refactor: update execution history API and tests

Author: shuoweil

bigframes/session/__init__.py

@@ -109,6 +109,39 @@
 logger = logging.getLogger(__name__)
 
 
+class _ExecutionHistory(pandas.DataFrame):
+    @property
+    def _constructor(self):
+        return _ExecutionHistory
+
+    def _repr_html_(self) -> str | None:
+        try:
+            import bigframes.formatting_helpers as formatter
+
+            if self.empty:
+                return "<div>No executions found.</div>"
+
+            cols = ["job_id", "status", "total_bytes_processed", "job_url"]
+            df_display = self[cols].copy()
+            df_display["total_bytes_processed"] = df_display[
+                "total_bytes_processed"
+            ].apply(formatter.get_formatted_bytes)
+
+            def format_url(url):
+                return f'<a target="_blank" href="{url}">Open Job</a>' if url else ""
+
+            df_display["job_url"] = df_display["job_url"].apply(format_url)
+
+            # Rename job_id to query_id to match user expectations
+            df_display = df_display.rename(columns={"job_id": "query_id"})
+
+            compact_html = df_display.to_html(escape=False, index=False)
+
+            return compact_html
+        except Exception:
+            return super()._repr_html_()  # type: ignore
+
+
 @log_adapter.class_logger
 class Session(
     third_party_pandas_gbq.GBQIOMixin,
@@ -373,7 +406,7 @@ def slot_millis_sum(self):
 
     def execution_history(self) -> pandas.DataFrame:
         """Returns a list of underlying BigQuery executions initiated by BigFrames in the current session."""
-        return pandas.DataFrame([job.__dict__ for job in self._metrics.jobs])
+        return _ExecutionHistory([job.__dict__ for job in self._metrics.jobs])
 
     @property
     def _allows_ambiguity(self) -> bool:

bigframes/session/metrics.py

@@ -60,6 +60,11 @@ def from_job(
         if query_text and len(query_text) > 1024:
             query_text = query_text[:1021] + "..."
 
+        job_id = getattr(query_job, "job_id", None)
+        job_url = None
+        if job_id:
+            job_url = f"https://console.cloud.google.com/bigquery?project={query_job.project}&j=bq:{query_job.location}:{job_id}&page=queryresults"
+
         metadata = cls(
             job_id=query_job.job_id,
             location=query_job.location,
@@ -72,9 +77,7 @@ def from_job(
             job_type=query_job.job_type,
             error_result=query_job.error_result,
             query=query_text,
-            job_url=f"https://console.cloud.google.com/bigquery?project={query_job.project}&j=bq:{query_job.location}:{query_job.job_id}&page=queryresults"
-            if getattr(query_job, "job_id", None)
-            else None,
+            job_url=job_url,
         )
         if isinstance(query_job, QueryJob):
             metadata.cached = getattr(query_job, "cache_hit", None)
@@ -111,8 +114,15 @@ def from_row_iterator(
         if query_text and len(query_text) > 1024:
             query_text = query_text[:1021] + "..."
 
+        job_id = getattr(row_iterator, "job_id", None)
+        job_url = None
+        if job_id:
+            project = getattr(row_iterator, "project", "")
+            location = getattr(row_iterator, "location", "")
+            job_url = f"https://console.cloud.google.com/bigquery?project={project}&j=bq:{location}:{job_id}&page=queryresults"
+
         return cls(
-            job_id=getattr(row_iterator, "job_id", None),
+            job_id=job_id,
             query_id=getattr(row_iterator, "query_id", None),
             location=getattr(row_iterator, "location", None),
             project=getattr(row_iterator, "project", None),
@@ -126,9 +136,7 @@ def from_row_iterator(
             job_type="query",
             cached=getattr(row_iterator, "cache_hit", None),
             query=query_text,
-            job_url=f"https://console.cloud.google.com/bigquery?project={getattr(row_iterator, 'project', '')}&j=bq:{getattr(row_iterator, 'location', '')}:{getattr(row_iterator, 'job_id', '')}&page=queryresults"
-            if getattr(row_iterator, "job_id", None)
-            else None,
+            job_url=job_url,
         )