Initial SQL-only 1C adapter baseline

scripts/convert_1c_metadata_to_rag.py

@@ -0,0 +1,114 @@
+from __future__ import annotations
+
+import argparse
+import json
+from pathlib import Path
+
+
+ROOT = Path(__file__).resolve().parents[1]
+DEFAULT_INPUT = ROOT / "plugins" / "1c" / "metadata" / "examples" / "metadata.example.json"
+DEFAULT_OUTPUT = ROOT / "plugins" / "1c" / "rag" / "sources" / "metadata.generated.md"
+
+KIND_TITLES = {
+    "catalog": "Справочник",
+    "document": "Документ",
+    "register": "Регистр",
+    "common_module": "Общий модуль",
+    "enum": "Перечисление",
+    "report": "Отчет",
+    "processing": "Обработка",
+    "other": "Объект",
+}
+
+
+def load_snapshot(path: Path) -> dict:
+    with path.open("r", encoding="utf-8") as handle:
+        data = json.load(handle)
+    if not isinstance(data, dict):
+        raise ValueError("metadata snapshot must be a JSON object")
+    if data.get("schema_version") != 1:
+        raise ValueError("metadata snapshot schema_version must be 1")
+    if not isinstance(data.get("objects"), list):
+        raise ValueError("metadata snapshot objects must be a list")
+    return data
+
+
+def object_to_markdown(obj: dict) -> str:
+    kind = obj.get("kind") or "other"
+    title = KIND_TITLES.get(kind, "Объект")
+    name = obj.get("name") or "БезИмени"
+    synonym = obj.get("synonym")
+    description = obj.get("description")
+
+    lines = [f"## {title}: {name}", ""]
+    if synonym:
+        lines.extend([f"Синоним: {synonym}", ""])
+    if description:
+        lines.extend([description, ""])
+
+    attributes = obj.get("attributes") or []
+    if attributes:
+        lines.extend(["### Реквизиты", ""])
+        for attr in attributes:
+            attr_name = attr.get("name") or "БезИмени"
+            attr_type = attr.get("type") or "не указан"
+            attr_synonym = attr.get("synonym")
+            suffix = f" ({attr_synonym})" if attr_synonym else ""
+            lines.append(f"- {attr_name}{suffix}: {attr_type}")
+        lines.append("")
+
+    tabular_sections = obj.get("tabular_sections") or []
+    if tabular_sections:
+        lines.extend(["### Табличные части", ""])
+        for section in tabular_sections:
+            section_name = section.get("name") or "БезИмени"
+            section_synonym = section.get("synonym")
+            suffix = f" ({section_synonym})" if section_synonym else ""
+            lines.append(f"- {section_name}{suffix}")
+            for attr in section.get("attributes") or []:
+                attr_name = attr.get("name") or "БезИмени"
+                attr_type = attr.get("type") or "не указан"
+                lines.append(f"  - {attr_name}: {attr_type}")
+        lines.append("")
+
+    return "\n".join(lines).strip()
+
+
+def snapshot_to_markdown(snapshot: dict) -> str:
+    source = snapshot.get("source") or {}
+    source_name = source.get("name") or "unknown"
+    created_at = snapshot.get("created_at") or "unknown"
+
+    sections = [
+        "# 1C Metadata Snapshot",
+        "",
+        f"Источник: {source_name}",
+        f"Дата снимка: {created_at}",
+        "",
+        "Этот файл сгенерирован из metadata snapshot и предназначен для RAG-поиска.",
+        "",
+    ]
+
+    for obj in snapshot.get("objects") or []:
+        sections.append(object_to_markdown(obj))
+        sections.append("")
+
+    return "\n".join(sections).strip() + "\n"
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser(description="Convert a 1C metadata snapshot to a Markdown RAG source.")
+    parser.add_argument("--input", type=Path, default=DEFAULT_INPUT)
+    parser.add_argument("--output", type=Path, default=DEFAULT_OUTPUT)
+    args = parser.parse_args()
+
+    snapshot = load_snapshot(args.input)
+    markdown = snapshot_to_markdown(snapshot)
+    args.output.parent.mkdir(parents=True, exist_ok=True)
+    args.output.write_text(markdown, encoding="utf-8")
+    print(f"Wrote RAG source to {args.output}")
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())