Update streamlit_app.py

streamlit_app.py

@@ -138,6 +138,13 @@ def normalize_summary_payload(summary: dict) -> dict:
     return normalized
 
 
+def is_valid_url(value: object) -> bool:
+    if not isinstance(value, str):
+        return False
+
+    return value.startswith(("http://", "https://"))
+
+
 def apply_filters(df: pd.DataFrame) -> pd.DataFrame:
     st.sidebar.header("Filters")
 
@@ -231,25 +238,52 @@ def render_bullet_list(items: list[str], empty_message: str) -> None:
         st.markdown(f"- {item}")
 
 
+def get_summary_source_articles(
+    df: pd.DataFrame,
+    summary: dict,
+    fallback_limit: int = 15,
+) -> pd.DataFrame:
+    stories = summary.get("top_stories", [])
+
+    story_ids = {
+        str(story.get("article_id"))
+        for story in stories
+        if isinstance(story, dict) and story.get("article_id")
+    }
+
+    if story_ids:
+        matched_df = df[df["article_id"].astype(str).isin(story_ids)]
+
+        if not matched_df.empty:
+            return matched_df
+
+    relevant_df = df[df["label"] != "not relevant to field"].copy()
+
+    if "published_at" in relevant_df:
+        relevant_df = relevant_df.sort_values("published_at", ascending=False)
+
+    return relevant_df.head(fallback_limit)
+
+
 def render_daily_summary_source_basis(
     df: pd.DataFrame,
     summary: dict,
 ) -> pd.DataFrame:
-    summary_date = summary.get("summary_date")
+    summary_df = get_summary_source_articles(df, summary)
 
-    if summary_date and "published_day" in df:
-        summary_df = df[df["published_day"] == summary_date]
-    else:
-        summary_df = df
+    generated_at = summary.get("generated_at")
+    top_story_count = len(summary.get("top_stories", []))
 
-    if summary_date:
+    if generated_at:
         st.caption(
-            f"Summary is based on {len(summary_df)} classified articles "
-            f"published on {summary_date}."
+            f"Summary is based on the latest relevant classified articles available "
+            f"when it was generated at {generated_at}. "
+            f"{top_story_count} top stories are shown in the summary."
         )
     else:
         st.caption(
-            f"Summary is based on {len(summary_df)} classified articles."
+            "Summary is based on the latest relevant classified articles available "
+            f"when it was generated. {top_story_count} top stories are shown in the summary."
         )
 
     return summary_df
@@ -345,7 +379,7 @@ def render_daily_summary(summary: dict) -> None:
                 st.markdown("**Decision relevance**")
                 st.write(decision_relevance)
 
-            if url:
+            if is_valid_url(url):
                 st.link_button("Open article", url)
 
             if article_id:
@@ -411,8 +445,8 @@ def render_article_browser(df: pd.DataFrame) -> None:
                 st.write(description)
 
             url = row.get("url")
-            if pd.notnull(url) and str(url).strip():
-                st.link_button("Open article", str(url))
+            if is_valid_url(url):
+                st.link_button("Open article", url)
 
             st.markdown("**More details**")
 
@@ -432,10 +466,6 @@ def main() -> None:
         "with filters for action categories, dates, sources, and search terms."
     )
 
-    if st.sidebar.button("Refresh data"):
-        st.cache_data.clear()
-        st.rerun()
-
     df = load_classified_articles()
     summary = load_daily_summary()