default source-location identity and opt-in content-digest identity for file move/rename reconciliation, PDF/DOCX-style placeholder ingestion

tests/test_markitect_ingestion_adapter.py

@@ -0,0 +1,97 @@
+import sys
+from pathlib import Path
+from types import SimpleNamespace
+
+import pytest
+
+from kontextual_engine import SourcePayload, SourceReference, content_digest
+from kontextual_engine.adapters.markitect_tool import MarkitectMarkdownExtractor
+from kontextual_engine.errors import AdapterUnavailableError
+
+
+def test_markitect_markdown_extractor_missing_dependency_is_structured(monkeypatch: pytest.MonkeyPatch) -> None:
+    monkeypatch.setitem(sys.modules, "markitect_tool", None)
+    extractor = MarkitectMarkdownExtractor()
+    payload = markdown_payload("# Missing Adapter\n")
+
+    with pytest.raises(AdapterUnavailableError) as exc_info:
+        extractor.extract(payload)
+
+    assert exc_info.value.details == {
+        "adapter": "markitect-tool",
+        "media_type": "text/markdown",
+    }
+
+
+def test_markitect_markdown_extractor_delegates_to_markitect_tool(
+    monkeypatch: pytest.MonkeyPatch,
+    tmp_path: Path,
+) -> None:
+    source = tmp_path / "decision.md"
+    source.write_text("# Decision\n\nUse Markitect.\n", encoding="utf-8")
+    calls: list[tuple[str, str]] = []
+
+    def parse_markdown_file(path: Path) -> SimpleNamespace:
+        calls.append(("parse_markdown_file", str(path)))
+        return SimpleNamespace(
+            to_dict=lambda: {
+                "frontmatter": {"status": "accepted"},
+                "headings": [{"level": 1, "text": "Decision", "line": 1}],
+                "sections": [
+                    {
+                        "heading": {"level": 1, "text": "Decision", "line": 1},
+                        "blocks": [{"type": "paragraph", "text": "Use Markitect.", "line_start": 3}],
+                    }
+                ],
+            }
+        )
+
+    def snapshot_identity_for_file(path: Path, *, parse_options: dict) -> SimpleNamespace:
+        calls.append(("snapshot_identity_for_file", f"{path}:{parse_options['profile']}"))
+        return SimpleNamespace(
+            to_dict=lambda: {
+                "snapshot_id": "snapshot:decision",
+                "content_hash": "sha256:decision",
+                "parser": "markdown-it-py/commonmark",
+            }
+        )
+
+    monkeypatch.setitem(
+        sys.modules,
+        "markitect_tool",
+        SimpleNamespace(
+            parse_markdown_file=parse_markdown_file,
+            parse_markdown=lambda text, source_path=None: None,
+            snapshot_identity_for_file=snapshot_identity_for_file,
+        ),
+    )
+
+    result = MarkitectMarkdownExtractor().extract(markdown_payload(source.read_text(encoding="utf-8"), source))
+
+    assert calls == [
+        ("parse_markdown_file", str(source)),
+        ("snapshot_identity_for_file", f"{source}:default"),
+    ]
+    assert result.normalized.structure["frontmatter"] == {"status": "accepted"}
+    assert result.normalized.fields["heading_count"] == 1
+    assert result.normalized.fields["section_count"] == 1
+    assert result.metadata["snapshot"]["snapshot_id"] == "snapshot:decision"
+    assert result.normalized.extractor_metadata["snapshot"]["parser"] == "markdown-it-py/commonmark"
+
+
+def markdown_payload(markdown: str, path: Path | None = None) -> SourcePayload:
+    data = markdown.encode("utf-8")
+    source_ref = SourceReference(
+        source_system="local_file",
+        path=str(path) if path else None,
+        checksum=content_digest(data),
+        connector_ref=f"local_file:{path}" if path else None,
+    )
+    return SourcePayload(
+        connector_name="local_file",
+        source_uri=str(path) if path else "memory://markdown",
+        source_ref=source_ref,
+        media_type="text/markdown",
+        content=data,
+        title=path.stem if path else "Markdown",
+    )