file node class in default schema

4 files changed, 12 insertions, 13 deletions

diff --git a/bsie/base/extractor.py b/bsie/base/extractor.py
index a5c7846..678dcec 100644
--- a/bsie/base/extractor.py
+++ b/bsie/base/extractor.py
@@ -35,6 +35,7 @@ SCHEMA_PREAMBLE = '''
 
     # essential nodes
     bsfs:Entity rdfs:subClassOf bsfs:Node .
+    bsfs:File rdfs:subClassOf bsfs:Entity .
 
     # common definitions
     xsd:string rdfs:subClassOf bsfs:Literal .
@@ -77,15 +78,13 @@ class Extractor(abc.ABC):
         return hash((type(self), self.CONTENT_READER, self.schema))
 
     def predicates(self) -> typing.Iterator[_schema.Predicate]:
-        """Return the predicates that may be part of extracted triples."""
-        # NOTE: Some predicates in the schema might not occur in actual triples,
-        # but are defined due to predicate class hierarchy. E.g., bsfs:Predicate
-        # is part of every schema but should not be used in triples.
-        # Announcing all predicates might not be the most efficient way, however,
-        # it is the most safe one. Concrete extractors that produce additional
-        # predicates (e.g. auxiliary nodes with their own predicates) should
-        # overwrite this method to only include the principal predicates.
-        return self.schema.predicates()
+        ent = self.schema.node(ns.bsfs.Entity)
+        return (
+            pred
+            for pred
+            in self.schema.predicates()
+            if pred.domain <= ent or (pred.range is not None and pred.range <= ent)
+            )
 
     @abc.abstractmethod
     def extract(
diff --git a/bsie/extractor/generic/path.py b/bsie/extractor/generic/path.py
index 2cc592a..00165e3 100644
--- a/bsie/extractor/generic/path.py
+++ b/bsie/extractor/generic/path.py
@@ -32,7 +32,7 @@ class Path(extractor.Extractor):
     def __init__(self):
         super().__init__(bsfs.schema.Schema.from_string(extractor.SCHEMA_PREAMBLE + '''
             bse:filename rdfs:subClassOf bsfs:Predicate ;
-                rdfs:domain bsfs:Entity ;
+                rdfs:domain bsfs:File ;
                 rdfs:range xsd:string ;
                 rdfs:label "File name"^^xsd:string ;
                 schema:description "Filename of entity in some filesystem."^^xsd:string ;
diff --git a/bsie/extractor/generic/stat.py b/bsie/extractor/generic/stat.py
index dfde7d2..0f4267f 100644
--- a/bsie/extractor/generic/stat.py
+++ b/bsie/extractor/generic/stat.py
@@ -32,7 +32,7 @@ class Stat(extractor.Extractor):
     def __init__(self):
         super().__init__(bsfs.schema.Schema.from_string(extractor.SCHEMA_PREAMBLE + '''
             bse:filesize rdfs:subClassOf bsfs:Predicate ;
-                rdfs:domain bsfs:Entity ;
+                rdfs:domain bsfs:File ;
                 rdfs:range xsd:integer ;
                 rdfs:label "File size"^^xsd:string ;
                 schema:description "File size of entity in some filesystem."^^xsd:string ;
diff --git a/bsie/tools/pipeline.py b/bsie/tools/pipeline.py
index 7fdd935..3d08993 100644
--- a/bsie/tools/pipeline.py
+++ b/bsie/tools/pipeline.py
@@ -97,8 +97,8 @@ class Pipeline():
             rdr2ext[rdr].add(ext)
 
         # create subject for file
-        uuid = _uuid.UCID.from_path(path)
-        subject = Node(ns.bsfs.Entity, self._prefix + uuid)
+        uuid = bsfs.uuid.UCID.from_path(path)
+        subject = node.Node(ns.bsfs.File, self._prefix + 'file#' + uuid)
 
         # extract information
         for rdr, extrs in rdr2ext.items():