Adds custom functions support

lib/src/sparql/eval.rs

@@ -1,7 +1,7 @@
 use crate::model::vocab::{rdf, xsd};
 use crate::model::xsd::*;
-use crate::model::Triple;
 use crate::model::{BlankNode, LiteralRef, NamedNodeRef};
+use crate::model::{NamedNode, Term, Triple};
 use crate::sparql::algebra::{Query, QueryDataset};
 use crate::sparql::dataset::DatasetView;
 use crate::sparql::error::EvaluationError;
@@ -38,6 +38,7 @@ pub struct SimpleEvaluator {
     base_iri: Option<Rc<Iri<String>>>,
     now: DateTime,
     service_handler: Rc<dyn ServiceHandler<Error = EvaluationError>>,
+    custom_functions: Rc<HashMap<NamedNode, Rc<dyn Fn(&[Term]) -> Option<Term>>>>,
 }
 
 impl SimpleEvaluator {
@@ -45,12 +46,14 @@ impl SimpleEvaluator {
         dataset: Rc<DatasetView>,
         base_iri: Option<Rc<Iri<String>>>,
         service_handler: Rc<dyn ServiceHandler<Error = EvaluationError>>,
+        custom_functions: Rc<HashMap<NamedNode, Rc<dyn Fn(&[Term]) -> Option<Term>>>>,
     ) -> Self {
         Self {
             dataset,
             base_iri,
             now: DateTime::now().unwrap(),
             service_handler,
+            custom_functions,
         }
     }
 
@@ -2016,6 +2019,24 @@ impl SimpleEvaluator {
                     _ => None,
                 })
             }
+            PlanExpression::CustomFunction(function_name, args) => {
+                if let Some(function) = self.custom_functions.get(function_name).cloned() {
+                    let args = args
+                        .iter()
+                        .map(|e| self.expression_evaluator(e))
+                        .collect::<Vec<_>>();
+                    let dataset = self.dataset.clone();
+                    Rc::new(move |tuple| {
+                        let args = args
+                            .iter()
+                            .map(|f| dataset.decode_term(&f(tuple)?).ok())
+                            .collect::<Option<Vec<_>>>()?;
+                        Some(dataset.encode_term(&function(&args)?))
+                    })
+                } else {
+                    Rc::new(|_| None)
+                }
+            }
         }
     }
 

lib/src/sparql/mod.rs

@@ -16,6 +16,7 @@ mod service;
 mod update;
 mod xml_results;
 
+use crate::model::{NamedNode, Term};
 pub use crate::sparql::algebra::{Query, Update};
 use crate::sparql::dataset::DatasetView;
 pub use crate::sparql::error::EvaluationError;
@@ -32,6 +33,7 @@ use crate::sparql::service::{EmptyServiceHandler, ErrorConversionServiceHandler}
 pub(crate) use crate::sparql::update::evaluate_update;
 use crate::storage::Storage;
 pub use spargebra::ParseError;
+use std::collections::HashMap;
 use std::rc::Rc;
 use std::time::Duration;
 
@@ -47,11 +49,13 @@ pub(crate) fn evaluate_query(
         spargebra::Query::Select {
             pattern, base_iri, ..
         } => {
-            let (plan, variables) = PlanBuilder::build(&dataset, &pattern, true)?;
+            let (plan, variables) =
+                PlanBuilder::build(&dataset, &pattern, true, &options.custom_functions)?;
             Ok(SimpleEvaluator::new(
                 Rc::new(dataset),
                 base_iri.map(Rc::new),
                 options.service_handler(),
+                Rc::new(options.custom_functions),
             )
             .evaluate_select_plan(
                 &plan,
@@ -66,11 +70,13 @@ pub(crate) fn evaluate_query(
         spargebra::Query::Ask {
             pattern, base_iri, ..
         } => {
-            let (plan, _) = PlanBuilder::build(&dataset, &pattern, false)?;
+            let (plan, _) =
+                PlanBuilder::build(&dataset, &pattern, false, &options.custom_functions)?;
             SimpleEvaluator::new(
                 Rc::new(dataset),
                 base_iri.map(Rc::new),
                 options.service_handler(),
+                Rc::new(options.custom_functions),
             )
             .evaluate_ask_plan(&plan)
         }
@@ -80,23 +86,32 @@ pub(crate) fn evaluate_query(
             base_iri,
             ..
         } => {
-            let (plan, variables) = PlanBuilder::build(&dataset, &pattern, false)?;
-            let construct = PlanBuilder::build_graph_template(&dataset, &template, variables);
+            let (plan, variables) =
+                PlanBuilder::build(&dataset, &pattern, false, &options.custom_functions)?;
+            let construct = PlanBuilder::build_graph_template(
+                &dataset,
+                &template,
+                variables,
+                &options.custom_functions,
+            );
             Ok(SimpleEvaluator::new(
                 Rc::new(dataset),
                 base_iri.map(Rc::new),
                 options.service_handler(),
+                Rc::new(options.custom_functions),
             )
             .evaluate_construct_plan(&plan, construct))
         }
         spargebra::Query::Describe {
             pattern, base_iri, ..
         } => {
-            let (plan, _) = PlanBuilder::build(&dataset, &pattern, false)?;
+            let (plan, _) =
+                PlanBuilder::build(&dataset, &pattern, false, &options.custom_functions)?;
             Ok(SimpleEvaluator::new(
                 Rc::new(dataset),
                 base_iri.map(Rc::new),
                 options.service_handler(),
+                Rc::new(options.custom_functions),
             )
             .evaluate_describe_plan(&plan))
         }
@@ -110,7 +125,8 @@ pub(crate) fn evaluate_query(
 /// a simple HTTP 1.1 client is used to execute [SPARQL 1.1 Federated Query](https://www.w3.org/TR/sparql11-federated-query/) SERVICE calls.
 #[derive(Clone, Default)]
 pub struct QueryOptions {
-    pub(crate) service_handler: Option<Rc<dyn ServiceHandler<Error = EvaluationError>>>,
+    service_handler: Option<Rc<dyn ServiceHandler<Error = EvaluationError>>>,
+    custom_functions: HashMap<NamedNode, Rc<dyn Fn(&[Term]) -> Option<Term>>>,
     http_timeout: Option<Duration>,
 }
 
@@ -138,6 +154,36 @@ impl QueryOptions {
         self
     }
 
+    /// Adds a custom SPARQL evaluation function
+    ///
+    /// ```
+    /// use oxigraph::store::Store;
+    /// use oxigraph::model::*;
+    /// use oxigraph::sparql::{QueryOptions, QueryResults};
+    ///
+    /// let store = Store::new()?;
+    ///
+    /// if let QueryResults::Solutions(mut solutions) = store.query_opt(
+    ///     "SELECT (<http://www.w3.org/ns/formats/N-Triples>(1) AS ?nt) WHERE {}",
+    ///     QueryOptions::default().with_custom_function(
+    ///         NamedNode::new("http://www.w3.org/ns/formats/N-Triples")?,
+    ///         |args| args.get(0).map(|t| Literal::from(t.to_string()).into())
+    ///     )
+    /// )? {
+    ///     assert_eq!(solutions.next().unwrap()?.get("nt"), Some(&Literal::from("\"1\"^^<http://www.w3.org/2001/XMLSchema#integer>").into()));
+    /// }
+    /// # Result::<_,Box<dyn std::error::Error>>::Ok(())
+    /// ```
+    #[inline]
+    pub fn with_custom_function(
+        mut self,
+        name: NamedNode,
+        evaluator: impl Fn(&[Term]) -> Option<Term> + 'static,
+    ) -> Self {
+        self.custom_functions.insert(name, Rc::new(evaluator));
+        self
+    }
+
     fn service_handler(&self) -> Rc<dyn ServiceHandler<Error = EvaluationError>> {
         self.service_handler.clone().unwrap_or_else(|| {
             if cfg!(feature = "http_client") {

lib/src/sparql/plan.rs

@@ -1,3 +1,4 @@
+use crate::model::NamedNode;
 use crate::sparql::model::Variable;
 use crate::storage::numeric_encoder::EncodedTerm;
 use spargebra::algebra::GraphPattern;
@@ -454,6 +455,7 @@ pub enum PlanExpression {
     YearMonthDurationCast(Box<Self>),
     DayTimeDurationCast(Box<Self>),
     StringCast(Box<Self>),
+    CustomFunction(NamedNode, Vec<Self>),
 }
 
 impl PlanExpression {
@@ -557,8 +559,9 @@ impl PlanExpression {
                 c.lookup_used_variables(callback);
                 d.lookup_used_variables(callback);
             }
-
-            PlanExpression::Concat(es) | PlanExpression::Coalesce(es) => {
+            PlanExpression::Concat(es)
+            | PlanExpression::Coalesce(es)
+            | PlanExpression::CustomFunction(_, es) => {
                 for e in es {
                     e.lookup_used_variables(callback);
                 }

lib/src/sparql/plan_builder.rs

@@ -1,4 +1,5 @@
-use crate::model::{LiteralRef, NamedNodeRef};
+use crate::error::invalid_data_error;
+use crate::model::{LiteralRef, NamedNode as OxNamedNode, NamedNodeRef, Term as OxTerm};
 use crate::sparql::dataset::DatasetView;
 use crate::sparql::error::EvaluationError;
 use crate::sparql::model::Variable as OxVariable;
@@ -7,12 +8,13 @@ use crate::storage::numeric_encoder::{EncodedTerm, EncodedTriple};
 use rand::random;
 use spargebra::algebra::*;
 use spargebra::term::*;
-use std::collections::{BTreeSet, HashSet};
+use std::collections::{BTreeSet, HashMap, HashSet};
 use std::mem::swap;
 use std::rc::Rc;
 
 pub struct PlanBuilder<'a> {
     dataset: &'a DatasetView,
+    custom_functions: &'a HashMap<OxNamedNode, Rc<dyn Fn(&[OxTerm]) -> Option<OxTerm>>>,
 }
 
 impl<'a> PlanBuilder<'a> {
@@ -20,9 +22,14 @@ impl<'a> PlanBuilder<'a> {
         dataset: &'a DatasetView,
         pattern: &GraphPattern,
         is_cardinality_meaningful: bool,
+        custom_functions: &'a HashMap<OxNamedNode, Rc<dyn Fn(&[OxTerm]) -> Option<OxTerm>>>,
     ) -> Result<(PlanNode, Vec<Variable>), EvaluationError> {
         let mut variables = Vec::default();
-        let plan = PlanBuilder { dataset }.build_for_graph_pattern(
+        let plan = PlanBuilder {
+            dataset,
+            custom_functions,
+        }
+        .build_for_graph_pattern(
             pattern,
             &mut variables,
             &PatternValue::Constant(EncodedTerm::DefaultGraph),
@@ -43,8 +50,13 @@ impl<'a> PlanBuilder<'a> {
         dataset: &'a DatasetView,
         template: &[TriplePattern],
         mut variables: Vec<Variable>,
+        custom_functions: &'a HashMap<OxNamedNode, Rc<dyn Fn(&[OxTerm]) -> Option<OxTerm>>>,
     ) -> Vec<TripleTemplate> {
-        PlanBuilder { dataset }.build_for_graph_template(template, &mut variables)
+        PlanBuilder {
+            dataset,
+            custom_functions,
+        }
+        .build_for_graph_template(template, &mut variables)
     }
 
     fn build_for_graph_pattern(
@@ -628,7 +640,16 @@ impl<'a> PlanBuilder<'a> {
                     self.build_for_expression(&parameters[0], variables, graph_name)?,
                 )),
                 Function::Custom(name) => {
-                    if name.iri == "http://www.w3.org/2001/XMLSchema#boolean" {
+                    let ox_name = OxNamedNode::new(&name.iri).map_err(invalid_data_error)?;
+                    if self.custom_functions.contains_key(&ox_name) {
+                        PlanExpression::CustomFunction(
+                            ox_name,
+                            parameters
+                                .iter()
+                                .map(|p| self.build_for_expression(p, variables, graph_name))
+                                .collect::<Result<Vec<_>, EvaluationError>>()?,
+                        )
+                    } else if name.iri == "http://www.w3.org/2001/XMLSchema#boolean" {
                         self.build_cast(
                             parameters,
                             PlanExpression::BooleanCast,

lib/src/sparql/update.rs

@@ -138,11 +138,17 @@ impl SimpleUpdateEvaluator<'_> {
         algebra: &GraphPattern,
     ) -> Result<(), EvaluationError> {
         let dataset = Rc::new(DatasetView::new(self.transaction.reader(), using));
-        let (plan, variables) = PlanBuilder::build(dataset.as_ref(), algebra, false)?;
+        let (plan, variables) = PlanBuilder::build(
+            dataset.as_ref(),
+            algebra,
+            false,
+            &self.options.query_options.custom_functions,
+        )?;
         let evaluator = SimpleEvaluator::new(
             dataset.clone(),
             self.base_iri.clone(),
             self.options.query_options.service_handler(),
+            Rc::new(self.options.query_options.custom_functions.clone()),
         );
         let mut bnodes = HashMap::new();
         for tuple in evaluator.plan_evaluator(&plan)(EncodedTuple::with_capacity(variables.len())) {