Merge pull request #11 from white-gecko/feature/improvements

More improvements

Taskfile.yml

@@ -3,28 +3,31 @@ version: '3'
 tasks:
 
   install_dependencies:
-    cmds:
-      - pip install poetry
+    - pip install poetry
 
   install:
-    cmds:
-      - poetry install
+    - poetry install
 
   code_formating:
-    cmds:
-      - poetry run black shaclgen tests
+    - poetry run black shaclgen tests
 
   code_formating_check:
-    cmds:
-      - poetry run black --check shaclgen tests
+    - poetry run black --check shaclgen tests
 
   test:
-    cmds:
-      - poetry run pytest -rP --cov=shaclgen --cov-report=term-missing
+    - poetry run pytest -rP --cov=shaclgen --cov-report=term-missing
 
   ci:
     cmds:
       - task: install_dependencies
       - task: install
       - task: test
       - task: code_formating_check
+
+# Aliases
+#
+# Should be new syntax onece v4 is out:
+# https://github.com/go-task/task/issues/675
+
+  black:
+    - task: code_formating

pyproject.toml

@@ -17,11 +17,12 @@ classifiers = [
 python = "^3.7"
 rdflib = "^6.1.1"
 click = "^8.1.2"
-pytest = "^7.1.1"
+pytest = "^7.4.0"
+loguru = "^0.6.0"
 
 [tool.poetry.dev-dependencies]
 black = "^22.3.0"
-pytest-cov = "^3.0.0"
+pytest-cov = "^4.1.0"
 
 [build-system]
 requires = ["poetry-core>=1.0.0"]

shaclgen/__main__.py

@@ -3,9 +3,13 @@
 from .schema import schema
 
 import click
-from sys import stdout
+from sys import stdout, stderr
 from rdflib import Graph
 from rdflib.util import guess_format
+from rdflib.namespace import NamespaceManager
+from loguru import logger
+import json
+import pkg_resources
 
 
 @click.command()
@@ -51,7 +55,33 @@
     help="use implicit class targets with RDFS",
     default=False,
 )
-def main(graphs, ontology, output, serial, prefixes, namespace, implicit_class_target):
+@click.option(
+    "--logfile",
+    help="Specify a logfile",
+    default=None,
+)
+@click.option(
+    "--loglevel",
+    help="Specify the standard output loglevel",
+    default="INFO",
+)
+@click.option(
+    "--logfilelevel",
+    help="Specify the loglevel for the logfile",
+    default="DEBUG",
+)
+def main(
+    graphs,
+    ontology,
+    output,
+    serial,
+    prefixes,
+    namespace,
+    implicit_class_target,
+    logfile,
+    loglevel,
+    logfilelevel,
+):
     """
     ---------------------------Shaclgen---------------------------
 
@@ -89,16 +119,35 @@ def main(graphs, ontology, output, serial, prefixes, namespace, implicit_class_t
 
     """
 
+    logger.remove()
+    logger.add(stderr, level=loglevel)
+    if logfile:
+        logger.add(logfile, level=logfilelevel)
+
     source_graph = Graph()
 
     for graph in graphs:
         source_graph.parse(graph, format=guess_format(graph))
 
+    path = "prefixes/namespaces.json"
+    filepath = pkg_resources.resource_filename(__name__, path)
+
+    namespaces = NamespaceManager(graph=Graph())
+
+    with open(filepath, "r", encoding="utf-8") as fin:
+        for prefix, namespace_iri in json.load(fin).items():
+            namespaces.bind(prefix, namespace_iri)
+
+    if prefixes:
+        with open(prefixes, "r", encoding="utf-8") as fin:
+            for prefix, namespace_iri in json.load(fin).items():
+                namespaces.bind(prefix, namespace_iri)
+
     g = None
     if ontology:
-        g = schema(source_graph, prefixes)
+        g = schema(source_graph, namespaces)
     else:
-        g = data_graph(source_graph, prefixes)
+        g = data_graph(source_graph, namespaces)
     shape_graph = g.gen_graph(
         namespace=namespace, implicit_class_target=implicit_class_target
     )

shaclgen/generator.py

@@ -3,7 +3,7 @@
 
 
 class Generator:
-    def __init__(self, graph: Graph, prefixes=None):
+    def __init__(self, graph: Graph, namespaces=None):
         pass
 
     def sh_label_gen(self, uri):

shaclgen/schema.py

@@ -1,11 +1,10 @@
+from loguru import logger
 from rdflib import Graph, Namespace
 from rdflib.namespace import XSD, RDF, RDFS, OWL, SH
 from rdflib.namespace import NamespaceManager
 from rdflib.term import URIRef, Literal, BNode
 import collections
-import json
 from rdflib.collection import Collection
-import pkg_resources
 from .generator import Generator
 
 """
@@ -17,7 +16,7 @@
 
 
 class schema(Generator):
-    def __init__(self, graph: Graph, prefixes=None):
+    def __init__(self, graph: Graph, namespaces=None):
         self.G = graph
 
         self.CLASSES = collections.OrderedDict()
@@ -39,21 +38,12 @@ def __init__(self, graph: Graph, prefixes=None):
             XSD.nonPositiveInteger,
         ]
 
-        path = "prefixes/namespaces.json"
-        filepath = pkg_resources.resource_filename(__name__, path)
-
-        self.namespaces = NamespaceManager(graph=Graph())
+        if namespaces:
+            self.namespaces = namespaces
+        else:
+            self.namespaces = NamespaceManager(graph=Graph())
         self.namespaces.bind("sh", SH)
 
-        with open(filepath, "r", encoding="utf-8") as fin:
-            for prefix, namespace in json.load(fin).items():
-                self.namespaces.bind(prefix, namespace)
-
-        if prefixes:
-            with open(prefixes, "r", encoding="utf-8") as fin:
-                for prefix, namespace in json.load(fin).items():
-                    self.namespaces.bind(prefix, namespace)
-
     def extract_props(self):
         properties = []
         self.PROPS = {}
@@ -207,9 +197,13 @@ def extract_restrictions(self):
             self.REST[rest]["value"] = rest_val[0]
 
     def gen_graph(self, namespace=None, implicit_class_target=False):
+        logger.info("Start Extraction of the Ontology")
+        logger.info("Properties …")
         self.extract_props()
+        logger.info("Classes …")
         self.extract_classes()
         self.extract_restrictions()
+        logger.info("Write resulting SHACL Graph …")
         ng = Graph(namespace_manager=self.namespaces)
 
         if namespace is not None: