fix: loading examples from raw.githubusercontent.com fails with 429 errors (#33354)

superset/examples/helpers.py

@@ -14,6 +14,34 @@
 # KIND, either express or implied.  See the License for the
 # specific language governing permissions and limitations
 # under the License.
+"""Helpers for loading Superset example datasets.
+
+All Superset example data files (CSV, JSON, etc.) are fetched via the
+jsDelivr CDN instead of raw.githubusercontent.com to avoid GitHub API
+rate limits (60 anonymous requests/hour/IP).
+
+jsDelivr is a multi‑CDN front for public GitHub repos and supports
+arbitrary paths including nested folders. It doesn’t use the GitHub REST API
+and advertises unlimited bandwidth for open-source use.
+
+Example URL::
+
+    https://cdn.jsdelivr.net/gh/apache-superset/examples-data@master/datasets/examples/slack/messages.csv
+
+Environment knobs
+-----------------
+``SUPERSET_EXAMPLES_DATA_REF``  (default: ``master``)
+    Tag / branch / SHA to pin so builds remain reproducible.
+
+``SUPERSET_EXAMPLES_BASE_URL``
+    Override the base completely if you want to host the files elsewhere
+    (internal mirror, S3 bucket, ASF downloads, …).  **Include any query
+    string required by your hosting (e.g. ``?raw=true`` if you point back
+    to a GitHub *blob* URL).**
+"""
+
+from __future__ import annotations
+
 import os
 from typing import Any
 
@@ -22,27 +50,41 @@ from superset.connectors.sqla.models import SqlaTable
 from superset.models.slice import Slice
 from superset.utils import json
 
-BASE_URL = "https://github.com/apache-superset/examples-data/blob/master/"
+# ---------------------------------------------------------------------------
+# Public sample‑data mirror configuration
+# ---------------------------------------------------------------------------
+BASE_COMMIT: str = os.getenv("SUPERSET_EXAMPLES_DATA_REF", "master")
+BASE_URL: str = os.getenv(
+    "SUPERSET_EXAMPLES_BASE_URL",
+    f"https://cdn.jsdelivr.net/gh/apache-superset/examples-data@{BASE_COMMIT}/",
+)
 
-misc_dash_slices: set[str] = set()  # slices assembled in a 'Misc Chart' dashboard
+# Slices assembled into a 'Misc Chart' dashboard
+misc_dash_slices: set[str] = set()
+
+# ---------------------------------------------------------------------------
+# Utility functions
+# ---------------------------------------------------------------------------
 
 
 def get_table_connector_registry() -> Any:
+    """Return the SqlaTable registry so we can mock it in unit tests."""
     return SqlaTable
 
 
 def get_examples_folder() -> str:
+    """Return local path to the examples folder (when vendored)."""
     return os.path.join(app.config["BASE_DIR"], "examples")
 
 
 def update_slice_ids(pos: dict[Any, Any]) -> list[Slice]:
-    """Update slice ids in position_json and return the slices found."""
+    """Update slice ids in ``position_json`` and return the slices found."""
     slice_components = [
         component
         for component in pos.values()
         if isinstance(component, dict) and component.get("type") == "CHART"
     ]
-    slices = {}
+    slices: dict[str, Slice] = {}
     for name in {component["meta"]["sliceName"] for component in slice_components}:
         slc = db.session.query(Slice).filter_by(slice_name=name).first()
         if slc:
@@ -56,17 +98,24 @@ def update_slice_ids(pos: dict[Any, Any]) -> list[Slice]:
 
 
 def merge_slice(slc: Slice) -> None:
-    o = db.session.query(Slice).filter_by(slice_name=slc.slice_name).first()
-    if o:
-        db.session.delete(o)
+    """Upsert a Slice by name."""
+    existing = db.session.query(Slice).filter_by(slice_name=slc.slice_name).first()
+    if existing:
+        db.session.delete(existing)
     db.session.add(slc)
 
 
 def get_slice_json(defaults: dict[Any, Any], **kwargs: Any) -> str:
+    """Return JSON string for a chart definition, merging extra kwargs."""
     defaults_copy = defaults.copy()
     defaults_copy.update(kwargs)
     return json.dumps(defaults_copy, indent=4, sort_keys=True)
 
 
 def get_example_url(filepath: str) -> str:
-    return f"{BASE_URL}{filepath}?raw=true"
+    """Return an absolute URL to *filepath* under the examples‑data repo.
+
+    All calls are routed through jsDelivr unless overridden. Supports nested
+    paths like ``datasets/examples/slack/messages.csv``.
+    """
+    return f"{BASE_URL}{filepath}"

tests/unit_tests/migrations/viz/time_related_fields_test.py

@@ -21,7 +21,7 @@ from tests.unit_tests.migrations.viz.utils import migrate_and_assert
 
 SOURCE_FORM_DATA: dict[str, Any] = {
     "granularity_sqla": "ds",
-    "time_range": "100 years ago : today",
+    "time_range": "1925-04-24 : 2025-04-24",
     "viz_type": "pivot_table",
 }
 
@@ -29,7 +29,7 @@ TARGET_FORM_DATA: dict[str, Any] = {
     "form_data_bak": SOURCE_FORM_DATA,
     "granularity_sqla": "ds",
     "rowOrder": "value_z_to_a",
-    "time_range": "100 years ago : today",
+    "time_range": "1925-04-24 : 2025-04-24",
     "viz_type": "pivot_table_v2",
 }
 
@@ -40,7 +40,7 @@ def test_migration() -> None:
     target["adhoc_filters"] = [
         {
             "clause": "WHERE",
-            "comparator": "100 years ago : today",
+            "comparator": "1925-04-24 : 2025-04-24",
             "expressionType": "SIMPLE",
             "operator": "TEMPORAL_RANGE",
             "subject": "ds",