feat: add rdf n-tripples graph load source by VanyaGlazunov · Pull Request #7 · SparseLinearAlgebra/pathrex

VanyaGlazunov · 2026-03-17T10:10:30Z

Adds support for loading graphs from RDF N-Triples format as a new edge source.

closes #3 I think.

suvorovrain · 2026-03-22T14:33:45Z

src/formats/nt.rs

+#[derive(Debug, Clone, Default)]
+pub enum LabelExtraction {
+    /// Use only the local name: the fragment (`#name`) or last path segment.
+    /// For example, `http://example.org/ns/knows` → `"knows"`.


Suggested change

/// For example, `http://example.org/ns/knows` → `"knows"`.

/// For example, `http://example.org/ns/knows` → `knows`.

I don't think there should be any quotation marks here so as not to confuse it with literals.

suvorovrain · 2026-03-22T14:34:02Z

src/formats/nt.rs

+    #[default]
+    LocalName,
+    /// Use the full IRI string as the label.
+    /// For example, `http://example.org/ns/knows` → `"http://example.org/ns/knows"`.


suvorovrain · 2026-03-22T14:46:02Z

src/formats/nt.rs

+    }
+}
+
+impl<R: Read> Iterator for NTriples<R> {


<...> <http://a.org/knows> <...> <...> <http://b.org/knows> <...>

We lose info about difference between these two predicates if use LocalName. Is it good?

suvorovrain · 2026-03-22T14:47:05Z

src/formats/nt.rs

+                } else if let Some(pos) = iri.rfind('/') {
+                    iri[pos + 1..].to_owned()
+                } else {
+                    iri.to_owned()


can we obtain something like <...> <http://a.org/> <...>? In this case predicate will be empty "". Should we handle this case?

suvorovrain · 2026-03-22T14:48:00Z

@VanyaGlazunov Nice. Can you answer on my questions in review?

georgiy-belyanin · 2026-03-24T14:48:12Z

src/formats/nt.rs

+}
+
+#[cfg(test)]
+mod tests {


Could you, please, introduce a test involving non-ascii chars. Especially I am interested in Chinese and Cyrillic support.

georgiy-belyanin

Thank you for working on this! Consider, please, my comment and Rodion's one.

Copilot

Pull request overview

Adds a new RDF N-Triples format parser and wires it into the in-memory graph builder so graphs can be loaded directly from .nt sources (per Issue #3).

Changes:

Introduces formats::nt::NTriples<R> iterator (backed by oxttl) to parse N-Triples into Edge items.
Adds GraphSource<InMemoryBuilder> support for NTriples<R> and a corresponding in-memory loading test.
Updates formats module exports and repo documentation to include the new format.

Reviewed changes

Copilot reviewed 4 out of 5 changed files in this pull request and generated 5 comments.

Show a summary per file

File	Description
src/graph/mod.rs	Minor formatting adjustment in `check_graph` signature.
src/graph/inmemory.rs	Imports `NTriples`, implements `GraphSource` for it, and adds an in-memory load test.
src/formats/nt.rs	New N-Triples parser/iterator implementation with label-extraction strategy and unit tests.
src/formats/mod.rs	Exposes `nt` module + `NTriples` re-export; extends `FormatError` with N-Triples-related variants.
AGENTS.md	Documents the new `nt.rs` format module and usage details.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-24T14:56:15Z

AGENTS.md

+[`NTriples<R>`](src/formats/nt.rs:57) parses [W3C N-Triples](https://www.w3.org/TR/n-triples/)
+RDF files using `oxttl`. Each triple `(subject, predicate, object)` becomes an
+[`Edge`](src/graph/mod.rs:154) where:
+
+- `source` — subject IRI or blank-node ID (`_:label`).
+- `target` — object IRI or blank-node ID; triples whose object is an RDF
+  literal yield `Err(FormatError::LiteralAsNode)` (callers may filter these out).
+- `label` — predicate IRI, transformed by [`LabelExtraction`](src/formats/nt.rs:36):
+
+| Variant | Behaviour |
+|---|---|
+| `LocalName` (default) | Fragment (`#name`) or last path segment of the predicate IRI |
+| `FullIri` | Full predicate IRI string |
+
+Constructors:
+
+- [`NTriples::new(reader)`](src/formats/nt.rs:72) — uses `LabelExtraction::LocalName`.
+- [`NTriples::with_label_extraction(reader, strategy)`](src/formats/nt.rs:76) — explicit strategy.


Several Markdown source links here use hard-coded line numbers that don’t match the newly added src/formats/nt.rs (e.g., NTriples<R> is at line 64, LabelExtraction at 38, and the constructors at 70/74). Updating these references will prevent broken links in GitHub’s rendered documentation.

Copilot · 2026-03-24T14:56:15Z

src/formats/mod.rs

+
+    /// An RDF literal appeared as a subject or object where a node IRI or
+    /// blank node was expected.
+    #[error("RDF literal cannot be used as a graph node (triple skipped)")]


FormatError::LiteralAsNode's display text says “(triple skipped)”, but nothing in the iterator / GraphBuilder::load path actually skips the triple—item? will propagate the error and abort loading unless the caller filters it out. Either remove the “triple skipped” wording from the error message, or implement default skipping at the GraphSource/builder layer for this specific variant so the message matches behavior.

Suggested change

#[error("RDF literal cannot be used as a graph node (triple skipped)")]

#[error("RDF literal cannot be used as a graph node")]

Copilot · 2026-03-24T14:56:15Z

src/formats/mod.rs

+    #[error("N-Triples parse error: {0}")]
+    NTriples(String),
+
+    /// An RDF literal appeared as a subject or object where a node IRI or


The doc comment for LiteralAsNode says “subject or object”, but in RDF/N-Triples a literal cannot appear in the subject position (and the parser type for subject here is NamedOrBlankNode). Consider adjusting the wording to just “object” (or “term position where an IRI/blank node was expected”) to avoid implying an impossible case.

Suggested change

/// An RDF literal appeared as a subject or object where a node IRI or

/// An RDF literal appeared in a term position where an IRI or

Copilot · 2026-03-24T14:56:16Z

src/formats/nt.rs

+    fn extract_label(iri: &str, strategy: &LabelExtraction) -> String {
+        match strategy {
+            LabelExtraction::FullIri => iri.to_owned(),
+            LabelExtraction::LocalName => {
+                // Fragment takes priority, then last path segment.
+                if let Some(pos) = iri.rfind('#') {
+                    iri[pos + 1..].to_owned()
+                } else if let Some(pos) = iri.rfind('/') {
+                    iri[pos + 1..].to_owned()
+                } else {
+                    iri.to_owned()
+                }
+            }


extract_label can return an empty string for IRIs that end with # or / (e.g., http://example.org/ns#). That would create edges with an empty label and store them under the empty-string key in label_buffers, which is likely unintended. Consider falling back to the full IRI (or returning a parse error) when the computed local name is empty.

Copilot · 2026-03-24T14:56:16Z

src/graph/inmemory.rs

+impl<R: Read> GraphSource<InMemoryBuilder> for NTriples<R> {
+    fn apply_to(self, mut builder: InMemoryBuilder) -> Result<InMemoryBuilder, GraphError> {
+        for item in self {
+            builder.push_edge(item?)?;
+        }
+        Ok(builder)
+    }


The GraphSource impl for NTriples<R> duplicates the same for item in self { builder.push_edge(item?)?; } loop used by the existing Csv<R> impl. Since InMemoryBuilder already has with_stream (which handles IntoIterator<Item = Result<Edge, E>>), consider delegating to builder.with_stream(self) to keep the loading logic in one place and avoid divergence between format loaders over time.

VanyaGlazunov force-pushed the feat-add-rdf-graph-load branch 2 times, most recently from e470334 to 9b98673 Compare March 19, 2026 02:54

VanyaGlazunov force-pushed the main branch from eb29a4e to 1561f87 Compare March 19, 2026 04:25

VanyaGlazunov added 2 commits March 19, 2026 07:30

feat: add rdf n-tripples graph load source

22141e5

doc: update AGENTS.md

7fbc869

VanyaGlazunov force-pushed the feat-add-rdf-graph-load branch from 9b98673 to 7fbc869 Compare March 19, 2026 04:31

VanyaGlazunov requested review from georgiy-belyanin and suvorovrain March 19, 2026 04:34

suvorovrain reviewed Mar 22, 2026

View reviewed changes

georgiy-belyanin reviewed Mar 24, 2026

View reviewed changes

georgiy-belyanin requested a review from Copilot March 24, 2026 14:51

georgiy-belyanin reviewed Mar 24, 2026

View reviewed changes

Copilot started reviewing on behalf of georgiy-belyanin March 24, 2026 14:52 View session

Copilot AI reviewed Mar 24, 2026

View reviewed changes

	/// For example, `http://example.org/ns/knows` → `"knows"`.
	/// For example, `http://example.org/ns/knows` → `knows`.

	#[error("RDF literal cannot be used as a graph node (triple skipped)")]
	#[error("RDF literal cannot be used as a graph node")]

	/// An RDF literal appeared as a subject or object where a node IRI or
	/// An RDF literal appeared in a term position where an IRI or

Conversation

VanyaGlazunov commented Mar 17, 2026

Uh oh!

suvorovrain Mar 22, 2026

Choose a reason for hiding this comment

Uh oh!

suvorovrain Mar 22, 2026

Choose a reason for hiding this comment

Uh oh!

suvorovrain Mar 22, 2026

Choose a reason for hiding this comment

Uh oh!

suvorovrain Mar 22, 2026

Choose a reason for hiding this comment

Uh oh!

suvorovrain commented Mar 22, 2026

Uh oh!

georgiy-belyanin Mar 24, 2026

Choose a reason for hiding this comment

Uh oh!

georgiy-belyanin left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Mar 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 24, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 24, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants