first usable curator UI

src/repo_registry/repo_ingestion/metadata.py

@@ -0,0 +1,86 @@
+from __future__ import annotations
+
+import json
+import tomllib
+from dataclasses import dataclass
+from pathlib import Path
+from urllib.parse import urlparse
+
+
+@dataclass(frozen=True)
+class RepositoryMetadata:
+    name: str
+    description: str | None
+
+
+class RepositoryMetadataExtractor:
+    def extract(self, source_path: str | Path, url: str) -> RepositoryMetadata:
+        root = Path(source_path)
+        pyproject = self._from_pyproject(root)
+        package = self._from_package_json(root)
+        readme = self._from_readme(root)
+        fallback_name = self._name_from_url_or_path(url)
+
+        return RepositoryMetadata(
+            name=pyproject.name or package.name or readme.name or fallback_name,
+            description=(
+                pyproject.description
+                or package.description
+                or readme.description
+            ),
+        )
+
+    def _from_pyproject(self, root: Path) -> RepositoryMetadata:
+        path = root / "pyproject.toml"
+        if not path.exists():
+            return RepositoryMetadata(name="", description=None)
+        try:
+            project = tomllib.loads(path.read_text(encoding="utf-8")).get("project", {})
+        except (OSError, tomllib.TOMLDecodeError):
+            return RepositoryMetadata(name="", description=None)
+        return RepositoryMetadata(
+            name=str(project.get("name") or ""),
+            description=project.get("description"),
+        )
+
+    def _from_package_json(self, root: Path) -> RepositoryMetadata:
+        path = root / "package.json"
+        if not path.exists():
+            return RepositoryMetadata(name="", description=None)
+        try:
+            package = json.loads(path.read_text(encoding="utf-8"))
+        except (OSError, json.JSONDecodeError):
+            return RepositoryMetadata(name="", description=None)
+        return RepositoryMetadata(
+            name=str(package.get("name") or ""),
+            description=package.get("description"),
+        )
+
+    def _from_readme(self, root: Path) -> RepositoryMetadata:
+        for readme in sorted(root.glob("README*")):
+            if not readme.is_file():
+                continue
+            try:
+                lines = readme.read_text(encoding="utf-8", errors="ignore").splitlines()
+            except OSError:
+                continue
+            title = ""
+            for line in lines:
+                stripped = line.strip()
+                cleaned = stripped.strip("#").strip()
+                if stripped.startswith("#") and cleaned and not title:
+                    title = cleaned
+                    continue
+                if cleaned:
+                    return RepositoryMetadata(name=title, description=cleaned)
+            if title:
+                return RepositoryMetadata(name=title, description=None)
+        return RepositoryMetadata(name="", description=None)
+
+    def _name_from_url_or_path(self, value: str) -> str:
+        parsed = urlparse(value)
+        path = parsed.path if parsed.scheme else value
+        name = Path(path.rstrip("/")).name or "repository"
+        if name.endswith(".git"):
+            name = name[:-4]
+        return name or "repository"