Fix scheduled news coverage and run date reporting (#10)

tests/test_report_localization.py

@@ -10,6 +10,8 @@ from tradingagents.reporting import save_report_bundle
 class ReportLocalizationTests(unittest.TestCase):
     def test_save_report_bundle_uses_korean_labels(self):
         final_state = {
+            "analysis_date": "2026-04-06",
+            "trade_date": "2026-04-02",
             "market_report": "시장 보고서 본문",
             "sentiment_report": "소셜 보고서 본문",
             "news_report": "뉴스 보고서 본문",
@@ -39,6 +41,8 @@ class ReportLocalizationTests(unittest.TestCase):
 
         self.assertIn("트레이딩 분석 리포트", report_text)
         self.assertIn("생성 시각", report_text)
+        self.assertIn("분석 기준일: 2026-04-06", report_text)
+        self.assertIn("시장 데이터 기준일: 2026-04-02", report_text)
         self.assertIn("애널리스트 팀 리포트", report_text)
         self.assertIn("포트폴리오 매니저 최종 판단", report_text)
         self.assertIn("시장 애널리스트", report_text)

tests/test_scheduled_analysis.py

@@ -24,13 +24,14 @@ class _FakeTradingAgentsGraph:
         self.config = config or {}
         self.callbacks = callbacks or []
 
-    def propagate(self, ticker, trade_date):
+    def propagate(self, ticker, trade_date, analysis_date=None):
         if ticker == "FAIL":
             raise RuntimeError("synthetic failure")
 
         final_state = {
             "company_of_interest": ticker,
             "trade_date": trade_date,
+            "analysis_date": analysis_date or trade_date,
             "market_report": f"## Market\n{ticker} market analysis",
             "sentiment_report": f"## Sentiment\n{ticker} sentiment analysis",
             "news_report": f"## News\n{ticker} news analysis",
@@ -104,6 +105,7 @@ subtitle = "Automated"
             self.assertEqual(manifest["settings"]["provider"], "codex")
             self.assertEqual(manifest["settings"]["deep_model"], "gpt-5.4")
             self.assertEqual(manifest["settings"]["quick_model"], "gpt-5.4")
+            self.assertEqual(manifest["tickers"][0]["analysis_date"], manifest["started_at"][:10])
 
             run_dir = archive_dir / "runs" / manifest["started_at"][:4] / manifest["run_id"]
             self.assertTrue((run_dir / "run.json").exists())
@@ -118,6 +120,7 @@ subtitle = "Automated"
             self.assertIn("partial failure", index_html)
             self.assertIn("NVDA", run_html)
             self.assertIn("Rendered report", ticker_html)
+            self.assertIn("Analysis date", ticker_html)
             self.assertTrue((site_dir / "downloads" / manifest["run_id"] / "NVDA" / "complete_report.md").exists())
 
     def test_main_site_only_rebuilds_from_existing_archive(self):
@@ -162,6 +165,7 @@ subtitle = "Automated"
                             {
                                 "ticker": "NVDA",
                                 "status": "success",
+                                "analysis_date": "2026-04-05",
                                 "trade_date": "2026-04-04",
                                 "decision": "BUY",
                                 "started_at": "2026-04-05T09:13:00+09:00",

tests/test_yfinance_news.py

@@ -0,0 +1,68 @@
+import unittest
+from unittest.mock import patch
+
+from tradingagents.dataflows.yfinance_news import get_news_yfinance
+
+
+def _article(date_value: str, title: str, link: str) -> dict:
+    return {
+        "content": {
+            "title": title,
+            "summary": f"Summary for {title}",
+            "provider": {"displayName": "Unit Test"},
+            "canonicalUrl": {"url": link},
+            "pubDate": f"{date_value}T12:00:00Z",
+        }
+    }
+
+
+class _FakeTicker:
+    def __init__(self, full_news: list[dict]):
+        self.full_news = list(full_news)
+
+    def get_news(self, count=20):
+        return self.full_news[:count]
+
+
+class YFinanceNewsTests(unittest.TestCase):
+    def test_get_news_yfinance_expands_feed_depth_to_cover_requested_window(self):
+        recent_articles = [
+            _article(f"2026-04-{day:02d}", f"Recent article {day}", f"https://example.com/recent-{day}")
+            for day in range(6, 2, -1)
+            for _ in range(15)
+        ]
+        older_articles = [
+            _article("2026-04-02", "Alphabet April 2 article", "https://example.com/apr2"),
+            _article("2026-04-01", "Alphabet April 1 article", "https://example.com/apr1"),
+        ]
+        fake_ticker = _FakeTicker(recent_articles + older_articles)
+
+        with (
+            patch("tradingagents.dataflows.yfinance_news.yf.Ticker", return_value=fake_ticker),
+            patch("tradingagents.dataflows.yfinance_news.yf_retry", side_effect=lambda fn: fn()),
+        ):
+            result = get_news_yfinance("GOOGL", "2026-03-26", "2026-04-02")
+
+        self.assertIn("Alphabet April 2 article", result)
+        self.assertIn("[2026-04-02]", result)
+
+    def test_get_news_yfinance_reports_feed_coverage_when_window_is_unavailable(self):
+        fake_ticker = _FakeTicker(
+            [
+                _article("2026-04-06", "Fresh article", "https://example.com/fresh"),
+                _article("2026-04-05", "Fresh article 2", "https://example.com/fresh-2"),
+            ]
+        )
+
+        with (
+            patch("tradingagents.dataflows.yfinance_news.yf.Ticker", return_value=fake_ticker),
+            patch("tradingagents.dataflows.yfinance_news.yf_retry", side_effect=lambda fn: fn()),
+        ):
+            result = get_news_yfinance("GOOGL", "2026-03-26", "2026-04-02")
+
+        self.assertIn("No news found for GOOGL between 2026-03-26 and 2026-04-02", result)
+        self.assertIn("2026-04-05 to 2026-04-06", result)
+
+
+if __name__ == "__main__":
+    unittest.main()

tradingagents/agents/analysts/news_analyst.py

@@ -10,7 +10,7 @@ from tradingagents.dataflows.config import get_config
 
 def create_news_analyst(llm):
     def news_analyst_node(state):
-        current_date = state["trade_date"]
+        current_date = state.get("analysis_date") or state["trade_date"]
         instrument_context = build_instrument_context(state["company_of_interest"])
 
         tools = [

tradingagents/agents/analysts/social_media_analyst.py

@@ -5,7 +5,7 @@ from tradingagents.dataflows.config import get_config
 
 def create_social_media_analyst(llm):
     def social_media_analyst_node(state):
-        current_date = state["trade_date"]
+        current_date = state.get("analysis_date") or state["trade_date"]
         instrument_context = build_instrument_context(state["company_of_interest"])
 
         tools = [

tradingagents/agents/utils/agent_states.py

@@ -46,6 +46,7 @@ class RiskDebateState(TypedDict):
 class AgentState(MessagesState):
     company_of_interest: Annotated[str, "Company that we are interested in trading"]
     trade_date: Annotated[str, "What date we are trading at"]
+    analysis_date: Annotated[str, "What date the full analysis is being generated on"]
 
     sender: Annotated[str, "Agent that sent this message"]
 

tradingagents/dataflows/yfinance_news.py

@@ -1,12 +1,46 @@
 """yfinance-based news data fetching functions."""
 
-import yfinance as yf
-from datetime import datetime
+from datetime import datetime, timezone
+
 from dateutil.relativedelta import relativedelta
+import yfinance as yf
 
 from .stockstats_utils import yf_retry
 
 
+_TICKER_NEWS_FETCH_COUNTS = (20, 50, 100)
+_MAX_FILTERED_TICKER_ARTICLES = 25
+
+
+def _parse_pub_date(raw_value) -> datetime | None:
+    """Normalize yfinance pub date values into a timezone-aware datetime."""
+    if raw_value in (None, ""):
+        return None
+
+    if isinstance(raw_value, datetime):
+        return raw_value
+
+    if isinstance(raw_value, (int, float)):
+        try:
+            return datetime.fromtimestamp(raw_value, tz=timezone.utc)
+        except (OverflowError, OSError, ValueError):
+            return None
+
+    if isinstance(raw_value, str):
+        normalized = raw_value.strip()
+        if not normalized:
+            return None
+        try:
+            return datetime.fromisoformat(normalized.replace("Z", "+00:00"))
+        except ValueError:
+            try:
+                return datetime.fromtimestamp(float(normalized), tz=timezone.utc)
+            except (OverflowError, OSError, ValueError):
+                return None
+
+    return None
+
+
 def _extract_article_data(article: dict) -> dict:
     """Extract article data from yfinance news format (handles nested 'content' structure)."""
     # Handle nested content structure
@@ -22,13 +56,7 @@ def _extract_article_data(article: dict) -> dict:
         link = url_obj.get("url", "")
 
         # Get publish date
-        pub_date_str = content.get("pubDate", "")
-        pub_date = None
-        if pub_date_str:
-            try:
-                pub_date = datetime.fromisoformat(pub_date_str.replace("Z", "+00:00"))
-            except (ValueError, AttributeError):
-                pass
+        pub_date = _parse_pub_date(content.get("pubDate", ""))
 
         return {
             "title": title,
@@ -44,10 +72,79 @@ def _extract_article_data(article: dict) -> dict:
             "summary": article.get("summary", ""),
             "publisher": article.get("publisher", "Unknown"),
             "link": article.get("link", ""),
-            "pub_date": None,
+            "pub_date": _parse_pub_date(article.get("providerPublishTime")),
         }
 
 
+def _article_identity(article: dict) -> str:
+    """Return a stable identity key for deduplicating news articles."""
+    link = article.get("link", "").strip()
+    if link:
+        return link
+
+    title = article.get("title", "").strip()
+    publisher = article.get("publisher", "").strip()
+    pub_date = article.get("pub_date")
+    stamp = pub_date.isoformat() if isinstance(pub_date, datetime) else ""
+    return f"{publisher}::{title}::{stamp}"
+
+
+def _collect_ticker_news(
+    ticker: str,
+    start_dt: datetime,
+) -> tuple[list[dict], datetime | None, datetime | None]:
+    """Fetch increasingly larger ticker feeds until the requested window is covered."""
+    collected: list[dict] = []
+    seen: set[str] = set()
+    oldest_pub_date = None
+    newest_pub_date = None
+
+    for count in _TICKER_NEWS_FETCH_COUNTS:
+        news = yf_retry(lambda batch_size=count: yf.Ticker(ticker).get_news(count=batch_size))
+        if not news:
+            continue
+
+        for article in news:
+            data = _extract_article_data(article)
+            identity = _article_identity(data)
+            if identity in seen:
+                continue
+            seen.add(identity)
+            collected.append(data)
+
+            pub_date = data.get("pub_date")
+            if pub_date:
+                if newest_pub_date is None or pub_date > newest_pub_date:
+                    newest_pub_date = pub_date
+                if oldest_pub_date is None or pub_date < oldest_pub_date:
+                    oldest_pub_date = pub_date
+
+        if oldest_pub_date and oldest_pub_date.replace(tzinfo=None) <= start_dt:
+            break
+        if len(news) < count:
+            break
+
+    collected.sort(
+        key=lambda article: article["pub_date"].timestamp() if article.get("pub_date") else float("-inf"),
+        reverse=True,
+    )
+    return collected, oldest_pub_date, newest_pub_date
+
+
+def _format_coverage_note(oldest_pub_date: datetime | None, newest_pub_date: datetime | None) -> str:
+    """Describe the yfinance coverage window when no article matches the requested range."""
+    if oldest_pub_date and newest_pub_date:
+        return (
+            "; the current yfinance ticker feed only covered "
+            f"{oldest_pub_date.strftime('%Y-%m-%d')} to {newest_pub_date.strftime('%Y-%m-%d')} at query time"
+        )
+    if oldest_pub_date:
+        return f"; the current yfinance ticker feed only reached back to {oldest_pub_date.strftime('%Y-%m-%d')}"
+    if newest_pub_date:
+        return f"; the current yfinance ticker feed only returned articles up to {newest_pub_date.strftime('%Y-%m-%d')}"
+    return ""
+
+
 def get_news_yfinance(
     ticker: str,
     start_date: str,
@@ -65,38 +162,40 @@ def get_news_yfinance(
         Formatted string containing news articles
     """
     try:
-        stock = yf.Ticker(ticker)
-        news = yf_retry(lambda: stock.get_news(count=20))
-
-        if not news:
-            return f"No news found for {ticker}"
-
-        # Parse date range for filtering
         start_dt = datetime.strptime(start_date, "%Y-%m-%d")
         end_dt = datetime.strptime(end_date, "%Y-%m-%d")
+        articles, oldest_pub_date, newest_pub_date = _collect_ticker_news(ticker, start_dt)
+
+        if not articles:
+            return f"No news found for {ticker}"
 
         news_str = ""
         filtered_count = 0
 
-        for article in news:
-            data = _extract_article_data(article)
-
+        for data in articles:
             # Filter by date if publish time is available
             if data["pub_date"]:
                 pub_date_naive = data["pub_date"].replace(tzinfo=None)
                 if not (start_dt <= pub_date_naive <= end_dt + relativedelta(days=1)):
                     continue
 
-            news_str += f"### {data['title']} (source: {data['publisher']})\n"
+            date_prefix = ""
+            if data["pub_date"]:
+                date_prefix = f"[{data['pub_date'].strftime('%Y-%m-%d')}] "
+
+            news_str += f"### {date_prefix}{data['title']} (source: {data['publisher']})\n"
             if data["summary"]:
                 news_str += f"{data['summary']}\n"
             if data["link"]:
                 news_str += f"Link: {data['link']}\n"
             news_str += "\n"
             filtered_count += 1
+            if filtered_count >= _MAX_FILTERED_TICKER_ARTICLES:
+                break
 
         if filtered_count == 0:
-            return f"No news found for {ticker} between {start_date} and {end_date}"
+            coverage_note = _format_coverage_note(oldest_pub_date, newest_pub_date)
+            return f"No news found for {ticker} between {start_date} and {end_date}{coverage_note}"
 
         return f"## {ticker} News, from {start_date} to {end_date}:\n\n{news_str}"
 

tradingagents/graph/propagation.py

@@ -16,13 +16,17 @@ class Propagator:
         self.max_recur_limit = max_recur_limit
 
     def create_initial_state(
-        self, company_name: str, trade_date: str
+        self,
+        company_name: str,
+        trade_date: str,
+        analysis_date: str | None = None,
     ) -> Dict[str, Any]:
         """Create the initial state for the agent graph."""
         return {
             "messages": [("human", company_name)],
             "company_of_interest": company_name,
             "trade_date": str(trade_date),
+            "analysis_date": str(analysis_date or trade_date),
             "investment_debate_state": InvestDebateState(
                 {
                     "bull_history": "",

tradingagents/graph/trading_graph.py

@@ -202,14 +202,14 @@ class TradingAgentsGraph:
             ),
         }
 
-    def propagate(self, company_name, trade_date):
+    def propagate(self, company_name, trade_date, analysis_date=None):
         """Run the trading agents graph for a company on a specific date."""
 
         self.ticker = company_name
 
         # Initialize state
         init_agent_state = self.propagator.create_initial_state(
-            company_name, trade_date
+            company_name, trade_date, analysis_date=analysis_date
         )
         args = self.propagator.get_graph_args()
 
@@ -245,6 +245,7 @@ class TradingAgentsGraph:
         self.log_states_dict[str(trade_date)] = {
             "company_of_interest": final_state["company_of_interest"],
             "trade_date": final_state["trade_date"],
+            "analysis_date": final_state.get("analysis_date", final_state["trade_date"]),
             "market_report": final_state["market_report"],
             "sentiment_report": final_state["sentiment_report"],
             "news_report": final_state["news_report"],

tradingagents/reporting.py

@@ -19,6 +19,8 @@ def save_report_bundle(
     save_path = Path(save_path)
     save_path.mkdir(parents=True, exist_ok=True)
     labels = _labels_for(language)
+    analysis_date = _coerce_text(final_state.get("analysis_date"))
+    trade_date = _coerce_text(final_state.get("trade_date"))
 
     sections: list[str] = []
 
@@ -104,10 +106,13 @@ def save_report_bundle(
             f"### {labels['portfolio_manager']}\n{portfolio_decision}"
         )
 
-    header = (
-        f"# {labels['report_title']}: {ticker}\n\n"
-        f"{labels['generated_at']}: {generated_at.strftime('%Y-%m-%d %H:%M:%S')}\n\n"
-    )
+    metadata_lines = [f"{labels['generated_at']}: {generated_at.strftime('%Y-%m-%d %H:%M:%S')}"]
+    if analysis_date:
+        metadata_lines.append(f"{labels['analysis_date']}: {analysis_date}")
+    if trade_date:
+        metadata_lines.append(f"{labels['trade_date']}: {trade_date}")
+
+    header = f"# {labels['report_title']}: {ticker}\n\n" + "\n".join(metadata_lines) + "\n\n"
     complete_report = save_path / "complete_report.md"
     _write_text(complete_report, header + "\n\n".join(sections))
     return complete_report
@@ -132,6 +137,8 @@ def _labels_for(language: str) -> dict[str, str]:
         return {
             "report_title": "트레이딩 분석 리포트",
             "generated_at": "생성 시각",
+            "analysis_date": "분석 기준일",
+            "trade_date": "시장 데이터 기준일",
             "section_analysts": "I. 애널리스트 팀 리포트",
             "section_research": "II. 리서치 팀 판단",
             "section_trading": "III. 트레이딩 팀 계획",
@@ -154,6 +161,8 @@ def _labels_for(language: str) -> dict[str, str]:
     return {
         "report_title": "Trading Analysis Report",
         "generated_at": "Generated",
+        "analysis_date": "Analysis date",
+        "trade_date": "Market data date",
         "section_analysts": "I. Analyst Team Reports",
         "section_research": "II. Research Team Decision",
         "section_trading": "III. Trading Team Plan",

tradingagents/scheduled/runner.py

@@ -159,6 +159,7 @@ def _run_single_ticker(
 
     ticker_started = datetime.now(ZoneInfo(config.run.timezone))
     timer_start = perf_counter()
+    analysis_date = ticker_started.date().isoformat()
 
     try:
         trade_date = resolve_trade_date(ticker, config)
@@ -169,7 +170,11 @@ def _run_single_ticker(
             config=_graph_config(config, engine_results_dir),
             callbacks=[stats_handler],
         )
-        final_state, decision = graph.propagate(ticker, trade_date)
+        final_state, decision = graph.propagate(
+            ticker,
+            trade_date,
+            analysis_date=analysis_date,
+        )
 
         report_dir = ticker_dir / "report"
         report_file = save_report_bundle(
@@ -198,6 +203,7 @@ def _run_single_ticker(
             "ticker": ticker,
             "status": "success",
             "trade_date": trade_date,
+            "analysis_date": analysis_date,
             "decision": str(decision),
             "started_at": ticker_started.isoformat(),
             "finished_at": datetime.now(ZoneInfo(config.run.timezone)).isoformat(),
@@ -216,6 +222,7 @@ def _run_single_ticker(
             "ticker": ticker,
             "status": "success",
             "trade_date": trade_date,
+            "analysis_date": analysis_date,
             "decision": str(decision),
             "started_at": ticker_started.isoformat(),
             "finished_at": analysis_payload["finished_at"],
@@ -232,6 +239,7 @@ def _run_single_ticker(
         error_payload = {
             "ticker": ticker,
             "status": "failed",
+            "analysis_date": analysis_date,
             "error": str(exc),
             "traceback": traceback.format_exc(),
             "started_at": ticker_started.isoformat(),
@@ -244,6 +252,7 @@ def _run_single_ticker(
         return {
             "ticker": ticker,
             "status": "failed",
+            "analysis_date": analysis_date,
             "trade_date": None,
             "decision": None,
             "error": str(exc),
@@ -285,6 +294,7 @@ def _serialize_final_state(final_state: dict[str, Any]) -> dict[str, Any]:
     return {
         "company_of_interest": final_state.get("company_of_interest"),
         "trade_date": final_state.get("trade_date"),
+        "analysis_date": final_state.get("analysis_date"),
         "market_report": final_state.get("market_report"),
         "sentiment_report": final_state.get("sentiment_report"),
         "news_report": final_state.get("news_report"),

tradingagents/scheduled/site.py

@@ -163,6 +163,7 @@ def _render_run_page(manifest: dict[str, Any], settings: SiteSettings) -> str:
                 <a href="{_escape(ticker_summary['ticker'])}.html">{_escape(ticker_summary['ticker'])}</a>
                 <span class="status {ticker_summary['status']}">{_escape(ticker_summary['status'])}</span>
               </div>
+              <p><strong>Analysis date</strong><span>{_escape(ticker_summary.get('analysis_date') or '-')}</span></p>
               <p><strong>Trade date</strong><span>{_escape(ticker_summary.get('trade_date') or '-')}</span></p>
               <p><strong>Duration</strong><span>{ticker_summary.get('duration_seconds', 0):.1f}s</span></p>
               <p><strong>Decision</strong><span>{_escape(ticker_summary.get('decision') or ticker_summary.get('error') or '-')}</span></p>
@@ -239,10 +240,12 @@ def _render_ticker_page(
       <div>
         <p class="eyebrow">Ticker report</p>
         <h1>{_escape(ticker_summary['ticker'])}</h1>
-        <p class="subtitle">{_escape(ticker_summary.get('trade_date') or '-')} / {_escape(ticker_summary['status'])}</p>
+        <p class="subtitle">Analysis {_escape(ticker_summary.get('analysis_date') or '-')} / Market {_escape(ticker_summary.get('trade_date') or '-')} / {_escape(ticker_summary['status'])}</p>
       </div>
       <div class="hero-card">
         <div class="status {ticker_summary['status']}">{_escape(ticker_summary['status'])}</div>
+        <p><strong>Analysis date</strong><span>{_escape(ticker_summary.get('analysis_date') or '-')}</span></p>
+        <p><strong>Trade date</strong><span>{_escape(ticker_summary.get('trade_date') or '-')}</span></p>
         <p><strong>Decision</strong><span>{_escape(ticker_summary.get('decision') or '-')}</span></p>
         <p><strong>Duration</strong><span>{ticker_summary.get('duration_seconds', 0):.1f}s</span></p>
         <p><strong>LLM calls</strong><span>{ticker_summary.get('metrics', {}).get('llm_calls', 0)}</span></p>