python-re3data-v2 0.1.0__py3-none-any.whl

re3data/_serializer.py

@@ -0,0 +1,82 @@
+"""Serialize parsed re3data resources into dict, JSON and CSV representations.
+
+Dict conventions (matching the contract established by the original
+python-re3data library):
+
+- Keys are the XML element/attribute names (camelCase).
+- Elements with text and attributes become ``{"value": ..., "language": ...}``.
+- Dates are rendered as ISO-8601 strings (``2023-12-06``).
+- Absent optional scalars are ``None``; absent repeats are empty lists.
+"""
+
+from __future__ import annotations
+
+import csv
+import io
+import json
+from dataclasses import asdict, is_dataclass
+from datetime import date
+from typing import TYPE_CHECKING, Any
+
+from re3data._resources import RepositorySummary
+
+if TYPE_CHECKING:
+    from re3data._resources import Repository
+
+
+def _normalize(node: Any) -> Any:
+    """Recursively convert non-JSON-native leaves into JSON-native ones."""
+    if isinstance(node, date):
+        return node.isoformat()
+    if is_dataclass(node):
+        return {key: _normalize(value) for key, value in asdict(node).items()}
+    if isinstance(node, dict):
+        return {key: _normalize(value) for key, value in node.items()}
+    if isinstance(node, (list, tuple)):
+        return [_normalize(item) for item in node]
+    return node
+
+
+def _to_dict(parsed: Repository | list[RepositorySummary]) -> dict[str, Any] | list[dict[str, Any]]:
+    """Serialize a parsed resource into a dictionary."""
+    return _normalize(parsed)
+
+
+def _to_json(parsed: Repository | list[RepositorySummary]) -> str:
+    """Serialize a parsed resource into a JSON string."""
+    return json.dumps(_normalize(parsed), indent=2, ensure_ascii=False)
+
+
+def _repository_summary_row(summary: RepositorySummary) -> dict[str, Any]:
+    return {
+        "id": summary.id,
+        "doi": summary.doi,
+        "name": summary.name,
+        "href": summary.href,
+    }
+
+
+def _to_csv(parsed: Repository | list[RepositorySummary]) -> str:
+    """Serialize repository summaries into a CSV string.
+
+    Only list responses produce tabular output; a single repository detail is
+    rendered as one summary row.
+    """
+    summaries: list[RepositorySummary]
+    if isinstance(parsed, list):
+        summaries = parsed
+    else:
+        summaries = [
+            RepositorySummary(
+                id=parsed.id,
+                doi=parsed.identifiers.doi if parsed.identifiers else None,
+                name=parsed.repositoryName.value if parsed.repositoryName else None,
+                link=None,
+            )
+        ]
+    buffer = io.StringIO()
+    writer = csv.DictWriter(buffer, fieldnames=["id", "doi", "name", "href"], lineterminator="\n")
+    writer.writeheader()
+    for summary in summaries:
+        writer.writerow(_repository_summary_row(summary))
+    return buffer.getvalue().rstrip("\n")