Creating SHACL-based knowledge graph subsets

.gitmodules

@@ -5,3 +5,6 @@
 [submodule "shacl_testsuite/data-shapes"]
 	path = shacl_testsuite/data-shapes
 	url = https://github.com/w3c/data-shapes.git
+[submodule "shacl_validation/tests/data-shapes"]
+	path = shacl_validation/tests/data-shapes
+	url = https://github.com/w3c/data-shapes

Cargo.toml

@@ -16,7 +16,6 @@ members = [
     "shapemap",
     "shacl_ast",
     "shacl_validation",
-    "shacl_testsuite",
     "shapes_converter",
     "sparql_service",
     "python",
@@ -26,8 +25,9 @@ exclude = [
     "shex_compact_winnow"
 ]
 
-default-members = [
-    "rudof_cli"
+default-members = [ # TODO: this can possibly be removed when changing shex_testsuite
+    "rudof_cli",
+    "shacl_validation",
 ]
 
 [workspace.package]
@@ -58,13 +58,11 @@ shapemap = { version = "0.1.0", path = "./shapemap" }
 shacl_ast = { version = "0.1.35", path = "./shacl_ast" }
 shacl_validation = { version = "0.1.37", path = "./shacl_validation" }
 shapes_converter = { version = "0.1.33", path = "./shapes_converter" }
-shex_testsuite = { version = "0.1.0", path = "./shex_testsuite" }
 shex_validation = { version = "0.1.0", path = "./shex_validation" }
 shex_compact = { version = "0.1.0", path = "./shex_compact" }
 srdf = { version = "0.1.35", path = "./srdf" }
 sparql_service = { version = "0.1.37", path = "./sparql_service" }
 
-# [dependencies]
 # External dependencies
 anyhow = "1.0"
 clap = { version = "4.2.1", features = ["derive"] }

python/src/pyrudof_lib.rs

@@ -4,8 +4,8 @@ use pyo3::{exceptions::PyValueError, pyclass, pymethods, PyErr, PyResult, Python
 use rudof_lib::{
     iri, DCTAPFormat, PrefixMap, QueryShapeMap, RDFFormat, ReaderMode, ResultShapeMap, Rudof,
     RudofConfig, RudofError, ShExFormat, ShExFormatter, ShExSchema, ShaclFormat, ShaclSchema,
-    ShaclValidationMode, ShapeMapFormat, ShapeMapFormatter, ShapesGraphSource, UmlGenerationMode,
-    ValidationReport, ValidationStatus, DCTAP,
+    ShaclValidationMode, ShapeMapFormat, ShapeMapFormatter, ShapesGraphSource, Subsetting,
+    UmlGenerationMode, ValidationReport, ValidationStatus, DCTAP,
 };
 use std::{ffi::OsStr, fs::File, io::BufReader, path::Path};
 
@@ -309,17 +309,18 @@ impl PyRudof {
     /// which can be extracted from the current RDF data,
     /// or from the current SHACL schema.
     /// If there is no current SHACL schema, it tries to get it from the current RDF data
-    #[pyo3(signature = (mode = &PyShaclValidationMode::Native, shapes_graph_source = &PyShapesGraphSource::CurrentSchema ))]
+    #[pyo3(signature = (mode = &PyShaclValidationMode::Native, shapes_graph_source = &PyShapesGraphSource::CurrentSchema, subsetting = &PySubsetting::None ))]
     pub fn validate_shacl(
         &mut self,
         mode: &PyShaclValidationMode,
         shapes_graph_source: &PyShapesGraphSource,
+        subsetting: &PySubsetting,
     ) -> PyResult<PyValidationReport> {
         let mode = cnv_shacl_validation_mode(mode);
         let shapes_graph_source = cnv_shapes_graph_source(shapes_graph_source);
         let result = self
             .inner
-            .validate_shacl(&mode, &shapes_graph_source)
+            .validate_shacl(&mode, &shapes_graph_source, cnv_subsetting(subsetting))
             .map_err(cnv_err)?;
         Ok(PyValidationReport { inner: result })
     }
@@ -657,6 +658,14 @@ pub enum PyShaclValidationMode {
     Sparql,
 }
 
+#[pyclass(eq, eq_int, name = "Subsetting")]
+#[derive(PartialEq)]
+pub enum PySubsetting {
+    None,
+    Full,
+    Provenance,
+}
+
 #[pyclass(eq, eq_int, name = "ShapesGraphSource")]
 #[derive(PartialEq)]
 pub enum PyShapesGraphSource {
@@ -800,6 +809,14 @@ fn cnv_shacl_validation_mode(mode: &PyShaclValidationMode) -> ShaclValidationMod
     }
 }
 
+fn cnv_subsetting(subsetting: &PySubsetting) -> Subsetting {
+    match subsetting {
+        PySubsetting::None => Subsetting::None,
+        PySubsetting::Full => Subsetting::Full,
+        PySubsetting::Provenance => Subsetting::Provenance,
+    }
+}
+
 fn cnv_shapes_graph_source(sgs: &PyShapesGraphSource) -> ShapesGraphSource {
     match sgs {
         PyShapesGraphSource::CurrentData => ShapesGraphSource::CurrentData,

rudof_cli/src/cli.rs

@@ -2,6 +2,7 @@ use crate::input_spec::InputSpec;
 use crate::{InputConvertFormat, OutputConvertFormat};
 use clap::{Parser, Subcommand, ValueEnum};
 use shacl_validation::shacl_processor::ShaclValidationMode;
+use shacl_validation::Subsetting;
 use srdf::{RDFFormat, ReaderMode};
 use std::fmt::Display;
 use std::{fmt::Formatter, path::PathBuf};
@@ -208,6 +209,16 @@ pub enum Command {
         /// Config file path, if unset it assumes default config
         #[arg(short = 'c', long = "config-file", value_name = "Config file name")]
         config: Option<PathBuf>,
+
+        /// Subsetting mode
+        #[arg(
+            short = 'u',
+            long = "subsetting",
+            value_name = "Subsetting mode",
+            default_value_t = Subsetting::None,
+            value_enum
+        )]
+        subsetting: Subsetting,
     },
 
     /// Validate RDF using ShEx schemas
@@ -347,6 +358,16 @@ pub enum Command {
         /// Config file path, if unset it assumes default config
         #[arg(short = 'c', long = "config-file", value_name = "Config file name")]
         config: Option<PathBuf>,
+
+        /// Subsetting mode
+        #[arg(
+            short = 'u',
+            long = "subsetting",
+            value_name = "Subsetting mode",
+            default_value_t = Subsetting::None,
+            value_enum
+        )]
+        subsetting: Subsetting,
     },
 
     /// Show information about RDF data

rudof_cli/src/main.rs

@@ -28,6 +28,7 @@ use rudof_lib::{
     Rudof, RudofConfig, ShExFormat, ShExFormatter, ShaclFormat, ShaclValidationMode,
     ShapeMapFormatter, ShapeMapParser, ShapesGraphSource,
 };
+use shacl_validation::Subsetting;
 use shapemap::{NodeSelector, ShapeMapFormat as ShapemapFormat, ShapeSelector};
 use shapes_converter::ShEx2Sparql;
 use shapes_converter::{ImageFormat, ShEx2Html, ShEx2Uml, Shacl2ShEx, Tap2ShEx, UmlGenerationMode};
@@ -147,6 +148,7 @@ fn main() -> Result<()> {
             output,
             config,
             force_overwrite,
+            subsetting,
         }) => {
             let config = get_config(config)?;
             match validation_mode {
@@ -186,6 +188,7 @@ fn main() -> Result<()> {
                         output,
                         &config,
                         *force_overwrite,
+                        *subsetting,
                     )
                 }
             }
@@ -234,6 +237,7 @@ fn main() -> Result<()> {
             output,
             force_overwrite,
             config,
+            subsetting,
         }) => {
             let config = get_config(config)?;
             run_validate_shacl(
@@ -248,6 +252,7 @@ fn main() -> Result<()> {
                 output,
                 &config,
                 *force_overwrite,
+                *subsetting,
             )
         }
         Some(Command::Data {
@@ -615,20 +620,23 @@ fn run_validate_shacl(
     output: &Option<PathBuf>,
     config: &RudofConfig,
     force_overwrite: bool,
+    subsetting: Subsetting,
 ) -> Result<()> {
     let (mut writer, _color) = get_writer(output, force_overwrite)?;
     let mut rudof = Rudof::new(config);
+
     get_data_rudof(&mut rudof, data, data_format, endpoint, reader_mode, config)?;
-    let result = if let Some(schema) = schema {
+
+    if let Some(schema) = schema {
         let reader_mode = reader_mode_convert(*reader_mode);
         let shapes_format = shapes_format.unwrap_or_default();
         add_shacl_schema_rudof(&mut rudof, schema, &shapes_format, &reader_mode, config)?;
-        rudof.validate_shacl(&mode, &ShapesGraphSource::current_schema())
-    } else {
-        rudof.validate_shacl(&mode, &ShapesGraphSource::current_data())
-    }?;
+    };
+
+    let result = rudof.validate_shacl(&mode, &ShapesGraphSource::current_schema(), subsetting)?;
 
     writeln!(writer, "Result:\n{}", result)?;
+
     Ok(())
 }
 

rudof_lib/src/rudof.rs

@@ -1,8 +1,7 @@
 use crate::{RudofConfig, RudofError, ShapesGraphSource};
 use iri_s::IriS;
 use shacl_ast::{ShaclParser, ShaclWriter};
-use shacl_validation::shacl_processor::{GraphValidation, ShaclProcessor};
-use shacl_validation::store::graph::Graph;
+use shacl_validation::shacl_processor::ShaclProcessor;
 
 use shapemap::{NodeSelector, ShapeSelector};
 use shapes_converter::{ShEx2Uml, Tap2ShEx};
@@ -21,8 +20,8 @@ pub use dctap::{DCTAPFormat, DCTap as DCTAP};
 pub use iri_s::iri;
 pub use prefixmap::PrefixMap;
 pub use shacl_ast::ShaclFormat;
-pub use shacl_validation::shacl_processor::ShaclValidationMode;
 pub use shacl_validation::validation_report::report::ValidationReport;
+pub use shacl_validation::{shacl_processor::ShaclValidationMode, Subsetting};
 pub use shapemap::{QueryShapeMap, ResultShapeMap, ShapeMapFormat, ValidationStatus};
 pub use shex_compact::{ShExFormatter, ShapeMapParser, ShapemapFormatter as ShapeMapFormatter};
 pub use shex_validation::Validator as ShExValidator;
@@ -436,6 +435,7 @@ impl Rudof {
         &mut self,
         mode: &ShaclValidationMode,
         shapes_graph_source: &ShapesGraphSource,
+        subsetting: Subsetting,
     ) -> Result<ValidationReport> {
         let (compiled_schema, shacl_schema) = match shapes_graph_source {
             ShapesGraphSource::CurrentSchema if self.shacl_schema.is_some() => {
@@ -459,7 +459,7 @@ impl Rudof {
                 Ok((compiled_schema, ast_schema))
             }
         }?;
-        let validator = GraphValidation::from_graph(Graph::from_data(self.rdf_data.clone()), *mode);
+        let validator = ShaclProcessor::new(self.rdf_data.clone(), *mode, subsetting);
         let result = ShaclProcessor::validate(&validator, &compiled_schema).map_err(|e| {
             RudofError::SHACLValidationError {
                 error: format!("{e}"),
@@ -657,7 +657,7 @@ fn shacl_format2rdf_format(shacl_format: &ShaclFormat) -> Result<RDFFormat> {
 mod tests {
     use iri_s::iri;
     use shacl_ast::ShaclFormat;
-    use shacl_validation::shacl_processor::ShaclValidationMode;
+    use shacl_validation::{shacl_processor::ShaclValidationMode, Subsetting};
     use shapemap::ShapeMapFormat;
     use shex_ast::{compiled::shape_label::ShapeLabel, Node};
     use shex_validation::ShExFormat;
@@ -760,6 +760,7 @@ mod tests {
             .validate_shacl(
                 &ShaclValidationMode::Native,
                 &crate::ShapesGraphSource::CurrentSchema,
+                Subsetting::None,
             )
             .unwrap();
         assert!(result.results().is_empty())
@@ -805,6 +806,7 @@ mod tests {
             .validate_shacl(
                 &ShaclValidationMode::Native,
                 &crate::ShapesGraphSource::CurrentSchema,
+                Subsetting::None,
             )
             .unwrap();
         assert!(!result.conforms())
@@ -840,6 +842,7 @@ mod tests {
             .validate_shacl(
                 &ShaclValidationMode::Native,
                 &crate::ShapesGraphSource::CurrentData,
+                Subsetting::None,
             )
             .unwrap();
         assert!(!result.conforms())
@@ -875,6 +878,7 @@ mod tests {
             .validate_shacl(
                 &ShaclValidationMode::Native,
                 &crate::ShapesGraphSource::CurrentData,
+                Subsetting::None,
             )
             .unwrap();
         assert!(result.conforms())