merge: chore/cleanup-remove-bloat-and-secrets into main

app/domain/labels/sources/etherscan_csv.py

@@ -0,0 +1,70 @@
+"""Etherscan combined labels CSV source — 51K EVM labels via DuckDB."""
+from __future__ import annotations
+
+import asyncio
+import logging
+from datetime import UTC, datetime
+
+from app.domain.labels.models import Label, LabelSource
+
+log = logging.getLogger(__name__)
+
+CSV_PATH = "/home/dev/rmi/backend/data/etherscan_combined_labels.csv"
+
+
+async def query_etherscan_csv(address: str, chain: str = "ethereum") -> list[Label]:
+    """Query the Etherscan combined labels CSV via DuckDB.
+
+    Schema:
+      address,chain,source,name,label_type,label_subtype,project,entity_type,entity_types
+
+    Returns:
+        List of labels (usually 1 per matching row).
+    """
+    def _query() -> list[Label]:
+        from app.core.duckdb_analytics import DuckDBAnalytics
+
+        try:
+            d = DuckDBAnalytics()
+            try:
+                rows = d.query_parquet(
+                    CSV_PATH,
+                    """
+                    SELECT address, chain, source, name, label_type, label_subtype, project, entity_type
+                    FROM data
+                    WHERE LOWER(address) = ?
+                    LIMIT 20
+                    """,
+                    [address.lower()],
+                )
+                labels = []
+                for row in rows:
+                    label_value = row.get("name") or row.get("label_type") or row.get("source")
+                    if not label_value:
+                        continue
+                    chain_name = row.get("chain", chain).lower()
+                    labels.append(
+                        Label(
+                            address=row["address"].lower(),
+                            chain=chain_name,
+                            label_type="entity",
+                            label_value=label_value,
+                            source=LabelSource.ETHERSCAN_CSV,
+                            confidence=0.85,
+                            verified_at=datetime.now(UTC),
+                            attributes={
+                                "raw_label_type": row.get("label_type"),
+                                "label_subtype": row.get("label_subtype"),
+                                "project": row.get("project"),
+                                "entity_type": row.get("entity_type"),
+                            },
+                        )
+                    )
+                return labels
+            finally:
+                d.close()
+        except Exception as e:
+            log.warning("etherscan_csv_query_fail address=%s err=%s", address[:10], e)
+            raise
+
+    return await asyncio.to_thread(_query)