okfgraph 0.2.4__tar.gz → 0.2.6__tar.gz

{okfgraph-0.2.4 → okfgraph-0.2.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: okfgraph
-Version: 0.2.4
+Version: 0.2.6
 Summary: Ladybug-backed OKF knowledge graph with ONNX + Jina v5 embeddings
 License-Expression: Apache-2.0 OR MIT
 Requires-Python: >=3.11

{okfgraph-0.2.4 → okfgraph-0.2.6}/README.md

@@ -1,4 +1,4 @@
-# OKFgraph 0.2.4
+# OKFgraph 0.2.5
 
 [![PyPI](https://img.shields.io/pypi/v/okfgraph)](https://pypi.org/project/okfgraph/)
 [![Python](https://img.shields.io/badge/python-%3E%3D3.11-blue)](https://www.python.org/)
@@ -49,7 +49,7 @@ a swappable `DocumentConverter` seam. What isn't needed isn't installed.
 Requires Python ≥ 3.11.
 
 ```bash
-pip install "okfgraph[pdf,omni]"   # PyPI release (okf-embed ships platform wheels)
+pip install "okfgraph[pdf,omni]"   # PyPI (okf-embed ships platform wheels; published)
 ```
 
 Or from source with `uv` (also builds the `okf-embed` Rust wheel from
@@ -273,6 +273,7 @@ Key design decisions:
 ```bash
 uv run --project . pytest tests/ -q        # full suite (model loads; takes a while)
 uv run --project . pytest tests/test_ranking.py tests/test_mcp_server.py -q   # fast subset
+cd rust/okf-embed && cargo test --locked   # Rust unit tests (pure, no model)
 ```
 
 Golden fixtures under `tests/fixtures/` (`ppr_graph`, `diff_a`/`diff_b`,
@@ -297,7 +298,7 @@ okfgraph/
 │   ├── images.py          # IngestMode (text|optional|omni), planning helpers
 │   ├── security.py        # SSRF/domain guards for remote images
 │   ├── tools.py           # legacy tool definitions (superseded by mcp_server)
-│   └── components/        # ranking, links, search, import_, export, diff, doctor,
+│   └── components/        # ranking, links, search, lint, import_, export, diff, doctor,
 │                          # embedding, converters, image_assets, delta, purge, schema, ingest
 ├── rust/okf-embed/        # Jina v5 Rust loader (ort, CUDA-opportunistic) + Python wheel
 ├── skills/                # okfgraph-mcp, okfgraph-cli, okfgraph-ingest (harness-neutral source)

{okfgraph-0.2.4 → okfgraph-0.2.6}/okfgraph/cli.py

@@ -695,7 +695,10 @@ def _ingest(args):
     """
     logger = logging.getLogger("cli")
     router = _router(args)
-    kind = getattr(args, "kind", "pdf") or "pdf"
+    kind = getattr(args, "kind", None)
+    if kind not in ("md", "pdf", "thoughts"):
+        print("[ERROR] --kind is required: md|pdf|thoughts")
+        return
     tags = args.tags.split(",") if getattr(args, "tags", None) else None
     if kind == "md":
         md_file = getattr(args, "md_file", None)
@@ -1171,8 +1174,8 @@ def build_parser():
     p = sub.add_parser("ingest", help="Add content: markdown file, PDF, or thoughts")
     _add_global(p)
     _add_logging_flags(p)
-    p.add_argument("--kind", default="pdf", choices=["md", "pdf", "thoughts"],
-                   help="What to ingest (default: pdf)")
+    p.add_argument("--kind", required=True, choices=["md", "pdf", "thoughts"],
+                   help="What to ingest")
     p.add_argument("--md-file", default=None, help="Markdown file (--kind md)")
     p.add_argument("--pdf-file", default=None, help="PDF file (--kind pdf)")
     p.add_argument("--thoughts", default=None, help="Raw reasoning text (--kind thoughts)")

{okfgraph-0.2.4 → okfgraph-0.2.6}/okfgraph/components/import_.py

@@ -145,6 +145,36 @@ class ImportManager:
         self.image_mgr = image_mgr
         self.purge_mgr = purge_mgr
 
+    def _merge_directory_contains(self, parent_id: str, child_id: str):
+        """Create a Directory→Directory CONTAINS edge without crashing Ladybug.
+
+        Ladybug 0.20.3 segfaults when one Cypher statement combines an
+        existing-node ``MERGE``, a new-node ``MERGE``, and a relationship
+        ``MERGE`` on a long-lived connection. Separate idempotent statements
+        preserve the same graph while avoiding that planner path.
+        """
+        self.conn.execute("MERGE (p:Directory {id: $id})", {"id": parent_id})
+        self.conn.execute("MERGE (d:Directory {id: $id})", {"id": child_id})
+        self.conn.execute("""
+            MATCH (p:Directory {id: $parent}), (d:Directory {id: $child})
+            MERGE (p)-[:CONTAINS]->(d)
+        """, {"parent": parent_id, "child": child_id})
+
+    def _merge_concept_contains(self, parent_id: str, child_id: str):
+        """Create a Directory→Concept CONTAINS edge without crashing Ladybug.
+
+        See ``_merge_directory_contains``: a single three-clause ``MERGE`` can
+        segfault ladybug 0.20.3 after matching an existing Directory and
+        creating the next sibling Concept. The separate statements below are
+        semantically equivalent.
+        """
+        self.conn.execute("MERGE (d:Directory {id: $id})", {"id": parent_id})
+        self.conn.execute("MERGE (c:Concept {id: $id})", {"id": child_id})
+        self.conn.execute("""
+            MATCH (d:Directory {id: $parent}), (c:Concept {id: $child})
+            MERGE (d)-[:CONTAINS]->(c)
+        """, {"parent": parent_id, "child": child_id})
+
     def _batch_build_directories(self, cids: List[str]):
         """Build directory hierarchy for a batch of concept IDs.
 
@@ -166,11 +196,7 @@ class ImportManager:
         for d in sorted_dirs:
             parent = "/".join(d.split("/")[:-1]) if "/" in d else None
             if parent and parent in dir_paths:
-                self.conn.execute("""
-                    MERGE (p:Directory {id: $parent})
-                    MERGE (d:Directory {id: $child})
-                    MERGE (p)-[:CONTAINS]->(d)
-                """, {"parent": parent, "child": d})
+                self._merge_directory_contains(parent, d)
             elif parent:
                 # Parent is root (not a directory node)
                 self.conn.execute("""
@@ -186,11 +212,7 @@ class ImportManager:
             parts = cid.split("/")
             if len(parts) > 1:
                 parent_dir = "/".join(parts[:-1])
-                self.conn.execute("""
-                    MERGE (d:Directory {id: $parent})
-                    MERGE (c:Concept {id: $child})
-                    MERGE (d)-[:CONTAINS]->(c)
-                """, {"parent": parent_dir, "child": cid})
+                self._merge_concept_contains(parent_dir, cid)
 
 
     def _load_link_index(self):
@@ -867,17 +889,9 @@ class ImportManager:
                 parent = "/".join(path_parts[:i])
                 child = "/".join(path_parts[: i + 1])
                 if i == len(path_parts) - 1:
-                    self.conn.execute("""
-                        MERGE (d:Directory {id: $parent})
-                        MERGE (c:Concept {id: $child})
-                        MERGE (d)-[:CONTAINS]->(c)
-                    """, {"parent": parent, "child": child})
+                    self._merge_concept_contains(parent, child)
                 else:
-                    self.conn.execute("""
-                        MERGE (p:Directory {id: $parent})
-                        MERGE (d:Directory {id: $child})
-                        MERGE (p)-[:CONTAINS]->(d)
-                    """, {"parent": parent, "child": child})
+                    self._merge_directory_contains(parent, child)
 
         # NOTE: link extraction intentionally lives outside _insert_concept
         # (callers run _extract_links_for_concept after the node exists), so

{okfgraph-0.2.4 → okfgraph-0.2.6}/okfgraph.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: okfgraph
-Version: 0.2.4
+Version: 0.2.6
 Summary: Ladybug-backed OKF knowledge graph with ONNX + Jina v5 embeddings
 License-Expression: Apache-2.0 OR MIT
 Requires-Python: >=3.11

{okfgraph-0.2.4 → okfgraph-0.2.6}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "okfgraph"
-version = "0.2.4"
+version = "0.2.6"
 description = "Ladybug-backed OKF knowledge graph with ONNX + Jina v5 embeddings"
 requires-python = ">=3.11"
 license = "Apache-2.0 OR MIT"

{okfgraph-0.2.4 → okfgraph-0.2.6}/tests/test_chunking.py

@@ -449,26 +449,27 @@ class TestPhase5_CLI:
     actual commands with real data and verify output formatting.
     """
 
-    def test_cli_has_search_chunks_command(self):
+    def test_removed_granular_commands_gone(self, capsys):
+        # search-chunks/chunks/reconstruct were folded into search/read in the
+        # 0.2.0 16→5 consolidation; assert argparse rejects them outright.
         from okfgraph.cli import build_parser
         parser = build_parser()
-        args = parser.parse_args(["search-chunks", "test"])
-        assert args.command == "search-chunks"
-        assert args.query == "test"
-
-    def test_cli_has_chunks_command(self):
-        from okfgraph.cli import build_parser
-        parser = build_parser()
-        args = parser.parse_args(["chunks", "some_id"])
-        assert args.command == "chunks"
-        assert args.concept_id == "some_id"
-
-    def test_cli_has_reconstruct_command(self):
+        for dead in ("search-chunks", "chunks", "reconstruct"):
+            with pytest.raises(SystemExit):
+                parser.parse_args([dead, "test"])
+        capsys.readouterr()  # swallow argparse usage noise
+
+    def test_replacement_verbs_cover_removed_commands(self):
+        # search --target chunks ≙ search-chunks; read --include chunks ≙ chunks;
+        # read --include document ≙ reconstruct.
         from okfgraph.cli import build_parser
         parser = build_parser()
-        args = parser.parse_args(["reconstruct", "some_id"])
-        assert args.command == "reconstruct"
-        assert args.concept_id == "some_id"
+        args = parser.parse_args(["search", "test", "--target", "chunks"])
+        assert args.command == "search" and args.target == "chunks"
+        args = parser.parse_args(["read", "--include", "chunks", "some_id"])
+        assert args.concept_id == "some_id" and args.include == "chunks"
+        args = parser.parse_args(["read", "--include", "document", "some_id"])
+        assert args.concept_id == "some_id" and args.include == "document"
 
     def test_cli_has_chunking_options(self):
         from okfgraph.cli import build_parser

{okfgraph-0.2.4 → okfgraph-0.2.6}/tests/test_cli.py

@@ -74,6 +74,14 @@ class TestCLIHelp:
         assert result.returncode == 0
         assert "--kind" in result.stdout
 
+    def test_ingest_kind_required(self):
+        result = subprocess.run(
+            [sys.executable, "-m", "okfgraph.cli", "ingest", "--md-file", "note.md"],
+            capture_output=True, text=True,
+        )
+        assert result.returncode == 2
+        assert "--kind" in result.stderr
+
     def test_export_help(self):
         result = subprocess.run(
             [sys.executable, "-m", "okfgraph.cli", "export", "--help"],

{okfgraph-0.2.4 → okfgraph-0.2.6}/tests/test_router.py

@@ -166,17 +166,21 @@ class TestDeviceSelection:
         assert r.device == "cpu"
         assert r.encoder.used_cuda is False
 
-    def test_device_cuda_uses_gpu_when_available(self, tmp_path, capsys):
+    def test_device_cuda_uses_gpu_when_available(self, tmp_path, caplog):
+        # The fallback warning travels via logging (bound to the real stderr
+        # fd at handler creation), so capsys can never see it — assert on the
+        # log record instead. Deterministic on GPU and CPU machines alike.
+        import logging
         from okfgraph.router import OKFRouter
-        r = OKFRouter(db_path=str(tmp_path / "test.db"), bundle_root=str(tmp_path), device="cuda")
+        with caplog.at_level(logging.WARNING, logger="okfgraph.router"):
+            r = OKFRouter(db_path=str(tmp_path / "test.db"), bundle_root=str(tmp_path), device="cuda")
         assert r.device == "cuda"
-        err = capsys.readouterr().err
         if r.encoder.used_cuda:
             # Happy path: GPU active, no fallback warning.
-            assert "CUDA" not in err
+            assert not any("CUDA" in m for m in caplog.messages)
         else:
-            # CPU fallback: the Rust core warns on stderr.
-            assert "CUDA" in err
+            # CPU fallback: the router warns about the missing EP.
+            assert any("no CUDA execution provider" in m for m in caplog.messages)
 
     def test_device_cuda_warning_only_once(self, tmp_path, capsys):
         from okfgraph.router import OKFRouter