Initial commit

us2kr/summarize.py

@@ -0,0 +1,30 @@
+"""Glue between the pipeline and the LLM layer."""
+
+from __future__ import annotations
+
+import logging
+
+from us2kr.config import Settings
+from us2kr.llm.base import LLMProvider
+from us2kr.llm.minimax import MiniMaxProvider
+from us2kr.models import MarketPayload, Report
+
+log = logging.getLogger(__name__)
+
+
+def build_provider(settings: Settings) -> LLMProvider:
+    """Construct the configured LLM provider (MiniMax only for now)."""
+    return MiniMaxProvider(settings)
+
+
+def summarize_payload(payload: MarketPayload, provider: LLMProvider) -> tuple[Report, str]:
+    """Summarise via the provider and guarantee `report.data_gaps` includes every payload gap."""
+    report, model_used = provider.summarize(payload)
+    merged: list[str] = []
+    for gap in [*payload.data_gaps, *report.data_gaps]:
+        if gap and gap not in merged:
+            merged.append(gap)
+    if merged != list(report.data_gaps):
+        report = report.model_copy(update={"data_gaps": merged})
+    log.info("summary produced by %s (%d data gaps)", model_used, len(merged))
+    return report, model_used