Initial curated itworx-knowledge-hub source from 3b1751d0d9dee551aa6994eee0093704dc334817

tools/leafwiki-inspector/tests/test_inspect.py

@@ -0,0 +1,120 @@
+from __future__ import annotations
+
+import hashlib
+import json
+import sqlite3
+import subprocess
+import sys
+import tempfile
+import unittest
+from pathlib import Path
+
+
+REPOSITORY = Path(__file__).resolve().parents[3]
+INSPECTOR = REPOSITORY / "tools" / "leafwiki-inspector" / "inspect.py"
+MARKDOWN_FIXTURE = REPOSITORY / "fixtures" / "markdown-import"
+
+
+def tree_hash(root: Path) -> str:
+    digest = hashlib.sha256()
+    for path in sorted(root.rglob("*"), key=lambda item: item.as_posix()):
+        relative = path.relative_to(root).as_posix().encode("utf-8")
+        digest.update(relative)
+        if path.is_symlink():
+            digest.update(b"symlink")
+            digest.update(path.readlink().as_posix().encode("utf-8"))
+        elif path.is_file():
+            digest.update(b"file")
+            digest.update(path.read_bytes())
+        else:
+            digest.update(b"dir")
+    return digest.hexdigest()
+
+
+class InspectorTests(unittest.TestCase):
+    def run_inspector(self, source: Path, output: Path | None = None):
+        command = [sys.executable, str(INSPECTOR), str(source)]
+        if output is not None:
+            command.extend(["--output", str(output)])
+        return subprocess.run(
+            command,
+            check=False,
+            text=True,
+            capture_output=True,
+        )
+
+    def test_markdown_fixture_is_inventoried_without_source_changes(self):
+        before = tree_hash(MARKDOWN_FIXTURE)
+
+        result = self.run_inspector(MARKDOWN_FIXTURE)
+
+        self.assertEqual(0, result.returncode, result.stderr)
+        report = json.loads(result.stdout)
+        self.assertGreaterEqual(report["summary"]["countsByKind"]["markdown"], 3)
+        pages = [
+            entry
+            for entry in report["entries"]
+            if entry["kind"] == "markdown"
+        ]
+        self.assertTrue(
+            any(
+                link["target"] == "Shell-operators"
+                for page in pages
+                for link in page["markdown"]["wikiLinks"]
+            )
+        )
+        self.assertEqual(before, tree_hash(MARKDOWN_FIXTURE))
+
+    def test_sqlite_is_opened_read_only_and_schema_is_reported(self):
+        with tempfile.TemporaryDirectory() as temporary:
+            root = Path(temporary) / "source"
+            root.mkdir()
+            database = root / "leafwiki.sqlite"
+
+            connection = sqlite3.connect(database)
+            connection.executescript(
+                """
+                CREATE TABLE pages (
+                    id INTEGER PRIMARY KEY,
+                    title TEXT NOT NULL,
+                    body TEXT NOT NULL
+                );
+                INSERT INTO pages (title, body) VALUES ('Test', '# Test');
+                """
+            )
+            connection.commit()
+            connection.close()
+
+            before = tree_hash(root)
+            result = self.run_inspector(root)
+            after = tree_hash(root)
+
+            self.assertEqual(0, result.returncode, result.stderr)
+            report = json.loads(result.stdout)
+            sqlite_entry = next(
+                entry for entry in report["entries"] if entry["kind"] == "sqlite"
+            )
+            pages = next(
+                table for table in sqlite_entry["sqlite"]["tables"]
+                if table["name"] == "pages"
+            )
+            self.assertEqual(1, pages["rowCount"])
+            self.assertEqual(before, after)
+            self.assertFalse((root / "leafwiki.sqlite-journal").exists())
+            self.assertFalse((root / "leafwiki.sqlite-wal").exists())
+
+    def test_report_cannot_be_written_inside_source(self):
+        with tempfile.TemporaryDirectory() as temporary:
+            root = Path(temporary) / "source"
+            root.mkdir()
+            (root / "page.md").write_text("# Page\n", encoding="utf-8")
+
+            result = self.run_inspector(root, root / "report.json")
+
+            self.assertEqual(2, result.returncode)
+            self.assertIn("Refusing", result.stderr)
+            self.assertFalse((root / "report.json").exists())
+
+
+if __name__ == "__main__":
+    unittest.main()