Improve detection of RDF/XML files when loading unknown content - Fixes

#98

Imported type stubs and resolved ALL MyPy issues! (this was a big effort)
Logic fixes in the dataset loader (thanks to inconsistencies exposed by MyPy)
Added ability to pass a TextIO or TextIOWrapper object into the dataset loader
Bump black version
Bump version to v0.17.1

CHANGELOG.md

@@ -4,6 +4,30 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Python PEP 440 Versioning](https://www.python.org/dev/peps/pep-0440/).
 
+## [Unreleased]
+
+- Nothing yet
+
+
+## [0.17.1] - 2021-10-11
+
+## Fixes
+- Handle transitive subclasses when evaluating sh:targetClass - @gtfierro
+  - Fixes #96
+- Improve detection of RDF/XML files when loading unknown content
+  - Fixes #98
+- Imported type stubs and resolved ALL MyPy issues! (this was a big effort)
+- Logic fixes in the dataset loader (thanks to inconsistencies exposed by MyPy)
+
+## Changed
+- Add special cases to sh:dataclass constraint, when the given shape uses rdfs:Literal or rdfs:Dataclass as the dataclass to match on
+  - Fixes #71
+
+## Added
+- Add datashapes.org/schema as a built-in graph
+  - Fixes #98
+- Added ability to pass a TextIO or TextIOWrapper object into the dataset loader
+
 ## [0.17.0.post1] - 2021-09-15
 
 ## Notice
@@ -828,7 +852,8 @@ just leaves the files open. Now it is up to the command-line client to close the
 
 - Initial version, limited functionality
 
-[Unreleased]: https://github.com/RDFLib/pySHACL/compare/v0.17.0.post1...HEAD
+[Unreleased]: https://github.com/RDFLib/pySHACL/compare/v0.17.1...HEAD
+[0.17.1]: https://github.com/RDFLib/pySHACL/compare/v0.17.0.post1...v0.17.1
 [0.17.0.post1]: https://github.com/RDFLib/pySHACL/compare/v0.17.0...v0.17.0.post1
 [0.17.0]: https://github.com/RDFLib/pySHACL/compare/v0.16.2...v0.17.0
 [0.16.2]: https://github.com/RDFLib/pySHACL/compare/v0.16.1...v0.16.2

CITATION.cff

@@ -8,8 +8,8 @@ authors:
   given-names: "Nicholas"
   orcid: "http://orcid.org/0000-0002-8742-7730"
 title: "pySHACL"
-version: 0.17.0
+version: 0.17.1
 doi: 10.5281/zenodo.4750840
 license: Apache-2.0
-date-released: 2021-07-20
+date-released: 2021-10-11
 url: "https://github.com/RDFLib/pySHACL"

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
 
 [tool.poetry]
 name = "pyshacl"
-version = "0.17.0.post2"
+version = "0.17.1"
 # Don't forget to change the version number in __init__.py and CITATION.cff along with this one
 description = "Python SHACL Validator"
 license = "Apache-2.0"
@@ -39,6 +39,7 @@ packages = [
 
 include = [
     { path = "pyshacl/assets/*.ttl", format = "sdist" },
+    { path = "pyshacl/assets/*.py", format = "sdist" },
     { path = "hooks/*", format = "sdist" },
     { path = "MANIFEST.in", format = "sdist" },
     { path = "pyproject.toml", format = "sdist" },
@@ -54,28 +55,31 @@ include = [
 [tool.poetry.dependencies]
 python = "^3.7.0"  # Compatible python versions must be declared here
 rdflib = ">=6.0.0,<7"
-owlrl = "^5.2.3"
+owlrl = "^5.2.3,<7"
 rdflib-jsonld = { version=">=0.4.0,<0.6", optional=true}
 pyduktape2 = {version="^0.4.1", optional=true}
 flake8 = {version="^3.8.0", optional=true}
 isort = {version="^5.7.0", optional=true}
-black = {version="21.8b0", optional=true}
+black = {version="21.9b0", optional=true}
+mypy = {version="^0.800", optional=true}
+types-setuptools = {version="*", optional=true}
 
 [tool.poetry.dev-dependencies]
 coverage = "^4.5"
 pytest = "^5.0"
 pytest-cov = "^2.8.1"
 flake8 = {version="^3.8.0", optional=true}
 isort = {version="^5.7.0", optional=true}
-black = {version="21.8b0", optional=true}
+black = {version="21.9b0", optional=true}
 mypy = {version="^0.800", optional=true}
+types-setuptools = {version="*", optional=true}
 
 
 [tool.poetry.extras]
 jsonld = ["rdflib-jsonld"]
 js = ["pyduktape2"]
 dev-lint = ["isort", "black", "flake8"]
-dev-type-checking = ["mypy"]
+dev-type-checking = ["mypy", "types-setuptools"]
 
 [tool.poetry.scripts]
 pyshacl = "pyshacl.cli:main"
@@ -85,7 +89,7 @@ from = {format = "poetry", path = "pyproject.toml"}
 to = {format = "setuppy", path = "setup.py"}
 
 [tool.black]
-required-version = "21.8b0"
+required-version = "21.9b0"
 line-length = "119"
 skip-string-normalization = true
 target-version = ['py37']
@@ -163,7 +167,7 @@ commands =
 [testenv:type-checking]
 commands_pre =
     poetry install -vv -n --no-root --extras "dev-type-checking"
-    poetry run pip3 install "mypy>=0.800"
+    poetry run pip3 install "mypy>=0.800" "types-setuptools"
 commands =
     - poetry show
     poetry run python3 -m mypy --ignore-missing-imports pyshacl

pyshacl/__init__.py

@@ -6,7 +6,7 @@
 
 
 # version compliant with https://www.python.org/dev/peps/pep-0440/
-__version__ = '0.17.0.post2'
+__version__ = '0.17.1'
 # Don't forget to change the version number in pyproject.toml and CITATION.cff along with this one
 
 __all__ = ['validate', 'Validator', '__version__', 'Shape', 'ShapesGraph']

pyshacl/helper/expression_helper.py

@@ -134,7 +134,7 @@ def value_nodes_from_path(sg, focus, path_val, target_graph, recursion=0):
 
 def nodes_from_node_expression(
     expr, focus_node, data_graph: 'GraphLike', sg: 'ShapesGraph', recurse_depth=0
-) -> Union[Set['RDFNode'], List['RDFNode']]:
+) -> Union[Set[Union['RDFNode', None]], List[Union['RDFNode', None]]]:
     # https://www.w3.org/TR/shacl-af/#node-expressions
     if expr == SH_this:
         return [focus_node]
@@ -151,15 +151,15 @@ def nodes_from_node_expression(
         if len(unions):
             union_list = next(iter(unions))
             parts = list(sg.graph.items(union_list))
-            all_nodes: Set['RDFNode'] = set()
+            all_nodes: Set[Union['RDFNode', None]] = set()
             for p in parts:
                 new_parts = nodes_from_node_expression(p, focus_node, data_graph, sg, recurse_depth=recurse_depth + 1)
                 all_nodes = all_nodes.union(new_parts)
             return all_nodes
         if len(intersections):
             inter_list = next(iter(intersections))
             parts = list(data_graph.items(inter_list))
-            inter_nodes: Set[RDFNode] = set()
+            inter_nodes: Set[Union['RDFNode', None]] = set()
             new = True
             for p in parts:
                 new_parts = nodes_from_node_expression(p, focus_node, data_graph, sg, recurse_depth=recurse_depth + 1)
@@ -221,7 +221,7 @@ def nodes_from_node_expression(
                 "The SHACLFunction {} was not defined in this SHACL Shapes file.".format(fnexpr)
             )
         argslist_parts = list(sg.graph.items(fnargslist))
-        args_sets = [
+        args_sets: List[Union[List[Union['RDFNode', None]], Set[Union['RDFNode', None]]]] = [
             nodes_from_node_expression(p, focus_node, data_graph, sg, recurse_depth=recurse_depth + 1)
             for p in argslist_parts
         ]

pyshacl/monkey/__init__.py

@@ -11,6 +11,7 @@
 RDFLIB_421 = LooseVersion("4.2.1")
 RDFLIB_500 = LooseVersion("5.0.0")
 RDFLIB_600 = LooseVersion("6.0.0")
+RDFLIB_602 = LooseVersion("6.0.2")
 
 
 def rdflib_bool_patch():
@@ -55,6 +56,12 @@ def __ge__(term, other):
     setattr(rdflib.term.Literal, "__le__", __le__)
 
 
+def empty_iterator():
+    if False:
+        # noinspection PyUnreachableCode
+        yield None  # type: ignore[unreachable]
+
+
 def apply_patches():
     if RDFLIB_421 >= RDFLIB_VERSION:
         rdflib_term_ge_le_patch()
@@ -63,4 +70,7 @@ def apply_patches():
     if RDFLIB_421 <= RDFLIB_VERSION < RDFLIB_600:
         # RDFLib 6.0.0+ comes with its own Memory2 store (called "Memory") by default
         plugin.register("default", store.Store, "pyshacl.monkey.memory2", "Memory2")
+    if RDFLIB_602 == RDFLIB_VERSION:
+        # Fixes https://github.com/RDFLib/rdflib/pull/1432
+        setattr(store.Store, "namespaces", empty_iterator)
     return True

pyshacl/rdfutil/clone.py

@@ -80,9 +80,11 @@ def mix_datasets(
     base_named_graphs = list(base_ds.contexts())
     if target_ds is None:
         target_ds = rdflib.Dataset(default_union=default_union)
+    elif isinstance(target_ds, rdflib.ConjunctiveGraph):
+        raise RuntimeError("Cannot mix new graphs into a ConjunctiveGraph, use Dataset instead.")
     elif target_ds == "inplace":
         pass  # do nothing here
-    elif not isinstance(target_ds, (rdflib.Dataset, rdflib.ConjunctiveGraph)):
+    elif not isinstance(target_ds, rdflib.Dataset):
         raise RuntimeError("Cannot mix datasets if target_ds passed in is not a Dataset itself.")
     if isinstance(extra_ds, (rdflib.Dataset, rdflib.ConjunctiveGraph)):
         mixin_graphs = list(extra_ds.contexts())
@@ -126,7 +128,9 @@ def mix_graphs(base_graph: GraphLike, extra_graph: GraphLike, target_graph: Opti
     :return: The cloned graph with mixed in triples from extra_graph
     :rtype: rdflib.Graph
     """
-    if isinstance(base_graph, (rdflib.ConjunctiveGraph, rdflib.Dataset)):
+    if isinstance(base_graph, (rdflib.ConjunctiveGraph, rdflib.Dataset)) and isinstance(
+        target_graph, (rdflib.ConjunctiveGraph, rdflib.Dataset)
+    ):
         return mix_datasets(base_graph, extra_graph, target_ds=target_graph)
     if target_graph is None:
         g = clone_graph(base_graph, target_graph=None, identifier=base_graph.identifier)