Merge pull request #6 from eccenca/develop

v1.0.0beta4

CHANGELOG.md

@@ -4,6 +4,20 @@ All notable changes to this project will be documented in this file.
 
 The format is based on [Keep a Changelog](http://keepachangelog.com/) and this project adheres to [Semantic Versioning](https://semver.org/)
 
+## [1.0.0beta4] 2024-07-12
+
+### Fixed
+
+- fixed errors on CMEM instances with self-signed/invalid certificates
+
+### Added
+
+- valid OWL profiles can be read on the Reason plugin input instead of validating the ontology in the plugin
+
+### Changed
+
+- use DCMI Metadata Terms for provenance
+- new icons
 
 ## [1.0.0beta3] 2024-07-09
 

README.md

@@ -86,6 +86,14 @@ parameters to include inferred axiom generators:
 
 Validate the input ontology against OWL profiles (DL, EL, QL, RL, and Full). The ontology is annotated in the output graph. 
 
+### Process valid OWL profiles from input
+
+If enabled along with the "Validate OWL2 profiles" parameter, the list of valid profiles is taken from the plugin input, 
+without validating the ontology against the profiles in the plugin. The inputs need to include the entity paths "profile"
+for the valid profiles, and "ontology" for the ontology IRI. If the "Validate OWL2 profiles" parameter is enabled in the
+"Validate" plugin, it can be directly connected to the input of the "Reason" plugin.
+
+
 ### Maximum RAM Percentage
 
 Maximum heap size for the Java virtual machine in the DI container running the reasoning process.
@@ -138,7 +146,8 @@ Raise an error if inconsistencies are found. If enabled, the plugin does not out
 ### Validate OWL2 profiles
 
 Validate the input ontology against OWL profiles (DL, EL, QL, RL, and Full). The valid profiles are added to the output 
-Markdown file and the ontology is annotated in the output graph. The plugin outputs the profiles using the path "profile". 
+Markdown file and the ontology is annotated in the output graph. The plugin outputs the profiles with path "profile",
+and the ontology IRI with path "ontology".
 
 ### Maximum RAM Percentage
 

cmem_plugin_reason/plugin_reason.py

@@ -1,6 +1,8 @@
 """Reasoning workflow plugin module"""
 
+from collections.abc import Sequence
 from datetime import UTC, datetime
+from os import environ
 from pathlib import Path
 from tempfile import TemporaryDirectory
 from time import time
@@ -9,6 +11,7 @@
 from cmem.cmempy.dp.proxy.graph import get
 from cmem_plugin_base.dataintegration.context import ExecutionContext
 from cmem_plugin_base.dataintegration.description import Icon, Plugin, PluginParameter
+from cmem_plugin_base.dataintegration.entity import Entities
 from cmem_plugin_base.dataintegration.parameter.graph import GraphParameterType
 from cmem_plugin_base.dataintegration.plugins import WorkflowPlugin
 from cmem_plugin_base.dataintegration.types import BoolParameterType, StringParameterType
@@ -26,16 +29,17 @@
     get_provenance,
     post_profiles,
     post_provenance,
-    remove_temp,
     robot,
     send_result,
     validate_profiles,
 )
 
+environ["SSL_VERIFY"] = "false"
+
 
 @Plugin(
     label="Reason",
-    icon=Icon(file_name="reason.png", package=__package__),
+    icon=Icon(file_name="fluent--brain-circuit-24-regular.svg", package=__package__),
     description="Performs OWL reasoning.",
     documentation="""A task performing OWL reasoning. With an OWL ontology and a data graph as input
     the reasoning result is written to a specified graph. The following reasoners are supported:
@@ -162,6 +166,16 @@
             description="",
             default_value=False,
         ),
+        PluginParameter(
+            param_type=BoolParameterType(),
+            name="input_profiles",
+            label="Process valid OWL profiles from input",
+            description="""If the "validate OWL profiles" parameter is enabled, take values from the
+            input (paths "profile" and "ontology") instead of running the validation in the plugin.
+            """,
+            default_value=False,
+            advanced=True,
+        ),
     ],
 )
 class ReasonPlugin(WorkflowPlugin):
@@ -188,6 +202,7 @@ def __init__(  # noqa: PLR0913
         sub_data_property: bool = False,
         sub_object_property: bool = False,
         validate_profile: bool = False,
+        input_profiles: bool = False,
         max_ram_percentage: int = MAX_RAM_PERCENTAGE_DEFAULT,
     ) -> None:
         self.axioms = {
@@ -244,6 +259,7 @@ def __init__(  # noqa: PLR0913
         self.output_graph_iri = output_graph_iri
         self.reasoner = reasoner
         self.validate_profile = validate_profile
+        self.input_profiles = input_profiles
         self.max_ram_percentage = max_ram_percentage
 
     def get_graphs(self, graphs: dict, context: ExecutionContext) -> None:
@@ -282,9 +298,8 @@ def reason(self, graphs: dict) -> None:
             f"--language-annotation rdfs:comment "
             f'"Reasoning result set of <{self.data_graph_iri}> and '
             f'<{self.ontology_graph_iri}>" en '
-            f'--link-annotation prov:wasDerivedFrom "{self.data_graph_iri}" '
-            f"--link-annotation prov:wasDerivedFrom "
-            f'"{self.ontology_graph_iri}" '
+            f'--link-annotation dc:source "{self.data_graph_iri}" '
+            f'--link-annotation dc:source "{self.ontology_graph_iri}" '
             f'--typed-annotation dc:created "{utctime}" xsd:dateTime '
             f'--output "{self.temp}/result.ttl"'
         )
@@ -296,8 +311,32 @@ def reason(self, graphs: dict) -> None:
                 raise OSError(response.stderr.decode())
             raise OSError("ROBOT error")
 
-    def _execute(self, context: ExecutionContext) -> None:
+    def post_valid_profiles(self, inputs: Sequence[Entities], graphs: dict) -> None:
+        """Post valid profiles. Optionally get valid profiles from input."""
+        if self.input_profiles:
+            values = next(inputs[0].entities).values
+            paths = [p.path for p in inputs[0].schema.paths]
+            validated_ontology = values[paths.index("ontology")][0]
+            valid_profiles = values[paths.index("profile")]
+            if validated_ontology != self.ontology_graph_iri:
+                raise ValueError(
+                    "The ontology IRI validated with Validate differs from the input ontology IRI."
+                )
+        else:
+            valid_profiles = validate_profiles(self, graphs)
+        post_profiles(self, valid_profiles)
+
+    def _execute(self, inputs: Sequence[Entities], context: ExecutionContext) -> None:
         """`Execute plugin"""
+        if self.input_profiles:
+            if not inputs:
+                raise OSError(
+                    'Input entities needed if "Process valid OWL profiles from input" is enabled'
+                )
+            paths = [p.path for p in inputs[0].schema.paths]
+            if "profile" not in paths or "ontology" not in paths:
+                raise ValueError("Invalid input for processing OWL profiles")
+
         setup_cmempy_user_access(context.user)
         graphs = get_graphs_tree((self.data_graph_iri, self.ontology_graph_iri))
         self.get_graphs(graphs, context)
@@ -306,11 +345,10 @@ def _execute(self, context: ExecutionContext) -> None:
         setup_cmempy_user_access(context.user)
         send_result(self.output_graph_iri, Path(self.temp) / "result.ttl")
         if self.validate_profile:
-            post_profiles(self, validate_profiles(self, graphs))
+            self.post_valid_profiles(inputs, graphs)
         post_provenance(self, get_provenance(self, context))
-        remove_temp(self)
 
-    def execute(self, inputs: tuple, context: ExecutionContext) -> None:  # noqa: ARG002
+    def execute(self, inputs: Sequence[Entities], context: ExecutionContext) -> None:
         """Remove temp files on error"""
         with TemporaryDirectory() as self.temp:
-            self._execute(context)
+            self._execute(inputs, context)

cmem_plugin_reason/plugin_validate.py

@@ -1,6 +1,7 @@
 """Ontology consistency validation workflow plugin module"""
 
 from datetime import UTC, datetime
+from os import environ
 from pathlib import Path
 from tempfile import TemporaryDirectory
 from time import time
@@ -38,16 +39,18 @@
     validate_profiles,
 )
 
+environ["SSL_VERIFY"] = "false"
+
 
 @Plugin(
-    label="Validate",
+    label="Validate OWL consistency",
     description="Validates the consistency of an OWL ontology.",
     documentation="""A task validating the consistency of an OWL ontology and generating an
     explanation if inconsistencies are found. The explanation can be written to the project as a
     Markdown file and/or to a specified graph. The Markdown string is also provided as an output
     entity using the path "text". The following reasoners are supported: ELK, Expression
     Materializing Reasoner, HermiT, JFact, Structural Reasoner and Whelk.""",
-    icon=Icon(package=__package__, file_name="validate.png"),
+    icon=Icon(file_name="file-icons--owl.svg", package=__package__),
     parameters=[
         REASONER_PARAMETER,
         ONTOLOGY_GRAPH_IRI_PARAMETER,
@@ -156,7 +159,7 @@ def explain(self, graphs: dict) -> None:
                 f'--language-annotation rdfs:label "Ontology Validation Result {utctime}" en '
                 f"--language-annotation rdfs:comment "
                 f'"Ontology validation of <{self.ontology_graph_iri}>" en '
-                f'--link-annotation prov:wasDerivedFrom "{self.ontology_graph_iri}" '
+                f'--link-annotation dc:source "{self.ontology_graph_iri}" '
                 f'--typed-annotation dc:created "{utctime}" xsd:dateTime '
                 f'--output "{self.temp}/output.ttl"'
             )
@@ -191,8 +194,8 @@ def add_profiles(self, valid_profiles: list) -> list:
 
     def make_entities(self, text: str, valid_profiles: list) -> Entities:
         """Make entities"""
-        values = [[text]]
-        paths = [EntityPath(path="markdown")]
+        values = [[text], [self.ontology_graph_iri]]
+        paths = [EntityPath(path="markdown"), EntityPath(path="ontology")]
         if self.validate_profile:
             values.append(valid_profiles)
             paths.append(EntityPath(path="profile"))

cmem_plugin_reason/utils.py

@@ -1,13 +1,10 @@
 """Common constants and functions"""
 
 import json
-import re
 import shlex
-import unicodedata
 from collections import OrderedDict
 from pathlib import Path
 from secrets import token_hex
-from shutil import rmtree
 from subprocess import CompletedProcess, run
 from xml.etree.ElementTree import Element, SubElement, tostring
 
@@ -71,16 +68,6 @@
 )
 
 
-def convert_iri_to_filename(value: str) -> str:
-    """Convert IRI to filename"""
-    value = unicodedata.normalize("NFKD", value).encode("ascii", "ignore").decode("ascii")
-    value = re.sub(r"\.", "_", value.lower())
-    value = re.sub(r"/", "_", value.lower())
-    value = re.sub(r"[^\w\s-]", "", value.lower())
-    value = re.sub(r"[-\s]+", "-", value).strip("-_")
-    return value + ".nt"
-
-
 def create_xml_catalog_file(dir_: str, graphs: dict) -> None:
     """Create XML catalog file"""
     file_name = Path(dir_) / "catalog-v001.xml"
@@ -103,12 +90,12 @@ def get_graphs_tree(graph_iris: tuple) -> dict:
     graphs = {}
     for graph_iri in graph_iris:
         if graph_iri not in graphs:
-            graphs[graph_iri] = convert_iri_to_filename(graph_iri)
+            graphs[graph_iri] = f"{token_hex(8)}.nt"
             tree = get_graph_import_tree(graph_iri)
             for value in tree["tree"].values():
                 for iri in value:
                     if iri not in graphs:
-                        graphs[iri] = convert_iri_to_filename(iri)
+                        graphs[iri] = f"{token_hex(8)}.nt"
     return graphs
 
 
@@ -122,14 +109,6 @@ def send_result(iri: str, filepath: Path) -> None:
     )
 
 
-def remove_temp(plugin: WorkflowPlugin) -> None:
-    """Remove temporary files"""
-    try:
-        rmtree(plugin.temp)
-    except (OSError, FileNotFoundError) as err:
-        plugin.log.warning(f"Cannot remove directory {plugin.temp} ({err})")
-
-
 def post_provenance(plugin: WorkflowPlugin, prov: dict | None) -> None:
     """Post provenance"""
     if not prov:
@@ -141,7 +120,7 @@ def post_provenance(plugin: WorkflowPlugin, prov: dict | None) -> None:
     insert_query = f"""
         INSERT DATA {{
             GRAPH <{plugin.output_graph_iri}> {{
-                <{plugin.output_graph_iri}> <http://www.w3.org/ns/prov#wasGeneratedBy>
+                <{plugin.output_graph_iri}> <http://purl.org/dc/terms/creator>
                     <{prov["plugin_iri"]}> .
                 <{prov["plugin_iri"]}> a <{prov["plugin_type"]}>,
                     <https://vocab.eccenca.com/di/CustomTask> .