feat: implement query checking

The change introduces a check_query callable which runs an extensible
compose pipeline of query checkers.

Note regarding QueryParseException: This custom exception is intended
to be a thin wrapper around a pyparsing ParseException that RDFLib
raises. This avoids introducing pyparsing as a dependency just to be able to
test against this exception. I feel like RDFLib should not raise a
pyparsing exception but provide a thin wrapper itself.
See RDFLib/rdflib#3057.

The check_query function runs in SPARQLModelAdapter to enable fast
failures on inapplicable queries. Note that this somewhat couples
QueryConstructor to SPARQLModelAdapter; QueryConstructor should be
marked private for this reason.

Closes #116. Closes #126.

Author: lu-pl

rdfproxy/adapter.py

@@ -9,6 +9,7 @@
 from rdfproxy.mapper import _ModelBindingsMapper
 from rdfproxy.sparql_strategies import HttpxStrategy, SPARQLStrategy
 from rdfproxy.utils._types import _TModelInstance
+from rdfproxy.utils.checkers.query_checker import check_query
 from rdfproxy.utils.models import Page, QueryParameters
 
 
@@ -40,7 +41,7 @@ def __init__(
         sparql_strategy: type[SPARQLStrategy] = HttpxStrategy,
     ) -> None:
         self._target = target
-        self._query = query
+        self._query = check_query(query)
         self._model = model
 
         self.sparql_strategy = sparql_strategy(self._target)

rdfproxy/constructor.py

@@ -1,4 +1,5 @@
 from rdfproxy.utils._types import _TModelInstance
+from rdfproxy.utils.checkers.query_checker import check_query
 from rdfproxy.utils.models import QueryParameters
 from rdfproxy.utils.sparql_utils import (
     add_solution_modifier,

rdfproxy/utils/_exceptions.py

@@ -11,3 +11,18 @@ class InvalidGroupingKeyException(Exception):
 
 class QueryConstructionException(Exception):
     """Exception for indicating failed SPARQL query construction."""
+
+
+class UnsupportedQueryException(Exception):
+    """Exception for indicating that a given SPARQL query is not supported."""
+
+
+class QueryParseException(Exception):
+    """Exception for indicating that a given SPARQL query raised a parse error.
+
+    This exception is intended to wrap and re-raise all exceptions
+    raised from parsing a SPARQL query with RDFLib's parseQuery function.
+
+    parseQuery raises a pyparsing.exceptions.ParseException,
+    which would require to introduce pyparsing as a dependency just for testing.
+    """

rdfproxy/utils/_types.py

@@ -1,9 +1,13 @@
 """Type definitions for rdfproxy."""
 
+from collections import UserString
 from collections.abc import Iterable
-from typing import Protocol, TypeAlias, TypeVar, runtime_checkable
+from typing import Generic, Protocol, TypeAlias, TypeVar, runtime_checkable
 
 from pydantic import BaseModel, ConfigDict as PydanticConfigDict
+from rdflib.plugins.sparql.parser import parseQuery
+from rdflib.plugins.sparql.parserutils import CompValue
+from rdfproxy.utils._exceptions import QueryParseException
 
 
 _TModelInstance = TypeVar("_TModelInstance", bound=BaseModel)
@@ -49,3 +53,24 @@ class ConfigDict(PydanticConfigDict, total=False):
 
     group_by: str
     model_bool: _TModelBoolValue
+
+
+_TQuery = TypeVar("_TQuery", bound=str)
+
+
+class ParsedSPARQL(Generic[_TQuery], UserString):
+    """UserString for encapsulating parsed SPARQL queries."""
+
+    def __init__(self, query: _TQuery) -> None:
+        self.data: _TQuery = query
+        self.parse_object: CompValue = self._get_parse_object(query)
+
+    @staticmethod
+    def _get_parse_object(query: str) -> CompValue:
+        try:
+            _parsed = parseQuery(query)
+        except Exception as e:
+            raise QueryParseException(e) from e
+        else:
+            _, parse_object = _parsed
+            return parse_object

rdfproxy/utils/checkers/query_checker.py

@@ -0,0 +1,64 @@
+"""Functionality for performing checks on SPARQL queries."""
+
+import logging
+
+from rdfproxy.utils._exceptions import UnsupportedQueryException
+from rdfproxy.utils._types import ParsedSPARQL, _TQuery
+from rdfproxy.utils.utils import compose_left
+
+
+logger = logging.getLogger(__name__)
+
+
+def _check_select_query(parsed_sparql: ParsedSPARQL) -> ParsedSPARQL:
+    """Check if a SPARQL query is a SELECT query.
+
+    This is meant to run as a component in check_query.
+    """
+    logger.debug("Running SELECT query check.")
+
+    if parsed_sparql.parse_object.name != "SelectQuery":
+        raise UnsupportedQueryException("Only SELECT queries are applicable.")
+    return parsed_sparql
+
+
+def _check_solution_modifiers(parsed_sparql: ParsedSPARQL) -> ParsedSPARQL:
+    """Check if a SPARQL query has a solution modifier.
+
+    This is meant to run as a component in check_query.
+    """
+    logger.debug("Running solution modifier check.")
+
+    def _has_modifier():
+        for mod_name in ["limitoffset", "groupby", "having", "orderby"]:
+            if (mod := getattr(parsed_sparql.parse_object, mod_name)) is not None:
+                return mod
+        return False
+
+    if mod := _has_modifier():
+        logger.critical("Detected solution modifier '%s' in outer query.", mod)
+        raise UnsupportedQueryException(
+            "Solution modifiers for top-level queries are currently not supported."
+        )
+
+    return parsed_sparql
+
+
+def check_query(query: _TQuery) -> _TQuery:
+    """Check a SPARQL query by running a compose pipeline of checks.
+
+    The pipeline expects a SPARQL query string and
+    will return that string if all checks pass.
+
+    _parse_query is meant to be the first component
+    and _get_query_string is meant to be the last component.
+    """
+    logger.debug("Running query check pipeline on '%s'", query)
+    parsed_sparql = ParsedSPARQL(query=query)
+
+    result: ParsedSPARQL = compose_left(
+        _check_select_query,
+        _check_solution_modifiers,
+    )(parsed_sparql)
+
+    return result.data