iolanta 1.2.5__py3-none-any.whl → 1.2.7__py3-none-any.whl

iolanta/cli/main.py

@@ -1,17 +1,20 @@
 import locale
 import logging
+from pathlib import Path
 from typing import Annotated
 
 import loguru
 import platformdirs
 from documented import DocumentedError
-from rdflib import Literal
+from rdflib import Literal, URIRef
 from rich.console import Console
 from rich.markdown import Markdown
 from typer import Argument, Exit, Option, Typer
+from yarl import URL
 
 from iolanta.cli.models import LogLevel
 from iolanta.iolanta import Iolanta
+from iolanta.models import NotLiteralNode
 
 DEFAULT_LANGUAGE = locale.getlocale()[0].split('_')[0]
 
@@ -38,8 +41,24 @@ def construct_app() -> Typer:
 app = construct_app()
 
 
+def string_to_node(name: str) -> NotLiteralNode:
+    """
+    Parse a string into a node identifier.
+
+    String might be:
+      * a URL,
+      * or a local disk path.
+    """
+    url = URL(name)
+    if url.scheme:
+        return URIRef(name)
+
+    path = Path(name).absolute()
+    return URIRef(f'file://{path}')
+
+
 @app.command(name='browse')
-def render_command(   # noqa: WPS231, WPS238
+def render_command(   # noqa: WPS231, WPS238, WPS210, C901
     url: Annotated[str, Argument()],
     as_datatype: Annotated[
         str, Option(
@@ -79,12 +98,18 @@ def render_command(   # noqa: WPS231, WPS238
         logger=logger,
     )
 
-    node = iolanta.string_to_node(url)
+    node_url = URL(url)
+    if node_url.scheme:
+        node = URIRef(url)
+    else:
+        path = Path(url).absolute()
+        node = URIRef(f'file://{path}')
+        iolanta.add(path)
 
     try:
         renderable, stack = iolanta.render(
             node=node,
-            as_datatype=iolanta.string_to_node(as_datatype),
+            as_datatype=URIRef(as_datatype),
         )
 
     except DocumentedError as documented_error:

iolanta/cyberspace/processor.py

@@ -8,14 +8,18 @@ from threading import Lock
 from types import MappingProxyType
 from typing import Any, Iterable, Mapping
 
+import diskcache
 import funcy
 import loguru
+import platformdirs
 import reasonable
 import yaml_ld
+from nanopub import NanopubClient
 from rdflib import ConjunctiveGraph, Namespace, URIRef, Variable
 from rdflib.plugins.sparql.algebra import translateQuery
 from rdflib.plugins.sparql.evaluate import evalQuery
 from rdflib.plugins.sparql.parser import parseQuery
+from rdflib.plugins.sparql.parserutils import CompValue
 from rdflib.plugins.sparql.sparql import Query
 from rdflib.query import Processor
 from rdflib.term import BNode, Literal, Node
@@ -53,7 +57,7 @@ REDIRECTS = MappingProxyType({
     #   - either find a way to resolve these URLs automatically,
     #   - or create a repository of those redirects online.
     'https://purl.org/vocab/vann/': URIRef(
-        'https://vocab.org/vann/vann-vocab-20100607.rdf'
+        'https://vocab.org/vann/vann-vocab-20100607.rdf',
     ),
     URIRef(DC): URIRef(DCTERMS),
     URIRef(RDF): URIRef(RDF),
@@ -69,6 +73,34 @@ REDIRECTS = MappingProxyType({
 })
 
 
+@diskcache.Cache(
+    directory=str(
+        platformdirs.user_cache_path(
+            appname='iolanta',
+        ) / 'find_retractions_for',
+    ),
+).memoize(expire=datetime.timedelta(days=8).total_seconds())
+def find_retractions_for(nanopublication: URIRef) -> set[URIRef]:
+    """Find nanopublications that retract the given one."""
+    # See https://github.com/fair-workflows/nanopub/issues/168 for
+    # context of this dirty hack.
+    use_server = 'http://grlc.nanopubs.lod.labs.vu.nl/api/local/local/'
+
+    client = NanopubClient(
+        use_server=use_server,
+    )
+    client.grlc_urls = [use_server]
+
+    retractions = client.find_retractions_of(
+        str(nanopublication).replace(
+            'https://',
+            'http://',
+        ),
+    )
+
+    return {URIRef(retraction) for retraction in retractions}
+
+
 def _extract_from_mapping(  # noqa: WPS213
     algebra: Mapping[str, Any],
 ) -> Iterable[URIRef | Variable]:
@@ -188,15 +220,12 @@ def resolve_variables(
 
 
 def extract_mentioned_urls_from_query(
-    query: str,
+    query: Query,
     bindings: dict[str, Node],
     base: str | None,
     namespaces: dict[str, Namespace],
 ) -> tuple[Query, set[URIRef]]:
     """Extract URLs that a SPARQL query somehow mentions."""
-    parse_tree = parseQuery(query)
-    query = translateQuery(parse_tree, base, namespaces)
-
     return query, set(
         resolve_variables(
             extract_mentioned_urls(query.algebra),
@@ -218,6 +247,35 @@ class Skipped:
 LoadResult = Loaded | Skipped
 
 
+def _extract_nanopublication_uris(
+    algebra: CompValue,
+) -> Iterable[URIRef]:
+    """Extract nanopublications to get retracting information for."""
+    match algebra.name:
+        case 'SelectQuery' | 'Project' | 'Distinct' | 'Graph':
+            yield from _extract_nanopublication_uris(algebra['p'])
+
+        case 'BGP':
+            for retractor, retracts, retractee in algebra['triples']:
+                if retracts == URIRef(
+                    'https://purl.org/nanopub/x/retracts',
+                ) and isinstance(retractor, Variable):
+                    yield retractee
+
+        case 'LeftJoin' | 'Join':
+            yield from _extract_nanopublication_uris(algebra['p1'])
+            yield from _extract_nanopublication_uris(algebra['p2'])
+
+        case 'Filter' | 'OrderBy':
+            return
+
+        case unknown_name:
+            raise ValueError(
+                f'Unknown algebra name: {unknown_name}, '
+                f'content: {algebra}',
+            )
+
+
 @dataclasses.dataclass(frozen=True)
 class GlobalSPARQLProcessor(Processor):  # noqa: WPS338, WPS214
     """
@@ -325,8 +383,18 @@ class GlobalSPARQLProcessor(Processor):  # noqa: WPS338, WPS214
             query = strOrQuery
 
         else:
+            parse_tree = parseQuery(strOrQuery)
+            query = translateQuery(parse_tree, base, initNs)
+
+            self.load_retracting_nanopublications_by_query(
+                query=query,
+                bindings=initBindings,
+                base=base,
+                namespaces=initNs,
+            )
+
             query, urls = extract_mentioned_urls_from_query(
-                query=strOrQuery,
+                query=query,
                 bindings=initBindings,
                 base=base,
                 namespaces=initNs,
@@ -395,6 +463,8 @@ class GlobalSPARQLProcessor(Processor):  # noqa: WPS338, WPS214
         )
         if existing_triple is not None:
             return Skipped()
+        else:
+            self.logger.warning(f'Existing triples not found for {source_uri}')
 
         # FIXME This is definitely inefficient. However, python-yaml-ld caches
         #   the document, so the performance overhead is not super high.
@@ -441,7 +511,7 @@ class GlobalSPARQLProcessor(Processor):  # noqa: WPS338, WPS214
             return Loaded()
 
         except Exception as err:
-            self.logger.info('%s | Failed: %s', source, err)
+            self.logger.info(f'{source} | Failed: {err}')
 
             self.graph.add((
                 source_uri,
@@ -536,32 +606,13 @@ class GlobalSPARQLProcessor(Processor):  # noqa: WPS338, WPS214
         self.graph.addN(quad_tuples)
         self.graph.last_not_inferred_source = source
 
-        created_graphs = {
-            normalize_term(quad.graph)
+        into_graphs = ', '.join({
+            quad.graph
             for quad in quads
-            if quad.graph != source_uri
-        }
-        self.graph.addN(
-            itertools.chain.from_iterable(
-                [
-                    (
-                        source_uri,
-                        IOLANTA['has-sub-graph'],
-                        created_graph,
-                        source_uri,
-                    ),
-                    (
-                        created_graph,
-                        RDF.type,
-                        IOLANTA.Graph,
-                        source_uri,
-                    ),
-                ]
-                for created_graph in created_graphs
-            ),
+        })
+        self.logger.info(
+            f'{source} | loaded successfully into graphs: {into_graphs}',
         )
-
-        self.logger.info(f'{source} | loaded successfully.')
         return Loaded()
 
     def resolve_term(self, term: Node, bindings: dict[str, Node]):
@@ -573,3 +624,32 @@ class GlobalSPARQLProcessor(Processor):  # noqa: WPS338, WPS214
             )
 
         return term
+
+    def load_retracting_nanopublications_by_query(    # noqa: WPS231
+        self,
+        query: Query,
+        bindings: dict[str, Node],
+        base: str | None,
+        namespaces: dict[str, Namespace],
+    ):
+        """
+        If the query requires information about retracting nanopubs, load them.
+
+        FIXME: This function presently does nothing because `nanopub` library
+          has problems: https://github.com/fair-workflows/nanopub/issues/168
+
+        TODO: Generalize this mechanism to allow for plugins which analyze
+          SPARQL queries and load information into the graph based on their
+          content.
+        """
+        nanopublications = list(
+            resolve_variables(
+                terms=_extract_nanopublication_uris(query.algebra),
+                bindings=bindings,
+            ),
+        )
+
+        for nanopublication in nanopublications:
+            retractions = find_retractions_for(nanopublication)
+            for retraction in retractions:
+                self.load(retraction)

iolanta/iolanta.py

@@ -104,15 +104,6 @@ class Iolanta:   # noqa: WPS214
         init=False,
     )
 
-    @functools.cached_property
-    def loader(self) -> Loader[SourceType]:
-        """
-        Construct loader.
-
-        FIXME: Delete this.
-        """
-        return construct_root_loader(logger=self.logger)
-
     @property
     def plugin_classes(self) -> List[Type[Plugin]]:
         """Installed Iolanta plugins."""
@@ -162,19 +153,14 @@ class Iolanta:   # noqa: WPS214
         context: Optional[LDContext] = None,
         graph_iri: Optional[URIRef] = None,
     ) -> 'Iolanta':
-        """
-        Parse & load information from given URL into the graph.
-
-        FIXME:
-          * Maybe implement context.json/context.yaml files support.
-        """
+        """Parse & load information from given URL into the graph."""
         self.logger.info(f'Adding to graph: {source}')
         self.sources_added_not_yet_inferred.append(source)
 
         if not isinstance(source, Path):
             source = Path(source)
 
-        for source_file in source.rglob('*'):
+        for source_file in list(source.rglob('*')) or [source]:
             if source_file.is_dir():
                 continue
 
@@ -319,31 +305,6 @@ class Iolanta:   # noqa: WPS214
         self.bind_namespaces()
         self.add_files_from_plugins()
 
-    def string_to_node(self, name: str | Node) -> NotLiteralNode:
-        """
-        Parse a string into a node identifier.
-
-        String might be:
-          * a full IRI,
-          * a qname,
-          * a qname with implied `local:` part,
-          * or a blank node identifier.
-        """
-        if isinstance(name, Node):
-            return name
-
-        if ':' in name:
-            # This is either a full IRI, a qname, or a blank node identifier.
-            try:
-                # Trying to interpret this as QName.
-                return self.graph.namespace_manager.expand_curie(name)
-            except ValueError:
-                # If it is not a QName then it is an IRI, let's return it.
-                return URIRef(name)
-
-        # This string does not include an ":", so we imply `local:`.
-        return URIRef(f'local:{name}')
-
     def render(
         self,
         node: Node,

iolanta/parse_quads.py

@@ -7,6 +7,7 @@ from rdflib.term import Node
 
 from iolanta.errors import UnresolvedIRI
 from iolanta.models import Quad
+from iolanta.namespaces import IOLANTA, RDF
 from iolanta.parsers.errors import SpaceInProperty
 
 
@@ -66,6 +67,20 @@ def parse_quads(
         else:
             graph_name = URIRef(graph_name)
 
+            yield Quad(
+                graph,
+                IOLANTA['has-sub-graph'],
+                graph_name,
+                graph,
+            )
+
+        yield Quad(
+            graph_name,
+            RDF.type,
+            IOLANTA.Graph,
+            IOLANTA.meta,
+        )
+
         for quad in quads:
             try:
                 yield Quad(

{iolanta-1.2.5.dist-info → iolanta-1.2.7.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: iolanta
-Version: 1.2.5
+Version: 1.2.7
 Summary: Semantic Web browser
 License: MIT
 Author: Anatoly Scherbakov
@@ -15,12 +15,14 @@ Provides-Extra: all
 Requires-Dist: boltons (>=24.0.0)
 Requires-Dist: classes (>=0.4.0)
 Requires-Dist: deepmerge (>=0.1.1)
+Requires-Dist: diskcache (>=5.6.3)
 Requires-Dist: documented (>=0.1.1)
 Requires-Dist: dominate (>=2.6.0)
 Requires-Dist: funcy (>=2.0)
 Requires-Dist: iolanta-tables (>=0.1.7,<0.2.0) ; extra == "all"
 Requires-Dist: loguru (>=0.7.3)
 Requires-Dist: more-itertools (>=9.0.0)
+Requires-Dist: nanopub (>=2.0.1)
 Requires-Dist: owlrl (>=6.0.2)
 Requires-Dist: oxrdflib (>=0.4.0)
 Requires-Dist: python-frontmatter (>=0.5.0)

{iolanta-1.2.5.dist-info → iolanta-1.2.7.dist-info}/RECORD

@@ -6,7 +6,7 @@ iolanta/cli/formatters/choose.py,sha256=Ac4hNoptvnhuJB77K9KOn5oMF7j2RxmNuaLko28W
 iolanta/cli/formatters/csv.py,sha256=OVucZxhcMjihUli0wkbSOKo500-i7DNV0Zdf6oIougU,753
 iolanta/cli/formatters/json.py,sha256=jkNldFApSWw0kcMkeIPvI2Vt4JTE-Rvx5mWqKJb3Sj4,741
 iolanta/cli/formatters/pretty.py,sha256=Ik75CR5GMBDJCvES9eF0bQPj64ZJD40iFDSSZH0s9aA,2920
-iolanta/cli/main.py,sha256=xuVm5kVYEvx_BbfRbIODoikHdum-g6BK_c0dvlV3HZ4,2436
+iolanta/cli/main.py,sha256=e-2VfSEG0rw34Qe3UTBOW0A1en77PErZOUdGtH29PzI,2992
 iolanta/cli/models.py,sha256=cjbpowdzI4wAP0DUk3qoVHyimk6AZwlXi9CGmusZTuM,159
 iolanta/cli/pretty_print.py,sha256=M6E3TmhzA6JY5GeUVmDZLmOh5u70-393PVit4voFKDI,977
 iolanta/context.py,sha256=bZR-tbZIrDQ-Vby01PMDZ6ifxM-0YMK68RJvAsyqCTs,507
@@ -14,7 +14,7 @@ iolanta/conversions.py,sha256=hbLwRF1bAbOxy17eMWLHhYksbdCWN-v4-0y0wn3XSSg,1185
 iolanta/cyberspace/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 iolanta/cyberspace/inference/wikibase-claim.sparql,sha256=JSawj3VTc9ZPoU9mvh1w1AMYb3cWyZ3x1rEYWUsKZ6A,209
 iolanta/cyberspace/inference/wikibase-statement-property.sparql,sha256=SkSHZZlxWVDwBM3aLo0Q7hLuOj9BsIQnXtcuAuwaxqU,240
-iolanta/cyberspace/processor.py,sha256=RX96E9OLQL_fJMxsgmVn0XCwQfb8phnxYEiQgLY9akg,17365
+iolanta/cyberspace/processor.py,sha256=uy4dMVjCdQbYJUTwYJcxontkhiWoEEwoQkItHOXhqr0,20065
 iolanta/data/cli.yaml,sha256=TsnldYXoY5GIzoNuPDvwBKGw8eAEForZW1FCKqKI0Kg,1029
 iolanta/data/context.yaml,sha256=OULEeDkSqshabaXF_gMujgwFLDJvt9eQn_9FftUlSUw,1424
 iolanta/data/foaf-meta.yaml,sha256=BhRTFY_uue5FsJEDJnSgnA8hCr927RHnET9ZrQnFpdU,2647
@@ -98,7 +98,7 @@ iolanta/facets/title/sparql/title.sparql,sha256=4rz47tjwX2OJavWMzftaYKil1-ZHB76Z
 iolanta/facets/wikibase_statement_title/__init__.py,sha256=_yk1akxgSJOiUBJIc8QGrD2vovvmx_iw_vJDuv1rD7M,91
 iolanta/facets/wikibase_statement_title/facets.py,sha256=mUH7twlAgoeX7DgLQuRBQv4ORT6GWbN-0eJ1aliSfiQ,724
 iolanta/facets/wikibase_statement_title/sparql/statement-title.sparql,sha256=n07DQWxKqB5c3CA4kacq2HSN0R0dLgnMnLP1AxMo5YA,320
-iolanta/iolanta.py,sha256=nCDhbolzqYVCvi9ld8uaVOSD3NfvNnwpxyIsXOX34q0,13810
+iolanta/iolanta.py,sha256=eUFmtme7ZuFYPc1_KLO5ZLFWRjcVsPpYnGf2WzVkJzA,12662
 iolanta/loaders/__init__.py,sha256=QTiKCsQc1BTS-IlY2CQsN9iVpEIPqYFvI9ERMYVZCbU,99
 iolanta/loaders/base.py,sha256=-DxYwqG1bfDXB2p_S-mKpkc_3Sh14OHhePbe65Iq3-s,3381
 iolanta/loaders/data_type_choice.py,sha256=zRUXBIzjvuW28P_dhMDVevE9C8EFEIx2_X39WydWrtM,1982
@@ -111,7 +111,7 @@ iolanta/loaders/scheme_choice.py,sha256=GHA4JAD-Qq3uNdej4vs_bCrN1WMRshVRvOMxaYyP
 iolanta/models.py,sha256=L0iFaga4-PCaWP18WmT03NLK_oT7q49V0WMTQskoffI,2824
 iolanta/namespaces.py,sha256=fyWDCGPwU8Cs8d-Vmci4H0-fuIe7BMSevvEKFvG7Gf4,1153
 iolanta/node_to_qname.py,sha256=a82_qpgT87cbekY_76tTkl4Z-6Rz6am4UGIQChUf9Y0,794
-iolanta/parse_quads.py,sha256=rIzpy6zt-ad63pvT5nX0CG6v89HgUzPFDl31jklHkOw,2504
+iolanta/parse_quads.py,sha256=yOI154hGRCNv4UjEuhicP0cQWNWid_52mE86VWYfH8w,2833
 iolanta/parsers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 iolanta/parsers/base.py,sha256=ZFjj0BLzW4985BdC6PwbngenhMuSYW5mNLpprZRWjhA,1048
 iolanta/parsers/dict_parser.py,sha256=t_OK2meUh49DqSaOYkSgEwxMKUNNgjJY8rZAyL4NQKI,4546
@@ -130,7 +130,7 @@ iolanta/widgets/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 iolanta/widgets/mixin.py,sha256=nDRCOc-gizCf1a5DAcYs4hW8eZEd6pHBPFsfm0ncv7E,251
 ldflex/__init__.py,sha256=8IELqR55CQXuI0BafjobXSK7_kOc-qKVTrEtEwXnZRA,33
 ldflex/ldflex.py,sha256=omKmOo5PUyn8Evw4q_lqKCJOq6yqVOcLAYxnYkdwOUs,3332
-iolanta-1.2.5.dist-info/METADATA,sha256=bVTLK79ikFHEPuag6C6wDKfQSLOlTB2U6MzRDt73GDA,1250
-iolanta-1.2.5.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-iolanta-1.2.5.dist-info/entry_points.txt,sha256=fIp9g4kzjSNcTsTSjUCk4BIG3laHd3b3hUZlkjgFAGU,179
-iolanta-1.2.5.dist-info/RECORD,,
+iolanta-1.2.7.dist-info/METADATA,sha256=1yhC21w4KtNTH5COJ9cpUWoHioHzBV3mzWZMbg6IfyU,1318
+iolanta-1.2.7.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+iolanta-1.2.7.dist-info/entry_points.txt,sha256=fIp9g4kzjSNcTsTSjUCk4BIG3laHd3b3hUZlkjgFAGU,179
+iolanta-1.2.7.dist-info/RECORD,,