prohibit certain characters in URI and ensure URIs in bsfs.graph

5 files changed, 12 insertions, 19 deletions

diff --git a/bsfs/graph/nodes.py b/bsfs/graph/nodes.py
index c3530c1..84996c7 100644
--- a/bsfs/graph/nodes.py
+++ b/bsfs/graph/nodes.py
@@ -52,9 +52,8 @@ class Nodes():
         self._backend = backend
         self._ac = access_control
         self._node_type = node_type
-        self._guids = set(guids)
-        # create helper instances
-        # FIXME: Assumes that the schema does not change while the instance is in use!
+        # convert to URI since this is not guaranteed by Graph
+        self._guids = {URI(guid) for guid in guids}
 
     def __eq__(self, other: typing.Any) -> bool:
         return isinstance(other, Nodes) \
diff --git a/bsfs/schema/types.py b/bsfs/schema/types.py
index 54adffb..104580d 100644
--- a/bsfs/schema/types.py
+++ b/bsfs/schema/types.py
@@ -98,7 +98,7 @@ class _Type():
             parent: typing.Optional['_Type'] = None,
             **annotations: typing.Any,
             ):
-        self.uri = uri
+        self.uri = URI(uri)
         self.parent = parent
         self.annotations = annotations
 
diff --git a/bsfs/triple_store/sparql/parse_filter.py b/bsfs/triple_store/sparql/parse_filter.py
index 8959b2c..bf19a02 100644
--- a/bsfs/triple_store/sparql/parse_filter.py
+++ b/bsfs/triple_store/sparql/parse_filter.py
@@ -154,7 +154,7 @@ class Filter():
         puri = f'<{puri}>' # type: ignore [assignment] # variable re-use confuses mypy
         # apply reverse flag
         if node.reverse:
-            puri = URI('^' + puri)
+            puri = '^' + puri
             dom, rng = rng, dom # type: ignore [assignment] # variable re-use confuses mypy
         # check path consistency
         if not node_type <= dom:
@@ -267,9 +267,7 @@ class Filter():
         """
         if not isinstance(node_type, bsc.Node):
             raise errors.BackendError(f'expected Node, found {node_type}')
-        if not rdflib.term._is_valid_uri(node.value): # pylint: disable=protected-access
-            raise errors.BackendError(f'<{node.value}> is not a serializable uri')
-        return f'VALUES {head} {{ <{node.value}> }}'
+        return f'VALUES {head} {{ <{URI(node.value)}> }}'
 
     def _equals(self, node_type: bsc.Vertex, node: ast.filter.Equals, head: str) -> str:
         """
diff --git a/bsfs/triple_store/sparql/sparql.py b/bsfs/triple_store/sparql/sparql.py
index bd98f46..68c0027 100644
--- a/bsfs/triple_store/sparql/sparql.py
+++ b/bsfs/triple_store/sparql/sparql.py
@@ -284,10 +284,7 @@ class SparqlStore(base.TripleStoreBase):
             raise errors.ConsistencyError(f'{node_type} is not defined in the schema')
         # check and create guids
         for guid in guids:
-            # check convert to rdflib.URIRef
-            if not rdflib.term._is_valid_uri(guid): # pylint: disable=protected-access
-                raise ValueError(guids)
-            subject = rdflib.URIRef(guid)
+            subject = rdflib.URIRef(URI(guid))
             # check node existence
             if (subject, rdflib.RDF.type, None) in self._graph:
                 # FIXME: node exists and may have a different type! ignore? raise? report?
@@ -326,10 +323,7 @@ class SparqlStore(base.TripleStoreBase):
                 raise errors.InstanceError(inconsistent)
         # check guids
         # FIXME: Fail or skip inexistent nodes?
-        guids = set(guids)
-        invalid = {guid for guid in guids if not rdflib.term._is_valid_uri(guid)} # pylint: disable=protected-access
-        if len(invalid) > 0:
-            raise ValueError(invalid)
+        guids = {URI(guid) for guid in guids}
         inconsistent = {guid for guid in guids if not self._has_type(guid, node_type)}
         if len(inconsistent) > 0:
             raise errors.InstanceError(inconsistent)
diff --git a/bsfs/utils/uri.py b/bsfs/utils/uri.py
index 0693017..5755a6e 100644
--- a/bsfs/utils/uri.py
+++ b/bsfs/utils/uri.py
@@ -4,6 +4,8 @@ import re
 import typing
 
 # constants
+RX_CHARS = re.compile(r'[<>" {}|\\^]')
+
 RX_URI = re.compile(r'''
     ^
     (?:(?P<scheme>[^:/?#]+):)?      # scheme, ://-delimited
@@ -77,6 +79,9 @@ class URI(str):
         no claim about the validity of an URI!
 
         """
+        # check characters
+        if RX_CHARS.search(query) is not None:
+            return False
         # check uri
         parts = RX_URI.match(query)
         if parts is not None:
@@ -227,9 +232,6 @@ class URI(str):
 
     # overload formatting methods
 
-    def format(self, *args, **kwargs) -> 'URI':
-        return URI(super().format(*args, **kwargs))
-
     def __mod__(self, *args) -> 'URI':
         return URI(super().__mod__(*args))