diff --git a/CHANGELOG.md b/CHANGELOG.md
index 7c50fb1b69907ffb23b05fa8829d01a34b6c9165..77ab1a2bbd479a18a883ec210236e48170513e28 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -9,6 +9,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
### Added ###
+- Validation module for checking a list of generated records against a list of json schemas
+ that can be generated from a yaml data model file.
- DictElementConverters can now make use of `match_properties` which
works analogous to `match_properties` in ROCrateEntityConverter and
`match_attrib` in XMLConverter.
diff --git a/src/caoscrawler/validator.py b/src/caoscrawler/validator.py
new file mode 100644
index 0000000000000000000000000000000000000000..33e29b02db429e3382248bbd80d2d00cd7b07c6b
--- /dev/null
+++ b/src/caoscrawler/validator.py
@@ -0,0 +1,163 @@
+#!/usr/bin/env python3
+# encoding: utf-8
+#
+# This file is a part of the LinkAhead Project.
+#
+# Copyright (C) 2024 Indiscale GmbH <info@indiscale.com>
+# Copyright (C) 2024 Alexander Schlemmer
+#
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as
+# published by the Free Software Foundation, either version 3 of the
+# License, or (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program. If not, see <https://www.gnu.org/licenses/>.
+#
+# ** end header
+#
+
+"""
+This module contains functions to validate the output of a scanner run with a
+json schema.
+"""
+
+import jsonschema
+import linkahead as db
+# from caosadvancedtools.models.parser import parse_model_from_string
+from caosadvancedtools.json_schema_exporter import recordtype_to_json_schema
+from caosadvancedtools.models.parser import parse_model_from_yaml
+from jsonschema import ValidationError
+from linkahead.high_level_api import convert_to_python_object
+
+
+def load_json_schema_from_datamodel_yaml(filename: str) -> dict[str, dict]:
+ """
+ Load a data model yaml file (using caosadvancedtools) and convert
+ all record types into a json schema using the json_schema_exporter module.
+
+ Arguments
+ ---------
+ filename: str
+ The filename of the yaml file to load.
+
+ Returns
+ -------
+ A dict of json schema objects. The keys are the record types for which the schemas
+ are generated.
+ """
+
+ model = parse_model_from_yaml(filename)
+
+ rt_schemas = {}
+ for el_key, el in model.items():
+ if isinstance(el, db.RecordType):
+ rt_schemas[el_key] = recordtype_to_json_schema(el)
+
+ return rt_schemas
+
+
+def representer_ordereddict(dumper, data):
+ """
+ Helper function to be able to represent the converted json schema objects correctly as yaml.
+ This representer essentially replaced OrderedDict objects with simple dict objects.
+
+ Since Python 3.7 dicts are ordered by default, see e.g.:
+ https://softwaremaniacs.org/blog/2020/02/05/dicts-ordered/en/
+
+ Example how to use the representer:
+ ```python
+ yaml.add_representer(OrderedDict, caoscrawler.validator.representer_ordereddict)
+ ```
+ """
+ return dumper.represent_data(dict(data))
+
+
+def _apply_schema_patches(pobj: dict):
+ """
+ Changes applied:
+ - properties are moved vom subitem "proeprties" to top-level.
+ - The following keys are deleted: parents, role, name, description, metadata, properties
+ """
+ if "properties" not in pobj:
+ # this is probably a file
+ return pobj
+ for prop in pobj["properties"]:
+ if isinstance(pobj["properties"][prop], dict):
+ pobj[prop] = _apply_schema_patches(pobj["properties"][prop])
+ else:
+ pobj[prop] = pobj["properties"][prop]
+
+ for keyd in ("parents", "role", "name",
+ "description", "metadata", "properties"):
+ if keyd in pobj:
+ del pobj[keyd]
+
+ return pobj
+
+
+def convert_record(record: db.Record):
+ """
+ Convert a record into a form suitable for validation with jsonschema.
+
+ Uses `high_level_api.convert_to_python_object`
+ Afterwards `_apply_schema_patches` is called recursively to refactor the dictionary
+ to match the current form of the jsonschema.
+
+ Arguments:
+ ----------
+ record: db.Record
+ The record that is supposed to be converted.
+ """
+ pobj = convert_to_python_object(record).serialize()
+ return _apply_schema_patches(pobj)
+
+
+def validate(records: list[db.Record], schemas: dict[str, dict]) -> list[tuple]:
+ """
+ Validate a list of records against a dictionary of schemas.
+ The keys of the dictionary are record types and the corresponding values are json schemata
+ associated with that record type. The current implementation assumes that each record that is
+ checked has exactly one parent and raises an error if that is not the case.
+ The schema belonging to a record is identified using the name of the first (and only) parent
+ of the record.
+
+ Arguments:
+ ----------
+
+ records: list[db.Record]
+ List of records that will be validated.
+
+ schemas: dict[str, dict]
+ A dictionary of JSON schemas generated using `load_json_schema_from_datamodel_yaml`.
+
+ Returns:
+ --------
+ A list of tuples, one element for each record:
+
+ - Index 0: A boolean that determines whether the schema belonging to the record type of the
+ record matched.
+ - Index 1: A validation error if the schema did not match or None otherwise.
+ """
+
+ retval = []
+ for r in records:
+ if len(r.parents) != 1:
+ raise NotImplementedError(
+ "Schema validation is only supported if records have exactly one parent.")
+ parname = r.parents[0].name
+ if parname not in schemas:
+ raise RuntimeError(
+ "No schema for record type {} in schema dictionary.".format(parname))
+ try:
+ jsonschema.validate(convert_record(r), schemas[parname])
+ retval.append((True, None))
+ except ValidationError as ex:
+ retval.append((False, ex))
+ return retval
diff --git a/unittests/datamodels/datamodel.yaml b/unittests/datamodels/datamodel.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..2759ecba7f2967062937d9b2f4805a9b501ab6c4
--- /dev/null
+++ b/unittests/datamodels/datamodel.yaml
@@ -0,0 +1,6 @@
+Dataset:
+ obligatory_properties:
+ keywords:
+ datatype: TEXT
+ dateModified:
+ datatype: DATETIME
diff --git a/unittests/test_validation.py b/unittests/test_validation.py
new file mode 100644
index 0000000000000000000000000000000000000000..a3215963f67b61241b321a0eb7345f9fe6fde1f2
--- /dev/null
+++ b/unittests/test_validation.py
@@ -0,0 +1,86 @@
+#!/usr/bin/env python3
+# encoding: utf-8
+#
+# This file is a part of the LinkAhead Project.
+#
+# Copyright (C) 2024 Indiscale GmbH <info@indiscale.com>
+# Copyright (C) 2024 Alexander Schlemmer <a.schlemmer@indiscale.com>
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU Affero General Public License as
+# published by the Free Software Foundation, either version 3 of the
+# License, or (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU Affero General Public License for more details.
+#
+# You should have received a copy of the GNU Affero General Public License
+# along with this program. If not, see <https://www.gnu.org/licenses/>.
+#
+
+"""
+test validation
+"""
+from os.path import join
+from pathlib import Path
+
+import jsonschema
+import linkahead as db
+import pytest
+from caoscrawler.validator import (convert_record,
+ load_json_schema_from_datamodel_yaml,
+ validate)
+from jsonschema import ValidationError
+
+UNITTESTDIR = Path(__file__).parent
+
+
+def test_create_json_schema():
+ json = load_json_schema_from_datamodel_yaml(join(UNITTESTDIR, "datamodels", "datamodel.yaml"))
+ r = db.Record()
+ r.add_parent(name="Dataset")
+ r.add_property(name="keywords", value="jakdlfjakdf")
+ r.add_property(name="dateModified", value="2024-11-16")
+
+ pobj = convert_record(r)
+ # print(yaml.dump(pobj))
+ # print(yaml.dump(json[0]))
+ assert "Dataset" in json
+ jsonschema.validate(pobj, json["Dataset"])
+
+ # Failing test:
+ r = db.Record()
+ r.add_parent(name="Dataset")
+ r.add_property(name="keywordss", value="jakdlfjakdf")
+ r.add_property(name="dateModified", value="2024-11-16")
+
+ pobj = convert_record(r)
+
+ with pytest.raises(ValidationError, match=".*'keywords' is a required property.*"):
+ jsonschema.validate(pobj, json["Dataset"])
+
+
+def test_validation():
+ """
+ Test for the main validation API function `validate`
+ """
+ json = load_json_schema_from_datamodel_yaml(
+ join(UNITTESTDIR, "datamodels", "datamodel.yaml"))
+ r1 = db.Record()
+ r1.add_parent(name="Dataset")
+ r1.add_property(name="keywords", value="jakdlfjakdf")
+ r1.add_property(name="dateModified", value="2024-11-16")
+
+ r2 = db.Record()
+ r2.add_parent(name="Dataset")
+ r2.add_property(name="keywordss", value="jakdlfjakdf")
+ r2.add_property(name="dateModified", value="2024-11-16")
+
+ valres = validate([r1, r2], json)
+ assert valres[0][0] is True
+ assert valres[0][1] is None
+ assert not valres[1][0]
+ assert isinstance(valres[1][1], ValidationError)
+ assert valres[1][1].message == "'keywords' is a required property"