareable URL. Navigate to the tab you want to share and copy the URL from your browser's address bar — the tab anchor is encoded in the URL so that anyone opening the link lands directly on that tab.

docs/docs/using-superset/embedding.mdx

@@ -88,6 +88,21 @@ embedDashboard({
 
 If the callback returns `null` or is not provided, Superset uses its own permalink URL as a fallback.
 
+### Permalink origin rewriting
+
+This rewrite only applies to the non-embedded permalink path — it has no effect on embedded dashboards. When Superset is not embedded, it rewrites the origin of any permalink URL it generates to `window.location.origin` before showing it to the user, which keeps a proxied or subdirectory-deployed Superset from handing out a permalink that points at an internal hostname the user's browser can't reach.
+
+When Superset **is** embedded, this rewrite is skipped entirely regardless of the flag below: a `resolvePermalinkUrl` callback's return value is used as-is, and if no callback is provided (or it fails), the backend-supplied URL is also returned as-is.
+
+If your reverse proxy correctly forwards `X-Forwarded-Host` and you'd rather non-embedded permalinks carry the backend's literal origin, opt out of the rewrite with `EMBEDDED_DISABLE_PERMALINK_ORIGIN_REWRITE`:
+
+```python
+# superset_config.py
+EMBEDDED_DISABLE_PERMALINK_ORIGIN_REWRITE = True
+```
+
+This defaults to `False` (rewrite enabled) and only affects non-embedded permalinks. Flipping the default would regress the common proxied/subdirectory deployment by exposing an unreachable internal host in copied permalinks.
+
 ---
 
 ## Feature Flags for Embedded Mode

docs/docs/using-superset/recently-archived.mdx

@@ -11,6 +11,11 @@ When soft-delete is enabled, deleting a chart, dashboard, or dataset archives it
 instead of removing it permanently. The **Recently Archived** view lets owners
 and admins find archived objects and restore them.
 
+A chart used by an alert or report cannot be archived while that dependency
+exists. In the chart list view, the archive confirmation lists the alerts and
+reports that use the chart; a blocked attempt names them and asks you to
+detach or delete them first.
+
 :::note
 
 This view is gated by the `SOFT_DELETE` feature flag. When the flag is off the
@@ -21,7 +26,8 @@ page and its menu entry are hidden, and deletes are permanent as before.
 ## Finding archived objects
 
 Open **Recently Archived** and pick a type — **Chart**, **Dashboard**, or
-**Dataset** — from the Type selector. The view shows one type at a time; each
+**Dataset** (shown as **Datasource** when semantic layers are enabled) — from
+the Type selector. The view shows one type at a time; each
 type is read from its own list endpoint, so the same row-level access rules that
 govern the normal lists apply here.
 

docs/docs/using-superset/using-ai-with-superset.mdx

@@ -248,11 +248,12 @@ Ask your admin for the MCP server URL and any authentication tokens you need.
 
 ### Datasets
 
-| Tool                     | Description                                      |
-| ------------------------ | ------------------------------------------------ |
-| `list_datasets`          | List datasets with filtering and search          |
-| `get_dataset_info`       | Get dataset metadata (columns, metrics, filters) |
-| `create_virtual_dataset` | Create a virtual dataset from a SQL query        |
+| Tool                     | Description                                                                                                                  |
+| ------------------------ | ---------------------------------------------------------------------------------------------------------------------------- |
+| `list_datasets`          | List datasets with filtering and search                                                                                      |
+| `get_dataset_info`       | Get dataset metadata (columns, metrics, schema details)                                                                      |
+| `create_virtual_dataset` | Create a virtual dataset from a SQL query                                                                                    |
+| `update_dataset_metric`  | Update a saved metric's expression, name, verbose_name, or format (affects every chart using it; requires dataset ownership) |
 
 ### Charts
 

docs/docs/using-superset/version-history.mdx

@@ -15,29 +15,29 @@ description of what changed — "Chart renamed to Q3 Revenue", "Added filter on
 'Region'" — rather than a raw diff. You can search the history and filter it
 down to changes on the entity itself or on the things it depends on.
 
-## Enabling it
-
-Two switches are involved, and both matter.
+## Enabling and disabling it
 
 | Setting | Type | Effect |
 | --- | --- | --- |
 | `VERSION_HISTORY` | Feature flag | Shows the version history UI |
 | `ENABLE_VERSIONING_CAPTURE` | Config value | Records versions as entities are saved |
 
+Both default to on. To turn the feature off:
+
 ```python
 # superset_config.py
-FEATURE_FLAGS = {"VERSION_HISTORY": True}
-ENABLE_VERSIONING_CAPTURE = True
+FEATURE_FLAGS = {"VERSION_HISTORY": False}
+ENABLE_VERSIONING_CAPTURE = False
 ```
 
-Both default to off. They are separate because capture is the expensive half:
-an operator may want to start recording history before exposing the UI, so that
-there is something to show when they do.
+Restart Superset and its workers for the capture change to take effect. Existing
+history remains readable while capture is off, but **Restore** is unavailable
+(404).
 
-Turning the UI on without capture gives a panel that reports "No history yet"
-and never fills, so enable capture first — or at the same time. History only
-accrues from the moment capture is switched on; earlier edits are not
-reconstructed.
+Disable them together: capture off with the UI left on gives a panel that
+stops filling — an empty or stale history misrepresents the entity as
+unchanged. History only accrues while capture is on; edits made while it was
+off are not reconstructed.
 
 ## Viewing history
 

docs/package.json

@@ -58,15 +58,15 @@
     "@fontsource/inter": "^5.3.0",
     "@mdx-js/react": "^3.1.1",
     "@saucelabs/theme-github-codeblock": "^0.3.0",
-    "@storybook/addon-docs": "^10.5.7",
+    "@storybook/addon-docs": "^10.5.10",
     "@superset-ui/core": "^0.20.4",
-    "@swc/core": "^1.15.47",
-    "antd": "^6.5.4",
-    "baseline-browser-mapping": "^2.11.12",
-    "caniuse-lite": "^1.0.30001807",
-    "docusaurus-plugin-openapi-docs": "^5.1.3",
-    "docusaurus-theme-openapi-docs": "^5.1.3",
-    "js-yaml": "^5.2.3",
+    "@swc/core": "^1.16.1",
+    "antd": "^6.6.1",
+    "baseline-browser-mapping": "^2.11.16",
+    "caniuse-lite": "^1.0.30001809",
+    "docusaurus-plugin-openapi-docs": "^5.2.0",
+    "docusaurus-theme-openapi-docs": "^5.2.0",
+    "js-yaml": "^5.3.0",
     "json-bigint": "^1.0.0",
     "prism-react-renderer": "^2.4.1",
     "react": "^18.3.1",
@@ -77,8 +77,8 @@
     "react-table": "^7.8.0",
     "remark-import-partial": "^0.0.2",
     "reselect": "^5.2.0",
-    "storybook": "^10.5.7",
-    "swagger-ui-react": "^5.32.12",
+    "storybook": "^10.5.10",
+    "swagger-ui-react": "^5.32.14",
     "swc-loader": "^0.2.7",
     "tinycolor2": "^1.4.2",
     "unist-util-visit": "^5.1.0"
@@ -89,14 +89,14 @@
     "@eslint/js": "^9.39.2",
     "@types/js-yaml": "^4.0.9",
     "@types/react": "^19.1.8",
-    "@typescript-eslint/eslint-plugin": "^8.66.0",
-    "@typescript-eslint/parser": "^8.66.0",
+    "@typescript-eslint/eslint-plugin": "^8.67.0",
+    "@typescript-eslint/parser": "^8.67.0",
     "eslint": "^9.39.2",
     "eslint-plugin-react": "^7.37.5",
-    "globals": "^17.9.0",
-    "oxfmt": "^0.62.0",
+    "globals": "^17.11.0",
+    "oxfmt": "^0.64.0",
     "typescript": "~6.0.3",
-    "typescript-eslint": "^8.66.0",
+    "typescript-eslint": "^8.67.0",
     "webpack": "^5.109.2"
   },
   "browserslist": {

docs/sip/median-stddev-variance-aggregates.md

@@ -0,0 +1,236 @@
+<!--
+Licensed to the Apache Software Foundation (ASF) under one
+or more contributor license agreements.  See the NOTICE file
+distributed with this work for additional information
+regarding copyright ownership.  The ASF licenses this file
+to you under the Apache License, Version 2.0 (the
+"License"); you may not use this file except in compliance
+with the License.  You may obtain a copy of the License at
+
+  http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing,
+software distributed under the License is distributed on an
+"AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+KIND, either express or implied.  See the License for the
+specific language governing permissions and limitations
+under the License.
+-->
+
+# SIP: System-wide MEDIAN, Sample Standard Deviation, and Sample Variance metric aggregates
+
+## [DRAFT — proposal for discussion]
+
+This document now has an accompanying implementation in this PR, for the
+proposed mechanism plus a first, empirically-verified engine set (Postgres,
+MySQL, DuckDB, Redshift by inheritance). It is intended to seed discussion on
+whether this is the right shape and scope before it goes up for a formal SIP
+vote, not to pre-empt that discussion, the code exists so reviewers have a
+concrete design to react to rather than a description of one.
+
+## Motivation
+
+Before #41184 (SIP-216, the non-additive-totals fix), the Pivot Table chart
+exposed an "Aggregation function" control with 18 choices, including
+`Median`, `Sample Standard Deviation`, `Sample Variance`, `First`, `Last`,
+`Count Unique Values`, and `List Unique Values`. #41184 deleted that control
+wholesale, and deliberately so: it re-aggregated already-aggregated cell
+values to compute totals/subtotals, which is exactly the class of bug
+SIP-216 fixed (summing per-group averages, averaging per-group medians, etc.
+produces silently wrong totals). #42761 subsequently restored the one piece
+of that control's functionality that was cleanly separable from the
+correctness bug, the "show as % of row/column/total" display option,
+redesigned as a decoupled, post-hoc-only `showValuesAs` control.
+
+A user has since noticed that several of the other pre-#41184 options never
+came back. Checking today's metric aggregate list (`AVG, COUNT,
+COUNT_DISTINCT, MAX, MIN, SUM`, see
+`superset-frontend/packages/superset-ui-core/src/query/types/Metric.ts`),
+most of these have a reasonable equivalent already: `Count Unique Values`
+maps to `COUNT_DISTINCT`; `Count`/`Average`/`Max`/`Min` are already standard
+aggregates; the two "fraction of" variants are already covered by
+`showValuesAs`. But `Median`, `Sample Standard Deviation`, and `Sample
+Variance` have no equivalent today anywhere in Superset, not just in Pivot
+Table, in any chart type, since the aggregate list is shared across the
+whole app.
+
+This is a real, currently-live gap, not a hypothetical one:
+`superset/mcp_service/chart/chart_utils.py`, `schemas.py`, and
+`prompts/create_chart_guided.py` already treat `STDDEV`, `VAR`, `MEDIAN`,
+and `PERCENTILE` as valid aggregate values in their own validation and
+documentation, but those values are never recognized by
+`superset/connectors/sqla/models.py`'s `sqla_aggregations` dict (the actual
+mapping from aggregate name to SQL), so an AI agent using the MCP tool to
+build a chart with `"aggregate": "STDDEV"` today creates a chart that
+**errors at query time** with "Adhoc metric aggregate is invalid." This SIP
+proposes closing that gap for real, at the source, rather than patching
+around it in MCP.
+
+## Proposed change
+
+Add `MEDIAN`, `STDDEV_SAMP`, and `VAR_SAMP` as first-class, system-wide
+metric aggregates, available anywhere a metric aggregate is chosen (every
+chart type, SQL Lab metric picker, MCP), not as a Pivot-Table-specific
+control.
+
+**Why this is safe with respect to SIP-216, and needs no Pivot-Table-specific
+code at all:** Pivot Table's non-additive-totals machinery
+(`superset-frontend/plugins/plugin-chart-pivot-table/src/plugin/utilities.ts`)
+already classifies any metric aggregate not in `ADDITIVE_AGGREGATES = {SUM,
+COUNT, MIN, MAX}` as non-additive, which routes totals/subtotals through the
+correct DB-`GROUPING SETS`-rollup path rather than client-side
+re-aggregation (`AVG` and `COUNT_DISTINCT` already go through this path
+today). `MEDIAN`/`STDDEV_SAMP`/`VAR_SAMP` fall into that bucket
+automatically, with zero changes needed to the additivity logic. So once
+these are valid, buildable SQL aggregates, Pivot Table (and every other
+chart) gets correct behavior for free. This is the version of "restore the
+control" that does not reopen the bug that was just fixed.
+
+**Where the actual change needs to land, and what this PR does:**
+
+1. **Done.** `superset-frontend/packages/superset-ui-core/src/query/types/Metric.ts`,
+   extended the `Aggregate` type.
+2. **Done.** `superset-frontend/src/explore/constants.ts`, added to `AGGREGATES`
+   (drives `AGGREGATES_OPTIONS`, the dropdown in `AdhocMetricEditPopover`).
+3. **Done**, but not consolidated. `superset/connectors/sqla/models.py`
+   (`sqla_aggregations`) and `superset/models/helpers.py`
+   (`ExploreMixin.sqla_aggregations`) are both wired to consult the new
+   `BaseEngineSpec.get_extended_aggregation_func`, in addition to their
+   existing 6-aggregate dict, so neither's original, already-tested behavior
+   changed. They remain two separate dicts, consolidating them into one
+   source of truth is left as a follow-up (see Open questions).
+4. **Done**, and it surfaced a second, smaller bug on top of the one this SIP
+   opened with: MCP's own aggregate names (`STDDEV`, `VAR`) never matched any
+   real Superset aggregate, before or after this PR, they were always going
+   to error regardless of what this SIP does. `superset/mcp_service/chart/*`
+   now accepts the old shorthand as an alias, normalized to the real,
+   unambiguous names (`STDDEV_SAMP`, `VAR_SAMP`) this PR ships, and the guided
+   prompt text points at the correct names going forward. `MEDIAN`/
+   `PERCENTILE` were already spelled correctly in MCP; `PERCENTILE` remains
+   unimplemented (it needs a parameter this schema has no field for) and is
+   unchanged by this PR, out of scope here.
+
+**The part that needs real engineering care, this must not be a blind
+`sa.func.MEDIAN` / `sa.func.STDDEV_SAMP` / `sa.func.VAR_SAMP`:**
+
+`sqla_aggregations` today is a flat, engine-unaware dict (`sa.func.AVG`,
+etc., SQLAlchemy emits whatever function name it is given, with zero
+validation that the target dialect actually has it). Superset already has
+precedent for exactly this class of per-engine capability difference:
+`BaseEngineSpec.supports_grouping_sets` and `_time_grain_expressions`, both
+introduced by #41184 itself. This SIP proposes the same shape, a new
+per-engine-overridable mechanism (for example
+`BaseEngineSpec.get_aggregate_sql(aggregate, column)` with a sensible
+default, overridden per engine spec where the default does not hold),
+rather than a single hardcoded dict.
+
+Verified findings so far (via `sqlglot.transpile`, cross-checked against
+known engine docs; **not** exhaustively tested against live databases, that
+is necessary follow-up work this SIP alone cannot complete):
+
+| Engine | `MEDIAN(x)` | `STDDEV_SAMP(x)` | `VAR_SAMP(x)` |
+|---|---|---|---|
+| Postgres | `PERCENTILE_CONT(0.5) WITHIN GROUP (ORDER BY x)` | native | native |
+| MySQL | no native equivalent, needs explicit "unsupported" handling, not a blind emit | native | MySQL's `VARIANCE()` is an alias for `VAR_POP` (population), not `VAR_SAMP` (sample); a naive dialect-name substitution would silently compute the wrong statistic and needs an explicit, verified expression instead |
+| SQLite | only if the specific build was compiled with the (non-default) `SQLITE_ENABLE_PERCENTILE` extension (added in SQLite 3.43, 2023), cannot be assumed available | not available in core SQLite | not available in core SQLite |
+| BigQuery / Snowflake / DuckDB / Redshift / Oracle / T-SQL / Databricks / Spark | native `MEDIAN(x)` | native | native on BigQuery/Snowflake/Databricks/Spark, where `VARIANCE` is correctly sample variance; T-SQL has no function named `VARIANCE` at all and needs `VAR(x)` instead |
+| Trino / Presto / Hive | `PERCENTILE_CONT` / `approx_percentile` (dialect- and exactness-dependent) | native | `variance` is correctly sample variance per Trino/Presto docs |
+
+This table is deliberately not exhaustive, Superset has roughly 75
+`db_engine_specs` files. The proposed default (`BaseEngineSpec`) should be
+the safe choice (mark unsupported, surface a clear user-facing error) rather
+than an optimistic one, with individual engine specs opting in once
+verified. Ship for the handful of engines above first, extend
+opportunistically.
+
+**`Count Unique Values`, `First`, `Last`, `List Unique Values`, explicitly
+out of scope for this SIP:**
+
+- `Count Unique Values` needs no work, it is already `COUNT_DISTINCT`.
+- `First`/`Last` have no well-defined, unambiguous meaning as a plain
+  `GROUP BY` aggregate without an explicit ordering; most engines only
+  support this via window functions (`FIRST_VALUE`/`LAST_VALUE` `OVER
+  (ORDER BY ...)`) or do not support it as a simple aggregate at all
+  (Postgres has neither built in). Restoring this properly would mean
+  designing an "order by" sub-control on the metric, a real, separate
+  feature, not a one-line aggregate addition. Proposed as a follow-up SIP if
+  there is demand.
+- `List Unique Values` maps to the `STRING_AGG`/`GROUP_CONCAT`/`LISTAGG`/
+  `ARRAY_AGG(DISTINCT ...)` family, real dialect differences, plus an open
+  UX question (unbounded cell content for high-cardinality columns).
+  Proposed as a follow-up SIP.
+
+## New or changed public interfaces
+
+- New `Aggregate` values (`MEDIAN`, `STDDEV_SAMP`, `VAR_SAMP`) selectable
+  anywhere the standard metric control appears, every chart type, not just
+  Pivot Table.
+- New `BaseEngineSpec` extensibility point for per-engine aggregate SQL
+  generation (exact shape TBD in implementation, likely mirrors
+  `_time_grain_expressions`).
+- No REST API surface changes beyond the existing metric aggregate field
+  accepting new values.
+
+## Migration plan and compatibility
+
+No new tables/columns needed for the aggregate addition itself.
+
+Restoring prior chart settings, the way #42761 restored `show_values_as` for
+charts that had it before #41184, is murkier here than it was for that PR
+and needs its own design pass: the old `aggregate_function` was a single
+Pivot-Table-level setting applied uniformly to every metric on the chart,
+not a per-metric property. A chart that had `aggregate_function: Median`
+before #41184, with a metric of `SUM(sales)`, was already silently wrong
+under the old architecture (that is the bug that was fixed); mechanically
+rewriting its metric to `MEDIAN(sales)` on upgrade would change what the
+chart's leaf cells display, not just its totals, which may not match user
+intent. This SIP proposes a best-effort, flagged-for-review migration
+(surface affected charts to admins rather than silently rewriting them)
+rather than a fully automatic one-to-one restoration.
+
+## Rejected alternatives
+
+- **Restoring the old `aggregateFunction` Pivot-Table control as-is.**
+  Rejected: this is the literal mechanism SIP-216 removed because it
+  reintroduces incorrect totals for non-additive metrics. Any fix has to go
+  through the metric's own aggregate, not a separate pivot-level override.
+- **Routing all metric SQL generation through `sqlglot` expression-building
+  instead of SQLAlchemy's `sa.func`.** More architecturally thorough (would
+  give correct dialect syntax for free across more of the roughly 75 engine
+  specs), but a much larger, more invasive change to a hot path used by
+  every chart query. Noted as a possible future direction, not this SIP's
+  scope; this SIP proposes the smaller, `supports_grouping_sets`-shaped
+  extensibility point instead.
+
+## Open questions
+
+- **Resolved for this PR, worth confirming as the community's preferred
+  shape:** implemented as `BaseEngineSpec._extended_aggregations` (a
+  `{aggregate_name: sqla_column -> sqla_column}` dict) plus a
+  `get_extended_aggregation_func` accessor, set on the concrete or shared
+  base engine spec class per engine (e.g. on `PostgresBaseEngineSpec` so
+  Redshift inherits it, but *not* on `PrestoBaseEngineSpec` so Hive/Spark/
+  Databricks don't silently inherit unverified behavior, mirroring how
+  `supports_grouping_sets` is opted into per-concrete-engine there today).
+  Did not route through the `superset/sql/dialects/` sqlglot-based layer;
+  that layer is for SQL Lab parsing, wiring it into chart-metric query
+  building felt like a separate, larger change from this SIP's scope.
+- **Still open, not addressed in this PR:** how aggressively should
+  `MEDIAN` degrade on engines without a native or exact equivalent?
+  Trino/Presto/Hive were left unimplemented (unsupported) specifically to
+  avoid silently answering this with an approximate function
+  (`approx_percentile`) that changes the semantics of what a user asked
+  for. If someone wants `MEDIAN` on those engines, this needs a real
+  decision: require explicit opt-in, show a UI warning, or keep it
+  disallowed.
+- **Resolved for this PR:** left the two `sqla_aggregations` dicts
+  (`connectors/sqla/models.py` and `models/helpers.py`) unconsolidated,
+  both now separately wired to the same new `get_extended_aggregation_func`
+  hook. Consolidating them into one source of truth is real but unrelated
+  cleanup, not bundled here to keep the diff reviewable.
+- **New, from implementation:** only Postgres, MySQL (partial), DuckDB, and
+  Redshift (by inheritance, unverified) ship enabled. BigQuery, Snowflake,
+  Trino, Presto, Hive, Spark, Databricks, Oracle, and T-SQL all have
+  documented (not live-verified) support per the table above but are not
+  yet wired up, each needs the same live-instance verification treatment
+  before being enabled, this PR intentionally didn't guess.

docs/src/pages/community.tsx

@@ -67,7 +67,7 @@ const communityLinks = [
       'Join our monthly virtual meetups and register for any upcoming events on Meetup',
   },
   {
-    url: 'https://github.com/apache/superset/blob/master/RESOURCES/INTHEWILD.md',
+    url: 'https://superset.apache.org/inTheWild/',
     title: 'Organizations',
     description:
       'A list of some of the organizations using Superset in production.',

docs/static/feature-flags.json

@@ -93,12 +93,6 @@
         "lifecycle": "development",
         "description": "Enable semantic layers and show semantic views alongside datasets"
       },
-      {
-        "name": "SOFT_DELETE",
-        "default": true,
-        "lifecycle": "development",
-        "description": "Temporary rollout / kill-switch gate for soft delete (off = legacy hard delete). An emergency stop, not a clean rollback: flipping ON->OFF resurrects already-soft-deleted rows. Retained through this release as the move-back lever; removed (along with its two gate points \u2014 BaseDAO.delete routing and the do_orm_execute visibility listener) once post-flip confidence is established."
-      },
       {
         "name": "TABLE_V2_TIME_COMPARISON_ENABLED",
         "default": false,
@@ -110,12 +104,6 @@
         "default": false,
         "lifecycle": "development",
         "description": "Enables the tagging system for organizing assets"
-      },
-      {
-        "name": "VERSION_HISTORY",
-        "default": true,
-        "lifecycle": "development",
-        "description": "Enables the version history panel on Explore and Dashboard pages. History only accrues while ``ENABLE_VERSIONING_CAPTURE`` is also on; with capture off the panel renders but stays empty, so the two ship with matching defaults and should be changed together."
       }
     ],
     "testing": [
@@ -132,6 +120,12 @@
         "lifecycle": "testing",
         "description": "Enables filter functionality in Alerts and Reports"
       },
+      {
+        "name": "ALERT_REPORTS_RETRY",
+        "default": false,
+        "lifecycle": "testing",
+        "description": "Enables automatic retry functionality for failed report executions"
+      },
       {
         "name": "ALERT_REPORT_SLACK_V2",
         "default": true,
@@ -233,6 +227,12 @@
         "lifecycle": "testing",
         "description": "Apply RLS rules to SQL Lab queries. Requires query parsing/manipulation. May break queries or allow RLS bypass. Use with care!"
       },
+      {
+        "name": "SOFT_DELETE",
+        "default": true,
+        "lifecycle": "testing",
+        "description": "Temporary rollout / kill-switch gate for soft delete (off = legacy hard delete). An emergency stop, not a clean rollback: flipping ON->OFF resurrects already-soft-deleted rows. Retained through this release as the move-back lever; removed (along with its two gate points \u2014 BaseDAO.delete routing and the do_orm_execute visibility listener) once post-flip confidence is established."
+      },
       {
         "name": "SSH_TUNNELING",
         "default": false,
@@ -245,6 +245,12 @@
         "default": false,
         "lifecycle": "testing",
         "description": "Use analogous colors in charts"
+      },
+      {
+        "name": "VERSION_HISTORY",
+        "default": true,
+        "lifecycle": "testing",
+        "description": "Enables the version history panel on Explore and Dashboard pages. History only accrues while ``ENABLE_VERSIONING_CAPTURE`` is also on; with capture off the panel renders empty or stale history, so the two ship with matching defaults and should be changed together."
       }
     ],
     "stable": [

docs/static/resources/openapi.json

@@ -3407,22 +3407,26 @@
             "nullable": true,
             "type": "string"
           },
+          "description": {
+            "nullable": true,
+            "type": "string"
+          },
+          "editors": {
+            "items": {
+              "$ref": "#/components/schemas/DashboardRestApi.get_list.Subject"
+            },
+            "type": "array"
+          },
           "id": {
             "type": "integer"
           },
           "is_managed_externally": {
             "type": "boolean"
           },
-          "owners": {
-            "$ref": "#/components/schemas/DashboardRestApi.get_list.User2"
-          },
           "published": {
             "nullable": true,
             "type": "boolean"
           },
-          "roles": {
-            "$ref": "#/components/schemas/DashboardRestApi.get_list.Role"
-          },
           "slug": {
             "maxLength": 255,
             "nullable": true,
@@ -3432,10 +3436,10 @@
             "readOnly": true
           },
           "tags": {
-            "$ref": "#/components/schemas/DashboardRestApi.get_list.Tag"
-          },
-          "thumbnail_url": {
-            "readOnly": true
+            "items": {
+              "$ref": "#/components/schemas/DashboardRestApi.get_list.Tag"
+            },
+            "type": "array"
           },
           "url": {
             "readOnly": true
@@ -3444,21 +3448,46 @@
             "format": "uuid",
             "nullable": true,
             "type": "string"
+          },
+          "viewers": {
+            "items": {
+              "$ref": "#/components/schemas/DashboardRestApi.get_list.Subject1"
+            },
+            "type": "array"
           }
         },
         "type": "object"
       },
-      "DashboardRestApi.get_list.Role": {
+      "DashboardRestApi.get_list.Subject": {
         "properties": {
           "id": {
             "type": "integer"
           },
-          "name": {
-            "maxLength": 64,
+          "label": {
+            "maxLength": 255,
             "type": "string"
+          },
+          "type": {
+            "type": "integer"
           }
         },
-        "required": ["name"],
+        "required": ["label", "type"],
+        "type": "object"
+      },
+      "DashboardRestApi.get_list.Subject1": {
+        "properties": {
+          "id": {
+            "type": "integer"
+          },
+          "label": {
+            "maxLength": 255,
+            "type": "string"
+          },
+          "type": {
+            "type": "integer"
+          }
+        },
+        "required": ["label", "type"],
         "type": "object"
       },
       "DashboardRestApi.get_list.Tag": {
@@ -3511,23 +3540,6 @@
         "required": ["first_name", "last_name"],
         "type": "object"
       },
-      "DashboardRestApi.get_list.User2": {
-        "properties": {
-          "first_name": {
-            "maxLength": 64,
-            "type": "string"
-          },
-          "id": {
-            "type": "integer"
-          },
-          "last_name": {
-            "maxLength": 64,
-            "type": "string"
-          }
-        },
-        "required": ["first_name", "last_name"],
-        "type": "object"
-      },
       "DashboardRestApi.post": {
         "properties": {
           "certification_details": {
@@ -16506,7 +16518,7 @@
                     },
                     "result": {
                       "items": {
-                        "type": "object"
+                        "$ref": "#/components/schemas/DashboardRestApi.get_list"
                       },
                       "type": "array"
                     }

pyproject.toml

@@ -42,7 +42,7 @@ dependencies = [
     # ``google-auth`` 2.53+ dropped it, so Superset must declare it
     # explicitly to keep fresh ``pip install apache-superset`` working
     # without the ``base.txt`` lock file (#40962).
-    "cachetools>=7.1.6, <8",
+    "cachetools>=7.1.7, <8",
     "celery>=5.6.3, <6.0.0",
     "click>=8.4.2",
     "click-option-group",
@@ -75,7 +75,7 @@ dependencies = [
     "humanize",
     "isodate",
     "jsonpath-ng>=1.8.0, <2",
-    "Mako>=1.2.2",
+    "Mako>=1.4.1",
     "markdown>=3.10.3",
     # marshmallow 4 compatibility: see superset/marshmallow_compatibility.py for a
     # Flask-AppBuilder workaround. Tracking issue:
@@ -94,16 +94,16 @@ dependencies = [
     "parsedatetime",
     "paramiko>=3.4.0, <4.0", # 4.0 removed DSSKey, still referenced by sshtunnel
     "pgsanity",
-    "Pillow>=11.0.0, <13",
+    "Pillow>=12.3.0, <13", # raise floor to match resolved pin; closes SCA false-positive on 11.x-range CVEs already fixed in 12.3.0
     "polyline>=2.0.4, <3.0",
     "pydantic>=2.8.0",
     "pyparsing>=3.3.2, <4",
     "python-dateutil",
     "python-dotenv", # optional dependencies for Flask but required for Superset, see https://flask.palletsprojects.com/en/stable/installation/#optional-dependencies
     "pygeohash",
-    "pyarrow>=24.0.0, <26", # before upgrading pyarrow, check that all db dependencies support this, see e.g. https://github.com/apache/superset/pull/34693
+    "pyarrow>=25.0.1, <26", # before upgrading pyarrow, check that all db dependencies support this, see e.g. https://github.com/apache/superset/pull/34693
     "pyyaml>=6.0.3, <7.0.0",
-    "PyJWT>=2.4.0, <3.0",
+    "PyJWT>=2.13.0, <3.0", # raise floor to match resolved pin; closes SCA false-positive on 2.4.x-range CVEs already fixed in 2.13.0
     "redis>=5.0.0, <9.0",
     "rison>=2.0.1, <3.0",
 
@@ -111,10 +111,10 @@ dependencies = [
     "sshtunnel>=0.4.0, <0.5",
     "simplejson>=4.1.1",
     "slack_sdk>=3.43.0, <4",
-    "sqlalchemy>=2.0.0, <2.1",
+    "sqlalchemy>=2.0.52, <2.1",
     "sqlalchemy-continuum>=1.6.0, <2.0.0",
     "sqlalchemy-utils>=0.42.1, <0.43", # expanding lowerbound to work with pydoris
-    "sqlglot>=30.16.0, <31", # 30.16.0 adds Trino inline UDF IF/CASE routine statement parsing
+    "sqlglot>=30.17.0, <31", # 30.16.0 adds Trino inline UDF IF/CASE routine statement parsing
     # newer pandas needs 0.9+
     "tabulate>=0.10.0, <1.0",
     "typing-extensions>=4.16.0, <5",
@@ -135,14 +135,23 @@ athena = ["pyathena[pandas]>=3.35.4, <4"]
 # superset/db_engine_specs/aurora.py's known_incompatibilities metadata.
 aurora-data-api = ["preset-sqlalchemy-aurora-data-api>=0.2.8,<0.3"]
 bigquery = [
-    "pandas-gbq>=0.35.0",
+    "pandas-gbq>=0.35.1",
     # 1.17.1 is likely the final release: googleapis/python-bigquery-sqlalchemy
     # was archived 2026-05-16. Both 1.17.0 and 1.17.1 support SQLAlchemy 1.4/2.0.
-    "sqlalchemy-bigquery>=1.17.1",
+    "sqlalchemy-bigquery>=1.17.2",
     "google-cloud-bigquery>=3.42.3",
 ]
-clickhouse = ["clickhouse-connect>=1.6.0, <2.0"]
-cockroachdb = ["cockroachdb>=0.3.5, <0.4"]
+clickhouse = ["clickhouse-connect>=1.7.2, <2.0"]
+# The `cockroachdb` PyPI package (last released 2021) is abandoned and its
+# SQLAlchemy dialect cannot even import under SQLAlchemy 2.0 (it references
+# sqlalchemy.dialects.postgresql.psycopg2.PGCompiler_psycopg2, removed in
+# 2.0). sqlalchemy-cockroachdb is the actively maintained replacement,
+# already linked from CockroachDbEngineSpec.metadata's docs_url, and
+# registers the same `cockroachdb` SQLAlchemy dialect entry point.
+# sqlalchemy-cockroachdb itself declares no DBAPI dependency (its own docs
+# require picking one), so pull in the same psycopg2-binary pin as the
+# `postgres` extra -- CockroachDB speaks the Postgres wire protocol.
+cockroachdb = ["sqlalchemy-cockroachdb>=2.0.0, <3", "psycopg2-binary==2.9.12"]
 crate = ["sqlalchemy-cratedb>=0.43.1, <1"]
 # sqlalchemy-d1's only release (0.1.0, Nov 2025) pins sqlalchemy<2,>=1.4,
 # explicitly excluding SQLAlchemy 2.0. See superset/db_engine_specs/d1.py's
@@ -185,11 +194,12 @@ excel = ["xlrd>=2.0.2, <2.1"]
 # installing this extra is only required to actually run exports.
 excel-export = ["boto3"]
 fastmcp = [
-    "fastmcp>=3.4.5,<4.0",
+    "fastmcp>=3.4.7,<4.0",
+    "mcp>=1.29.1,<2.0",
     # tiktoken backs the response-size-guard token estimator. Without
     # it, the middleware falls back to a coarser character-based
     # heuristic that under-counts JSON-heavy MCP responses.
-    "tiktoken>=0.13.0,<1.0",
+    "tiktoken>=0.14.0,<1.0",
 ]
 # sqlalchemy-firebird >=2.0.0 unconditionally requires SQLAlchemy 2.0 on
 # Python >=3.8 (which covers Superset's >=3.11 floor), with no dual-compat
@@ -197,7 +207,7 @@ fastmcp = [
 # landed (discussion #40273).
 firebird = ["sqlalchemy-firebird>=2.2.0"]
 firebolt = ["firebolt-sqlalchemy>=1.1.2, <2"]
-gevent = ["gevent>=26.7.0"]
+gevent = ["gevent>=26.8.0"]
 gsheets = ["shillelagh[gsheetsapi]>=1.4.5, <2"]
 hana = ["hdbcli==2.29.25", "sqlalchemy_hana==3.0.3"]
 hive = [
@@ -212,16 +222,25 @@ impala = ["impyla>=0.24.0, <0.25"]
 # superset/db_engine_specs/kusto.py's known_incompatibilities metadata.
 kusto = ["sqlalchemy-kusto>=3.1.2, <4"]
 kylin = ["kylinpy>=2.8.4, <2.9"]
+# MariaDB is a MySQL fork implementing the same wire protocol - connects via
+# the plain mysql dialect, same driver as mysql.
+mariadb = ["apache-superset[mysql]"]
+monetdb = ["sqlalchemy-monetdb>=2.1.0, <3", "pymonetdb>=1.9.1, <2"]
+mongodb = ["pymongosql>=0.7.3, <1"]
 mssql = ["pymssql>=2.3.13, <3"]
 # motherduck is an alias for duckdb - MotherDuck works via the duckdb driver
 motherduck = ["apache-superset[duckdb]"]
 mysql = ["mysqlclient>=2.2.8, <3"]
 ocient = [
     # Closed-source vendor package with no public changelog; permissive
-    # unpinned sqlalchemy>=1.4 declared, but SQLAlchemy 2.0 support is
-    # unverified. Lower confidence than the other bumps in this PR.
+    # unpinned sqlalchemy>=1.4 declared. Verified compatible with SQLAlchemy
+    # 2.0 against pyocient>=3.9.0 (discussion #40273): dialect construction,
+    # error extraction, and GIS-type sanitization all pass under 2.0.52. Note
+    # pyocient 3.9.0 relocated its geo-type classes from private top-level
+    # names (pyocient._STPoint) to public ones under pyocient.api
+    # (pyocient.api.STPoint), which is unrelated to the SQLAlchemy bump.
     "sqlalchemy-ocient>=3.0.0, <4",
-    "pyocient>=1.0.15, <4",
+    "pyocient>=3.9.0, <4",
     "shapely",
     "geojson",
 ]
@@ -231,8 +250,8 @@ pinot = ["pinotdb>=5.0.0, <10.0.0"]
 playwright = ["playwright>=1.62.0, <2"]
 postgres = ["psycopg2-binary==2.9.12"]
 presto = ["pyhive[presto]>=0.6.5"]
-trino = ["trino>=0.338.0"]
-prophet = ["prophet>=1.3.0, <2"]
+trino = ["trino>=0.339.0"]
+prophet = ["prophet>=1.4.0, <2"]
 # sqlalchemy-redshift cuts hard from SQLAlchemy 1.4-only (0.8.x) to 2.0-only
 # (>=1.0.0) with no dual-compat release. Bumped now that Superset's own
 # SQLAlchemy 2.0 core bump has landed (discussion #40273).
@@ -245,7 +264,7 @@ risingwave = ["sqlalchemy-risingwave>=2.0.0"]
 shillelagh = ["shillelagh[all]>=1.4.5, <2"]
 singlestore = ["sqlalchemy-singlestoredb>=1.2.1, <2"]
 snowflake = ["snowflake-sqlalchemy>=1.11.0, <2"]
-sqlite = ["syntaqlite>=0.7.0,<0.8.0"]
+sqlite = ["syntaqlite>=0.9.0,<0.10.0"]
 spark = [
     "pyhive[hive_pure_sasl]>=0.7",
     "tableschema",
@@ -255,7 +274,10 @@ tdengine = [
     "taospy>=2.8.10",
     "taos-ws-py>=0.7.0"
 ]
-teradata = ["teradatasql>=20.0.0.64"]
+teradata = ["teradatasql>=20.0.0.66"]
+# TimescaleDB is a genuine Postgres extension, not a fork - connects via the
+# plain postgresql dialect, same driver as postgres.
+timescaledb = ["apache-superset[postgres]"]
 thumbnails = [] # deprecated, will be removed in 7.0
 vertica = ["sqlalchemy-vertica-python>= 0.6.3, < 0.7"]
 netezza = ["nzalchemy>= 11.1.2, < 11.2"]
@@ -263,11 +285,18 @@ starrocks = ["starrocks>=1.3.4, <2"]
 doris = ["pydoris>=1.2.0, <2.0.0"]
 oceanbase = ["oceanbase_py>=0.0.1.2"]
 ydb = ["ydb-sqlalchemy>=0.1.22", "ydb-sqlglot-plugin>=0.2.8"]
+# YugabyteDB's YSQL layer is fully Postgres-wire compatible - connects via
+# the plain postgresql dialect, same driver as postgres.
+yugabytedb = ["apache-superset[postgres]"]
 development = [
     # no bounds for apache-superset-extensions-cli until a stable version
     "apache-superset-extensions-cli",
     "boto3",
-    "docker",
+    # 7.0.0 raises `docker.errors.DockerException: ... Not supported URL
+    # scheme http+docker` against the requests/urllib3 versions pinned
+    # elsewhere in this file -- breaks testcontainers (tests/testcontainers/)
+    # before any container even starts. 7.2.0 is confirmed working.
+    "docker>=7.2.0",
     "flask-testing",
     "freezegun",
     "grpcio>=1.82.1",
@@ -289,7 +318,7 @@ development = [
     "ruff",
     "sqloxide",
     "statsd",
-    "syntaqlite>=0.7.0,<0.8.0",
+    "syntaqlite>=0.9.0,<0.10.0",
 ]
 
 [project.urls]

pytest.ini

@@ -19,7 +19,13 @@ testpaths =
     tests
 python_files = *_test.py test_*.py *_tests.py *viz/utils.py
 # `-p no:warnings` temporarily disabled in favor of more finely tuned `filterwarnings`.
-#addopts = -p no:warnings
+# `not testcontainers` excludes tests/testcontainers/ by default: those spin up
+# real Docker containers, and `testpaths = tests` would otherwise pull them into
+# every plain `pytest` run. The dedicated CI job (testcontainers.yml) overrides
+# this with an explicit `-m testcontainers` to run them.
+addopts = -m "not testcontainers"
+markers =
+    testcontainers: exercises a real database via testcontainers-python (needs Docker); excluded by default, see .github/workflows/testcontainers.yml
 asyncio_mode = auto
 
 # `ignore` is effectively equivalent to `-p no:warnings`.

requirements/base.in

@@ -28,9 +28,9 @@ numexpr>=2.9.0
 # Security: CVE-2026-34073 (MEDIUM) - Improper Certificate Validation
 cryptography>=50.0.0,<51.0.0
 # Security: Snyk - XSS vulnerability in Mako templates
-mako>=1.3.11,<2.0.0
+mako>=1.4.1,<2.0.0
 # Security: CVE-2024-52338 (CRITICAL) - Deserialization of untrusted data in IPC/Parquet readers
-pyarrow>=24.0.0,<26.0.0
+pyarrow>=25.0.1,<26.0.0
 # Security: CVE-2026-27459 - pyopenssl certificate validation
 pyopenssl>=26.0.0,<27.0.0
 # Security: CVE-2026-25645 (MEDIUM) - Insecure Temporary File

requirements/base.txt

@@ -44,7 +44,7 @@ cachelib==0.13.0
     # via
     #   flask-caching
     #   flask-session
-cachetools==7.1.6
+cachetools==7.1.7
     # via apache-superset (pyproject.toml)
 cattrs==25.1.1
     # via requests-cache
@@ -168,11 +168,11 @@ greenlet==3.5.4
     #   apache-superset (pyproject.toml)
     #   shillelagh
     #   sqlalchemy
-gunicorn==26.0.0
+gunicorn==26.2.0
     # via apache-superset (pyproject.toml)
 hashids==1.3.1
     # via apache-superset (pyproject.toml)
-holidays==0.102
+holidays==0.103
     # via apache-superset (pyproject.toml)
 humanize==4.12.3
     # via apache-superset (pyproject.toml)
@@ -207,7 +207,7 @@ kombu==5.6.2
     # via celery
 limits==5.1.0
     # via flask-limiter
-mako==1.3.12
+mako==1.4.1
     # via
     #   -r requirements/base.in
     #   apache-superset (pyproject.toml)
@@ -222,7 +222,7 @@ markupsafe==3.0.2
     #   mako
     #   werkzeug
     #   wtforms
-marshmallow==4.3.0
+marshmallow==4.3.1
     # via
     #   apache-superset (pyproject.toml)
     #   flask-appbuilder
@@ -263,7 +263,6 @@ packaging==25.0
     #   apache-superset (pyproject.toml)
     #   apispec
     #   deprecation
-    #   gunicorn
     #   kombu
     #   limits
     #   shillelagh
@@ -287,7 +286,7 @@ prison==0.2.1
     # via flask-appbuilder
 prompt-toolkit==3.0.51
     # via click-repl
-pyarrow==25.0.0
+pyarrow==25.0.1
     # via
     #   -r requirements/base.in
     #   apache-superset (pyproject.toml)
@@ -381,7 +380,7 @@ six==1.17.0
     #   wtforms-json
 slack-sdk==3.43.0
     # via apache-superset (pyproject.toml)
-sqlalchemy==2.0.51
+sqlalchemy==2.0.52
     # via
     #   apache-superset (pyproject.toml)
     #   alembic
@@ -399,7 +398,7 @@ sqlalchemy-utils==0.42.1
     #   apache-superset (pyproject.toml)
     #   apache-superset-core
     #   flask-appbuilder
-sqlglot==30.16.0
+sqlglot==30.17.0
     # via
     #   apache-superset (pyproject.toml)
     #   apache-superset-core

requirements/development.in

@@ -16,5 +16,32 @@
 # specific language governing permissions and limitations
 # under the License.
 #
--e .[development,bigquery,druid,duckdb,fastmcp,gevent,gsheets,mysql,postgres,presto,prophet,trino,thumbnails]
+-e .[development,bigquery,clickhouse,cockroachdb,crate,databend,druid,duckdb,elasticsearch,fastmcp,firebird,gevent,gsheets,monetdb,mongodb,mssql,mysql,oracle,postgres,presto,prophet,risingwave,starrocks,trino,thumbnails,ydb]
 -e ./superset-extensions-cli[test]
+# testcontainers-backed db_engine_specs tests (tests/testcontainers/) --
+# see .github/workflows/testcontainers.yml
+#
+# `db2` (the `ibm-db-sa`/`ibm-db` driver) and `oceanbase` (the `oceanbase_py`
+# driver) are both deliberately left out of the baseline dev install above:
+# `ibm-db` ships no Linux arm64 wheel, breaking the multi-platform
+# (amd64+arm64) dev Docker image build; `oceanbase_py` pins
+# `sqlalchemy-utils>=0.38.3,<0.39`, which conflicts outright with Superset's
+# own `sqlalchemy-utils==0.42.1` pin -- there's no version of both that can
+# coexist in one resolved environment. Both testcontainers CI jobs install
+# their driver on demand instead, only for their own matrix leg -- see
+# .github/workflows/testcontainers.yml.
+#
+# mariadb/timescaledb/yugabytedb need no testcontainers extra of their own:
+# they reuse the postgres/mysql container classes pointed at a different
+# image, and psycopg2-binary/mysqlclient are already pulled in above via
+# the postgres/mysql extras. Plain postgres/mysql obviously need nothing
+# extra either. clickhouse and starrocks also need no testcontainers extra:
+# ClickHouseContainer has no driver import of its own (clickhouse-connect,
+# pulled in above via the clickhouse extra, is all the test needs), and
+# StarRocks has no dedicated testcontainers module at all -- its test uses
+# a generic DockerContainer plus the same mysqlclient the mysql extra
+# already provides. databend/risingwave/firebird/ydb are the same story:
+# none has a dedicated testcontainers module, so each test uses a generic
+# DockerContainer plus whatever driver its own extra above already
+# provides.
+testcontainers[cockroachdb,cratedb,mongodb,mssql,mysql,oracle,postgres,trino]>=4.15.0,<5

requirements/development.txt

@@ -12,10 +12,17 @@
     #   apache-superset
 aiofile==3.9.0
     # via py-key-value-aio
+aiohappyeyeballs==2.7.1
+    # via aiohttp
+aiohttp==3.14.3
+    # via ydb
+aiosignal==1.4.0
+    # via aiohttp
 alembic==1.15.2
     # via
     #   -c requirements/base-constraint.txt
     #   flask-migrate
+    #   starrocks
 amqp==5.3.1
     # via
     #   -c requirements/base-constraint.txt
@@ -24,6 +31,8 @@ annotated-types==0.7.0
     # via
     #   -c requirements/base-constraint.txt
     #   pydantic
+antlr4-python3-runtime==4.13.2
+    # via pymongosql
 anyio==4.11.0
     # via
     #   httpx
@@ -42,9 +51,12 @@ apsw==3.50.1.0
     #   shillelagh
 astroid==3.3.10
     # via pylint
+asyncmy2==0.2.21
+    # via starrocks
 attrs==25.3.0
     # via
     #   -c requirements/base-constraint.txt
+    #   aiohttp
     #   cattrs
     #   cyclopts
     #   jsonschema
@@ -65,6 +77,7 @@ backports-tarfile==1.2.0
 backports-zstd==1.6.0
     # via
     #   -c requirements/base-constraint.txt
+    #   clickhouse-connect
     #   flask-compress
 bcrypt==4.3.0
     # via
@@ -99,7 +112,7 @@ cachelib==0.13.0
     #   -c requirements/base-constraint.txt
     #   flask-caching
     #   flask-session
-cachetools==7.1.6
+cachetools==7.1.7
     # via
     #   -c requirements/base-constraint.txt
     #   apache-superset
@@ -117,8 +130,11 @@ celery==5.6.3
 certifi==2026.5.20
     # via
     #   -c requirements/base-constraint.txt
+    #   clickhouse-connect
+    #   elasticsearch
     #   httpcore
     #   httpx
+    #   opensearch-py
     #   requests
 cffi==2.0.0
     # via
@@ -160,6 +176,8 @@ click-repl==0.3.0
     # via
     #   -c requirements/base-constraint.txt
     #   celery
+clickhouse-connect==1.7.2
+    # via apache-superset
 cmdstanpy==1.1.0
     # via prophet
 colorama==0.4.6
@@ -171,6 +189,8 @@ contourpy==1.0.7
     # via matplotlib
 coverage==7.6.8
     # via pytest-cov
+crate==2.2.1
+    # via sqlalchemy-cratedb
 cron-descriptor==1.4.5
     # via
     #   -c requirements/base-constraint.txt
@@ -186,14 +206,20 @@ cryptography==50.0.0
     #   authlib
     #   google-auth
     #   joserfc
+    #   oracledb
     #   paramiko
     #   pyjwt
+    #   pymysql
     #   pyopenssl
     #   secretstorage
 cycler==0.12.1
     # via matplotlib
 cyclopts==4.2.4
     # via fastmcp-slim
+databend-driver==0.34.2
+    # via databend-sqlalchemy
+databend-sqlalchemy==0.5.5
+    # via apache-superset
 db-dtypes==1.3.1
     # via pandas-gbq
 defusedxml==0.7.1
@@ -216,8 +242,11 @@ dnspython==2.7.0
     # via
     #   -c requirements/base-constraint.txt
     #   email-validator
-docker==7.0.0
-    # via apache-superset
+    #   pymongo
+docker==7.2.0
+    # via
+    #   apache-superset
+    #   testcontainers
 docstring-parser==0.17.0
     # via cyclopts
 docutils==0.22.2
@@ -228,6 +257,10 @@ duckdb==1.5.5
     #   duckdb-engine
 duckdb-engine==0.17.0
     # via apache-superset
+elasticsearch==7.17.13
+    # via elasticsearch-dbapi
+elasticsearch-dbapi==0.2.13
+    # via apache-superset
 email-validator==2.2.0
     # via
     #   -c requirements/base-constraint.txt
@@ -237,16 +270,22 @@ et-xmlfile==2.0.0
     # via
     #   -c requirements/base-constraint.txt
     #   openpyxl
+events==0.5
+    # via opensearch-py
 exceptiongroup==1.3.0
     # via fastmcp-slim
-fastmcp==3.4.5
+fastmcp==3.4.7
     # via apache-superset
-fastmcp-slim==3.4.5
+fastmcp-slim==3.4.7
     # via fastmcp
 filelock==3.20.3
     # via
     #   -c requirements/base-constraint.txt
     #   virtualenv
+firebird-base==2.0.3
+    # via firebird-driver
+firebird-driver==2.0.3
+    # via sqlalchemy-firebird
 flask==2.3.3
     # via
     #   -c requirements/base-constraint.txt
@@ -327,17 +366,23 @@ fonttools==4.60.2
     # via matplotlib
 freezegun==1.5.1
     # via apache-superset
+frozenlist==1.8.0
+    # via
+    #   aiohttp
+    #   aiosignal
 future==1.0.0
     # via pyhive
 geographiclib==2.0
     # via
     #   -c requirements/base-constraint.txt
     #   geopy
+geojson==3.3.0
+    # via sqlalchemy-cratedb
 geopy==2.4.1
     # via
     #   -c requirements/base-constraint.txt
     #   apache-superset
-gevent==26.7.0
+gevent==26.8.0
     # via apache-superset
 google-api-core==2.33.0
     # via
@@ -389,9 +434,10 @@ grpcio==1.83.0
     #   apache-superset
     #   google-api-core
     #   grpcio-status
+    #   ydb
 grpcio-status==1.60.1
     # via google-api-core
-gunicorn==26.0.0
+gunicorn==26.2.0
     # via
     #   -c requirements/base-constraint.txt
     #   apache-superset
@@ -403,7 +449,7 @@ hashids==1.3.1
     # via
     #   -c requirements/base-constraint.txt
     #   apache-superset
-holidays==0.102
+holidays==0.103
     # via
     #   -c requirements/base-constraint.txt
     #   apache-superset
@@ -414,6 +460,7 @@ httpx==0.28.1
     # via
     #   fastmcp-slim
     #   mcp
+    #   testcontainers
 httpx-sse==0.4.1
     # via mcp
 humanize==4.12.3
@@ -430,12 +477,11 @@ idna==3.15
     #   httpx
     #   requests
     #   url-normalize
+    #   yarl
 importlib-metadata==8.7.0
     # via
     #   keyring
     #   opentelemetry-api
-importlib-resources==6.5.2
-    # via prophet
 iniconfig==2.0.0
     # via pytest
 isodate==0.7.2
@@ -470,6 +516,7 @@ jmespath==1.1.0
     # via
     #   boto3
     #   botocore
+    #   pymongosql
 joserfc==1.7.2
     # via fastmcp-slim
 jsonpath-ng==1.8.0
@@ -502,6 +549,8 @@ kombu==5.6.2
     # via
     #   -c requirements/base-constraint.txt
     #   celery
+lark==1.3.1
+    # via starrocks
 lazy-object-proxy==1.10.0
     # via openapi-spec-validator
 limits==5.1.0
@@ -509,8 +558,10 @@ limits==5.1.0
     #   -c requirements/base-constraint.txt
     #   flask-limiter
 lz4==4.4.5
-    # via trino
-mako==1.3.12
+    # via
+    #   clickhouse-connect
+    #   trino
+mako==1.4.1
     # via
     #   -c requirements/base-constraint.txt
     #   alembic
@@ -530,7 +581,7 @@ markupsafe==3.0.2
     #   mako
     #   werkzeug
     #   wtforms
-marshmallow==4.3.0
+marshmallow==4.3.1
     # via
     #   -c requirements/base-constraint.txt
     #   apache-superset
@@ -549,8 +600,10 @@ matplotlib==3.9.0
     # via prophet
 mccabe==0.7.0
     # via pylint
-mcp==1.24.0
-    # via fastmcp-slim
+mcp==1.29.1
+    # via
+    #   apache-superset
+    #   fastmcp-slim
 mdurl==0.1.2
     # via
     #   -c requirements/base-constraint.txt
@@ -567,6 +620,10 @@ msgspec==0.19.0
     # via
     #   -c requirements/base-constraint.txt
     #   flask-session
+multidict==6.7.1
+    # via
+    #   aiohttp
+    #   yarl
 mysqlclient==2.2.8
     # via apache-superset
 nh3==0.3.6
@@ -605,14 +662,22 @@ openpyxl==3.1.5
     # via
     #   -c requirements/base-constraint.txt
     #   pandas
+opensearch-py==2.8.0
+    # via elasticsearch-dbapi
 opentelemetry-api==1.39.1
     # via fastmcp-slim
+oracledb==4.0.2
+    # via
+    #   apache-superset
+    #   testcontainers
 ordered-set==4.1.0
     # via
     #   -c requirements/base-constraint.txt
     #   flask-limiter
 orjson==3.11.9
-    # via trino
+    # via
+    #   crate
+    #   trino
 packaging==25.0
     # via
     #   -c requirements/base-constraint.txt
@@ -620,11 +685,10 @@ packaging==25.0
     #   apispec
     #   db-dtypes
     #   deprecation
-    #   docker
     #   duckdb-engine
+    #   elasticsearch-dbapi
     #   fastmcp-slim
     #   google-cloud-bigquery
-    #   gunicorn
     #   kombu
     #   limits
     #   matplotlib
@@ -632,6 +696,8 @@ packaging==25.0
     #   pytest
     #   shillelagh
     #   sqlalchemy-bigquery
+    #   sqlalchemy-firebird
+    #   ydb
 pandas==2.3.3
     # via
     #   -c requirements/base-constraint.txt
@@ -640,7 +706,7 @@ pandas==2.3.3
     #   db-dtypes
     #   pandas-gbq
     #   prophet
-pandas-gbq==0.35.0
+pandas-gbq==0.35.1
     # via apache-superset
 parameterized==0.9.0
     # via apache-superset
@@ -693,16 +759,22 @@ prompt-toolkit==3.0.51
     # via
     #   -c requirements/base-constraint.txt
     #   click-repl
-prophet==1.3.0
+propcache==0.5.2
+    # via
+    #   aiohttp
+    #   yarl
+prophet==1.4.0
     # via apache-superset
 proto-plus==1.25.0
     # via google-api-core
 protobuf==5.29.6
     # via
+    #   firebird-base
     #   google-api-core
     #   googleapis-common-protos
     #   grpcio-status
     #   proto-plus
+    #   ydb
 psutil==6.1.0
     # via
     #   apache-superset
@@ -711,7 +783,7 @@ psycopg2-binary==2.9.12
     # via apache-superset
 py-key-value-aio==0.4.4
     # via fastmcp-slim
-pyarrow==25.0.0
+pyarrow==25.0.1
     # via
     #   -c requirements/base-constraint.txt
     #   apache-superset
@@ -776,6 +848,24 @@ pyjwt==2.13.0
     #   mcp
 pylint==3.3.7
     # via apache-superset
+pymonetdb==1.9.1
+    # via
+    #   apache-superset
+    #   sqlalchemy-monetdb
+pymongo==4.17.0
+    # via
+    #   pymongosql
+    #   testcontainers
+pymongosql==0.7.3
+    # via apache-superset
+pymssql==2.3.13
+    # via
+    #   apache-superset
+    #   testcontainers
+pymysql==1.2.0
+    # via
+    #   starrocks
+    #   testcontainers
 pynacl==1.6.2
     # via
     #   -c requirements/base-constraint.txt
@@ -820,11 +910,13 @@ python-dateutil==2.9.0.post0
     #   botocore
     #   celery
     #   croniter
+    #   firebird-driver
     #   flask-appbuilder
     #   freezegun
     #   google-cloud-bigquery
     #   holidays
     #   matplotlib
+    #   opensearch-py
     #   pandas
     #   pyhive
     #   shillelagh
@@ -835,6 +927,7 @@ python-dotenv==1.2.2
     #   apache-superset
     #   fastmcp-slim
     #   pydantic-settings
+    #   testcontainers
 python-ldap==3.4.7
     # via apache-superset
 python-multipart==0.0.29
@@ -878,6 +971,7 @@ requests==2.33.0
     #   google-api-core
     #   google-cloud-bigquery
     #   jsonschema-path
+    #   opensearch-py
     #   pydruid
     #   pyhive
     #   requests-cache
@@ -950,37 +1044,60 @@ slack-sdk==3.43.0
     #   apache-superset
 sniffio==1.3.1
     # via anyio
-sqlalchemy==2.0.51
+sqlalchemy==2.0.52
     # via
     #   -c requirements/base-constraint.txt
     #   alembic
     #   apache-superset
     #   apache-superset-core
+    #   databend-sqlalchemy
     #   duckdb-engine
+    #   elasticsearch-dbapi
     #   flask-appbuilder
     #   flask-sqlalchemy
     #   marshmallow-sqlalchemy
     #   shillelagh
     #   sqlalchemy-bigquery
+    #   sqlalchemy-cockroachdb
     #   sqlalchemy-continuum
+    #   sqlalchemy-cratedb
+    #   sqlalchemy-firebird
+    #   sqlalchemy-monetdb
+    #   sqlalchemy-risingwave
     #   sqlalchemy-utils
-sqlalchemy-bigquery==1.17.1
+    #   starrocks
+    #   testcontainers
+    #   ydb-sqlalchemy
+sqlalchemy-bigquery==1.17.2
+    # via apache-superset
+sqlalchemy-cockroachdb==2.0.4
     # via apache-superset
 sqlalchemy-continuum==1.7.0
     # via
     #   -c requirements/base-constraint.txt
     #   apache-superset
+sqlalchemy-cratedb==0.43.1
+    # via
+    #   apache-superset
+    #   testcontainers
+sqlalchemy-firebird==2.2.0
+    # via apache-superset
+sqlalchemy-monetdb==2.1.0
+    # via apache-superset
+sqlalchemy-risingwave==2.1.0
+    # via apache-superset
 sqlalchemy-utils==0.42.1
     # via
     #   -c requirements/base-constraint.txt
     #   apache-superset
     #   apache-superset-core
     #   flask-appbuilder
-sqlglot==30.16.0
+sqlglot==30.17.0
     # via
     #   -c requirements/base-constraint.txt
     #   apache-superset
     #   apache-superset-core
+    #   ydb-sqlglot-plugin
 sqloxide==0.1.51
     # via apache-superset
 sse-starlette==3.0.2
@@ -993,15 +1110,19 @@ starlette==1.3.1
     # via
     #   fastmcp-slim
     #   mcp
+starrocks==1.3.4
+    # via apache-superset
 statsd==4.0.1
     # via apache-superset
-syntaqlite==0.7.1
+syntaqlite==0.9.0
     # via apache-superset
 tabulate==0.10.0
     # via
     #   -c requirements/base-constraint.txt
     #   apache-superset
-tiktoken==0.13.0
+testcontainers==4.15.0
+    # via -r requirements/development.in
+tiktoken==0.14.0
     # via apache-superset
 tomli-w==1.2.0
     # via apache-superset-extensions-cli
@@ -1011,11 +1132,15 @@ tqdm==4.67.1
     # via
     #   cmdstanpy
     #   prophet
-trino==0.338.0
-    # via apache-superset
+trino==0.339.0
+    # via
+    #   apache-superset
+    #   testcontainers
 typing-extensions==4.16.0
     # via
     #   -c requirements/base-constraint.txt
+    #   aiohttp
+    #   aiosignal
     #   alembic
     #   anyio
     #   apache-superset
@@ -1027,6 +1152,7 @@ typing-extensions==4.16.0
     #   limits
     #   mcp
     #   opentelemetry-api
+    #   oracledb
     #   py-key-value-aio
     #   pydantic
     #   pydantic-core
@@ -1035,6 +1161,7 @@ typing-extensions==4.16.0
     #   shillelagh
     #   sqlalchemy
     #   starlette
+    #   testcontainers
     #   typing-inspection
 typing-inspection==0.4.2
     # via
@@ -1062,13 +1189,22 @@ urllib3==2.7.0
     # via
     #   -c requirements/base-constraint.txt
     #   botocore
+    #   clickhouse-connect
+    #   crate
     #   docker
+    #   elasticsearch
+    #   opensearch-py
     #   requests
     #   requests-cache
+    #   testcontainers
 uvicorn==0.37.0
     # via
     #   fastmcp-slim
     #   mcp
+verlib2==0.3.2
+    # via
+    #   crate
+    #   sqlalchemy-cratedb
 vine==5.1.0
     # via
     #   -c requirements/base-constraint.txt
@@ -1102,6 +1238,7 @@ wrapt==1.17.2
     # via
     #   -c requirements/base-constraint.txt
     #   deprecated
+    #   testcontainers
 wtforms==3.2.2
     # via
     #   -c requirements/base-constraint.txt
@@ -1122,6 +1259,18 @@ xlsxwriter==3.2.9
     #   -c requirements/base-constraint.txt
     #   apache-superset
     #   pandas
+yarl==1.24.5
+    # via aiohttp
+ydb==3.31.4
+    # via
+    #   ydb-dbapi
+    #   ydb-sqlalchemy
+ydb-dbapi==0.1.23
+    # via ydb-sqlalchemy
+ydb-sqlalchemy==0.1.22
+    # via apache-superset
+ydb-sqlglot-plugin==0.2.8
+    # via apache-superset
 zipp==3.23.0
     # via importlib-metadata
 zope-event==5.0

scripts/benchmark_migration.py

@@ -93,7 +93,7 @@ def find_models(module: ModuleType) -> list[type[Model]]:  # noqa: C901
     # where the current model is out-of-sync with the existing table after a
     # downgrade
     sqlalchemy_uri = current_app.config["SQLALCHEMY_DATABASE_URI"]
-    engine = create_engine(sqlalchemy_uri, future=True)
+    engine = create_engine(sqlalchemy_uri)
     Base = automap_base()  # noqa: N806
     Base.prepare(engine, reflect=True)
     seen = set()

scripts/change_detector.py

@@ -42,6 +42,7 @@ RETRYABLE_STATUS_CODES: frozenset[int] = frozenset({429})
 PATTERNS = {
     "python": [
         r"^\.github/workflows/.*python",
+        r"^\.github/workflows/frontend-bundle-size-nightly\.yml$",
         r"^\.github/workflows/scheduled-docker-image-refresh\.yml$",
         r"^docker-compose-image-tag\.yml$",
         r"^tests/",

scripts/docker-build-extra-flags.sh

@@ -0,0 +1,48 @@
+#!/usr/bin/env bash
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+
+# Computes the `--extra-flags` value passed to `supersetbot docker` for a
+# given build preset. Factored out of .github/workflows/docker.yml so the
+# PY_VER override logic below can be exercised by an always-on CI check
+# (docker.yml's docker-build job only runs when the change detector's
+# docker/python/frontend outputs are true, and the PR build matrix never
+# includes py311/py312 at all, so a regression here would otherwise go
+# unnoticed until the fix actually runs on master) without duplicating -
+# and risking drift from - the logic used by the real build step.
+#
+# supersetbot's "py311"/"py312" presets pin their own --build-arg PY_VER,
+# which lands ahead of --extra-flags on the assembled buildx command line;
+# docker/buildx keeps the last value for a repeated --build-arg key, so
+# appending PY_VER here would override supersetbot's pin and silently make
+# "py311"/"py312" build the exact same image as "lean". Every other preset
+# gets the override so its build lands on the Dockerfile's own supported
+# Python version.
+#
+# Usage: docker-build-extra-flags.sh <build_preset> <image_tag>
+
+set -euo pipefail
+
+BUILD_PRESET="${1:?usage: docker-build-extra-flags.sh <build_preset> <image_tag>}"
+IMAGE_TAG="${2:?usage: docker-build-extra-flags.sh <build_preset> <image_tag>}"
+
+EXTRA_FLAGS="--build-arg INCLUDE_CHROMIUM=false --tag $IMAGE_TAG"
+if [ "$BUILD_PRESET" != "py311" ] && [ "$BUILD_PRESET" != "py312" ]; then
+  EXTRA_FLAGS="--build-arg PY_VER=3.11.14-slim-trixie $EXTRA_FLAGS"
+fi
+
+echo "$EXTRA_FLAGS"

superset-core/src/superset_core/semantic_layers/__init__.py

@@ -0,0 +1,18 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+
+"""Semantic layer contracts for extension authors."""

superset-core/src/superset_core/semantic_layers/types.py

@@ -18,8 +18,9 @@
 from __future__ import annotations
 
 import enum
-from dataclasses import dataclass
+from dataclasses import dataclass, field
 from datetime import date, datetime, time, timedelta
+from typing import Any
 
 import isodate
 import pyarrow as pa
@@ -90,6 +91,8 @@ class Dimension:
     definition: str | None = None
     description: str | None = None
     grain: Grain | None = None
+    verbose_name: str | None = field(default=None, compare=False)
+    metadata: dict[str, Any] = field(default_factory=dict, compare=False)
 
 
 class AggregationType(str, enum.Enum):
@@ -121,6 +124,9 @@ class Metric:
     definition: str
     description: str | None = None
     aggregation: AggregationType | None = None
+    verbose_name: str | None = field(default=None, compare=False)
+    d3format: str | None = field(default=None, compare=False)
+    metadata: dict[str, Any] = field(default_factory=dict, compare=False)
 
 
 @dataclass(frozen=True)

superset-core/tests/semantic_layers/test_types.py

@@ -0,0 +1,97 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+
+import pyarrow as pa
+from superset_core.semantic_layers.types import Dimension, Metric
+
+
+def test_dimension_metadata_is_not_part_of_identity() -> None:
+    first = Dimension(
+        "sales.region",
+        "region",
+        pa.utf8(),
+        verbose_name="Region",
+        metadata={"display_name": "Region"},
+    )
+    second = Dimension(
+        "sales.region",
+        "region",
+        pa.utf8(),
+        verbose_name="Sales region",
+        metadata={"display_name": "Sales region"},
+    )
+
+    assert first == second
+    assert {first, second} == {first}
+
+
+def test_metric_metadata_is_not_part_of_identity() -> None:
+    first = Metric(
+        "sales.total_revenue",
+        "total_revenue",
+        pa.float64(),
+        "SUM(revenue)",
+        verbose_name="Total revenue",
+        d3format="$,.2f",
+        metadata={"unit": {"kind": "currency", "code": "USD"}},
+    )
+    second = Metric(
+        "sales.total_revenue",
+        "total_revenue",
+        pa.float64(),
+        "SUM(revenue)",
+        verbose_name="Revenue",
+        d3format=",.0f",
+        metadata={"unit": {"kind": "currency", "code": "EUR"}},
+    )
+
+    assert first == second
+    assert {first, second} == {first}
+
+
+def test_metric_accepts_superset_presentation_fields() -> None:
+    metric = Metric(
+        "sales.total_revenue",
+        "total_revenue",
+        pa.float64(),
+        "SUM(revenue)",
+        verbose_name="Total revenue",
+        d3format="$,.2f",
+    )
+
+    assert metric.verbose_name == "Total revenue"
+    assert metric.d3format == "$,.2f"
+
+
+def test_dimension_accepts_superset_presentation_fields() -> None:
+    dimension = Dimension(
+        "sales.region",
+        "region",
+        pa.utf8(),
+        verbose_name="Region",
+    )
+
+    assert dimension.verbose_name == "Region"
+
+
+def test_metadata_defaults_are_not_shared() -> None:
+    first = Metric("first", "first", pa.int64(), "COUNT(*)")
+    second = Metric("second", "second", pa.int64(), "COUNT(*)")
+
+    first.metadata["display_name"] = "First"
+
+    assert second.metadata == {}

superset-embedded-sdk/package-lock.json

@@ -19,6 +19,7 @@
         "@babel/preset-typescript": "7.29.7",
         "@types/node": "^25.4.0",
         "babel-loader": "^9.1.3",
+        "jsdom": "^26.1.0",
         "tscw-config": "^1.1.2",
         "typescript": "^5.9.3",
         "vitest": "^4.0.18",
@@ -26,6 +27,27 @@
         "webpack-cli": "^5.1.4"
       }
     },
+    "node_modules/@asamuzakjp/css-color": {
+      "version": "3.2.0",
+      "resolved": "https://registry.npmjs.org/@asamuzakjp/css-color/-/css-color-3.2.0.tgz",
+      "integrity": "sha512-K1A6z8tS3XsmCMM86xoWdn7Fkdn9m6RSVtocUrJYIwZnFVkng/PvkEoWtOWmP+Scc6saYWHWZYbndEEXxl24jw==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "@csstools/css-calc": "^2.1.3",
+        "@csstools/css-color-parser": "^3.0.9",
+        "@csstools/css-parser-algorithms": "^3.0.4",
+        "@csstools/css-tokenizer": "^3.0.3",
+        "lru-cache": "^10.4.3"
+      }
+    },
+    "node_modules/@asamuzakjp/css-color/node_modules/lru-cache": {
+      "version": "10.4.3",
+      "resolved": "https://registry.npmjs.org/lru-cache/-/lru-cache-10.4.3.tgz",
+      "integrity": "sha512-JNAzZcXrCt42VGLuYz0zfAzDfAvJWW6AfYlDBQyDV5DClI2m5sAmK+OIO7s59XfsRsWHp02jAJrRadPRGTt6SQ==",
+      "dev": true,
+      "license": "ISC"
+    },
     "node_modules/@babel/cli": {
       "version": "7.29.7",
       "resolved": "https://registry.npmjs.org/@babel/cli/-/cli-7.29.7.tgz",
@@ -1656,6 +1678,121 @@
         "node": ">=6.9.0"
       }
     },
+    "node_modules/@csstools/color-helpers": {
+      "version": "5.1.0",
+      "resolved": "https://registry.npmjs.org/@csstools/color-helpers/-/color-helpers-5.1.0.tgz",
+      "integrity": "sha512-S11EXWJyy0Mz5SYvRmY8nJYTFFd1LCNV+7cXyAgQtOOuzb4EsgfqDufL+9esx72/eLhsRdGZwaldu/h+E4t4BA==",
+      "dev": true,
+      "funding": [
+        {
+          "type": "github",
+          "url": "https://github.com/sponsors/csstools"
+        },
+        {
+          "type": "opencollective",
+          "url": "https://opencollective.com/csstools"
+        }
+      ],
+      "license": "MIT-0",
+      "engines": {
+        "node": ">=18"
+      }
+    },
+    "node_modules/@csstools/css-calc": {
+      "version": "2.1.4",
+      "resolved": "https://registry.npmjs.org/@csstools/css-calc/-/css-calc-2.1.4.tgz",
+      "integrity": "sha512-3N8oaj+0juUw/1H3YwmDDJXCgTB1gKU6Hc/bB502u9zR0q2vd786XJH9QfrKIEgFlZmhZiq6epXl4rHqhzsIgQ==",
+      "dev": true,
+      "funding": [
+        {
+          "type": "github",
+          "url": "https://github.com/sponsors/csstools"
+        },
+        {
+          "type": "opencollective",
+          "url": "https://opencollective.com/csstools"
+        }
+      ],
+      "license": "MIT",
+      "engines": {
+        "node": ">=18"
+      },
+      "peerDependencies": {
+        "@csstools/css-parser-algorithms": "^3.0.5",
+        "@csstools/css-tokenizer": "^3.0.4"
+      }
+    },
+    "node_modules/@csstools/css-color-parser": {
+      "version": "3.1.0",
+      "resolved": "https://registry.npmjs.org/@csstools/css-color-parser/-/css-color-parser-3.1.0.tgz",
+      "integrity": "sha512-nbtKwh3a6xNVIp/VRuXV64yTKnb1IjTAEEh3irzS+HkKjAOYLTGNb9pmVNntZ8iVBHcWDA2Dof0QtPgFI1BaTA==",
+      "dev": true,
+      "funding": [
+        {
+          "type": "github",
+          "url": "https://github.com/sponsors/csstools"
+        },
+        {
+          "type": "opencollective",
+          "url": "https://opencollective.com/csstools"
+        }
+      ],
+      "license": "MIT",
+      "dependencies": {
+        "@csstools/color-helpers": "^5.1.0",
+        "@csstools/css-calc": "^2.1.4"
+      },
+      "engines": {
+        "node": ">=18"
+      },
+      "peerDependencies": {
+        "@csstools/css-parser-algorithms": "^3.0.5",
+        "@csstools/css-tokenizer": "^3.0.4"
+      }
+    },
+    "node_modules/@csstools/css-parser-algorithms": {
+      "version": "3.0.5",
+      "resolved": "https://registry.npmjs.org/@csstools/css-parser-algorithms/-/css-parser-algorithms-3.0.5.tgz",
+      "integrity": "sha512-DaDeUkXZKjdGhgYaHNJTV9pV7Y9B3b644jCLs9Upc3VeNGg6LWARAT6O+Q+/COo+2gg/bM5rhpMAtf70WqfBdQ==",
+      "dev": true,
+      "funding": [
+        {
+          "type": "github",
+          "url": "https://github.com/sponsors/csstools"
+        },
+        {
+          "type": "opencollective",
+          "url": "https://opencollective.com/csstools"
+        }
+      ],
+      "license": "MIT",
+      "engines": {
+        "node": ">=18"
+      },
+      "peerDependencies": {
+        "@csstools/css-tokenizer": "^3.0.4"
+      }
+    },
+    "node_modules/@csstools/css-tokenizer": {
+      "version": "3.0.4",
+      "resolved": "https://registry.npmjs.org/@csstools/css-tokenizer/-/css-tokenizer-3.0.4.tgz",
+      "integrity": "sha512-Vd/9EVDiu6PPJt9yAh6roZP6El1xHrdvIVGjyBsHR0RYwNHgL7FJPyIIW4fANJNG6FtyZfvlRPpFI4ZM/lubvw==",
+      "dev": true,
+      "funding": [
+        {
+          "type": "github",
+          "url": "https://github.com/sponsors/csstools"
+        },
+        {
+          "type": "opencollective",
+          "url": "https://opencollective.com/csstools"
+        }
+      ],
+      "license": "MIT",
+      "engines": {
+        "node": ">=18"
+      }
+    },
     "node_modules/@discoveryjs/json-ext": {
       "version": "0.5.6",
       "resolved": "https://registry.npmjs.org/@discoveryjs/json-ext/-/json-ext-0.5.6.tgz",
@@ -2501,6 +2638,16 @@
         "acorn": "^8.14.0"
       }
     },
+    "node_modules/agent-base": {
+      "version": "7.1.4",
+      "resolved": "https://registry.npmjs.org/agent-base/-/agent-base-7.1.4.tgz",
+      "integrity": "sha512-MnA+YT8fwfJPgBx3m60MNqakm30XOkyIoH1y6huTQvC0PwZG7ki8NacLBcrPbNoo8vEZy7Jpuk7+jMO+CUovTQ==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": ">= 14"
+      }
+    },
     "node_modules/ajv": {
       "version": "8.20.0",
       "resolved": "https://registry.npmjs.org/ajv/-/ajv-8.20.0.tgz",
@@ -2868,6 +3015,34 @@
         "node": ">= 8"
       }
     },
+    "node_modules/cssstyle": {
+      "version": "4.6.0",
+      "resolved": "https://registry.npmjs.org/cssstyle/-/cssstyle-4.6.0.tgz",
+      "integrity": "sha512-2z+rWdzbbSZv6/rhtvzvqeZQHrBaqgogqt85sqFNbabZOuFbCVFb8kPeEtZjiKkbrm395irpNKiYeFeLiQnFPg==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "@asamuzakjp/css-color": "^3.2.0",
+        "rrweb-cssom": "^0.8.0"
+      },
+      "engines": {
+        "node": ">=18"
+      }
+    },
+    "node_modules/data-urls": {
+      "version": "5.0.0",
+      "resolved": "https://registry.npmjs.org/data-urls/-/data-urls-5.0.0.tgz",
+      "integrity": "sha512-ZYP5VBHshaDAiVZxjbRVcFJpc+4xGgT0bK3vzy1HLN8jTO975HEbuYzZJcHoQEY5K1a0z8YayJkyVETa08eNTg==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "whatwg-mimetype": "^4.0.0",
+        "whatwg-url": "^14.0.0"
+      },
+      "engines": {
+        "node": ">=18"
+      }
+    },
     "node_modules/debug": {
       "version": "4.4.3",
       "resolved": "https://registry.npmjs.org/debug/-/debug-4.4.3.tgz",
@@ -2886,6 +3061,13 @@
         }
       }
     },
+    "node_modules/decimal.js": {
+      "version": "10.6.0",
+      "resolved": "https://registry.npmjs.org/decimal.js/-/decimal.js-10.6.0.tgz",
+      "integrity": "sha512-YpgQiITW3JXGntzdUmyUR1V812Hn8T1YVXhCu+wO3OpS4eU9l4YdD3qjyiKdV6mvV29zapkMeD390UVEf2lkUg==",
+      "dev": true,
+      "license": "MIT"
+    },
     "node_modules/detect-libc": {
       "version": "2.1.2",
       "resolved": "https://registry.npmjs.org/detect-libc/-/detect-libc-2.1.2.tgz",
@@ -2915,6 +3097,19 @@
         "node": ">=10.13.0"
       }
     },
+    "node_modules/entities": {
+      "version": "6.0.1",
+      "resolved": "https://registry.npmjs.org/entities/-/entities-6.0.1.tgz",
+      "integrity": "sha512-aN97NXWF6AWBTahfVOIrB/NShkzi5H7F9r1s9mD3cDj4Ko5f2qhhVoYMibXF7GlLveb/D2ioWay8lxI97Ven3g==",
+      "dev": true,
+      "license": "BSD-2-Clause",
+      "engines": {
+        "node": ">=0.12"
+      },
+      "funding": {
+        "url": "https://github.com/fb55/entities?sponsor=1"
+      }
+    },
     "node_modules/envinfo": {
       "version": "7.13.0",
       "resolved": "https://registry.npmjs.org/envinfo/-/envinfo-7.13.0.tgz",
@@ -3316,6 +3511,60 @@
         "node": ">= 0.4"
       }
     },
+    "node_modules/html-encoding-sniffer": {
+      "version": "4.0.0",
+      "resolved": "https://registry.npmjs.org/html-encoding-sniffer/-/html-encoding-sniffer-4.0.0.tgz",
+      "integrity": "sha512-Y22oTqIU4uuPgEemfz7NDJz6OeKf12Lsu+QC+s3BVpda64lTiMYCyGwg5ki4vFxkMwQdeZDl2adZoqUgdFuTgQ==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "whatwg-encoding": "^3.1.1"
+      },
+      "engines": {
+        "node": ">=18"
+      }
+    },
+    "node_modules/http-proxy-agent": {
+      "version": "7.0.2",
+      "resolved": "https://registry.npmjs.org/http-proxy-agent/-/http-proxy-agent-7.0.2.tgz",
+      "integrity": "sha512-T1gkAiYYDWYx3V5Bmyu7HcfcvL7mUrTWiM6yOfa3PIphViJ/gFPbvidQ+veqSOHci/PxBcDabeUNCzpOODJZig==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "agent-base": "^7.1.0",
+        "debug": "^4.3.4"
+      },
+      "engines": {
+        "node": ">= 14"
+      }
+    },
+    "node_modules/https-proxy-agent": {
+      "version": "7.0.6",
+      "resolved": "https://registry.npmjs.org/https-proxy-agent/-/https-proxy-agent-7.0.6.tgz",
+      "integrity": "sha512-vK9P5/iUfdl95AI+JVyUuIcVtd4ofvtrOr3HNtM2yxC9bnMbEdp3x01OhQNnjb8IJYi38VlTE3mBXwcfvywuSw==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "agent-base": "^7.1.2",
+        "debug": "4"
+      },
+      "engines": {
+        "node": ">= 14"
+      }
+    },
+    "node_modules/iconv-lite": {
+      "version": "0.6.3",
+      "resolved": "https://registry.npmjs.org/iconv-lite/-/iconv-lite-0.6.3.tgz",
+      "integrity": "sha512-4fCk79wshMdzMp2rH06qWrJE4iolqLhCUH+OiuIgU++RB0+94NlDL81atO7GX55uUKueo0txHNtvEyI6D7WdMw==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "safer-buffer": ">= 2.1.2 < 3.0.0"
+      },
+      "engines": {
+        "node": ">=0.10.0"
+      }
+    },
     "node_modules/import-local": {
       "version": "3.1.0",
       "resolved": "https://registry.npmjs.org/import-local/-/import-local-3.1.0.tgz",
@@ -3438,6 +3687,13 @@
         "node": ">=0.10.0"
       }
     },
+    "node_modules/is-potential-custom-element-name": {
+      "version": "1.0.1",
+      "resolved": "https://registry.npmjs.org/is-potential-custom-element-name/-/is-potential-custom-element-name-1.0.1.tgz",
+      "integrity": "sha512-bCYeRA2rVibKZd+s2625gGnGF/t7DSqDs4dP7CrLA1m7jKWz6pps0LpYLJN8Q64HtmPKJ1hrN3nzPNKFEKOUiQ==",
+      "dev": true,
+      "license": "MIT"
+    },
     "node_modules/isexe": {
       "version": "2.0.0",
       "resolved": "https://registry.npmjs.org/isexe/-/isexe-2.0.0.tgz",
@@ -3498,6 +3754,46 @@
       "dev": true,
       "license": "MIT"
     },
+    "node_modules/jsdom": {
+      "version": "26.1.0",
+      "resolved": "https://registry.npmjs.org/jsdom/-/jsdom-26.1.0.tgz",
+      "integrity": "sha512-Cvc9WUhxSMEo4McES3P7oK3QaXldCfNWp7pl2NNeiIFlCoLr3kfq9kb1fxftiwk1FLV7CvpvDfonxtzUDeSOPg==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "cssstyle": "^4.2.1",
+        "data-urls": "^5.0.0",
+        "decimal.js": "^10.5.0",
+        "html-encoding-sniffer": "^4.0.0",
+        "http-proxy-agent": "^7.0.2",
+        "https-proxy-agent": "^7.0.6",
+        "is-potential-custom-element-name": "^1.0.1",
+        "nwsapi": "^2.2.16",
+        "parse5": "^7.2.1",
+        "rrweb-cssom": "^0.8.0",
+        "saxes": "^6.0.0",
+        "symbol-tree": "^3.2.4",
+        "tough-cookie": "^5.1.1",
+        "w3c-xmlserializer": "^5.0.0",
+        "webidl-conversions": "^7.0.0",
+        "whatwg-encoding": "^3.1.1",
+        "whatwg-mimetype": "^4.0.0",
+        "whatwg-url": "^14.1.1",
+        "ws": "^8.18.0",
+        "xml-name-validator": "^5.0.0"
+      },
+      "engines": {
+        "node": ">=18"
+      },
+      "peerDependencies": {
+        "canvas": "^3.0.0"
+      },
+      "peerDependenciesMeta": {
+        "canvas": {
+          "optional": true
+        }
+      }
+    },
     "node_modules/jsesc": {
       "version": "3.1.0",
       "resolved": "https://registry.npmjs.org/jsesc/-/jsesc-3.1.0.tgz",
@@ -3977,6 +4273,13 @@
         "node": ">=0.10.0"
       }
     },
+    "node_modules/nwsapi": {
+      "version": "2.2.24",
+      "resolved": "https://registry.npmjs.org/nwsapi/-/nwsapi-2.2.24.tgz",
+      "integrity": "sha512-7YRhZ3jS45LwmSCT4b2sVFHt/WuovaktDU07QrtOBY2PXskss5a9jfmR9jptyumwXST+rFjrmppMY1KT/yn35A==",
+      "dev": true,
+      "license": "MIT"
+    },
     "node_modules/obug": {
       "version": "2.1.1",
       "resolved": "https://registry.npmjs.org/obug/-/obug-2.1.1.tgz",
@@ -4033,6 +4336,19 @@
         "node": ">=6"
       }
     },
+    "node_modules/parse5": {
+      "version": "7.3.0",
+      "resolved": "https://registry.npmjs.org/parse5/-/parse5-7.3.0.tgz",
+      "integrity": "sha512-IInvU7fabl34qmi9gY8XOVxhYyMyuH2xUNpb2q8/Y+7552KlejkRvqvD19nMoUW/uQGGbqNpA6Tufu5FL5BZgw==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "entities": "^6.0.0"
+      },
+      "funding": {
+        "url": "https://github.com/inikulin/parse5?sponsor=1"
+      }
+    },
     "node_modules/path-exists": {
       "version": "4.0.0",
       "resolved": "https://registry.npmjs.org/path-exists/-/path-exists-4.0.0.tgz",
@@ -4142,6 +4458,16 @@
         "node": "^10 || ^12 || >=14"
       }
     },
+    "node_modules/punycode": {
+      "version": "2.3.1",
+      "resolved": "https://registry.npmjs.org/punycode/-/punycode-2.3.1.tgz",
+      "integrity": "sha512-vYt7UD1U9Wg6138shLtLOvdAu+8DsC/ilFtEVHcH+wydcSpNE20AfSOduf6MkRFahL5FY7X1oU7nKVZFtfq8Fg==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": ">=6"
+      }
+    },
     "node_modules/readdirp": {
       "version": "3.6.0",
       "resolved": "https://registry.npmjs.org/readdirp/-/readdirp-3.6.0.tgz",
@@ -4314,6 +4640,33 @@
         "@rolldown/binding-win32-x64-msvc": "1.1.3"
       }
     },
+    "node_modules/rrweb-cssom": {
+      "version": "0.8.0",
+      "resolved": "https://registry.npmjs.org/rrweb-cssom/-/rrweb-cssom-0.8.0.tgz",
+      "integrity": "sha512-guoltQEx+9aMf2gDZ0s62EcV8lsXR+0w8915TC3ITdn2YueuNjdAYh/levpU9nFaoChh9RUS5ZdQMrKfVEN9tw==",
+      "dev": true,
+      "license": "MIT"
+    },
+    "node_modules/safer-buffer": {
+      "version": "2.1.2",
+      "resolved": "https://registry.npmjs.org/safer-buffer/-/safer-buffer-2.1.2.tgz",
+      "integrity": "sha512-YZo3K82SD7Riyi0E1EQPojLz7kpepnSQI9IyPbHHg1XXXevb5dJI7tpyN2ADxGcQbHG7vcyRHk0cbwqcQriUtg==",
+      "dev": true,
+      "license": "MIT"
+    },
+    "node_modules/saxes": {
+      "version": "6.0.0",
+      "resolved": "https://registry.npmjs.org/saxes/-/saxes-6.0.0.tgz",
+      "integrity": "sha512-xAg7SOnEhrm5zI3puOOKyy1OMcMlIJZYNJY7xLBwSze0UjhPLnWfj2GF2EpT0jmzaJKIWKHLsaSSajf35bcYnA==",
+      "dev": true,
+      "license": "ISC",
+      "dependencies": {
+        "xmlchars": "^2.2.0"
+      },
+      "engines": {
+        "node": ">=v12.22.7"
+      }
+    },
     "node_modules/schema-utils": {
       "version": "4.3.3",
       "resolved": "https://registry.npmjs.org/schema-utils/-/schema-utils-4.3.3.tgz",
@@ -4447,6 +4800,13 @@
         "url": "https://github.com/sponsors/ljharb"
       }
     },
+    "node_modules/symbol-tree": {
+      "version": "3.2.4",
+      "resolved": "https://registry.npmjs.org/symbol-tree/-/symbol-tree-3.2.4.tgz",
+      "integrity": "sha512-9QNk5KwDF+Bvz+PyObkmSYjI5ksVUYtjW7AU22r2NKcfLJcXp96hkDWU3+XndOsUb+AQ9QhfzfCT2O+CNWT5Tw==",
+      "dev": true,
+      "license": "MIT"
+    },
     "node_modules/tapable": {
       "version": "2.3.0",
       "resolved": "https://registry.npmjs.org/tapable/-/tapable-2.3.0.tgz",
@@ -4592,6 +4952,26 @@
         "node": ">=14.0.0"
       }
     },
+    "node_modules/tldts": {
+      "version": "6.1.86",
+      "resolved": "https://registry.npmjs.org/tldts/-/tldts-6.1.86.tgz",
+      "integrity": "sha512-WMi/OQ2axVTf/ykqCQgXiIct+mSQDFdH2fkwhPwgEwvJ1kSzZRiinb0zF2Xb8u4+OqPChmyI6MEu4EezNJz+FQ==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "tldts-core": "^6.1.86"
+      },
+      "bin": {
+        "tldts": "bin/cli.js"
+      }
+    },
+    "node_modules/tldts-core": {
+      "version": "6.1.86",
+      "resolved": "https://registry.npmjs.org/tldts-core/-/tldts-core-6.1.86.tgz",
+      "integrity": "sha512-Je6p7pkk+KMzMv2XXKmAE3McmolOQFdxkKw0R8EYNr7sELW46JqnNeTX8ybPiQgvg1ymCoF8LXs5fzFaZvJPTA==",
+      "dev": true,
+      "license": "MIT"
+    },
     "node_modules/to-regex-range": {
       "version": "5.0.1",
       "resolved": "https://registry.npmjs.org/to-regex-range/-/to-regex-range-5.0.1.tgz",
@@ -4605,6 +4985,32 @@
         "node": ">=8.0"
       }
     },
+    "node_modules/tough-cookie": {
+      "version": "5.1.2",
+      "resolved": "https://registry.npmjs.org/tough-cookie/-/tough-cookie-5.1.2.tgz",
+      "integrity": "sha512-FVDYdxtnj0G6Qm/DhNPSb8Ju59ULcup3tuJxkFb5K8Bv2pUXILbf0xZWU8PX8Ov19OXljbUyveOFwRMwkXzO+A==",
+      "dev": true,
+      "license": "BSD-3-Clause",
+      "dependencies": {
+        "tldts": "^6.1.32"
+      },
+      "engines": {
+        "node": ">=16"
+      }
+    },
+    "node_modules/tr46": {
+      "version": "5.1.1",
+      "resolved": "https://registry.npmjs.org/tr46/-/tr46-5.1.1.tgz",
+      "integrity": "sha512-hdF5ZgjTqgAntKkklYw0R03MG2x/bSzTtkxmIRw/sTNV8YXsCJ1tfLAX23lhxhHJlEf3CRCOCGGWw3vI3GaSPw==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "punycode": "^2.3.1"
+      },
+      "engines": {
+        "node": ">=18"
+      }
+    },
     "node_modules/tscw-config": {
       "version": "1.1.2",
       "resolved": "https://registry.npmjs.org/tscw-config/-/tscw-config-1.1.2.tgz",
@@ -4931,6 +5337,19 @@
         "url": "https://github.com/sponsors/jonschlinkert"
       }
     },
+    "node_modules/w3c-xmlserializer": {
+      "version": "5.0.0",
+      "resolved": "https://registry.npmjs.org/w3c-xmlserializer/-/w3c-xmlserializer-5.0.0.tgz",
+      "integrity": "sha512-o8qghlI8NZHU1lLPrpi2+Uq7abh4GGPpYANlalzWxyWteJOCsr/P+oPBA49TOLu5FTZO4d3F9MnWJfiMo4BkmA==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "xml-name-validator": "^5.0.0"
+      },
+      "engines": {
+        "node": ">=18"
+      }
+    },
     "node_modules/watchpack": {
       "version": "2.5.1",
       "resolved": "https://registry.npmjs.org/watchpack/-/watchpack-2.5.1.tgz",
@@ -4944,6 +5363,16 @@
         "node": ">=10.13.0"
       }
     },
+    "node_modules/webidl-conversions": {
+      "version": "7.0.0",
+      "resolved": "https://registry.npmjs.org/webidl-conversions/-/webidl-conversions-7.0.0.tgz",
+      "integrity": "sha512-VwddBukDzu71offAQR975unBIGqfKZpM+8ZX6ySk8nYhVoo5CYaZyzt3YBvYtRtO+aoGlqxPg/B87NGVZ/fu6g==",
+      "dev": true,
+      "license": "BSD-2-Clause",
+      "engines": {
+        "node": ">=12"
+      }
+    },
     "node_modules/webpack": {
       "version": "5.105.0",
       "resolved": "https://registry.npmjs.org/webpack/-/webpack-5.105.0.tgz",
@@ -5070,6 +5499,44 @@
         "node": ">=10.13.0"
       }
     },
+    "node_modules/whatwg-encoding": {
+      "version": "3.1.1",
+      "resolved": "https://registry.npmjs.org/whatwg-encoding/-/whatwg-encoding-3.1.1.tgz",
+      "integrity": "sha512-6qN4hJdMwfYBtE3YBTTHhoeuUrDBPZmbQaxWAqSALV/MeEnR5z1xd8UKud2RAkFoPkmB+hli1TZSnyi84xz1vQ==",
+      "deprecated": "Use @exodus/bytes instead for a more spec-conformant and faster implementation",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "iconv-lite": "0.6.3"
+      },
+      "engines": {
+        "node": ">=18"
+      }
+    },
+    "node_modules/whatwg-mimetype": {
+      "version": "4.0.0",
+      "resolved": "https://registry.npmjs.org/whatwg-mimetype/-/whatwg-mimetype-4.0.0.tgz",
+      "integrity": "sha512-QaKxh0eNIi2mE9p2vEdzfagOKHCcj1pJ56EEHGQOVxp8r9/iszLUUV7v89x9O1p/T+NlTM5W7jW6+cz4Fq1YVg==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": ">=18"
+      }
+    },
+    "node_modules/whatwg-url": {
+      "version": "14.2.0",
+      "resolved": "https://registry.npmjs.org/whatwg-url/-/whatwg-url-14.2.0.tgz",
+      "integrity": "sha512-De72GdQZzNTUBBChsXueQUnPKDkg/5A5zp7pFDuQAj5UFoENpiACU0wlCvzpAGnTkj++ihpKwKyYewn/XNUbKw==",
+      "dev": true,
+      "license": "MIT",
+      "dependencies": {
+        "tr46": "^5.1.0",
+        "webidl-conversions": "^7.0.0"
+      },
+      "engines": {
+        "node": ">=18"
+      }
+    },
     "node_modules/which": {
       "version": "2.0.2",
       "resolved": "https://registry.npmjs.org/which/-/which-2.0.2.tgz",
@@ -5114,6 +5581,45 @@
       "integrity": "sha1-tSQ9jz7BqjXxNkYFvA0QNuMKtp8=",
       "dev": true
     },
+    "node_modules/ws": {
+      "version": "8.21.1",
+      "resolved": "https://registry.npmjs.org/ws/-/ws-8.21.1.tgz",
+      "integrity": "sha512-+0NTnW77fFN/DjQi6k/Sq/Yvk4Sgajw7urW8V+asjXnRgDs9gyGkdb7EzgfhA4goXsRIZKE28fzIXBHEzhuiWw==",
+      "dev": true,
+      "license": "MIT",
+      "engines": {
+        "node": ">=10.0.0"
+      },
+      "peerDependencies": {
+        "bufferutil": "^4.0.1",
+        "utf-8-validate": ">=5.0.2"
+      },
+      "peerDependenciesMeta": {
+        "bufferutil": {
+          "optional": true
+        },
+        "utf-8-validate": {
+          "optional": true
+        }
+      }
+    },
+    "node_modules/xml-name-validator": {
+      "version": "5.0.0",
+      "resolved": "https://registry.npmjs.org/xml-name-validator/-/xml-name-validator-5.0.0.tgz",
+      "integrity": "sha512-EvGK8EJ3DhaHfbRlETOWAS5pO9MZITeauHKJyb8wyajUfQUenkIg2MvLDTZ4T/TgIcm3HU0TFBgWWboAZ30UHg==",
+      "dev": true,
+      "license": "Apache-2.0",
+      "engines": {
+        "node": ">=18"
+      }
+    },
+    "node_modules/xmlchars": {
+      "version": "2.2.0",
+      "resolved": "https://registry.npmjs.org/xmlchars/-/xmlchars-2.2.0.tgz",
+      "integrity": "sha512-JZnDKK8B0RCDw84FNdDAIpZK+JuJw+s7Lz8nksI7SIuU3UXJJslUthsi+uWBUYOwPFwW7W7PRLRfUKpxjtjFCw==",
+      "dev": true,
+      "license": "MIT"
+    },
     "node_modules/yallist": {
       "version": "3.1.1",
       "resolved": "https://registry.npmjs.org/yallist/-/yallist-3.1.1.tgz",

superset-embedded-sdk/package.json

@@ -43,6 +43,7 @@
     "@babel/preset-typescript": "7.29.7",
     "@types/node": "^25.4.0",
     "babel-loader": "^9.1.3",
+    "jsdom": "^26.1.0",
     "tscw-config": "^1.1.2",
     "typescript": "^5.9.3",
     "vitest": "^4.0.18",

superset-embedded-sdk/src/index.test.ts

@@ -0,0 +1,213 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied.  See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+
+// @vitest-environment jsdom
+
+import { afterEach, beforeEach, describe, expect, test, vi } from "vitest";
+import { Switchboard } from "@superset-ui/switchboard";
+import { embedDashboard } from "./index";
+
+vi.mock("@superset-ui/switchboard");
+
+function makeFakeJWT(claims: any) {
+  // not a valid jwt, but close enough for this code
+  const tokenifiedClaims = Buffer.from(JSON.stringify(claims)).toString(
+    "base64",
+  );
+  return `abc.${tokenifiedClaims}.xyz`;
+}
+
+describe("embedDashboard", () => {
+  let mountPoint: HTMLElement;
+  let mockSwitchboard: Switchboard;
+
+  beforeEach(() => {
+    mountPoint = document.createElement("div");
+    document.body.appendChild(mountPoint);
+
+    mockSwitchboard = {
+      emit: vi.fn(),
+      get: vi.fn(),
+      start: vi.fn(),
+      defineMethod: vi.fn(),
+    } as any;
+
+    // Constructor mocks must use `function`, since arrow functions cannot be
+    // invoked with `new`.
+    vi.mocked(Switchboard).mockImplementation(function () {
+      return mockSwitchboard;
+    } as any);
+
+    // Mock MessageChannel API
+    globalThis.MessageChannel = vi.fn(function (this: any) {
+      this.port1 = {};
+      this.port2 = {};
+    }) as any;
+
+    // Mock iframe load event and sandbox
+    const originalCreateElement = document.createElement.bind(document);
+    vi.spyOn(document, "createElement").mockImplementation(tagName => {
+      const element = originalCreateElement(tagName);
+      if (tagName === "iframe") {
+        // Mock sandbox DOMTokenList
+        (element as any).sandbox = {
+          add: vi.fn(),
+        };
+        // Mock contentWindow for postMessage
+        Object.defineProperty(element, "contentWindow", {
+          writable: true,
+          value: {
+            postMessage: vi.fn(),
+          },
+        });
+        setTimeout(() => {
+          element.dispatchEvent(new Event("load"));
+        }, 0);
+      }
+      return element;
+    });
+  });
+
+  afterEach(() => {
+    document.body.removeChild(mountPoint);
+    vi.restoreAllMocks();
+  });
+
+  test("setDataMask sends dataMask to iframe", async () => {
+    const fakeToken = makeFakeJWT({ exp: Date.now() / 1000 + 300 });
+    const mockFetchGuestToken = vi.fn().mockResolvedValue(fakeToken);
+    const testDataMask = {
+      "NATIVE_FILTER-1": {
+        filterState: {
+          value: ["value1", "value2"],
+        },
+      },
+    };
+
+    const dashboard = await embedDashboard({
+      id: "test-id",
+      supersetDomain: "https://superset.example.com",
+      mountPoint,
+      fetchGuestToken: mockFetchGuestToken,
+    });
+
+    dashboard.setDataMask(testDataMask);
+
+    expect(mockSwitchboard.get).toHaveBeenCalledWith("setDataMask", {
+      dataMask: testDataMask,
+    });
+  });
+
+  test("setDataMask sends empty dataMask", async () => {
+    const fakeToken = makeFakeJWT({ exp: Date.now() / 1000 + 300 });
+    const mockFetchGuestToken = vi.fn().mockResolvedValue(fakeToken);
+    const emptyDataMask = {};
+
+    const dashboard = await embedDashboard({
+      id: "test-id",
+      supersetDomain: "https://superset.example.com",
+      mountPoint,
+      fetchGuestToken: mockFetchGuestToken,
+    });
+
+    dashboard.setDataMask(emptyDataMask);
+
+    expect(mockSwitchboard.get).toHaveBeenCalledWith("setDataMask", {
+      dataMask: emptyDataMask,
+    });
+  });
+
+  test("setDataMask drops the change-trigger flags observeDataMask adds", async () => {
+    const fakeToken = makeFakeJWT({ exp: Date.now() / 1000 + 300 });
+    const mockFetchGuestToken = vi.fn().mockResolvedValue(fakeToken);
+    const observedMask = {
+      "NATIVE_FILTER-1": {
+        filterState: {
+          value: ["CA"],
+        },
+      },
+      crossFiltersChanged: false,
+      nativeFiltersChanged: true,
+    };
+
+    const dashboard = await embedDashboard({
+      id: "test-id",
+      supersetDomain: "https://superset.example.com",
+      mountPoint,
+      fetchGuestToken: mockFetchGuestToken,
+    });
+
+    dashboard.setDataMask(observedMask);
+
+    expect(mockSwitchboard.get).toHaveBeenCalledWith("setDataMask", {
+      dataMask: {
+        "NATIVE_FILTER-1": observedMask["NATIVE_FILTER-1"],
+      },
+    });
+  });
+
+  test("setDataMask sends complex dataMask with multiple filters", async () => {
+    const fakeToken = makeFakeJWT({ exp: Date.now() / 1000 + 300 });
+    const mockFetchGuestToken = vi.fn().mockResolvedValue(fakeToken);
+    const complexDataMask = {
+      "NATIVE_FILTER-1": {
+        filterState: {
+          value: ["CA", "NY"],
+        },
+      },
+      "NATIVE_FILTER-2": {
+        filterState: {
+          value: [2023, 2024],
+        },
+      },
+    };
+
+    const dashboard = await embedDashboard({
+      id: "test-id",
+      supersetDomain: "https://superset.example.com",
+      mountPoint,
+      fetchGuestToken: mockFetchGuestToken,
+    });
+
+    dashboard.setDataMask(complexDataMask);
+
+    expect(mockSwitchboard.get).toHaveBeenCalledWith("setDataMask", {
+      dataMask: complexDataMask,
+    });
+  });
+
+  test("setDataMask rejects when the embedded page does not support it", async () => {
+    const fakeToken = makeFakeJWT({ exp: Date.now() / 1000 + 300 });
+    const mockFetchGuestToken = vi.fn().mockResolvedValue(fakeToken);
+    vi.mocked(mockSwitchboard.get).mockRejectedValue(
+      new Error('Method "setDataMask" is not defined'),
+    );
+
+    const dashboard = await embedDashboard({
+      id: "test-id",
+      supersetDomain: "https://superset.example.com",
+      mountPoint,
+      fetchGuestToken: mockFetchGuestToken,
+    });
+
+    await expect(dashboard.setDataMask({})).rejects.toThrow(
+      'Method "setDataMask" is not defined',
+    );
+  });
+});

superset-embedded-sdk/src/index.ts

@@ -115,6 +115,12 @@ export type EmbeddedDashboard = {
   getActiveTabs: () => Promise<string[]>;
   observeDataMask: (callbackFn: ObserveDataMaskCallbackFn) => void;
   getDataMask: () => Promise<Record<string, any>>;
+  /**
+   * Applies a data mask to the dashboard.
+   * Rejects if the embedded Superset page does not support `setDataMask`,
+   * so a version mismatch surfaces instead of silently doing nothing.
+   */
+  setDataMask: (dataMask: Record<string, any>) => Promise<void>;
   getChartStates: () => Promise<Record<string, any>>;
   getChartDataPayloads: (params?: {
     chartId?: number;
@@ -355,6 +361,21 @@ export async function embedDashboard({
     ourPort.get<string>("getDashboardPermalink", { anchor });
   const getActiveTabs = () => ourPort.get<string[]>("getActiveTabs");
   const getDataMask = () => ourPort.get<Record<string, any>>("getDataMask");
+  // `observeDataMask` hands the host a mask with the change-trigger booleans
+  // mixed in, so feeding that payload straight back into `setDataMask` is a
+  // natural thing for a host to do. Keep only the entries that look like a
+  // filter's mask, so those flags never reach the dashboard as filter ids.
+  // Sent with `get` rather than `emit` so the iframe acknowledges the call:
+  // an embedded page that predates `setDataMask` replies with an error instead
+  // of dropping the message silently.
+  const setDataMask = (dataMask: Record<string, any>) =>
+    ourPort.get<void>("setDataMask", {
+      dataMask: Object.fromEntries(
+        Object.entries(dataMask).filter(
+          ([, mask]) => typeof mask === "object" && mask !== null,
+        ),
+      ),
+    });
   const getChartStates = () =>
     ourPort.get<Record<string, any>>("getChartStates");
   const getChartDataPayloads = (params?: { chartId?: number }) =>
@@ -396,6 +417,7 @@ export async function embedDashboard({
     getActiveTabs,
     observeDataMask,
     getDataMask,
+    setDataMask,
     getChartStates,
     getChartDataPayloads,
     setThemeConfig,

superset-frontend/oxlint.json

@@ -113,7 +113,7 @@
     // === Import plugin rules ===
     "import/named": "error",
     "import/export": "error",
-    "import/no-named-as-default": "error",
+    "import/no-named-as-default": "warn",
     "import/no-named-as-default-member": "error",
     "import/no-mutable-exports": "error",
     "import/no-amd": "error",

superset-frontend/package.json

@@ -110,7 +110,7 @@
     "@fontsource/fira-code": "^5.3.0",
     "@fontsource/ibm-plex-mono": "^5.3.0",
     "@fontsource/inter": "^5.3.0",
-    "@googleapis/sheets": "^13.0.2",
+    "@googleapis/sheets": "^14.0.0",
     "@great-expectations/jsonforms-antd-renderers": "^2.2.10",
     "@jsonforms/core": "^3.7.0",
     "@jsonforms/react": "^3.7.0",
@@ -122,9 +122,9 @@
     "@luma.gl/shadertools": "~9.2.5",
     "@luma.gl/webgl": "~9.2.5",
     "@reduxjs/toolkit": "^1.9.3",
-    "@rjsf/core": "^6.7.1",
+    "@rjsf/core": "^6.8.0",
     "@rjsf/utils": "^6.6.2",
-    "@rjsf/validator-ajv8": "^6.7.1",
+    "@rjsf/validator-ajv8": "^6.8.0",
     "@scarf/scarf": "^1.4.0",
     "@superset-ui/chart-controls": "file:./packages/superset-ui-chart-controls",
     "@superset-ui/core": "file:./packages/superset-ui-core",
@@ -158,12 +158,12 @@
     "@visx/xychart": "^4.0.0",
     "ag-grid-community": "36.1.0",
     "ag-grid-react": "36.1.0",
-    "antd": "^6.5.4",
+    "antd": "^6.6.1",
     "chrono-node": "^2.10.1",
     "classnames": "^2.2.5",
     "content-disposition": "^2.0.1",
     "d3-scale": "^4.0.2",
-    "dayjs": "^1.11.21",
+    "dayjs": "^1.11.23",
     "dom-to-image-more": "^3.10.2",
     "dom-to-pdf": "^0.3.2",
     "echarts": "^6.1.0",
@@ -176,8 +176,8 @@
     "geostyler-openlayers-parser": "^5.7.1",
     "geostyler-style": "11.0.2",
     "geostyler-wfs-parser": "^3.0.1",
-    "google-auth-library": "^11.0.0",
-    "immer": "^11.1.16",
+    "google-auth-library": "^11.0.2",
+    "immer": "^11.1.18",
     "interweave": "^13.1.1",
     "jquery": "^4.0.0",
     "js-levenshtein": "^1.1.6",
@@ -230,7 +230,7 @@
     "use-event-callback": "^0.1.0",
     "use-immer": "^0.11.0",
     "use-query-params": "^2.2.2",
-    "uuid": "^14.0.1",
+    "uuid": "^14.0.2",
     "xlsx": "https://cdn.sheetjs.com/xlsx-0.20.3/xlsx-0.20.3.tgz",
     "yargs": "^18.1.0"
   },
@@ -257,16 +257,16 @@
     "@istanbuljs/nyc-config-typescript": "^1.0.1",
     "@playwright/test": "^1.62.1",
     "@pmmmwh/react-refresh-webpack-plugin": "^0.6.2",
-    "@storybook/addon-docs": "10.5.7",
-    "@storybook/addon-links": "10.5.7",
-    "@storybook/react-webpack5": "10.5.7",
+    "@storybook/addon-docs": "10.5.10",
+    "@storybook/addon-links": "10.5.10",
+    "@storybook/react-webpack5": "10.5.10",
     "@storybook/test-runner": "0.24.4",
     "@svgr/webpack": "^8.1.0",
-    "@swc/core": "^1.15.47",
-    "@swc/plugin-emotion": "^14.15.0",
-    "@swc/plugin-transform-imports": "^12.5.0",
+    "@swc/core": "^1.16.1",
+    "@swc/plugin-emotion": "^15.0.0",
+    "@swc/plugin-transform-imports": "^13.0.0",
     "@testing-library/dom": "^10.4.1",
-    "@testing-library/jest-dom": "^7.0.0",
+    "@testing-library/jest-dom": "^7.0.1",
     "@testing-library/react": "^15.0.0",
     "@testing-library/user-event": "^12.8.3",
     "@types/content-disposition": "^0.5.9",
@@ -277,7 +277,7 @@
     "@types/json-bigint": "^1.0.4",
     "@types/lodash-es": "^4.17.12",
     "@types/mousetrap": "^1.6.15",
-    "@types/node": "^26.1.2",
+    "@types/node": "^26.2.0",
     "@types/react": "^18.3.0",
     "@types/react-dom": "^18.3.0",
     "@types/react-loadable": "^5.5.11",
@@ -289,19 +289,19 @@
     "@types/rison": "0.1.0",
     "@types/tinycolor2": "^1.4.3",
     "@types/unzipper": "^0.10.11",
-    "@typescript-eslint/eslint-plugin": "^8.66.0",
+    "@typescript-eslint/eslint-plugin": "^8.67.0",
     "@typescript-eslint/parser": "^8.63.0",
     "babel-jest": "^30.4.1",
     "babel-loader": "^10.1.1",
     "babel-plugin-dynamic-import-node": "^2.3.3",
     "babel-plugin-jsx-remove-data-test-id": "^3.0.0",
-    "baseline-browser-mapping": "^2.11.12",
+    "baseline-browser-mapping": "^2.11.16",
     "cheerio": "1.2.0",
-    "concurrently": "^10.0.4",
+    "concurrently": "^10.0.5",
     "copy-webpack-plugin": "^14.0.0",
     "cross-env": "^10.1.0",
     "css-loader": "^7.1.4",
-    "eslint": "^10.8.0",
+    "eslint": "^10.8.1",
     "eslint-import-resolver-alias": "^1.1.2",
     "eslint-import-resolver-typescript": "^4.4.5",
     "eslint-plugin-i18n-strings": "file:eslint-rules/eslint-plugin-i18n-strings",
@@ -312,7 +312,7 @@
     "eslint-plugin-no-only-tests": "^3.4.0",
     "eslint-plugin-react-prefer-function-component": "^5.0.0",
     "eslint-plugin-react-you-might-not-need-an-effect": "^1.0.1",
-    "eslint-plugin-storybook": "10.5.7",
+    "eslint-plugin-storybook": "10.5.10",
     "eslint-plugin-testing-library": "^7.16.2",
     "eslint-plugin-theme-colors": "file:eslint-rules/eslint-plugin-theme-colors",
     "fetch-mock": "^12.6.0",
@@ -326,13 +326,13 @@
     "jest-websocket-mock": "^2.5.0",
     "js-yaml-loader": "^1.2.2",
     "jsdom": "^30.0.1",
-    "lerna": "^10.0.0",
+    "lerna": "^10.0.1",
     "lightningcss": "^1.33.0",
     "mini-css-extract-plugin": "^2.10.2",
     "minimizer-webpack-plugin": "^5.6.1",
     "open-cli": "^9.0.0",
-    "oxfmt": "^0.62.0",
-    "oxlint": "^1.77.0",
+    "oxfmt": "^0.64.0",
+    "oxlint": "^1.79.0",
     "po2json": "^0.4.5",
     "postcss-styled-syntax": "^0.7.2",
     "process": "^0.11.10",
@@ -343,13 +343,13 @@
     "source-map": "^0.8.0",
     "source-map-support": "^0.5.21",
     "speed-measure-webpack-plugin": "^1.6.0",
-    "storybook": "10.5.7",
+    "storybook": "10.5.10",
     "style-loader": "^4.0.0",
     "stylelint": "^17.14.1",
     "swc-loader": "^0.2.7",
     "ts-jest": "^29.4.12",
     "tscw-config": "^1.1.2",
-    "tsx": "^4.23.10",
+    "tsx": "^4.23.12",
     "typescript": "5.4.5",
     "unzipper": "^0.12.5",
     "wait-on": "^9.1.0",
@@ -415,6 +415,7 @@
     "minimatch@>=10": {
       "brace-expansion": ">=5.0.8"
     },
+    "nanoid@>=3 <4": "3.3.18",
     "nwsapi": "^2.2.13",
     "puppeteer": "^22.4.1",
     "tar": "^7.5.16",

superset-frontend/packages/superset-core/src/common/index.ts

@@ -130,6 +130,7 @@ export enum GenericDataType {
   String = 1,
   Temporal = 2,
   Boolean = 3,
+  MultiValue = 4,
 }
 
 /**

superset-frontend/packages/superset-ui-chart-controls/src/components/ColumnTypeLabel/ColumnTypeLabel.tsx

@@ -28,6 +28,7 @@ import {
   FieldBinaryOutlined,
   FieldStringOutlined,
   NumberOutlined,
+  UnorderedListOutlined,
 } from '@ant-design/icons';
 import { Icons } from '@superset-ui/core/components';
 
@@ -72,6 +73,10 @@ export function ColumnTypeLabel({ type }: ColumnTypeLabelProps) {
     typeIcon = <FieldBinaryOutlined aria-label={t('boolean type icon')} />;
   } else if (type === GenericDataType.Temporal) {
     typeIcon = <ClockCircleOutlined aria-label={t('temporal type icon')} />;
+  } else if (type === GenericDataType.MultiValue) {
+    typeIcon = (
+      <UnorderedListOutlined aria-label={t('multi-value type icon')} />
+    );
   }
 
   return <TypeIconWrapper>{typeIcon}</TypeIconWrapper>;

superset-frontend/packages/superset-ui-chart-controls/src/components/MetricOption.tsx

@@ -51,6 +51,20 @@ export interface MetricOptionProps {
   shouldShowTooltip?: boolean;
 }
 
+/**
+ * `url` is an arbitrary caller-supplied string rendered as an href. Only
+ * http(s) and relative URLs become links; other schemes degrade to plain
+ * text.
+ */
+function isSafeHref(url: string): boolean {
+  try {
+    const { protocol } = new URL(url, window.location.origin);
+    return protocol === 'http:' || protocol === 'https:';
+  } catch {
+    return false;
+  }
+}
+
 export function MetricOption({
   metric,
   labelRef,
@@ -70,7 +84,7 @@ export function MetricOption({
       `}
       ref={labelRef}
     >
-      {url ? (
+      {url && isSafeHref(url) ? (
         <Typography.Link
           href={url}
           target={openInNewWindow ? '_blank' : ''}

superset-frontend/packages/superset-ui-chart-controls/src/sections/timeComparison.tsx

@@ -122,6 +122,12 @@ export const timeComparisonControls: ({
             }
             return newState;
           },
+          // Re-run this control's validation whenever `time_compare` changes so
+          // the "date required" error clears once a non-custom shift is picked.
+          // Without it the stale error survives in Redux (see the
+          // dependantControls path in exploreReducer's SET_FIELD_VALUE handler)
+          // and blocks further chart updates until a page refresh.
+          validationDependencies: ['time_compare'],
         },
       },
     ],

superset-frontend/packages/superset-ui-chart-controls/src/utils/getTotalsMetrics.test.ts

@@ -17,7 +17,7 @@
  * under the License.
  */
 import { QueryFormMetric } from '@superset-ui/core';
-import { getTotalsMetrics } from './getTotalsMetrics';
+import { getTotalsMetrics, toTotalsAggregate } from './getTotalsMetrics';
 
 const simpleMetric = (aggregate: string): QueryFormMetric =>
   ({
@@ -76,4 +76,31 @@ describe('getTotalsMetrics', () => {
   test('returns an empty array when given no metrics', () => {
     expect(getTotalsMetrics([], 'AVG')).toEqual([]);
   });
+
+  test("ORIGINAL keeps each metric's own aggregate", () => {
+    const metrics = [
+      simpleMetric('COUNT_DISTINCT'),
+      sqlMetric(),
+      savedMetric(),
+    ];
+    const result = getTotalsMetrics(metrics, 'ORIGINAL');
+
+    expect(result).toBe(metrics);
+    expect(result[0]).toEqual(
+      expect.objectContaining({ aggregate: 'COUNT_DISTINCT' }),
+    );
+  });
+});
+
+describe('toTotalsAggregate', () => {
+  test.each(['SUM', 'AVG'] as const)('passes %s through', value => {
+    expect(toTotalsAggregate(value)).toBe(value);
+  });
+
+  test.each([undefined, null, '', 'MEDIAN', 'sum'])(
+    'falls back to ORIGINAL for %p',
+    value => {
+      expect(toTotalsAggregate(value)).toBe('ORIGINAL');
+    },
+  );
 });

superset-frontend/packages/superset-ui-chart-controls/src/utils/getTotalsMetrics.ts

@@ -18,26 +18,46 @@
  */
 import { isAdhocMetricSimple, QueryFormMetric } from '@superset-ui/core';
 
-export type TotalsAggregate = 'SUM' | 'AVG';
+/**
+ * How the "Show summary" totals row aggregates each metric.
+ *
+ * ``ORIGINAL`` keeps every metric's own aggregation. It is the default because
+ * overriding is not universally valid: ``SUM`` over a ``COUNT_DISTINCT`` of a
+ * non-numeric column (a uuid, say) is rejected outright by the database, and
+ * over a numeric id column it silently produces a meaningless number.
+ */
+export type TotalsAggregate = 'ORIGINAL' | 'SUM' | 'AVG';
 
 /**
- * Build the metrics for a chart's "Show summary" totals query, overriding
- * each Simple (adhoc) metric's aggregate function with the user-chosen
- * totals aggregate. The totals query has no GROUP BY, so the database
- * evaluates each metric fresh over all rows -- swapping the aggregate here
- * is a correct, independent computation, not a re-aggregation of
- * already-aggregated per-row values.
+ * Build the metrics for a chart's "Show summary" totals query.
  *
- * Custom-SQL metrics and saved (string) metrics pass through unchanged:
- * there is no safe way to rewrite an arbitrary SQL expression's aggregate
- * function without parsing it, so the totals row keeps their own native
- * aggregate for those.
+ * With SUM or AVG, each Simple (adhoc) metric is cloned with its aggregate
+ * replaced. The totals query has no GROUP BY, so the database evaluates each
+ * metric fresh over all rows -- that swap is an independent computation, not a
+ * re-aggregation of already-aggregated per-row values.
+ *
+ * Custom-SQL and saved (string) metrics always pass through unchanged: there is
+ * no safe way to rewrite an arbitrary SQL expression's aggregate without
+ * parsing it, so the totals row keeps their own native aggregate.
  */
 export function getTotalsMetrics(
   metrics: QueryFormMetric[],
   aggregate: TotalsAggregate,
 ): QueryFormMetric[] {
+  if (aggregate === 'ORIGINAL') {
+    return metrics;
+  }
   return metrics.map(metric =>
     isAdhocMetricSimple(metric) ? { ...metric, aggregate } : metric,
   );
 }
+
+/**
+ * Narrow a raw ``totals_aggregate`` form-data value to a TotalsAggregate.
+ *
+ * Anything other than an explicit SUM/AVG — including charts saved before the
+ * control existed — keeps each metric's own aggregation.
+ */
+export function toTotalsAggregate(value: unknown): TotalsAggregate {
+  return value === 'SUM' || value === 'AVG' ? value : 'ORIGINAL';
+}

superset-frontend/packages/superset-ui-chart-controls/test/components/ColumnTypeLabel.test.tsx

@@ -64,4 +64,21 @@ describe('ColumnOption', () => {
     renderColumnTypeLabel({ type: GenericDataType.Temporal });
     expect(screen.getByLabelText('temporal type icon')).toBeVisible();
   });
+  test('multi-value (array) type shows list icon', () => {
+    renderColumnTypeLabel({ type: GenericDataType.MultiValue });
+    expect(screen.getByLabelText('multi-value type icon')).toBeVisible();
+  });
+});
+
+describe('GenericDataType enum parity', () => {
+  // These numeric values are shared with the backend enum in
+  // superset/utils/core.py (GenericDataType). They must stay in sync because
+  // the backend serializes columns using these integers.
+  test('values match the backend contract', () => {
+    expect(GenericDataType.Numeric).toBe(0);
+    expect(GenericDataType.String).toBe(1);
+    expect(GenericDataType.Temporal).toBe(2);
+    expect(GenericDataType.Boolean).toBe(3);
+    expect(GenericDataType.MultiValue).toBe(4);
+  });
 });

superset-frontend/packages/superset-ui-chart-controls/test/components/MetricOption.test.tsx

@@ -112,3 +112,26 @@ test('shows a Tooltip for the verbose metric name', () => {
   const { getByTestId } = setup();
   expect(getByTestId('mock-tooltip')).toBeInTheDocument();
 });
+test('does not render javascript: URLs as links', () => {
+  // Regression test: the url prop can be creator-authored and must
+  // never become a script-bearing href for other viewers.
+  const { queryByRole, getByText } = setup({
+    url: 'javascript:alert(document.domain)', // eslint-disable-line no-script-url
+  });
+  expect(queryByRole('link')).not.toBeInTheDocument();
+  expect(getByText(defaultProps.metric.verbose_name)).toBeInTheDocument();
+});
+test('does not render data: URLs as links', () => {
+  const { queryByRole } = setup({
+    url: 'data:text/html,<script>alert(1)</script>',
+  });
+  expect(queryByRole('link')).not.toBeInTheDocument();
+});
+test('renders relative URLs as links', () => {
+  const { getByRole } = setup({
+    url: '/superset/dashboard/1/',
+  });
+  expect(
+    getByRole('link', { name: defaultProps.metric.verbose_name }),
+  ).toHaveAttribute('href', '/superset/dashboard/1/');
+});

superset-frontend/packages/superset-ui-core/package.json

@@ -67,8 +67,8 @@
     "d3-scale": "^4.0.2",
     "d3-time": "^3.1.0",
     "d3-time-format": "^4.1.0",
-    "dayjs": "^1.11.21",
-    "dompurify": "^3.4.13",
+    "dayjs": "^1.11.23",
+    "dompurify": "^3.4.14",
     "fetch-retry": "^6.0.0",
     "handlebars": "^4.7.9",
     "jed": "^1.1.1",
@@ -79,7 +79,7 @@
     "re-resizable": "^6.11.2",
     "react-ace": "^14.0.1",
     "react-draggable": "^4.7.1",
-    "react-error-boundary": "^6.1.2",
+    "react-error-boundary": "^6.1.3",
     "react-js-cron": "^6.0.2",
     "react-markdown": "^10.1.0",
     "react-resize-detector": "^7.1.2",
@@ -103,7 +103,7 @@
     "@types/d3-time-format": "^4.0.3",
     "@types/jquery": "^4.0.1",
     "@types/lodash": "^4.17.25",
-    "@types/node": "^26.1.2",
+    "@types/node": "^26.2.0",
     "@types/prop-types": "^15.7.15",
     "@types/react-syntax-highlighter": "^15.5.13",
     "@types/react-table": "^7.7.20",

superset-frontend/packages/superset-ui-core/src/components/Modal/Modal.test.tsx

@@ -0,0 +1,153 @@
+/**
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements.  See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership.  The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License.  You may obtain a copy of the License at
+ *
+ *   http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied.  See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+
+import { useState } from 'react';
+import { fireEvent, render, screen } from '@superset-ui/core/spec';
+import { Input } from '../Input';
+import { Modal } from './Modal';
+
+const drag = (
+  target: Element,
+  from: [number, number],
+  to: [number, number],
+) => {
+  fireEvent.mouseDown(target, { clientX: from[0], clientY: from[1] });
+  fireEvent.mouseMove(document, { clientX: to[0], clientY: to[1] });
+  fireEvent.mouseUp(document);
+};
+
+const isDragged = () => !!document.querySelector('.react-draggable-dragged');
+
+describe('Modal draggable', () => {
+  test('dragging from the title bar moves the modal', () => {
+    render(
+      <Modal show onHide={() => {}} title="Edit Dataset" draggable name="test">
+        <Input data-test="field" defaultValue="value" />
+      </Modal>,
+    );
+
+    const trigger = document.querySelector('.draggable-trigger') as HTMLElement;
+    drag(trigger, [100, 50], [150, 90]);
+
+    expect(isDragged()).toBe(true);
+  });
+
+  test('dragging inside modal content does not move the modal', () => {
+    render(
+      <Modal show onHide={() => {}} title="Edit Dataset" draggable name="test">
+        <Input data-test="field" defaultValue="first_view_event" />
+      </Modal>,
+    );
+
+    const input = screen.getByTestId('field');
+    drag(input, [200, 400], [260, 430]);
+
+    expect(isDragged()).toBe(false);
+  });
+
+  test('dragging inside modal content does not move the modal, even after an unrelated re-render while the title was hovered', () => {
+    // Regression test: the title bar used to gate dragging with a
+    // hover-tracked boolean (mouseover/mouseout on `.draggable-trigger`)
+    // instead of react-draggable's own `handle` prop. Because the title
+    // element was defined as an inline component recreated on every
+    // render, any unrelated state change while the cursor was over the
+    // title (e.g. typing in any field) force-remounted it without a real
+    // mouseout ever firing, leaving dragging permanently enabled -- so
+    // selecting text anywhere in the modal dragged the whole modal
+    // instead.
+    function Harness() {
+      const [tick, setTick] = useState(0);
+      return (
+        <Modal
+          show
+          onHide={() => {}}
+          title="Edit Dataset"
+          draggable
+          name="test"
+        >
+          <button
+            type="button"
+            data-test="rerender"
+            onClick={() => setTick(tick + 1)}
+          >
+            rerender
+          </button>
+          <Input data-test="field" defaultValue="first_view_event" />
+        </Modal>
+      );
+    }
+
+    render(<Harness />);
+
+    const trigger = document.querySelector('.draggable-trigger') as HTMLElement;
+    fireEvent.mouseOver(trigger);
+    fireEvent.click(screen.getByTestId('rerender'));
+
+    const input = screen.getByTestId('field');
+    drag(input, [200, 400], [260, 430]);
+
+    expect(isDragged()).toBe(false);
+  });
+
+  test('dragging is disabled entirely when draggable is not set', () => {
+    render(
+      <Modal show onHide={() => {}} title="Edit Dataset" name="test">
+        <Input data-test="field" defaultValue="value" />
+      </Modal>,
+    );
+
+    expect(document.querySelector('.draggable-trigger')).toBeNull();
+  });
+
+  test('draggableConfig cannot re-enable dragging on a non-draggable modal', () => {
+    render(
+      <Modal
+        show
+        onHide={() => {}}
+        title="Edit Dataset"
+        name="test"
+        draggableConfig={{ disabled: false }}
+      >
+        <Input data-test="field" defaultValue="value" />
+      </Modal>,
+    );
+
+    expect(document.querySelector('.draggable-trigger')).toBeNull();
+  });
+
+  test('draggableConfig can still opt a draggable modal out of dragging', () => {
+    render(
+      <Modal
+        show
+        onHide={() => {}}
+        title="Edit Dataset"
+        draggable
+        name="test"
+        draggableConfig={{ disabled: true }}
+      >
+        <Input data-test="field" defaultValue="value" />
+      </Modal>,
+    );
+
+    const trigger = document.querySelector('.draggable-trigger') as HTMLElement;
+    drag(trigger, [100, 50], [150, 90]);
+
+    expect(isDragged()).toBe(false);
+  });
+});

superset-frontend/packages/superset-ui-core/src/components/Modal/Modal.tsx

@@ -269,7 +269,6 @@ const CustomModal = ({
   );
   const draggableRef = useRef<HTMLDivElement>(null);
   const [bounds, setBounds] = useState<DraggableBounds>({});
-  const [dragDisabled, setDragDisabled] = useState<boolean>(true);
   const theme = useTheme();
 
   const handleOnHide = () => {
@@ -339,19 +338,7 @@ const CustomModal = ({
   }, [hideFooter, resizableConfig]);
 
   const ModalTitle = () =>
-    draggable ? (
-      <div
-        className="draggable-trigger"
-        onMouseOver={() => dragDisabled && setDragDisabled(false)}
-        onMouseOut={() => !dragDisabled && setDragDisabled(true)}
-        onFocus={() => dragDisabled && setDragDisabled(false)}
-        onBlur={() => !dragDisabled && setDragDisabled(true)}
-      >
-        {title}
-      </div>
-    ) : (
-      <>{title}</>
-    );
+    draggable ? <div className="draggable-trigger">{title}</div> : <>{title}</>;
 
   return (
     <StyledModal
@@ -378,13 +365,19 @@ const CustomModal = ({
       modalRender={modal =>
         resizable || draggable ? (
           <Draggable
-            disabled={!draggable || dragDisabled}
             bounds={bounds ?? false}
             onStart={(event, uiData) => onDragStart(event, uiData)}
+            {...draggableConfig}
+            // `disabled` and `handle` are applied after the spread so callers
+            // can't use `draggableConfig` to re-enable dragging on a
+            // non-draggable modal or move the drag handle off the title bar.
+            // A caller opting a draggable modal out via
+            // `draggableConfig.disabled` is still honored.
+            disabled={!draggable || !!draggableConfig?.disabled}
+            handle={draggable ? '.draggable-trigger' : undefined}
             // Pass nodeRef so react-draggable does not fall back to
             // ReactDOM.findDOMNode (deprecated in React 18+ Strict Mode).
             nodeRef={draggableRef}
-            {...draggableConfig}
           >
             {resizable ? (
               <Resizable className="resizable" {...getResizableConfig}>

superset-frontend/packages/superset-ui-core/src/components/Select/AsyncSelect.test.tsx

@@ -107,12 +107,14 @@ const getAllSelectOptions = () =>
 
 const findSelectOption = (text: string) =>
   waitFor(() =>
-    within(getElementByClassName('.rc-virtual-list')).getByText(text),
+    within(getElementByClassName('.ant-select-dropdown-list')).getByText(text),
   );
 
 const querySelectOption = (text: string) =>
   waitFor(() =>
-    within(getElementByClassName('.rc-virtual-list')).queryByText(text),
+    within(getElementByClassName('.ant-select-dropdown-list')).queryByText(
+      text,
+    ),
   );
 
 const findAllSelectOptions = () =>
@@ -644,7 +646,7 @@ test('does not add a new option if the option already exists', async () => {
   await type(option);
   await waitFor(() => {
     const array = within(
-      getElementByClassName('.rc-virtual-list'),
+      getElementByClassName('.ant-select-dropdown-list'),
     ).getAllByText(option);
     expect(array.length).toBe(1);
   });
@@ -1398,7 +1400,7 @@ test('appends page>1 results during an active search and discards them when sear
   // scrollTop via e.currentTarget in its onFallbackScroll handler, which
   // then forwards to onPopupScroll (handlePagination here).
   const holder = document.querySelector(
-    '.rc-virtual-list-holder',
+    '.ant-select-dropdown-list-holder',
   ) as HTMLElement | null;
   if (!holder) throw new Error('virtual-list holder not rendered');
   Object.defineProperty(holder, 'scrollHeight', {

superset-frontend/packages/superset-ui-core/src/components/Select/Select.test.tsx

@@ -93,12 +93,14 @@ const deselectAllButtonText = (length: number) =>
 
 const findSelectOption = (text: string) =>
   waitFor(() =>
-    within(getElementByClassName('.rc-virtual-list')).getByText(text),
+    within(getElementByClassName('.ant-select-dropdown-list')).getByText(text),
   );
 
 const querySelectOption = (text: string) =>
   waitFor(() =>
-    within(getElementByClassName('.rc-virtual-list')).queryByText(text),
+    within(getElementByClassName('.ant-select-dropdown-list')).queryByText(
+      text,
+    ),
   );
 
 const getAllSelectOptions = () =>

superset-frontend/packages/superset-ui-core/src/components/Select/types.ts

@@ -71,6 +71,9 @@ export type AntdExposedProps = Pick<
   | 'virtual'
   | 'getPopupContainer'
   | 'menuItemSelectedIcon'
+  // lets a caller with long option labels stop the popup inheriting the
+  // trigger's width, which otherwise truncates every option
+  | 'popupMatchSelectWidth'
 >;
 
 export type SelectOptionsType = Exclude<AntdProps['options'], undefined>;

superset-frontend/packages/superset-ui-core/src/components/ThemedAgGridReact/ThemedAgGridReact.test.tsx

@@ -246,6 +246,34 @@ test('wraps component with proper container div', () => {
   expect(wrapper).toHaveAttribute('data-themed-ag-grid', 'true');
 });
 
+test('applies non-transparent backgrounds to native menus, tooltips and overlays', () => {
+  const customTheme = {
+    ...supersetTheme,
+    colorBgElevated: '#f2f2f2',
+  };
+
+  render(
+    <ThemeProvider theme={customTheme}>
+      <ThemedAgGridReact rowData={mockRowData} columnDefs={mockColumnDefs} />
+    </ThemeProvider>,
+  );
+
+  const agGrid = screen.getByTestId('ag-grid-react');
+  const theme = JSON.parse(agGrid.getAttribute('data-theme') || '{}');
+
+  // ag-grid's own context/column menus, side bar, tooltips and overlays are
+  // rendered against these params rather than `backgroundColor` (which is
+  // intentionally 'transparent' so the surrounding app shows through the
+  // grid body). Without explicit values they inherit transparency too,
+  // making native menus/popups unreadable.
+  expect(theme.chromeBackgroundColor).toBe('#f2f2f2');
+  expect(theme.menuBackgroundColor).toBe('#f2f2f2');
+  expect(theme.menuBorder).toBe(true);
+  expect(theme.sideBarBackgroundColor).toBe('#f2f2f2');
+  expect(theme.tooltipBackgroundColor).toBe('#f2f2f2');
+  expect(theme.modalOverlayBackgroundColor).toBe('#f2f2f2');
+});
+
 test('handles missing theme gracefully', () => {
   const incompleteTheme = {
     ...supersetTheme,

superset-frontend/packages/superset-ui-core/src/components/ThemedAgGridReact/index.tsx

@@ -104,6 +104,17 @@ export const ThemedAgGridReact = forwardRef<
       foregroundColor: theme.colorText,
       browserColorScheme: isDarkMode ? 'dark' : 'light',
 
+      // Native menus, popups, side bar, tooltips and loading/no-rows overlays
+      // are rendered against these params rather than `backgroundColor`
+      // (which is intentionally transparent). Without explicit values they
+      // inherit transparency too, making them unreadable.
+      chromeBackgroundColor: theme.colorBgElevated,
+      menuBackgroundColor: theme.colorBgElevated,
+      menuBorder: true,
+      sideBarBackgroundColor: theme.colorBgElevated,
+      tooltipBackgroundColor: theme.colorBgElevated,
+      modalOverlayBackgroundColor: theme.colorBgElevated,
+
       // Header styling
       headerBackgroundColor: theme.colorFillTertiary,
       headerTextColor: theme.colorTextHeading,

superset-frontend/packages/superset-ui-core/src/number-format/factories/createSmartNumberFormatter.ts

@@ -64,9 +64,8 @@ export default function createSmartNumberFormatter(
     description,
     formatFunc: value => `${getSign(value)}${formatValue(value)}`,
     id:
-      id || signed
-        ? NumberFormats.SMART_NUMBER_SIGNED
-        : NumberFormats.SMART_NUMBER,
+      id ??
+      (signed ? NumberFormats.SMART_NUMBER_SIGNED : NumberFormats.SMART_NUMBER),
     label: label ?? 'Adaptive formatter',
   });
 }

superset-frontend/packages/superset-ui-core/src/query/types/Metric.ts

@@ -25,8 +25,11 @@ export type Aggregate =
   | 'COUNT'
   | 'COUNT_DISTINCT'
   | 'MAX'
+  | 'MEDIAN'
   | 'MIN'
-  | 'SUM';
+  | 'STDDEV_SAMP'
+  | 'SUM'
+  | 'VAR_SAMP';
 
 export interface AdhocMetricBase {
   hasCustomLabel?: boolean;

superset-frontend/packages/superset-ui-core/src/time-format/index.ts

@@ -49,5 +49,6 @@ export { default as finestTemporalGrainFormatter } from './formatters/finestTemp
 
 export { default as normalizeTimestamp } from './utils/normalizeTimestamp';
 export { default as denormalizeTimestamp } from './utils/denormalizeTimestamp';
+export { default as createTimeRangeFromGranularity } from './utils/createTimeRangeFromGranularity';
 
 export * from './types';

superset-frontend/packages/superset-ui-core/src/utils/featureFlags.ts

@@ -28,6 +28,7 @@ export enum FeatureFlag {
   AlertReportSlackV2 = 'ALERT_REPORT_SLACK_V2',
   AlertReportWebhook = 'ALERT_REPORT_WEBHOOK',
   AlertReportsFilter = 'ALERT_REPORTS_FILTER',
+  AlertReportsRetry = 'ALERT_REPORTS_RETRY',
   AllowFullCsvExport = 'ALLOW_FULL_CSV_EXPORT',
   ChartPluginsExperimental = 'CHART_PLUGINS_EXPERIMENTAL',
   ConfirmDashboardDiff = 'CONFIRM_DASHBOARD_DIFF',

superset-frontend/packages/superset-ui-core/src/utils/html.test.tsx

@@ -123,6 +123,25 @@ describe('isProbablyHTML', () => {
     expect(isProbablyHTML('<canvas></canvas>')).toBe(true);
     expect(isProbablyHTML('<iframe src="page.html"></iframe>')).toBe(true);
   });
+
+  test('should return true for script-capable and foreign-content tags', () => {
+    expect(isProbablyHTML('<svg onload="alert(1)"></svg>')).toBe(true);
+    expect(isProbablyHTML('<math><mi>x</mi></math>')).toBe(true);
+    expect(
+      isProbablyHTML('<details open ontoggle="alert(1)">x</details>'),
+    ).toBe(true);
+    expect(isProbablyHTML('<summary>x</summary>')).toBe(true);
+    expect(isProbablyHTML('<object data="x"></object>')).toBe(true);
+    expect(isProbablyHTML('<embed src="x">')).toBe(true);
+    expect(isProbablyHTML('<marquee>x</marquee>')).toBe(true);
+    expect(isProbablyHTML('<template>x</template>')).toBe(true);
+    expect(isProbablyHTML('<dialog open>x</dialog>')).toBe(true);
+  });
+
+  test('should return true for elements that parse into document.head', () => {
+    expect(isProbablyHTML('<style>body { display: none; }</style>')).toBe(true);
+    expect(isProbablyHTML('<title>injected</title>')).toBe(true);
+  });
 });
 
 describe('sanitizeHtmlIfNeeded', () => {
@@ -137,6 +156,24 @@ describe('sanitizeHtmlIfNeeded', () => {
     const sanitizedString = sanitizeHtmlIfNeeded(plainText);
     expect(sanitizedString).toEqual(plainText);
   });
+
+  test('should sanitize svg/details/style payloads instead of passing them through', () => {
+    const svgPayload = '<svg onload="alert(document.cookie)"></svg>';
+    const sanitizedSvg = sanitizeHtmlIfNeeded(svgPayload);
+    expect(sanitizedSvg).not.toContain('<svg');
+    expect(sanitizedSvg).not.toContain('onload');
+
+    // `details` (with its `open` attribute) is in js-xss's default
+    // whitelist, so the tag itself survives sanitization; the fix is that
+    // the payload is now routed through FilterXSS at all, which strips the
+    // non-whitelisted `ontoggle` handler instead of returning it verbatim.
+    const detailsPayload = '<details open ontoggle="alert(1)">x</details>';
+    const sanitizedDetails = sanitizeHtmlIfNeeded(detailsPayload);
+    expect(sanitizedDetails).toEqual('<details open>x</details>');
+
+    const stylePayload = '<style>body { display: none; }</style>';
+    expect(sanitizeHtmlIfNeeded(stylePayload)).not.toContain('<style');
+  });
 });
 
 describe('safeHtmlSpan', () => {

superset-frontend/packages/superset-ui-core/src/utils/html.tsx

@@ -154,6 +154,20 @@ const KNOWN_HTML_TAGS = new Set([
   'html',
   'head',
   'body',
+  // Script-capable elements and foreign-content roots (SVG/MathML). These
+  // must be classified as HTML so that downstream sanitization is applied;
+  // omitting them makes the heuristic fail open — payloads such as
+  // `<svg onload=...>` or `<details open ontoggle=...>` would be classified
+  // "not HTML" and returned verbatim by sanitizeHtmlIfNeeded.
+  'svg',
+  'math',
+  'details',
+  'summary',
+  'object',
+  'embed',
+  'marquee',
+  'template',
+  'dialog',
 ]);
 
 const HTML_TAG_PATTERN = new RegExp(
@@ -183,10 +197,15 @@ export function isProbablyHTML(text: string) {
   const parser = new DOMParser();
   const doc = parser.parseFromString(cleanedStr, 'text/html');
 
-  // Check if parsing created actual HTML elements (not just text nodes)
-  const elements = Array.from(doc.body.childNodes).filter(
-    node => node.nodeType === 1,
-  ) as Element[];
+  // Check if parsing created actual HTML elements (not just text nodes).
+  // Some elements (e.g. <style>, <title>, <meta>, <link>) parse into
+  // document.head rather than document.body, so both must be inspected —
+  // otherwise a bare <style> payload is classified "not HTML" and skips
+  // sanitization.
+  const elements = [
+    ...Array.from(doc.head.childNodes),
+    ...Array.from(doc.body.childNodes),
+  ].filter(node => node.nodeType === 1) as Element[];
 
   // If no elements were created, it's not HTML
   if (elements.length === 0) {

superset-frontend/packages/superset-ui-core/src/utils/tooltip.ts

@@ -19,19 +19,71 @@
 import { t } from '@apache-superset/core/translation';
 import { sanitizeHtml } from './html';
 
+export type TooltipTruncationMode = 'off' | 'end' | 'start' | 'middle';
+
+export const TRUNCATION_MAX_CHARS = 40;
+
 const TRUNCATION_STYLE = `
   max-width: 300px;
   overflow: hidden;
   text-overflow: ellipsis;
 `;
 
+const NOWRAP_STYLE = `
+  white-space: nowrap;
+`;
+
+/**
+ * Shortens plain text so a tooltip label stays readable, placing the ellipsis
+ * where the caller asked for it.
+ *
+ * Only 'start' and 'middle' slice. 'end' is handled by CSS in tooltipHtml, and
+ * 'off' means no truncation at all, so both return the input untouched.
+ *
+ * The input must be plain text. Callers are responsible for truncating before
+ * any markup (such as the ECharts series marker) is prepended, and before
+ * sanitization — slicing a string that already contains markup would cut into
+ * a tag.
+ */
+export function truncateLabel(
+  text: string,
+  mode: TooltipTruncationMode = 'end',
+): string {
+  if (
+    (mode !== 'start' && mode !== 'middle') ||
+    text.length <= TRUNCATION_MAX_CHARS
+  ) {
+    return text;
+  }
+  const budget = TRUNCATION_MAX_CHARS - 1;
+  if (mode === 'start') {
+    return `…${text.slice(-budget)}`;
+  }
+  const head = Math.ceil(budget / 2);
+  const tail = Math.floor(budget / 2);
+  return `${text.slice(0, head)}…${text.slice(-tail)}`;
+}
+
+function getTruncationStyle(mode: TooltipTruncationMode): string {
+  if (mode === 'end') {
+    return TRUNCATION_STYLE;
+  }
+  if (mode === 'off') {
+    return '';
+  }
+  // 'start' and 'middle' are already sliced upstream; keep them on one line.
+  return NOWRAP_STYLE;
+}
+
 export function tooltipHtml(
   data: string[][],
   title?: string,
   focusedRow?: number,
+  truncation: TooltipTruncationMode = 'end',
 ) {
+  const truncationStyle = getTruncationStyle(truncation);
   const titleRow = title
-    ? `<span style="font-weight: 700;${TRUNCATION_STYLE}">${title}</span>`
+    ? `<span style="font-weight: 700;${truncationStyle}">${title}</span>`
     : '';
   return sanitizeHtml(`
     <div>
@@ -46,7 +98,7 @@ export function tooltipHtml(
                 const cellStyle = `
                   text-align: ${j > 0 ? 'right' : 'left'};
                   padding-left: ${j === 0 ? 0 : 16}px;
-                  ${TRUNCATION_STYLE}
+                  ${truncationStyle}
                 `;
                 return `<td style="${cellStyle}">${cell}</td>`;
               });

superset-frontend/packages/superset-ui-core/test/number-format/factories/createSmartNumberFormatter.test.ts

@@ -24,6 +24,12 @@ describe('createSmartNumberFormatter(options)', () => {
     const formatter = createSmartNumberFormatter();
     expect(formatter).toBeInstanceOf(NumberFormatter);
   });
+  test('uses the supplied formatter id regardless of signed option', () => {
+    expect(createSmartNumberFormatter({ id: 'custom' }).id).toBe('custom');
+    expect(
+      createSmartNumberFormatter({ id: 'custom-signed', signed: true }).id,
+    ).toBe('custom-signed');
+  });
   describe('using default options', () => {
     const formatter = createSmartNumberFormatter();
     test('formats 0 correctly', () => {

superset-frontend/packages/superset-ui-core/test/utils/tooltip.test.ts

@@ -16,7 +16,12 @@
  * specific language governing permissions and limitations
  * under the License.
  */
-import { sanitizeHtml, tooltipHtml } from '@superset-ui/core';
+import {
+  sanitizeHtml,
+  tooltipHtml,
+  truncateLabel,
+  TRUNCATION_MAX_CHARS,
+} from '@superset-ui/core';
 
 const TITLE_STYLE =
   'style="font-weight: 700;max-width:300px;overflow:hidden;text-overflow:ellipsis;"';
@@ -182,3 +187,88 @@ test('should preserve table styling after sanitization (fixes ECharts tooltip fo
   expect(html).toContain('padding-left:16px');
   expect(html).toContain('max-width:300px');
 });
+
+describe('truncateLabel', () => {
+  const long = 'prod-us-east-1-service-checkout-latency-p99'; // 43 chars
+
+  test('returns text unchanged for off and end', () => {
+    expect(truncateLabel(long, 'off')).toBe(long);
+    expect(truncateLabel(long, 'end')).toBe(long);
+  });
+
+  test('defaults to end, which does not slice', () => {
+    expect(truncateLabel(long)).toBe(long);
+  });
+
+  test('truncates the start, keeping the distinguishing suffix', () => {
+    expect(truncateLabel(long, 'start')).toBe(
+      '…-us-east-1-service-checkout-latency-p99',
+    );
+    expect(truncateLabel(long, 'start')).toHaveLength(TRUNCATION_MAX_CHARS);
+  });
+
+  test('truncates the middle, keeping both ends', () => {
+    expect(truncateLabel(long, 'middle')).toBe(
+      'prod-us-east-1-servi…heckout-latency-p99',
+    );
+    expect(truncateLabel(long, 'middle')).toHaveLength(TRUNCATION_MAX_CHARS);
+  });
+
+  test('leaves text at or under the limit untouched', () => {
+    const atLimit = 'x'.repeat(TRUNCATION_MAX_CHARS);
+    expect(truncateLabel(atLimit, 'start')).toBe(atLimit);
+    expect(truncateLabel(atLimit, 'middle')).toBe(atLimit);
+    expect(truncateLabel('short', 'start')).toBe('short');
+    expect(truncateLabel('', 'middle')).toBe('');
+  });
+
+  test('truncates text one character over the limit', () => {
+    const overLimit = 'x'.repeat(TRUNCATION_MAX_CHARS + 1);
+    expect(truncateLabel(overLimit, 'start')).toBe(
+      `…${'x'.repeat(TRUNCATION_MAX_CHARS - 1)}`,
+    );
+  });
+});
+
+describe('tooltipHtml truncation modes', () => {
+  const rows = [['label', 'value']];
+
+  // sanitizeHtml normalizes spacing inside style attributes, and it does so
+  // differently across versions, so compare with whitespace stripped.
+  const styles = (
+    title: string | undefined,
+    truncation?: 'off' | 'end' | 'start' | 'middle',
+  ) => removeWhitespaces(tooltipHtml(rows, title, undefined, truncation));
+
+  test('emits the 300px cap for end and for the default', () => {
+    expect(styles('Title', 'end')).toContain('max-width:300px');
+    expect(tooltipHtml(rows, 'Title')).toBe(
+      tooltipHtml(rows, 'Title', undefined, 'end'),
+    );
+  });
+
+  test('emits no truncation style for off', () => {
+    const html = styles('Title', 'off');
+    expect(html).not.toContain('max-width');
+    expect(html).not.toContain('text-overflow');
+    expect(html).not.toContain('white-space');
+  });
+
+  test.each(['start', 'middle'] as const)(
+    'emits nowrap instead of a cap for %s',
+    mode => {
+      const html = styles('Title', mode);
+      expect(html).toContain('white-space:nowrap');
+      expect(html).not.toContain('max-width');
+    },
+  );
+
+  test('never slices cell text itself, whatever the mode', () => {
+    const longCell = 'y'.repeat(TRUNCATION_MAX_CHARS + 20);
+    (['off', 'end', 'start', 'middle'] as const).forEach(mode => {
+      expect(tooltipHtml([[longCell]], undefined, undefined, mode)).toContain(
+        longCell,
+      );
+    });
+  });
+});

superset-frontend/playwright.config.ts

@@ -47,6 +47,10 @@ export default defineConfig({
   // Retry logic - 2 retries in CI, 0 locally
   retries: process.env.CI ? 2 : 0,
 
+  // Disable capturing Git commit info as the project's history is increasingly dense
+  // and breach Playwright's default 3-seconds `git` command timeout limit
+  captureGitInfo: { commit: false, diff: false },
+
   // Reporter configuration - multiple reporters for better visibility
   reporter: process.env.CI
     ? [