test(query-context): update normalize_df mock signature for new offset param

normalize_df now accepts an optional third sql_shifted_temporal_labels parameter (passed positionally from processing_time_offsets). Three pre-existing mocks in this file stubbed normalize_df with a fixed two-argument lambda, which broke when the call site started passing a third positional argument.
test(dataset): regression guards for Hours Offset bound and grain bugs
2026-08-21 07:31:17 +00:00 · 2026-08-20 18:06:05 +00:00 · 2026-08-20 18:06:05 +00:00 · 2026-08-20 18:06:05 +00:00 · 2026-08-20 11:04:19 -07:00 · 2026-08-20 11:00:54 -07:00
193 changed files with 2077 additions and 349 deletions
@@ -1,6 +1,6 @@
 name: Bug report
 description: Report a bug to improve Superset's stability
-labels: ["bug"]
+labels: ["#bug"]
 body:
  - type: markdown
    attributes:
@@ -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
@@ -67,7 +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.
@@ -78,6 +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}}"
@@ -93,7 +93,7 @@
    "@typescript-eslint/parser": "^8.67.0",
    "eslint": "^9.39.2",
    "eslint-plugin-react": "^7.37.5",
-    "globals": "^17.10.0",
+    "globals": "^17.11.0",
    "oxfmt": "^0.63.0",
    "typescript": "~6.0.3",
    "typescript-eslint": "^8.67.0",
@@ -9174,10 +9174,10 @@ globals@^14.0.0:
  resolved "https://registry.yarnpkg.com/globals/-/globals-14.0.0.tgz#898d7413c29babcf6bafe56fcadded858ada724e"
  integrity sha512-oahGvuMGQlPw/ivIYBjVSrWAfWLBeku5tpPE2fOPLi+WHffIWbuh2tCjhyQhTBPMf5E9jDEH4FOmTYgYwbKwtQ==

-globals@^17.10.0:
-  version "17.10.0"
-  resolved "https://registry.yarnpkg.com/globals/-/globals-17.10.0.tgz#f9dbd847ae99e236f98b13095e2426ac3b25a45c"
-  integrity sha512-V0kztuWST2k8A/VbxAY8+L+7+Rgo3fyA24IHRLrZp7HOzJjV0gHSaZUjK9lpP/IrBSNite2tZ1prhRkinRu1CA==
+globals@^17.11.0:
+  version "17.11.0"
+  resolved "https://registry.yarnpkg.com/globals/-/globals-17.11.0.tgz#d643485bb30220d7751e511cf4f68c73d3870d87"
+  integrity sha512-Z2I8hM+PbJDXQDq3Icgpzv+mPdwr68iZUU9d5WW4FuXfDUQfkZaZuvjMv42/5crNyw154+9+VWXbYrUgDXbxNw==

 globalthis@^1.0.4:
  version "1.0.4"
@@ -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()
@@ -99,7 +99,7 @@
        "geostyler-openlayers-parser": "^5.7.1",
        "geostyler-style": "11.0.2",
        "geostyler-wfs-parser": "^3.0.1",
-        "google-auth-library": "^11.0.1",
+        "google-auth-library": "^11.0.2",
        "immer": "^11.1.16",
        "interweave": "^13.1.1",
        "jquery": "^4.0.0",
@@ -20612,7 +20612,7 @@
      "version": "0.8.0",
      "resolved": "https://registry.npmjs.org/expect-playwright/-/expect-playwright-0.8.0.tgz",
      "integrity": "sha512-+kn8561vHAY+dt+0gMqqj1oY+g5xWrsuGMk4QGxotT2WS545nVqqjs37z6hrYfIuucwqthzwJfCJUEYqixyljg==",
-      "deprecated": "\u26a0\ufe0f The 'expect-playwright' package is deprecated. The Playwright core assertions (via @playwright/test) now cover the same functionality. Please migrate to built-in expect. See https://playwright.dev/docs/test-assertions for migration.",
+      "deprecated": "⚠️ The 'expect-playwright' package is deprecated. The Playwright core assertions (via @playwright/test) now cover the same functionality. Please migrate to built-in expect. See https://playwright.dev/docs/test-assertions for migration.",
      "dev": true,
      "license": "MIT"
    },
@@ -22783,9 +22783,9 @@
      "license": "MIT"
    },
    "node_modules/google-auth-library": {
-      "version": "11.0.1",
-      "resolved": "https://registry.npmjs.org/google-auth-library/-/google-auth-library-11.0.1.tgz",
-      "integrity": "sha512-ZqfaYduu9ASUaFuUk5dF9g9QvufdhhSj7jFiEnCrTQcH57sFPKYetM0iU4dcKkQk6CqC1xpSrVr5uQ9NhqjNOg==",
+      "version": "11.0.2",
+      "resolved": "https://registry.npmjs.org/google-auth-library/-/google-auth-library-11.0.2.tgz",
+      "integrity": "sha512-vzpgPutxrghPsnjrjpzLX2bdv8IOL719Rh0oEjGnQu8YCIbnbMuTTQ5zU9LcKvLdOPgCxBwppbvnhgW90Qna5Q==",
      "license": "Apache-2.0",
      "dependencies": {
        "base64-js": "^1.3.0",
@@ -26023,7 +26023,7 @@
      "version": "0.4.0",
      "resolved": "https://registry.npmjs.org/jest-process-manager/-/jest-process-manager-0.4.0.tgz",
      "integrity": "sha512-80Y6snDyb0p8GG83pDxGI/kQzwVTkCxc7ep5FPe/F6JYdvRDhwr6RzRmPSP7SEwuLhxo80lBS/NqOdUIbHIfhw==",
-      "deprecated": "\u26a0\ufe0f The 'jest-process-manager' package is deprecated. Please migrate to Playwright's built-in test runner (@playwright/test) which now includes full Jest-style features and parallel testing. See https://playwright.dev/docs/intro for details.",
+      "deprecated": "⚠️ The 'jest-process-manager' package is deprecated. Please migrate to Playwright's built-in test runner (@playwright/test) which now includes full Jest-style features and parallel testing. See https://playwright.dev/docs/intro for details.",
      "dev": true,
      "license": "MIT",
      "dependencies": {
@@ -43073,15 +43073,6 @@
        "node": ">=12"
      }
    },
-    "packages/superset-ui-core/node_modules/dompurify": {
-      "version": "3.4.13",
-      "resolved": "https://registry.npmjs.org/dompurify/-/dompurify-3.4.13.tgz",
-      "integrity": "sha512-2vmYIoqjze2d+kakP8S/nS5shfsl587kzwEjcGlTdiksUVgFHnFCsLYDVj/JNqJVOQZGSYBTmuycv0PodwmnMQ==",
-      "license": "(MPL-2.0 OR Apache-2.0)",
-      "optionalDependencies": {
-        "@types/trusted-types": "^2.0.7"
-      }
-    },
    "packages/superset-ui-core/node_modules/react-ace": {
      "version": "14.0.1",
      "resolved": "https://registry.npmjs.org/react-ace/-/react-ace-14.0.1.tgz",
@@ -176,7 +176,7 @@
    "geostyler-openlayers-parser": "^5.7.1",
    "geostyler-style": "11.0.2",
    "geostyler-wfs-parser": "^3.0.1",
-    "google-auth-library": "^11.0.1",
+    "google-auth-library": "^11.0.2",
    "immer": "^11.1.16",
    "interweave": "^13.1.1",
    "jquery": "^4.0.0",
@@ -467,6 +467,14 @@ export function transformSeries(
          return formatter(numericValue);
        }
        if (!onlyTotal) {
+          // A stacked segment with no height begins and ends at the same
+          // coordinate as the top of the segment beneath it, so its label is
+          // drawn over that segment's label. Zero and null have no height, so
+          // they carry no label. The rich tooltip omits zero observations from
+          // a stacked series for the same reason.
+          if (stack && !numericValue) {
+            return '';
+          }
          if (
            numericValue >=
            (thresholdValues[dataIndex] || Number.MIN_SAFE_INTEGER)
@@ -20,13 +20,14 @@ import {
  CategoricalColorScale,
  ChartProps,
  TimeGranularity,
+  getNumberFormatter,
 } from '@superset-ui/core';
 import { GenericDataType } from '@apache-superset/core/common';
 import { supersetTheme } from '@apache-superset/core/theme';
 import type { SeriesOption } from 'echarts';
 import type { ScatterSeriesOption } from 'echarts/charts';
 import { EchartsTimeseriesSeriesType } from '../../src';
-import { TIMESERIES_CONSTANTS } from '../../src/constants';
+import { StackControlsValue, TIMESERIES_CONSTANTS } from '../../src/constants';
 import {
  LegendOrientation,
  EchartsTimeseriesChartProps,
@@ -566,3 +567,70 @@ test('getPadding should handle Left position with zero margin correctly', () =>
    getChartPaddingSpy.mockRestore();
  }
 });
+
+/**
+ * #42702: a stacked segment with no height starts and ends at the same
+ * coordinate as the top of the segment beneath it, so a value label on it is
+ * drawn over that segment's label. `percentage_threshold` does not filter these
+ * out: it defaults to 0, and `thresholdValues[dataIndex] || MIN_SAFE_INTEGER`
+ * turns a 0 threshold into "no filtering", which is intentional.
+ */
+const stackedLabel = (
+  numericValue: number | null,
+  opts: Record<string, unknown> = {},
+) => {
+  const series = transformSeries(
+    { id: 'B', name: 'B', data: [[1, numericValue]] } as SeriesOption,
+    mockColorScale,
+    'B',
+    {
+      seriesType: EchartsTimeseriesSeriesType.Bar,
+      stack: StackControlsValue.Stack,
+      showValue: true,
+      onlyTotal: false,
+      formatter: getNumberFormatter(),
+      thresholdValues: [0],
+      ...opts,
+    },
+  ) as SeriesOption & {
+    label: { formatter: (params: unknown) => string };
+  };
+  return series.label.formatter({
+    value: [1, numericValue],
+    dataIndex: 0,
+    seriesIndex: 1,
+    seriesName: 'B',
+  });
+};
+
+test('stacked value labels are omitted for a zero-height segment', () => {
+  expect(stackedLabel(0)).toBe('');
+  expect(stackedLabel(null)).toBe('');
+});
+
+test('stacked value labels are kept for segments that have height', () => {
+  expect(stackedLabel(32)).toBe('32');
+  expect(stackedLabel(-5)).toBe('-5');
+});
+
+test('a zero value keeps its label when the series is not stacked', () => {
+  // Without a stack the label sits on the bar itself, so there is nothing for
+  // it to collide with.
+  expect(stackedLabel(0, { stack: undefined })).toBe('0');
+});
+
+test('percentage_threshold still filters values below the threshold', () => {
+  // 10% of a 100 total. The zero-height guard must not swallow this rule.
+  expect(stackedLabel(5, { thresholdValues: [10] })).toBe('');
+  expect(stackedLabel(50, { thresholdValues: [10] })).toBe('50');
+});
+
+test('only-total labels are unaffected by the zero-height guard', () => {
+  expect(
+    stackedLabel(0, {
+      onlyTotal: true,
+      showValueIndexes: [1],
+      totalStackedValues: [32],
+    }),
+  ).toBe('32');
+});
@@ -223,6 +223,11 @@ export default function chartReducer(
  }

  if (action.type in actionHandlers) {
+    // ADD_CHART creates the entry, so it runs without prior state; every other
+    // handler reads state that is absent once the chart has been removed
+    if (action.type !== actions.ADD_CHART && !charts[action.key]) {
+      return charts;
+    }
    return {
      ...charts,
      [action.key]: actionHandlers[action.type](charts[action.key]),
@@ -91,4 +91,20 @@ describe('chart reducers', () => {
    expect(newState[chartKey].chartUpdateEndTime).toBeGreaterThan(0);
    expect(newState[chartKey].chartStatus).toEqual('failed');
  });
+
+  test('ignores an action for a chart that is no longer in state', () => {
+    const action = actions.chartUpdateStopped(999, new AbortController());
+    expect(() => chartReducer(charts, action)).not.toThrow();
+    expect(chartReducer(charts, action)).toEqual(charts);
+  });
+
+  test('still adds a chart that is not yet in state', () => {
+    const newChartKey = 2;
+    const newState = chartReducer(
+      charts,
+      actions.addChart({ ...chart, id: newChartKey }, newChartKey),
+    );
+    expect(newState[newChartKey].id).toEqual(newChartKey);
+    expect(newState[chartKey]).toEqual(testChart);
+  });
 });
@@ -105,10 +105,17 @@ export const SamplesPane = ({
        1,
      )
        .then(response => {
-          setData(ensureIsArray(response.data));
-          setColnames(ensureIsArray(response.colnames));
-          setColtypes(ensureIsArray(response.coltypes));
-          setRowCount(response.rowcount);
+          // A 200 that carries no `result` payload resolves to undefined here.
+          // Read through it so the pane falls back to its empty state instead
+          // of throwing a TypeError that surfaces as an internal error message.
+          const rows = ensureIsArray(response?.data);
+          setData(rows);
+          setColnames(ensureIsArray(response?.colnames));
+          setColtypes(ensureIsArray(response?.coltypes));
+          // Fall back to the rows actually returned rather than to zero: the
+          // controls only render when there are rows, and a hardcoded 0 would
+          // label a populated table as "0 rows".
+          setRowCount(response?.rowcount ?? rows.length);
          setResponseError('');
          cache.set(queryFormData, true);
          if (queryForce) {
@@ -60,6 +60,27 @@ describe('SamplesPane', () => {
    400,
  );

+  // A 200 response that carries no `result` payload, as reported in #36840.
+  fetchMock.post(
+    'end:/datasource/samples?force=false&datasource_type=table&datasource_id=37&per_page=100&page=1',
+    {},
+  );
+
+  // A 200 whose result carries rows but omits `rowcount`.
+  fetchMock.post(
+    'end:/datasource/samples?force=false&datasource_type=table&datasource_id=38&per_page=100&page=1',
+    {
+      result: {
+        data: [
+          { __timestamp: 1230768000000, genre: 'Action' },
+          { __timestamp: 1230768000010, genre: 'Horror' },
+        ],
+        colnames: ['__timestamp', 'genre'],
+        coltypes: [2, 1],
+      },
+    },
+  );
+
  const setForceQuery = jest.fn();

  afterAll(() => {
@@ -114,4 +135,29 @@ describe('SamplesPane', () => {
    expect(queryByText('Action')).toBeVisible();
    expect(queryByText('Horror')).toBeVisible();
  });
+
+  test('renders the empty state when the response carries no result payload', async () => {
+    const props = createSamplesPaneProps({ datasourceId: 37 });
+    const { findByText, queryByRole } = render(<SamplesPane {...props} />, {
+      useRedux: true,
+    });
+
+    expect(
+      await findByText('No samples were returned for this dataset'),
+    ).toBeVisible();
+    // The pane should not leak an internal TypeError through the error alert.
+    expect(queryByRole('alert')).not.toBeInTheDocument();
+  });
+
+  test('counts the returned rows when the response omits rowcount', async () => {
+    const props = createSamplesPaneProps({ datasourceId: 38 });
+    const { findByText, queryByText } = render(<SamplesPane {...props} />, {
+      useRedux: true,
+    });
+
+    expect(await findByText('Action')).toBeVisible();
+    // Falling back to 0 here would label a populated table as "0 rows".
+    expect(queryByText('0 rows')).not.toBeInTheDocument();
+    expect(queryByText('2 rows')).toBeVisible();
+  });
 });
@@ -28,7 +28,7 @@
        "@typescript-eslint/parser": "^8.67.0",
        "eslint": "^10.8.1",
        "eslint-config-prettier": "^10.1.8",
-        "globals": "^17.10.0",
+        "globals": "^17.11.0",
        "oxfmt": "^0.63.0",
        "tscw-config": "^1.1.2",
        "typescript": "^6.0.3",
@@ -2053,9 +2053,9 @@
      }
    },
    "node_modules/globals": {
-      "version": "17.10.0",
-      "resolved": "https://registry.npmjs.org/globals/-/globals-17.10.0.tgz",
-      "integrity": "sha512-V0kztuWST2k8A/VbxAY8+L+7+Rgo3fyA24IHRLrZp7HOzJjV0gHSaZUjK9lpP/IrBSNite2tZ1prhRkinRu1CA==",
+      "version": "17.11.0",
+      "resolved": "https://registry.npmjs.org/globals/-/globals-17.11.0.tgz",
+      "integrity": "sha512-Z2I8hM+PbJDXQDq3Icgpzv+mPdwr68iZUU9d5WW4FuXfDUQfkZaZuvjMv42/5crNyw154+9+VWXbYrUgDXbxNw==",
      "dev": true,
      "license": "MIT",
      "engines": {
@@ -36,7 +36,7 @@
    "@typescript-eslint/parser": "^8.67.0",
    "eslint": "^10.8.1",
    "eslint-config-prettier": "^10.1.8",
-    "globals": "^17.10.0",
+    "globals": "^17.11.0",
    "oxfmt": "^0.63.0",
    "tscw-config": "^1.1.2",
    "typescript": "^6.0.3",
@@ -280,9 +280,6 @@ def test_sqlalchemy_dialect(
    """
    Test the SQLAlchemy dialect, making sure it supports everything Superset needs.
    """
-    if "future" not in engine_kwargs:
-        engine_kwargs["future"] = True
-
    engine = create_engine(sqlalchemy_uri, **engine_kwargs)
    dialect = engine.dialect

@@ -227,7 +227,7 @@ class BaseStreamingCSVExportCommand(BaseCommand):
        delimiter = csv_export_config.get("sep", ",")
        decimal_separator = csv_export_config.get("decimal", ".")

-        with db.session(future=True) as session:
+        with db.session() as session:
            # Merge database to prevent DetachedInstanceError
            merged_database = session.merge(database)

@@ -291,19 +291,30 @@ class QueryContextFactory:  # pylint: disable=too-few-public-methods
                    ),
                    None,
                )
-                # Replaces x-axis column values with granularity
+                # Point the x-axis at the overridden Time Column (granularity).
                if x_axis_column:
                    if isinstance(x_axis_column, dict):
+                        # Only swap the underlying expression, keeping the
+                        # column's original label. The temporal offset join
+                        # (``processing_time_offsets``), the post-processing
+                        # pivot ``index`` and the frontend all reference this
+                        # column by its label; renaming it to the granularity
+                        # here desynchronizes those consumers from the label
+                        # the saved chart still advertises, which — with a Time
+                        # Comparison offset — collapses the series into a single
+                        # point.
                        x_axis_column["sqlExpression"] = granularity
-                        x_axis_column["label"] = granularity
                    else:
+                        # A bare string x-axis has no distinct label, so it is
+                        # replaced wholesale and the pivot ``index`` must be
+                        # realigned to the overridden column.
                        query_object.columns = [
                            granularity if column == x_axis_column else column
                            for column in query_object.columns
                        ]
-                    for post_processing in query_object.post_processing:
-                        if post_processing.get("operation") == "pivot":
-                            post_processing["options"]["index"] = [granularity]
+                        for post_processing in query_object.post_processing:
+                            if post_processing.get("operation") == "pivot":
+                                post_processing["options"]["index"] = [granularity]

            # If no temporal x-axis, then get the default temporal filter
            if not filter_to_remove:
@@ -17,6 +17,7 @@
 # pylint: disable=invalid-name
 from __future__ import annotations

+import inspect
 import logging
 from datetime import datetime
 from pprint import pformat
@@ -205,8 +206,86 @@ class QueryObject:  # pylint: disable=too-many-instance-attributes
    def _set_post_processing(
        self, post_processing: list[dict[str, Any] | None] | None
    ) -> None:
-        post_processing = post_processing or []
-        self.post_processing = [post_proc for post_proc in post_processing if post_proc]
+        self.post_processing = [
+            self._drop_unsupported_options(post_proc)
+            for post_proc in post_processing or []
+            if post_proc
+        ]
+
+    @staticmethod
+    def _drop_unsupported_options(post_proc: dict[str, Any]) -> dict[str, Any]:
+        """
+        Drop options that the post-processing operation no longer accepts.
+
+        A chart's ``query_context`` is written when the chart is saved and is
+        never rewritten afterwards, while Explore rebuilds the query from
+        ``form_data`` at every render. A chart saved by an older version of
+        Superset can therefore reference an option that has since been removed
+        from the operation. ``exec_post_processing`` passes the stored options
+        as keyword arguments, so that option raises a bare ``TypeError`` on
+        every path that replays the stored ``query_context`` -- the chart data
+        endpoint, alerts and reports, thumbnails, CSV export -- while the same
+        chart still renders correctly in Explore.
+
+        Comparing against the signature avoids a hard-coded list of removed
+        option names, which would need extending at each release.
+        """
+        operation = post_proc.get("operation")
+        function = (
+            getattr(pandas_postprocessing, operation, None)
+            if isinstance(operation, str)
+            else None
+        )
+        if function is None:
+            # A missing or unknown operation is left untouched, so that
+            # exec_post_processing reports it as InvalidPostProcessingError.
+            return post_proc
+
+        parameters = inspect.signature(function).parameters
+        if any(
+            parameter.kind is inspect.Parameter.VAR_KEYWORD
+            for parameter in parameters.values()
+        ):
+            return post_proc
+
+        # `exec_post_processing` calls the operation as `operation(df, **options)`,
+        # so an option can only reach a parameter that a caller may fill by
+        # keyword. That excludes the first parameter, which receives the
+        # DataFrame positionally, and any positional-only or `*args` parameter.
+        keyword_parameters = {
+            name
+            for position, (name, parameter) in enumerate(parameters.items())
+            if position > 0
+            and parameter.kind
+            in (
+                inspect.Parameter.POSITIONAL_OR_KEYWORD,
+                inspect.Parameter.KEYWORD_ONLY,
+            )
+        }
+
+        options = post_proc.get("options") or {}
+        unsupported = {key for key in options if key not in keyword_parameters}
+        if not unsupported:
+            return post_proc
+
+        # Logged at info: a chart saved before the option was removed hits this
+        # on every render, so a warning would repeat for as long as the chart
+        # is not resaved, without anything new to report.
+        logger.info(
+            "Dropping unsupported option(s) %s of post-processing operation "
+            "`%s`. The chart's stored query_context predates the current "
+            "signature of that operation.",
+            sorted(unsupported),
+            operation,
+        )
+        return {
+            **post_proc,
+            "options": {
+                key: value
+                for key, value in options.items()
+                if key in keyword_parameters
+            },
+        }

    def _init_series_columns(
        self,
@@ -99,6 +99,7 @@ from superset.models.helpers import (
    AuditMixinNullable,
    CertificationMixin,
    ExploreMixin,
+    get_effective_hours_offset,
    ImportExportMixin,
    QueryResult,
    SoftDeleteMixin,
@@ -1247,6 +1248,8 @@ class TableColumn(AuditMixinNullable, ImportExportMixin, CertificationMixin, Mod
        time_grain: str | None,
        label: str | None = None,
        template_processor: BaseTemplateProcessor | None = None,
+        apply_dataset_offset: bool = False,
+        sql_shifted_temporal_labels: set[str] | None = None,
    ) -> TimestampExpression | Label:
        """
        Return a SQLAlchemy Core element representation of self to be used in a query.
@@ -1254,6 +1257,8 @@ class TableColumn(AuditMixinNullable, ImportExportMixin, CertificationMixin, Mod
        :param time_grain: Optional time grain, e.g. P1Y
        :param label: alias/label that column is expected to have
        :param template_processor: template processor
+        :param apply_dataset_offset: shift the selected axis before truncation
+        :param sql_shifted_temporal_labels: labels shifted before truncation
        :return: A TimeExpression object wrapped in a Label if supported by db
        """
        label = label or utils.DTTM_ALIAS
@@ -1291,6 +1296,27 @@ class TableColumn(AuditMixinNullable, ImportExportMixin, CertificationMixin, Mod
            col = literal_column(expression, type_=type_)
        else:
            col = column(self.column_name, type_=type_)
+        if (
+            apply_dataset_offset
+            and time_grain
+            and self.table
+            and self.db_engine_spec.supports_temporal_column_shift
+            and (offset_hours := self.table.offset or 0)
+            and not self.table.get_dataset_timezone()
+        ):
+            effective_offset_hours = get_effective_hours_offset(
+                self.db_engine_spec,
+                self.type,
+                offset_hours,
+                db_extra=self.db_extra,
+            )
+            if effective_offset_hours:
+                col = self.db_engine_spec.get_temporal_column_shift_expr(
+                    col,
+                    effective_offset_hours,
+                )
+            if sql_shifted_temporal_labels is not None:
+                sql_shifted_temporal_labels.add(label)
        time_expr = self.db_engine_spec.get_timestamp_expr(col, pdf, time_grain)
        return self.database.make_sqla_column_compatible(time_expr, label)

@@ -1960,11 +1986,26 @@ class SqlaTable(
                )
            ) from ex

+    def _shift_temporal_column_if_needed(
+        self,
+        sqla_column: ColumnClause,
+        effective_offset_hours: int,
+    ) -> ColumnClause:
+        """Apply a nonzero effective dataset offset to a temporal expression."""
+        if not effective_offset_hours:
+            return sqla_column
+        return self.db_engine_spec.get_temporal_column_shift_expr(
+            sqla_column,
+            effective_offset_hours,
+        )
+
    def adhoc_column_to_sqla(  # pylint: disable=too-many-locals
        self,
        col: AdhocColumn,
        force_type_check: bool = False,
        template_processor: BaseTemplateProcessor | None = None,
+        apply_dataset_offset: bool = False,
+        sql_shifted_temporal_labels: set[str] | None = None,
    ) -> tuple[ColumnElement, utils.GenericDataType | None]:
        """
        Turn an adhoc column into a sqlalchemy column.
@@ -1974,6 +2015,8 @@ class SqlaTable(
               This is needed to validate if a filter with an adhoc column
               is applicable.
        :param template_processor: template_processor instance
+        :param apply_dataset_offset: shift the selected axis before truncation
+        :param sql_shifted_temporal_labels: labels shifted before truncation
        :returns: A tuple of (SQLAlchemy column, generic column type). The
            generic type is populated when the column type is resolved
            (either because the adhoc column matches a physical column, or
@@ -1990,6 +2033,7 @@ class SqlaTable(
        pdf = None
        is_column_reference = col.get("isColumnReference", False)
        generic_type: utils.GenericDataType | None = None
+        native_type: str | None = None

        metadata_lookup_key = self._render_adhoc_expression_for_metadata_lookup(
            sql_expression, template_processor
@@ -2004,6 +2048,7 @@ class SqlaTable(
            is_dttm = col_in_metadata.is_temporal
            pdf = col_in_metadata.python_date_format
            generic_type = col_in_metadata.type_generic
+            native_type = col_in_metadata.type
        else:
            # Column doesn't exist in metadata or is not a reference - treat as ad-hoc
            # expression Note: If isColumnReference=true but column not found, we still
@@ -2066,8 +2111,28 @@ class SqlaTable(
                # stay unquoted for numeric adhoc expressions like
                # CAST(... AS BIGINT)).
                generic_type = col_desc[0].get("type_generic")
+                probed_type = col_desc[0].get("type")
+                native_type = str(probed_type) if probed_type is not None else None

        if is_dttm and has_timegrain:
+            if (
+                apply_dataset_offset
+                and self.db_engine_spec.supports_temporal_column_shift
+                and (offset_hours := self.offset or 0)
+                and not self.get_dataset_timezone()
+            ):
+                effective_offset_hours = get_effective_hours_offset(
+                    self.db_engine_spec,
+                    native_type,
+                    offset_hours,
+                    db_extra=self.db_extra,
+                )
+                sqla_column = self._shift_temporal_column_if_needed(
+                    sqla_column,
+                    effective_offset_hours,
+                )
+                if sql_shifted_temporal_labels is not None:
+                    sql_shifted_temporal_labels.add(label)
            sqla_column = self.db_engine_spec.get_timestamp_expr(
                col=sqla_column,
                pdf=pdf,
@@ -539,6 +539,7 @@ class BaseEngineSpec:  # pylint: disable=too-many-public-methods
    # the ``array_*`` capability methods below must be implemented. Defaults to
    # False so engines that have not opted in keep treating arrays as strings.
    supports_multivalue_columns = False
+    supports_temporal_column_shift: bool = False
    allows_joins = True
    allows_subqueries = True
    allows_alias_in_select = True
@@ -1215,6 +1216,19 @@ class BaseEngineSpec:  # pylint: disable=too-many-public-methods

        return TimestampExpression(time_expr, col, type_=col.type)

+    @classmethod
+    def get_temporal_column_shift_expr(
+        cls,
+        col: ColumnClause,
+        offset_hours: int,
+    ) -> TimestampExpression:
+        """Shift a temporal SQL expression by a bounded number of hours."""
+        return TimestampExpression(
+            f"{{col}} + INTERVAL '{offset_hours}' HOUR",
+            col,
+            type_=col.type,
+        )
+
    @classmethod
    def _apply_year_to_dttm(cls, time_expr: str) -> str:
        """
@@ -389,7 +389,6 @@ class GSheetsEngineSpec(ShillelaghEngineSpec):
                    }
                }
            },
-            future=True,
        )
        conn = engine.connect()
        idx = 0
@@ -307,6 +307,7 @@ class PostgresEngineSpec(BasicParametersMixin, PostgresBaseEngineSpec):
    supports_catalog = True
    supports_dynamic_catalog = True
    supports_grouping_sets = True
+    supports_temporal_column_shift = True

    default_driver = "psycopg2"
    sqlalchemy_uri_placeholder = (
@@ -25,9 +25,14 @@ from typing import Any, TYPE_CHECKING
 from flask_babel import gettext as __
 from sqlalchemy import types
 from sqlalchemy.engine.reflection import Inspector
+from sqlalchemy.sql.elements import ColumnClause

 from superset.constants import TimeGrain
-from superset.db_engine_specs.base import BaseEngineSpec, DatabaseCategory
+from superset.db_engine_specs.base import (
+    BaseEngineSpec,
+    DatabaseCategory,
+    TimestampExpression,
+)
 from superset.errors import SupersetErrorType

 if TYPE_CHECKING:
@@ -43,6 +48,7 @@ class SqliteEngineSpec(BaseEngineSpec):

    disable_ssh_tunneling = True
    supports_multivalues_insert = True
+    supports_temporal_column_shift = True

    metadata = {
        "description": "SQLite is a self-contained, serverless SQL database engine.",
@@ -140,6 +146,20 @@ class SqliteEngineSpec(BaseEngineSpec):
            "ELSE printf('%04d-01-01', CAST({col} AS INTEGER)) END)"
        )

+    @classmethod
+    def get_temporal_column_shift_expr(
+        cls,
+        col: ColumnClause,
+        offset_hours: int,
+    ) -> TimestampExpression:
+        """Shift a temporal expression with SQLite's datetime modifier syntax."""
+        modifier = f"{offset_hours:+d} hours"
+        return TimestampExpression(
+            f"DATETIME({{col}}, '{modifier}')",
+            col,
+            type_=col.type,
+        )
+
    @classmethod
    def convert_dttm(
        cls, target_type: str, dttm: datetime, db_extra: dict[str, Any] | None = None
@@ -376,7 +376,7 @@ def upgrade_catalog_perms(engines: set[str] | None = None) -> None:

    """
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    # The Database model has an eager-loaded (``lazy="joined"``) ``ssh_tunnel``
    # backref. Eager-loading it here would SELECT every column on ``ssh_tunnels``,
@@ -581,7 +581,7 @@ def downgrade_catalog_perms(engines: set[str] | None = None) -> None:
    WARNING: models (datasets and charts) not in the default catalog are deleted!
    """
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    # See upgrade_catalog_perms: avoid eager-loading the ``ssh_tunnel`` backref so the
    # query stays schema-safe across migration revisions.
@@ -51,7 +51,7 @@ class Slice(Base):
 def upgrade():
    bind = op.get_bind()
    op.add_column("slices", sa.Column("perm", sa.String(length=2000), nullable=True))
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    # Use Slice class defined here instead of models.Slice
    for slc in session.query(Slice).all():
@@ -59,7 +59,7 @@ def upgrade():
    )

    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    # don't use models.DruidMetric
    # because it assumes the context is consistent with the application
@@ -94,7 +94,7 @@ class Dashboard(AuditMixin, Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    objects = session.query(Slice).all()
    objects += session.query(Dashboard).all()
@@ -50,7 +50,7 @@ class Slice(Base):
 def upgrade():
    bind = op.get_bind()
    op.add_column("slices", sa.Column("datasource_id", sa.Integer()))
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).all():
        if slc.druid_datasource_id:
@@ -63,7 +63,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)
    for slc in session.query(Slice).all():
        if slc.datasource_type == "druid":
            slc.druid_datasource_id = slc.datasource_id
@@ -45,7 +45,7 @@ class Database(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for obj in session.query(Database).all():
        obj.allow_run_sync = True
@@ -48,7 +48,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    slices = session.query(Slice).all()
    slice_len = len(slices)
@@ -61,7 +61,7 @@ class Url(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    urls = session.query(Url).all()
    urls_len = len(urls)
@@ -45,7 +45,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).filter(Slice.viz_type.like("deck_%")):
        params = json.loads(slc.params)
@@ -45,7 +45,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).filter(
        or_(Slice.viz_type.like("line"), Slice.viz_type.like("bar"))
@@ -75,7 +75,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).filter(
        or_(Slice.viz_type.like("line"), Slice.viz_type.like("bar"))
@@ -46,7 +46,7 @@ class Dashboard(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = session.query(Dashboard).all()
    for i, dashboard in enumerate(dashboards):
@@ -68,7 +68,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = session.query(Dashboard).all()
    for i, dashboard in enumerate(dashboards):
@@ -57,7 +57,7 @@ def upgrade():
        ),
    )

-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    # Use Slice class defined here instead of models.Slice
    for tbl in session.query(Table).all():
@@ -49,7 +49,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    slices = session.query(Slice).filter_by(viz_type="cal_heatmap").all()
    slice_len = len(slices)
@@ -45,7 +45,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).all():
        try:
@@ -63,7 +63,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).all():
        try:
@@ -45,7 +45,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).all():
        try:
@@ -68,7 +68,7 @@ class Database(Base):


 def replace(source, target):
-    with db.Session(bind=op.get_bind(), future=True) as session:
+    with db.Session(bind=op.get_bind()) as session:
        with session.begin():
            query = (
                session.query(Slice, Database)
@@ -80,7 +80,7 @@ def replace(source, target):

            for slc, database in query:
                try:
-                    engine = create_engine(database.sqlalchemy_uri, future=True)
+                    engine = create_engine(database.sqlalchemy_uri)

                    if engine.dialect.identifier_preparer._double_percents:
                        params = json.loads(slc.params)
@@ -50,7 +50,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).all():
        try:
@@ -66,7 +66,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).all():
        try:
@@ -47,7 +47,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).filter(Slice.viz_type == "pie").all():
        try:
@@ -68,7 +68,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).filter(Slice.viz_type == "pie").all():
        try:
@@ -134,7 +134,7 @@ def compute_time_compare(granularity, periods):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for chart in session.query(Slice):
        params = json.loads(chart.params or "{}")
@@ -163,7 +163,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for chart in session.query(Slice):
        params = json.loads(chart.params or "{}")
@@ -159,7 +159,7 @@ class TableColumn(BaseColumnMixin, Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    tables = [
        Annotation,
@@ -59,7 +59,7 @@ class TableColumn(BaseColumnMixin, Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    # Delete the orphaned columns records.
    for record in session.query(DruidColumn).all():
@@ -59,7 +59,7 @@ class SqlMetric(BaseMetricMixin, Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    # Delete the orphaned metrics records.
    for record in session.query(DruidMetric).all():
@@ -579,7 +579,7 @@ def scan_dashboard_positions_data(positions):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = session.query(Dashboard).all()
    for i, dashboard in enumerate(dashboards):
@@ -55,7 +55,7 @@ def is_v2_dash(positions):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = session.query(Dashboard).all()
    for i, dashboard in enumerate(dashboards):  # noqa: B007
@@ -46,7 +46,7 @@ class Dashboard(Base):

 def upgrade():  # noqa: C901
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = session.query(Dashboard).all()
    for i, dashboard in enumerate(dashboards):
@@ -75,7 +75,7 @@ def upgrade_slice(slc):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    filter_box_slices = session.query(Slice).filter_by(viz_type="filter_box")
    for slc in filter_box_slices.all():
@@ -90,7 +90,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    filter_box_slices = session.query(Slice).filter_by(viz_type="filter_box")
    for slc in filter_box_slices.all():
@@ -87,7 +87,7 @@ def upgrade():

    bind = op.get_bind()
    insp = sa.engine.reflection.Inspector.from_engine(bind)
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    tables = session.query(SqlaTable).all()
    for table in tables:
@@ -159,7 +159,7 @@ class TableColumn(BaseColumnMixin, Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    tables = [
        Annotation,
@@ -62,7 +62,7 @@ def add_parent_ids(node, layout):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = session.query(Dashboard).all()
    for i, dashboard in enumerate(dashboards):
@@ -88,7 +88,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = session.query(Dashboard).all()
    for i, dashboard in enumerate(dashboards):
@@ -47,7 +47,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).all():
        try:
@@ -94,7 +94,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).all():
        try:
@@ -46,7 +46,7 @@ class DashboardSlices(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    # find dup records in dashboard_slices tbl
    dup_records = (
@@ -71,7 +71,7 @@ def upgrade():
    op.add_column("tables", Column("schema_perm", String(length=1000), nullable=True))

    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)
    for t in session.query(Sqlatable).all():
        db_name = (
            t.database.verbose_name
@@ -161,7 +161,7 @@ down_revision = "11c737c17cc6"

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    tables = [
        Annotation,
@@ -48,7 +48,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).all():
        try:
@@ -63,7 +63,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).all():
        try:
@@ -53,7 +53,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).all():
        if slc.params:
@@ -69,7 +69,7 @@ class Dashboard(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = session.query(Dashboard).all()
    for i, dashboard in enumerate(dashboards):
@@ -56,7 +56,7 @@ def upgrade():
    """

    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    # Visualization types which support time granularity (hence negate).
    viz_types = [
@@ -74,7 +74,7 @@ def duration_by_name(database: Database):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    query = (
        session.query(Slice, Database)
@@ -89,7 +89,7 @@ def create_new_markdown_component(chart_position, url):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dash_to_migrate = defaultdict(list)
    iframe_urls = defaultdict(list)
@@ -152,7 +152,7 @@ def upgrade():  # noqa: C901
    """

    bind = op.get_bind()
-    session = orm.Session(bind=bind, future=True)
+    session = orm.Session(bind=bind)

    faulty_view_menus = (
        session.query(ViewMenu)
@@ -119,7 +119,7 @@ def update_dashboards(session, uuid_map):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for table_name, model in models.items():
        with op.batch_alter_table(table_name) as batch_op:
@@ -152,7 +152,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    # remove uuid from position_json
    update_dashboards(session, {})
@@ -135,7 +135,7 @@ def upgrade():
            ),
        )
    # Migrate data
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)
    alerts = session.query(Alert).all()
    for a in alerts:
        if a.sql_observer:
@@ -224,7 +224,7 @@ def downgrade():
    )

    # Migrate data
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)
    alerts = session.query(Alert).all()
    for a in alerts:
        if a.sql:
@@ -56,7 +56,7 @@ default_batch_size = int(os.environ.get("BATCH_SIZE", 200))

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    # Add uuid column
    try:
@@ -86,7 +86,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)  # noqa: F841
+    session = db.Session(bind=bind)  # noqa: F841

    # Remove uuid column
    with op.batch_alter_table("saved_query") as batch_op:
@@ -93,7 +93,7 @@ PVM_MAP = {

 def upgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the new permissions on the migration itself
    add_pvms(session, NEW_PVMS)
@@ -107,7 +107,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the old permissions on the migration itself
    add_pvms(session, get_reversed_new_pvms(PVM_MAP))
@@ -65,7 +65,7 @@ PVM_MAP = {

 def upgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the new permissions on the migration itself
    add_pvms(session, NEW_PVMS)
@@ -79,7 +79,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the old permissions on the migration itself
    add_pvms(session, get_reversed_new_pvms(PVM_MAP))
@@ -77,7 +77,7 @@ PVM_MAP = {

 def upgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the new permissions on the migration itself
    add_pvms(session, NEW_PVMS)
@@ -91,7 +91,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the old permissions on the migration itself
    add_pvms(session, get_reversed_new_pvms(PVM_MAP))
@@ -101,7 +101,7 @@ PVM_MAP = {

 def upgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the new permissions on the migration itself
    add_pvms(session, NEW_PVMS)
@@ -115,7 +115,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the old permissions on the migration itself
    add_pvms(session, get_reversed_new_pvms(PVM_MAP))
@@ -87,7 +87,7 @@ PVM_MAP = {

 def upgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the new permissions on the migration itself
    add_pvms(session, NEW_PVMS)
@@ -101,7 +101,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the old permissions on the migration itself
    add_pvms(session, get_reversed_new_pvms(PVM_MAP))
@@ -101,7 +101,7 @@ PVM_MAP = {

 def upgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the new permissions on the migration itself
    add_pvms(session, NEW_PVMS)
@@ -115,7 +115,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the old permissions on the migration itself
    add_pvms(session, get_reversed_new_pvms(PVM_MAP))
@@ -89,7 +89,7 @@ PVM_MAP = {

 def upgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the new permissions on the migration itself
    add_pvms(session, NEW_PVMS)
@@ -103,7 +103,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the old permissions on the migration itself
    add_pvms(session, get_reversed_new_pvms(PVM_MAP))
@@ -80,7 +80,7 @@ PVM_MAP = {

 def upgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the new permissions on the migration itself
    add_pvms(session, NEW_PVMS)
@@ -94,7 +94,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the old permissions on the migration itself
    add_pvms(session, get_reversed_new_pvms(PVM_MAP))
@@ -55,7 +55,7 @@ PVM_MAP = {

 def upgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the new permissions on the migration itself
    add_pvms(session, NEW_PVMS)
@@ -69,7 +69,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the old permissions on the migration itself
    add_pvms(session, get_reversed_new_pvms(PVM_MAP))
@@ -47,7 +47,7 @@ PVM_MAP = {

 def upgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the new permissions on the migration itself
    add_pvms(session, NEW_PVMS)
@@ -61,7 +61,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the old permissions on the migration itself
    add_pvms(session, get_reversed_new_pvms(PVM_MAP))
@@ -53,7 +53,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)
    x_dateunit_in_since = DateRangeMigration.x_dateunit_in_since
    x_dateunit_in_until = DateRangeMigration.x_dateunit_in_until

@@ -74,7 +74,7 @@ def upgrade():
    - If no dttm columns exist in the dataset, don't change the chart.
    """
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    slices_changed = 0

@@ -47,7 +47,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    slices = (
        session.query(Slice)
@@ -75,7 +75,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    slices = (
        session.query(Slice)
@@ -46,7 +46,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).filter(Slice.viz_type.like("directed_force")):
        params = json.loads(slc.params)
@@ -75,7 +75,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).filter(Slice.viz_type.like("graph_chart")):
        params = json.loads(slc.params)
@@ -62,7 +62,7 @@ def has_uuid_column(table_name, bind):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for table_name, model in models.items():
        # this script adds missing uuid columns
@@ -46,7 +46,7 @@ class Dashboard(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = (
        session.query(Dashboard)
@@ -74,7 +74,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = (
        session.query(Dashboard)
@@ -47,7 +47,7 @@ class SqlaTable(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for datasource in session.query(SqlaTable):
        if datasource.extra:
@@ -49,7 +49,7 @@ def upgrade():
    Convert all country names to lowercase
    """
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).filter(Slice.viz_type == "country_map").all():
        try:
@@ -69,7 +69,7 @@ def downgrade():
    Convert all country names to sentence case
    """
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).filter(Slice.viz_type == "country_map").all():
        try:
@@ -172,7 +172,7 @@ def downgrade_filter_set(filter_set: dict[str, Any]) -> int:

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = (
        session.query(Dashboard)
@@ -208,7 +208,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = (
        session.query(Dashboard)
@@ -94,7 +94,7 @@ def upgrade_dashboard(dashboard: dict[str, Any]) -> tuple[int, int]:

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = (
        session.query(Dashboard)
@@ -136,7 +136,7 @@ def downgrade_dashboard(dashboard: dict[str, Any]) -> tuple[int, int]:

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    dashboards = (
        session.query(Dashboard)
@@ -52,7 +52,7 @@ def upgrade():
    Fix databases with ``schemas_allowed_for_csv_upload`` stored as string.
    """
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for database in session.query(Database).all():
        try:
@@ -56,7 +56,7 @@ VALID_RENDERERS = (

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    slices = (
        session.query(Slice)
@@ -47,7 +47,7 @@ PVM_MAP = {

 def upgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the new permissions on the migration itself
    add_pvms(session, NEW_PVMS)
@@ -61,7 +61,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = Session(bind=bind, future=True)
+    session = Session(bind=bind)

    # Add the old permissions on the migration itself
    add_pvms(session, get_reversed_new_pvms(PVM_MAP))
@@ -50,7 +50,7 @@ def upgrade():

 def remove_value_if_too_long():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    # it will be easier for users to notice that their field has been deleted rather than truncated  # noqa: E501
    # so just remove it if it won't fit back into the 1000 string length column
@@ -49,7 +49,7 @@ class Dashboard(Base):
 def upgrade():
    logger.info("[AddTypeToNativeFilter] Starting upgrade")
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for dashboard in session.query(Dashboard).all():
        logger.info("[AddTypeToNativeFilter] Updating Dashboard<pk:%s> ", dashboard.id)
@@ -87,7 +87,7 @@ def upgrade():
 def downgrade():
    logger.info("[RemoveTypeToNativeFilter] Starting downgrade")
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for dashboard in session.query(Dashboard).all():
        logger.info(
@@ -47,7 +47,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    where_clause = and_(
        Slice.viz_type == "pivot_table_v2",
@@ -45,7 +45,7 @@ class Slice(Base):

 def migrate(mapping: dict[str, str]) -> None:
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for slc in session.query(Slice).all():
        try:
@@ -48,7 +48,7 @@ class Database(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for database in session.query(Database).all():
        try:
@@ -70,7 +70,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for database in session.query(Database).all():
        try:
@@ -49,7 +49,7 @@ def upgrade():
        batch_op.add_column(sa.Column("force_screenshot", sa.Boolean(), default=False))

    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for report in session.query(ReportSchedule).all():
        # Update existing alerts that send chart screenshots so that the cache is
@@ -50,7 +50,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    slices = session.query(Slice).filter(Slice.viz_type == "big_number_total").all()
    for slc in slices:
@@ -77,7 +77,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    slices = session.query(Slice).filter(Slice.viz_type == "big_number_total").all()
    for slc in slices:
@@ -51,7 +51,7 @@ class Slice(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    slices = session.query(Slice).filter(Slice.viz_type == "pivot_table_v2").all()
    for slc in slices:
@@ -75,7 +75,7 @@ def upgrade():

 def downgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    slices = session.query(Slice).filter(Slice.viz_type == "pivot_table_v2").all()
    for slc in slices:
@@ -51,7 +51,7 @@ class SavedQuery(Base):

 def upgrade():
    bind = op.get_bind()
-    session = db.Session(bind=bind, future=True)
+    session = db.Session(bind=bind)

    for model in (Query, SavedQuery):
        for record in session.query(model).filter(model.schema == "null"):
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
sadpandajoe	901cb3de9d	test(query-context): update normalize_df mock signature for new offset param normalize_df now accepts an optional third sql_shifted_temporal_labels parameter (passed positionally from processing_time_offsets). Three pre-existing mocks in this file stubbed normalize_df with a fixed two-argument lambda, which broke when the call site started passing a third positional argument.	2026-08-20 18:06:05 +00:00
sadpandajoe	41233bcbe3	test(dataset): regression guards for Hours Offset bound and grain bugs Covers both defects across an offset x grain x column-type matrix: - DATE-column filter bounds stay on the requested whole-day window for offsets 0/+1/+24/+25/-1/-25 (the bug admitted/dropped a day for any sub-24h offset); TIMESTAMP columns keep exact-hour bounds as a control. - Grained physical and adhoc BASE_AXIS temporal axes apply the offset exactly once, before grain truncation, matching the exact SQL literal and the resulting bucket for each combination -- including the case that previously applied the offset zero times for an ungrained axis, and negative sub-day DATE offsets that must not move the bucket. Physical and adhoc filter SQL are asserted byte-identical to pre-fix output for the explicitly out-of-scope grained-filter paths. - The new per-engine capability gate correctly excludes every engine that doesn't support the SQL-side shift (falls back to the pre-fix pandas-only offset, verified against the real code path with no shift SQL emitted) and correctly includes Postgres/SQLite (compiled SQL asserted exactly, across signed and >=24h offsets).	2026-08-20 18:06:05 +00:00
sadpandajoe	e023dbfa7c	fix(dataset): correct Hours Offset filter bounds and grain-truncation order The dataset "Hours offset" setting had two independent bugs: 1. Filter bounds shifted a DATE-typed column's literal bound by the raw offset before rendering it as a date-only value. A sub-24h offset could move which calendar day a bound truncated to, silently admitting an out-of-range day on one end and dropping the last requested day on the other. The offset is now quantized to whole days (truncating toward zero) before it's applied to a DATE column's bounds, so a sub-day offset never moves the selected day window. 2. The offset was applied only in pandas, after the database had already truncated a temporal axis to its time grain using the raw (unshifted) value. A row near a grain boundary could bucket to the wrong day, and the displayed label (computed via a different path) could disagree with the filter bounds. The offset is now applied in SQL, before grain truncation, at the axis-construction call sites -- gated behind a new `apply_dataset_offset` opt-in so filter construction and other unrelated callers are unaffected. A new `sql_shifted_temporal_labels` signal tells dataframe normalization which columns were already shifted in SQL so the legacy pandas-side offset isn't re-applied on top. Since the SQL-side shift depends on dialect-specific interval syntax, it's gated behind a new `BaseEngineSpec.supports_temporal_column_shift` capability flag (default False), enabled explicitly on the two engines it's been verified against (Postgres, SQLite). Engines that don't opt in keep the prior pandas-only behavior for the grain-truncation-order bug -- no new regression -- while gaining the DATE-bound quantization fix unconditionally (that part is engine-agnostic).	2026-08-20 18:06:05 +00:00
Parman MohammadalizadehandEvan Rusackas	148ffaff50	fix(plugin-chart-echarts): omit stacked value labels on zero-height segments (#42756 ) Co-authored-by: Evan Rusackas <evan@preset.io>	2026-08-20 11:04:19 -07:00
Ankit	83c6ea4e03	fix(explore): show the empty state when Samples returns no result payload (#43115 )	2026-08-20 11:00:54 -07:00
Hans Yu	1ca5e61f54	chore: Unset future flag in engines and sessions. (#43273 )	2026-08-20 10:58:37 -07:00
rlei	1ef12580a7	fix(chart): ignore chart actions for a chart no longer in state (#43228 )	2026-08-20 10:57:21 -07:00
yousoph Claude Opus 4.8 Joe Li	c0884c0f0c	fix(explore): keep x-axis label when overriding Time Column with time comparison (#42875 ) Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> Co-authored-by: Joe Li <joe@preset.io>	2026-08-20 10:52:22 -07:00
Sepuri Sai KrishnaandClaude Opus 5	2ff79bd495	fix(github): point the issue templates at labels that exist (#43357 ) Co-authored-by: Claude Opus 5 <noreply@anthropic.com>	2026-08-20 10:13:36 -07:00
Mafi Matt Fitzgerald Claude Opus 4.8 Amin Ghadersohi	e7dccd44a7	fix(reports): prevent blank/partial report PDFs from virtualized charts (#43348 ) Co-authored-by: Matt Fitzgerald <matt.fitzgerald@preset.io> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> Co-authored-by: Amin Ghadersohi <amin.ghadersohi@gmail.com>	2026-08-20 09:10:30 -04:00
dependabot[bot]	fdd3945dde	chore(deps): bump github/codeql-action/analyze from 4.37.6 to 4.37.7 (#43361 ) Signed-off-by: dependabot[bot] <support@github.com>	2026-08-20 13:58:31 +03:00
dependabot[bot]anddependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	8b67fb4d30	chore(deps-dev): bump globals from 17.10.0 to 17.11.0 in /superset-websocket (#43360 ) Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	2026-08-20 00:41:49 -07:00
dependabot[bot]anddependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	b288db91f9	chore(deps): bump github/codeql-action/init from 4.37.6 to 4.37.7 (#43362 ) Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	2026-08-20 00:41:43 -07:00
dependabot[bot]anddependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	80bba12d0f	chore(deps-dev): bump globals from 17.10.0 to 17.11.0 in /docs (#43363 ) Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	2026-08-20 00:41:38 -07:00
dependabot[bot]anddependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	5d4358a149	chore(deps): bump google-auth-library from 11.0.1 to 11.0.2 in /superset-frontend (#43364 ) Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	2026-08-20 00:41:25 -07:00
Arya Ketan Claude Opus 5 Evan Rusackas	f9cedf84e2	fix: drop post-processing options the operation no longer accepts (#42927 ) Signed-off-by: Arya Ketan <aryaketan@sharechat.co> Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com> Co-authored-by: Evan Rusackas <evan@preset.io>	2026-08-19 18:15:15 -07:00