fix(conftest_plugin): resolve default_ignore_datasets.yaml from the package

changelog/800.fix.md

@@ -0,0 +1,2 @@
+`default_ignore_datasets.yaml` is now shipped inside the `climate_ref` package and resolved with `importlib.resources`.
+The `config` and `solve_config` pytest fixtures previously located it relative to the monorepo source tree, so they raised `ValueError` for any project that installs `climate-ref` from a wheel.

packages/climate-ref/src/climate_ref/config.py

@@ -406,9 +406,8 @@ def _load_config(config_file: str | Path, doc: dict[str, Any]) -> "Config":
 
 
 DEFAULT_IGNORE_DATASETS_MAX_AGE = datetime.timedelta(hours=6)
-DEFAULT_IGNORE_DATASETS_URL = (
-    "https://raw.githubusercontent.com/Climate-REF/climate-ref/refs/heads/main/default_ignore_datasets.yaml"
-)
+DEFAULT_IGNORE_DATASETS_FILENAME = "default_ignore_datasets.yaml"
+DEFAULT_IGNORE_DATASETS_URL = f"https://raw.githubusercontent.com/Climate-REF/climate-ref/refs/heads/main/{DEFAULT_IGNORE_DATASETS_FILENAME}"
 
 
 def _get_default_ignore_datasets_file() -> Path:
@@ -423,7 +422,7 @@ def _get_default_ignore_datasets_file() -> Path:
     :
         Path to the ignore datasets file under the user cache directory.
     """
-    return platformdirs.user_cache_path("climate_ref") / "default_ignore_datasets.yaml"
+    return platformdirs.user_cache_path("climate_ref") / DEFAULT_IGNORE_DATASETS_FILENAME
 
 
 def refresh_ignore_datasets_file(config: "Config") -> None:

packages/climate-ref/src/climate_ref/conftest_plugin.py

@@ -26,6 +26,7 @@
 
 from __future__ import annotations
 
+import importlib.resources
 import os
 import re
 import tempfile
@@ -41,7 +42,7 @@
 from typer.testing import CliRunner
 
 from climate_ref import cli
-from climate_ref.config import Config, DiagnosticProviderConfig
+from climate_ref.config import DEFAULT_IGNORE_DATASETS_FILENAME, Config, DiagnosticProviderConfig
 from climate_ref.datasets.cmip6 import CMIP6DatasetAdapter
 from climate_ref.datasets.obs4mips import Obs4MIPsDatasetAdapter
 from climate_ref.models import Execution
@@ -231,17 +232,38 @@ def data_catalog(
     }
 
 
+def packaged_ignore_datasets_file() -> Path:
+    """
+    Locate the ``default_ignore_datasets.yaml`` shipped inside the ``climate_ref`` package.
+
+    Returns
+    -------
+    :
+        Path to the packaged ignore datasets file.
+
+    Raises
+    ------
+    ValueError
+        If the file is missing from the installed package.
+    """
+    resource = importlib.resources.files("climate_ref") / DEFAULT_IGNORE_DATASETS_FILENAME
+    local_ignore_file = Path(str(resource))
+    if not local_ignore_file.is_file():  # pragma: no cover - indicates a packaging error
+        raise ValueError(f"Could not find ignore file at {local_ignore_file}")
+    return local_ignore_file
+
+
 def _use_local_ignore_datasets_file(cfg: Config) -> None:
     """
-    Point the config at the in-tree default_ignore_datasets.yaml and disable fetching.
+    Point the config at the packaged default_ignore_datasets.yaml and disable fetching.
 
     This keeps tests deterministic and offline:
     they exercise the shipped default file rather than whatever copy happens to be cached on the host.
+
+    The file is resolved from the installed package rather than the source tree,
+    so the fixtures also work for provider packages that depend on a released ``climate-ref`` wheel.
     """
-    local_ignore_file = Path(__file__).parents[4] / "default_ignore_datasets.yaml"
-    if not local_ignore_file.is_file():
-        raise ValueError(f"Could not find ignore file at {local_ignore_file}")
-    cfg.ignore_datasets_file = local_ignore_file
+    cfg.ignore_datasets_file = packaged_ignore_datasets_file()
     cfg.ignore_datasets_url = ""
 
 

packages/climate-ref/src/climate_ref/default_ignore_datasets.yaml

@@ -0,0 +1,62 @@
+# This file is used to specify datasets that should be ignored by default.
+#
+# It can be used to ignore datasets that are known to have issues and keep
+# diagnostics that use multiple datasets from running.
+#
+# The format used in this configuration file is:
+# provider:
+#   diagnostic:
+#     source_type:
+#       - facet: value
+#       - other_facet: [other_value1, other_value2]
+#
+# The facets listed here will be used to create `climate_ref_core.constraints.IgnoreFacets`
+# instances that will be prepended to the constraints of the data requirements
+# of the diagnostics that the facets are listed under.
+#
+# If no `ignore_datasets_file` is specified in the REF configuration, this file
+# will be downloaded from GitHub and used. If the local copy of this file is
+# older than 6 hours, it will be updated.
+#
+esmvaltool:
+  sea-ice-sensitivity:
+      cmip6:
+        - experiment_id: historical
+          grid_label: gn
+          member_id: r1i1p1f1
+          source_id: CAS-ESM2-0
+          table_id: SImon
+          variable_id: siconc
+          version: v20201225
+        - experiment_id: historical
+          source_id: BCC-ESM1
+          table_id: SImon
+          variable_id: siconc
+        - experiment_id: historical
+          source_id: CMCC-CM2-HR4
+          table_id: SImon
+          variable_id: siconc
+        - experiment_id: historical
+          source_id: CMCC-CM2-SR5
+          table_id: SImon
+          variable_id: siconc
+        - experiment_id: historical
+          source_id: CMCC-ESM2
+          table_id: SImon
+          variable_id: siconc
+        - experiment_id: historical
+          source_id: ICON-ESM-LR
+          table_id: SImon
+          variable_id: siconc
+        - experiment_id: historical
+          source_id: NorESM2-LM
+          table_id: SImon
+          variable_id: siconc
+        - experiment_id: historical
+          source_id: NorESM2-MM
+          table_id: SImon
+          variable_id: siconc
+        - experiment_id: historical
+          source_id: SAM0-UNICON
+          table_id: SImon
+          variable_id: siconc

packages/climate-ref/tests/unit/test_conftest_plugin.py

@@ -0,0 +1,30 @@
+"""Tests for the shared pytest plugin shipped with ``climate_ref``."""
+
+from pathlib import Path
+
+from climate_ref.config import DEFAULT_IGNORE_DATASETS_FILENAME, Config
+from climate_ref.conftest_plugin import _use_local_ignore_datasets_file, packaged_ignore_datasets_file
+
+# The canonical copy, served over `DEFAULT_IGNORE_DATASETS_URL` from the default branch.
+REPO_ROOT_IGNORE_FILE = Path(__file__).parents[4] / DEFAULT_IGNORE_DATASETS_FILENAME
+
+
+def test_packaged_ignore_datasets_file_is_resolvable():
+    # Resolved from the installed package, so this also holds for a wheel install.
+    assert packaged_ignore_datasets_file().is_file()
+
+
+def test_packaged_ignore_datasets_file_matches_repo_root():
+    # The root copy is what `DEFAULT_IGNORE_DATASETS_URL` serves, so the two must not drift apart.
+    assert REPO_ROOT_IGNORE_FILE.is_file()
+    assert packaged_ignore_datasets_file().read_bytes() == REPO_ROOT_IGNORE_FILE.read_bytes()
+
+
+def test_use_local_ignore_datasets_file_disables_fetching():
+    cfg = Config.default()
+
+    _use_local_ignore_datasets_file(cfg)
+
+    assert cfg.ignore_datasets_file == packaged_ignore_datasets_file()
+    # An empty URL short-circuits `refresh_ignore_datasets_file`, keeping tests offline.
+    assert cfg.ignore_datasets_url == ""