Merge pull request #1276 from VisLab/fix_extras

Removed prerelease flag from schema loading

Author: VisLab

CHANGELOG.md

@@ -2,15 +2,18 @@
 
 ## Enhancements
 
-### Prerelease library schemas can now partner with prerelease standard schemas
+### Prerelease schemas are now always included in version lookups
 
-Loading a library schema with `withStandard` pointing to a prerelease version of the standard schema (e.g. `withStandard="8.5.0"`) would fail with a `BAD_WITH_STANDARD` error because the `withStandard` partner lookup was always restricted to released schemas, with no way to opt in to prerelease partner resolution.
+Previously, loading a schema whose version existed only in the prerelease cache required passing `check_prerelease=True` through every layer of the loading API. This flag has been **removed** from all public functions (`load_schema_version`, `load_schema`, `from_string`, `from_dataframes`), all schema loader classes, and all internal helpers. Prerelease schemas are now found automatically whenever they are present in the cache.
 
 **Changes:**
 
-- `load_schema()` and `from_string()` in `hed_schema_io.py` now accept a `check_prerelease=False` parameter. When `True`, the `withStandard` partner schema is also searched in the prerelease cache.
-- `SchemaLoader` (base class) and all subclasses (`SchemaLoaderXML`, `SchemaLoaderWiki`, `SchemaLoaderJSON`, `SchemaLoaderDF`) accept and forward `check_prerelease`.
-- `check_schema_loading.py` (`hed_check_schema_loading` script and `run_loading_check()`) now automatically passes `check_prerelease=True` when loading schemas from a prerelease directory, so `test_all_prerelease_schemas` in `spec_tests` works correctly for library prereleases partnered with a prerelease standard.
+- Removed the `check_prerelease` parameter from `load_schema_version()`, `load_schema()`, `from_string()`, and `from_dataframes()` in `hed_schema_io.py`.
+- Removed the parameter from `SchemaLoader` (base class) and all subclasses (`SchemaLoaderXML`, `SchemaLoaderWiki`, `SchemaLoaderJSON`, `SchemaLoaderDF`).
+- `get_hed_version_path()` in `hed_cache.py` now always searches both regular and prerelease directories (regular first).
+- `get_hed_versions()` in `hed_cache.py` now defaults to `check_prerelease=True`. **This is a silent API change**: external callers that omitted `check_prerelease` will now receive prerelease versions. Internal callers that need released-only versions (compliance checker, `deprecatedFrom` validation, hedId comparison) explicitly pass `check_prerelease=False`.
+- Default schema version is now resolved dynamically from the cache (highest released version) instead of being hardcoded, so new schema releases no longer require a code change.
+- `check_schema_loading.py` simplified — removed `_is_prerelease_partner()` helper.
 - `run_loading_check()` now raises `ValueError` immediately for mutually exclusive flag combinations (`prerelease_only` + `exclude_prereleases`, or `library_filter` + `standard_only`), consistent with the existing CLI-level validation.
 
 # Release 0.9.0 January 22, 2026

hed/schema/hed_cache.py

@@ -79,15 +79,16 @@ def get_cache_directory(cache_folder=None) -> str:
     return HED_CACHE_DIRECTORY
 
 
-def get_hed_versions(local_hed_directory=None, library_name=None, check_prerelease=False) -> Union[list, dict]:
+def get_hed_versions(local_hed_directory=None, library_name=None, check_prerelease=True) -> Union[list, dict]:
     """Get the HED versions in the HED directory.
 
     Parameters:
         local_hed_directory (str): Directory to check for versions which defaults to hed_cache.
         library_name (str or None): An optional schema library name.
                                     None retrieves the standard schema only.
                                     Pass "all" to retrieve all standard and library schemas as a dict.
-        check_prerelease (bool): If True, results can include prerelease schemas
+        check_prerelease (bool): If True, results can include prerelease schemas.
+                                 Pass False to get only released versions (used by compliance checks).
 
     Returns:
         Union[list, dict]: List of version numbers or dictionary {library_name: [versions]}.
@@ -101,7 +102,7 @@ def get_hed_versions(local_hed_directory=None, library_name=None, check_prerelea
 
     all_hed_versions = {}
     local_directories = [local_hed_directory]
-    if check_prerelease and not local_hed_directory.endswith(prerelease_suffix):
+    if check_prerelease and Path(local_hed_directory).name != "prerelease":
         local_directories.append(os.path.join(local_hed_directory, "prerelease"))
 
     hed_files = []
@@ -134,25 +135,22 @@ def get_hed_versions(local_hed_directory=None, library_name=None, check_prerelea
     return all_hed_versions
 
 
-def get_hed_version_path(
-    xml_version, library_name=None, local_hed_directory=None, check_prerelease=False
-) -> Union[str, None]:
+def get_hed_version_path(xml_version, library_name=None, local_hed_directory=None) -> Union[str, None]:
     """Get HED XML file path in a directory. Only returns filenames that exist.
 
     Parameters:
-        library_name (str or None): Optional the schema library name.
         xml_version (str): Returns this version if it exists
+        library_name (str or None): Optional the schema library name.
         local_hed_directory (str): Path to local HED directory. Defaults to HED_CACHE_DIRECTORY
-        check_prerelease (bool): Also check for prerelease schemas
 
     Returns:
-        Union[str, None]: The path to the latest HED version the HED directory.
+        Union[str, None]: The path to the requested HED version the HED directory.
 
     """
     if not local_hed_directory:
         local_hed_directory = HED_CACHE_DIRECTORY
 
-    hed_versions = get_hed_versions(local_hed_directory, library_name, check_prerelease)
+    hed_versions = get_hed_versions(local_hed_directory, library_name)
     if not hed_versions or not xml_version:
         return None
     if xml_version in hed_versions:
@@ -161,11 +159,10 @@ def get_hed_version_path(
         if os.path.exists(regular_path):
             return regular_path
 
-        # If check_prerelease is True, also check prerelease directory
-        if check_prerelease:
-            prerelease_path = _create_xml_filename(xml_version, library_name, local_hed_directory, True)
-            if os.path.exists(prerelease_path):
-                return prerelease_path
+        # Also check prerelease directory
+        prerelease_path = _create_xml_filename(xml_version, library_name, local_hed_directory, True)
+        if os.path.exists(prerelease_path):
+            return prerelease_path
     return None
 
 

hed/schema/hed_schema_io.py

@@ -22,9 +22,7 @@
 MAX_MEMORY_CACHE = 40
 
 
-def load_schema_version(
-    xml_version=None, xml_folder=None, check_prerelease=False
-) -> Union["HedSchema", "HedSchemaGroup"]:
+def load_schema_version(xml_version=None, xml_folder=None) -> Union["HedSchema", "HedSchemaGroup"]:
     """Return a HedSchema or HedSchemaGroup extracted from xml_version
 
     Parameters:
@@ -33,7 +31,6 @@ def load_schema_version(
                                            based on the output of HedSchema.get_formatted_version
                                            Basic format: `[schema_namespace:][library_name_]X.Y.Z`.
         xml_folder (str): Path to a folder containing schema.
-        check_prerelease (bool): If True, check the prerelease directory for schemas.
 
     Returns:
         Union[HedSchema, HedSchemaGroup]: The schema or schema group extracted.
@@ -53,19 +50,18 @@ def load_schema_version(
     if xml_version and isinstance(xml_version, list):
         xml_versions = parse_version_list(xml_version)
         schemas = [
-            _load_schema_version(xml_version=version, xml_folder=xml_folder, check_prerelease=check_prerelease)
-            for version in xml_versions.values()
+            _load_schema_version(xml_version=version, xml_folder=xml_folder) for version in xml_versions.values()
         ]
         if len(schemas) == 1:
             return schemas[0]
 
         name = ",".join([schema.version for schema in schemas])
         return HedSchemaGroup(schemas, name=name)
     else:
-        return _load_schema_version(xml_version=xml_version, xml_folder=xml_folder, check_prerelease=check_prerelease)
+        return _load_schema_version(xml_version=xml_version, xml_folder=xml_folder)
 
 
-def load_schema(hed_path, schema_namespace=None, schema=None, name=None, check_prerelease=False) -> "HedSchema":
+def load_schema(hed_path, schema_namespace=None, schema=None, name=None) -> "HedSchema":
     """Load a schema from the given file or URL path.
 
     Parameters:
@@ -77,7 +73,6 @@ def load_schema(hed_path, schema_namespace=None, schema=None, name=None, check_p
         schema (HedSchema or None): A HED schema to merge this new file into
                                    It must be a with-standard schema with the same value.
         name (str or None): User supplied identifier for this schema
-        check_prerelease (bool): If True, allow the partnered standard schema (withStandard) to be a prerelease version.
 
     Returns:
         HedSchema: The loaded schema.
@@ -103,22 +98,21 @@ def load_schema(hed_path, schema_namespace=None, schema=None, name=None, check_p
             file_as_string,
             schema_format=os.path.splitext(hed_path.lower())[1],
             name=name,
-            check_prerelease=check_prerelease,
         )
     elif hed_path.lower().endswith(".xml"):
-        hed_schema = SchemaLoaderXML.load(hed_path, schema=schema, name=name, check_prerelease=check_prerelease)
+        hed_schema = SchemaLoaderXML.load(hed_path, schema=schema, name=name)
     elif hed_path.lower().endswith(".mediawiki"):
-        hed_schema = SchemaLoaderWiki.load(hed_path, schema=schema, name=name, check_prerelease=check_prerelease)
+        hed_schema = SchemaLoaderWiki.load(hed_path, schema=schema, name=name)
     elif hed_path.lower().endswith(".json"):
-        hed_schema = SchemaLoaderJSON.load(hed_path, schema=schema, name=name, check_prerelease=check_prerelease)
+        hed_schema = SchemaLoaderJSON.load(hed_path, schema=schema, name=name)
     elif hed_path.lower().endswith(".tsv") or os.path.isdir(hed_path):
         if schema is not None:
             raise HedFileError(
                 HedExceptions.INVALID_HED_FORMAT,
                 "Cannot pass a schema to merge into spreadsheet loading currently.",
                 filename=name,
             )
-        hed_schema = SchemaLoaderDF.load_spreadsheet(filenames=hed_path, name=name, check_prerelease=check_prerelease)
+        hed_schema = SchemaLoaderDF.load_spreadsheet(filenames=hed_path, name=name)
     else:
         raise HedFileError(HedExceptions.INVALID_EXTENSION, "Unknown schema extension", filename=hed_path)
 
@@ -128,9 +122,7 @@ def load_schema(hed_path, schema_namespace=None, schema=None, name=None, check_p
     return hed_schema
 
 
-def from_string(
-    schema_string, schema_format=".xml", schema_namespace=None, schema=None, name=None, check_prerelease=False
-) -> "HedSchema":
+def from_string(schema_string, schema_format=".xml", schema_namespace=None, schema=None, name=None) -> "HedSchema":
     """Create a schema from the given string.
 
     Parameters:
@@ -141,7 +133,6 @@ def from_string(
         schema (HedSchema or None): A HED schema to merge this new file into
                                    It must be a with-standard schema with the same value.
         name (str or None): User supplied identifier for this schema
-        check_prerelease (bool): If True, allow the partnered standard schema (withStandard) to be a prerelease version.
 
     Returns:
         HedSchema: The loaded schema.
@@ -162,17 +153,11 @@ def from_string(
         schema_string = schema_string.replace("\r\n", "\n")
 
     if schema_format.endswith(".xml"):
-        hed_schema = SchemaLoaderXML.load(
-            schema_as_string=schema_string, schema=schema, name=name, check_prerelease=check_prerelease
-        )
+        hed_schema = SchemaLoaderXML.load(schema_as_string=schema_string, schema=schema, name=name)
     elif schema_format.endswith(".mediawiki"):
-        hed_schema = SchemaLoaderWiki.load(
-            schema_as_string=schema_string, schema=schema, name=name, check_prerelease=check_prerelease
-        )
+        hed_schema = SchemaLoaderWiki.load(schema_as_string=schema_string, schema=schema, name=name)
     elif schema_format.endswith(".json"):
-        hed_schema = SchemaLoaderJSON.load(
-            schema_as_string=schema_string, schema=schema, name=name, check_prerelease=check_prerelease
-        )
+        hed_schema = SchemaLoaderJSON.load(schema_as_string=schema_string, schema=schema, name=name)
     else:
         raise HedFileError(HedExceptions.INVALID_EXTENSION, f"Unknown schema extension {schema_format}", filename=name)
 
@@ -181,15 +166,14 @@ def from_string(
     return hed_schema
 
 
-def from_dataframes(schema_data, schema_namespace=None, name=None, check_prerelease=False) -> "HedSchema":
+def from_dataframes(schema_data, schema_namespace=None, name=None) -> "HedSchema":
     """Create a schema from the given string.
 
     Parameters:
         schema_data (dict of str or None): A dict of DF_SUFFIXES:file_as_string_or_df
                               Should have an entry for all values of DF_SUFFIXES.
         schema_namespace (str, None):  The name_prefix all tags in this schema will accept.
         name (str or None): User supplied identifier for this schema
-        check_prerelease (bool): If True, allow the partnered standard schema (withStandard) to be a prerelease version.
 
     Returns:
         HedSchema:  The loaded schema.
@@ -207,9 +191,7 @@ def from_dataframes(schema_data, schema_namespace=None, name=None, check_prerele
             HedExceptions.BAD_PARAMETERS, "Empty or non dict value passed to HedSchema.from_dataframes", filename=name
         )
 
-    hed_schema = SchemaLoaderDF.load_spreadsheet(
-        schema_as_strings_or_df=schema_data, name=name, check_prerelease=check_prerelease
-    )
+    hed_schema = SchemaLoaderDF.load_spreadsheet(schema_as_strings_or_df=schema_data, name=name)
 
     if schema_namespace:
         hed_schema.set_schema_prefix(schema_namespace=schema_namespace)
@@ -274,7 +256,7 @@ def parse_version_list(xml_version_list) -> dict:
 
 
 @functools.lru_cache(maxsize=MAX_MEMORY_CACHE)
-def _load_schema_version(xml_version=None, xml_folder=None, check_prerelease=False):
+def _load_schema_version(xml_version=None, xml_folder=None):
     """Return specified version
 
     Parameters:
@@ -284,7 +266,6 @@ def _load_schema_version(xml_version=None, xml_folder=None, check_prerelease=Fal
                            The schema namespace must be the same and not repeated if loading multiple merged schemas.
 
         xml_folder (str): Path to a folder containing schema.
-        check_prerelease (bool): If True, check the prerelease directory for schemas.
 
     Returns:
         Union[HedSchema, HedSchemaGroup]: The requested HedSchema object.
@@ -308,9 +289,7 @@ def _load_schema_version(xml_version=None, xml_folder=None, check_prerelease=Fal
     else:
         xml_versions = [""]
 
-    first_schema = _load_schema_version_sub(
-        xml_versions[0], schema_namespace, xml_folder=xml_folder, check_prerelease=check_prerelease, name=name
-    )
+    first_schema = _load_schema_version_sub(xml_versions[0], schema_namespace, xml_folder=xml_folder, name=name)
     filenames = [os.path.basename(first_schema.filename)]
 
     # Collect all duplicate issues for proper error reporting
@@ -321,7 +300,6 @@ def _load_schema_version(xml_version=None, xml_folder=None, check_prerelease=Fal
             version,
             schema_namespace,
             xml_folder=xml_folder,
-            check_prerelease=check_prerelease,
             schema=first_schema,
             name=name,
         )
@@ -357,16 +335,14 @@ def _load_schema_version(xml_version=None, xml_folder=None, check_prerelease=Fal
     return first_schema
 
 
-def _load_schema_version_sub(
-    xml_version, schema_namespace="", xml_folder=None, check_prerelease=False, schema=None, name=""
-):
-    """Return specified version(single version only for this one)
+def _load_schema_version_sub(xml_version, schema_namespace="", xml_folder=None, schema=None, name=""):
+    """Return specified version (single version only for this one).
 
     Parameters:
-        xml_version (str): HED version format string. Expected format: '[library_name_]X.Y.Z'
+        xml_version (str): HED version format string. Expected format: '[library_name_]X.Y.Z'.
+                           If empty, the latest released standard schema version from the cache is used.
         schema_namespace (str): The prefix this will have
         xml_folder (str): Path to a folder containing schema
-        check_prerelease (bool): If True, check the prerelease directory for schemas
         schema (HedSchema or None): A HED schema to merge this new file into.
         name (str): User supplied identifier for this schema
 
@@ -381,7 +357,15 @@ def _load_schema_version_sub(
         - The prefix is invalid
     """
     if not xml_version:
-        xml_version = "8.4.0"
+        versions = hed_cache.get_hed_versions(xml_folder, check_prerelease=False)
+        if isinstance(versions, list) and versions:
+            xml_version = versions[0]
+        else:
+            raise HedFileError(
+                HedExceptions.FILE_NOT_FOUND,
+                "No HED standard schema versions found in cache. Ensure schemas are installed or cached.",
+                "",
+            )
 
     # Parse library name from version string before validation
     library_name = ""
@@ -402,16 +386,13 @@ def _load_schema_version_sub(
         version_to_validate,
         library_name=library_name,
         local_hed_directory=xml_folder,
-        check_prerelease=check_prerelease,
     )
 
     if hed_file_path:
         hed_schema = load_schema(hed_file_path, schema_namespace=schema_namespace, schema=schema, name=name)
     else:
         library_string = f"for library '{library_name}'" if library_name else ""
-        known_versions = hed_cache.get_hed_versions(
-            xml_folder, library_name=library_name if library_name else "all", check_prerelease=check_prerelease
-        )
+        known_versions = hed_cache.get_hed_versions(xml_folder, library_name=library_name if library_name else "all")
         raise HedFileError(
             HedExceptions.FILE_NOT_FOUND,
             f"HED version {library_string}: '{version_to_validate}' not found. Check {hed_cache.get_cache_directory(xml_folder)} for cache or https://github.com/hed-standard/hed-schemas/tree/main/library_schemas. "