Add performance profiling and dataset title support

Introduces performance profiling utilities and integrates them into key data loading and map rendering functions in dash_app.py. Adds caching for datasets and map figures to improve efficiency. Enhances the UI to support an optional dataset title field and propagates this through callbacks and dataset publishing. Updates CSS for improved layout and panel scrolling. Adds a new RCH file and performance log files.

Author: wmobley

dash/assets/style.css

@@ -8,12 +8,16 @@ body {
 .app-root {
   min-height: 100vh;
   background: #f4f3ef;
+  display: flex;
+  flex-direction: column;
 }
 
 .app-shell {
   display: flex;
   gap: 16px;
   padding: 16px;
+  flex: 1;
+  align-items: stretch;
 }
 
 .header-bar {
@@ -108,6 +112,8 @@ body {
   border-radius: 12px;
   padding: 16px;
   box-shadow: 0 8px 20px rgba(0, 0, 0, 0.06);
+  max-height: calc(100vh - 140px);
+  overflow-y: auto;
 }
 
 .panel h2 {

dash/dash_app.py

@@ -39,6 +39,7 @@
     get_wel_period_keys,
     load_wel,
 )
+from flopy_interactive.utils.perf import perf_call, perf_note, perf_timer
 from flopy_interactive.viz.color_modes import apply_color_mode
 
 
@@ -47,6 +48,9 @@
 SUGGEST_TITLE_FILTER = "Barton Springs Edwards Aquifer"
 CKAN_URL = os.environ.get("FLOPY_CKAN_URL", CKAN_BASE_URL)
 
+_DATASET_CACHE: Dict[str, Dict] = {}
+_MAP_FIG_CACHE: Dict[tuple, go.Figure] = {}
+
 
 def get_datasets() -> List[Dict]:
     """Fetch CKAN datasets for the app session.
@@ -57,7 +61,7 @@ def get_datasets() -> List[Dict]:
     Returns:
         List of dataset metadata dicts.
     """
-    return search_ckan_datasets()
+    return perf_call("search_ckan_datasets", search_ckan_datasets)
 
 
 def _get_dataset_or_none(name: str | None) -> Dict | None:
@@ -87,6 +91,10 @@ def load_dataset(name: str) -> Dict:
     Returns:
         Dict with dataset, gdf, wel, rch, lookup, and nlay.
     """
+    cached = _DATASET_CACHE.get(name)
+    if cached is not None:
+        perf_note(f"load_dataset cache hit: {name}")
+        return cached
     dataset = _get_dataset_or_none(name)
     if not dataset:
         raise ValueError(f"Dataset not found: {name}")
@@ -96,8 +104,8 @@ def load_dataset(name: str) -> Dict:
     rch_resource = dataset["matches"]["rch"][0]
     wel_path = download_ckan_resource(wel_resource, base_dir / "wel")
     rch_path = download_ckan_resource(rch_resource, base_dir / "rch")
-    gdf = load_grid_resource(grid_resource, base_dir / "grid")
-    wel = load_wel(wel_path)
+    gdf = perf_call(f"load_grid_resource:{name}", load_grid_resource, grid_resource, base_dir / "grid")
+    wel = perf_call(f"load_wel:{name}", load_wel, wel_path)
     nrow = int(gdf["ROW"].max())
     ncol = int(gdf["COL"].max())
     try:
@@ -107,25 +115,27 @@ def load_dataset(name: str) -> Dict:
             keys = list(spd.data.keys())
             if keys:
                 nper = int(max(keys)) + 1
-        rch = load_rch(rch_path, nrow=nrow, ncol=ncol, nper=nper)
+        rch = perf_call(f"load_rch:{name}", load_rch, rch_path, nrow=nrow, ncol=ncol, nper=nper)
     except Exception:
         rch = None
-    cell_id_lookup = build_cell_id_lookup(gdf, wel)
+    cell_id_lookup = perf_call(f"build_cell_id_lookup:{name}", build_cell_id_lookup, gdf, wel)
     nlay = 1
     if hasattr(wel, "parent") and wel.parent is not None:
         nlay = int(getattr(wel.parent.dis, "nlay", 1))
     elif hasattr(wel, "model") and wel.model is not None:
         nlay = int(getattr(wel.model.dis, "nlay", 1))
     elif hasattr(wel, "_model") and wel._model is not None:
         nlay = int(getattr(wel._model.dis, "nlay", 1))
-    return {
+    data = {
         "dataset": dataset,
         "gdf": gdf,
         "wel": wel,
         "rch": rch,
         "cell_id_lookup": cell_id_lookup,
         "nlay": nlay,
     }
+    _DATASET_CACHE[name] = data
+    return data
 
 
 def _collect_wel_cells_for_periods(
@@ -191,8 +201,10 @@ def _build_map_figure(
     """
     gdf_map = gdf[["CELL_ID", "ROW", "COL", "geometry"]].copy()
     gdf_valid = gdf_map[gdf_map["geometry"].notna()].copy()
-    gdf_choro = _downsample_for_choropleth(gdf_valid, gdf, zoom)
-    grid_geojson = gdf_choro.set_index("CELL_ID").__geo_interface__
+    with perf_timer("downsample_for_choropleth"):
+        gdf_choro = _downsample_for_choropleth(gdf_valid, gdf, zoom)
+    with perf_timer("build_geojson"):
+        grid_geojson = gdf_choro.set_index("CELL_ID").__geo_interface__
     center_lat = float(gdf["_lat"].median())
     center_lon = float(gdf["_lon"].median())
 
@@ -305,15 +317,16 @@ def _build_map_figure(
     )
 
     color_gdf = gdf if show_grid else gdf_scatter
-    apply_color_mode(
-        fig,
-        color_gdf,
-        cells,
-        color_by,
-        normalize=False,
-        flux_label=flux_label,
-        force_linear=force_linear,
-    )
+    with perf_timer("apply_color_mode"):
+        apply_color_mode(
+            fig,
+            color_gdf,
+            cells,
+            color_by,
+            normalize=False,
+            flux_label=flux_label,
+            force_linear=force_linear,
+        )
 
     selected_ids = {int(cid) for cid in selected_ids}
     if selected_ids and len(fig.data) > 2:
@@ -334,6 +347,18 @@ def _build_map_figure(
     return fig
 
 
+def _apply_selection(fig: go.Figure, gdf, selected_ids: Iterable[int]) -> None:
+    """Apply selected cell overlay to an existing figure."""
+    if not fig.data or len(fig.data) <= 2:
+        return
+    selected_set = {int(cid) for cid in (selected_ids or [])}
+    if not selected_set:
+        fig.data[2].update(lon=[], lat=[])
+        return
+    selected_rows = gdf[gdf["CELL_ID"].isin(selected_set)]
+    fig.data[2].update(lon=selected_rows["_lon"], lat=selected_rows["_lat"])
+
+
 def _dataset_options() -> List[Dict[str, str]]:
     """Build dataset dropdown options from CKAN search.
 
@@ -485,7 +510,7 @@ def _slugify(value: str) -> str:
         dcc.Store(id="update-store", data=0),
         dcc.Store(id="loaded-dataset", data=default_dataset),
         dcc.Store(id="load-counter", data=0),
-        dcc.Store(id="ckan-jwt", data=""),
+        dcc.Store(id="ckan-jwt", data=os.environ.get("FLOPY_CKAN_JWT", "").strip()),
         dcc.Store(id="login-message", data=""),
         dcc.Store(id="tapis-username", data=""),
         dcc.Store(id="name-seed", data=str(uuid.uuid4())),
@@ -628,6 +653,13 @@ def _slugify(value: str) -> str:
                             options=[{"label": "Add missing wells", "value": "yes"}],
                             value=[],
                         ),
+                        html.Label("Dataset title"),
+                        dcc.Input(
+                            id="dataset-title",
+                            type="text",
+                            placeholder="Optional display title",
+                            className="login-input",
+                        ),
                         html.Label("New dataset name"),
                         dcc.Dropdown(
                             id="dataset-suggestions",
@@ -856,6 +888,15 @@ def login_ckan(n_clicks, username, password):
     return token, f"Logged in as {username}", "status status-ok", username, True, True, "Logged in"
 
 
+@app.callback(
+    Output("apply-rate", "disabled"),
+    Input("ckan-jwt", "data"),
+)
+def toggle_apply_rate(jwt_token):
+    """Enable Apply + Save only when authenticated."""
+    return not bool(jwt_token)
+
+
 @app.callback(
     Output("category-select", "options"),
     Output("category-select", "value"),
@@ -927,6 +968,7 @@ def update_dataset_suggestions(username, jwt_token):
     Output("output-wel", "value"),
     Output("source-url", "value"),
     Output("change-summary", "value"),
+    Output("dataset-title", "value"),
     Input("loaded-dataset", "data"),
     Input("flux-source", "value"),
     Input("rate-mode", "value"),
@@ -947,6 +989,7 @@ def update_dataset_suggestions(username, jwt_token):
     State("output-wel", "value"),
     State("source-url", "value"),
     State("change-summary", "value"),
+    State("dataset-title", "value"),
 )
 def suggest_names(
     loaded_dataset,
@@ -969,6 +1012,7 @@ def suggest_names(
     current_output_name,
     current_source_url,
     current_change_summary,
+    current_dataset_title,
 ):
     """Generate dataset/output names and change summary from UI state.
 
@@ -992,10 +1036,16 @@ def suggest_names(
         current_change_summary: Current change summary input.
 
     Returns:
-        Tuple of (dataset name, output filename, source URL, change summary).
+        Tuple of (dataset name, output filename, source URL, change summary, dataset title).
     """
     if not loaded_dataset:
-        return current_dataset_name, current_output_name, current_source_url, current_change_summary
+        return (
+            current_dataset_name,
+            current_output_name,
+            current_source_url,
+            current_change_summary,
+            current_dataset_title,
+        )
     triggered = ctx.triggered_id
     if suggested_name and suggested_name != "__new__":
         current_dataset_name = suggested_name
@@ -1049,7 +1099,18 @@ def suggest_names(
         f"{selection_desc}; {period_desc}; {layer_desc}; "
         f"Rate mode: {rate_mode}, New rate: {new_rate}; {add_desc}"
     )
-    return dataset_name, output_name, source_url, change_summary
+    dataset_title = current_dataset_title or ""
+    if triggered in ("loaded-dataset", "load-counter", "dataset-suggestions") or not dataset_title.strip():
+        if loaded_dataset == "gam-carrizo-wilcox-aquifer-central-portion-version-3-02":
+            dataset_title = f"Carrizo-Wilcox (v3.02) – updated-{name_seed}"
+        else:
+            dataset_title = (current_dataset_name or dataset_name or "").strip()
+            if not dataset_title and loaded_dataset:
+                try:
+                    dataset_title = str(load_dataset(loaded_dataset)["dataset"].get("title") or "").strip()
+                except Exception:
+                    dataset_title = ""
+    return dataset_name, output_name, source_url, change_summary, dataset_title
 
 
 @app.callback(
@@ -1199,6 +1260,7 @@ def update_output_label(flux_source):
     State("update-store", "data"),
     State("ckan-jwt", "data"),
     State("dataset-name", "value"),
+    State("dataset-title", "value"),
     State("output-wel", "value"),
     State("source-url", "value"),
     State("change-summary", "value"),
@@ -1217,6 +1279,7 @@ def apply_rate(
     update_counter,
     jwt_token,
     dataset_name,
+    dataset_title,
     output_wel,
     source_url,
     change_summary,
@@ -1237,6 +1300,7 @@ def apply_rate(
         update_counter: Current update counter.
         jwt_token: CKAN JWT token.
         dataset_name: Output dataset name.
+        dataset_title: Output dataset title.
         output_wel: Output WEL/RCH filename.
         source_url: Source URL string.
         change_summary: Change summary string.
@@ -1313,6 +1377,7 @@ def apply_rate(
                 provenance,
                 jwt_token=jwt_token or None,
                 new_dataset_name=dataset_name or None,
+                new_dataset_title=dataset_title or None,
                 source_url=source_url or None,
                 change_summary=change_summary or None,
                 maintainer_username=tapis_username or None,
@@ -1324,6 +1389,7 @@ def apply_rate(
                 provenance,
                 jwt_token=jwt_token or None,
                 new_dataset_name=dataset_name or None,
+                new_dataset_title=dataset_title or None,
                 source_url=source_url or None,
                 change_summary=change_summary or None,
                 maintainer_username=tapis_username or None,
@@ -1419,6 +1485,25 @@ def update_map(
             print(f"[grid] sample geometry: {sample}")
     except Exception as exc:
         print(f"[grid] geometry debug failed: {exc}")
+    zoom = None
+    if isinstance(relayout, dict):
+        zoom = relayout.get("map.zoom")
+        if zoom is None:
+            zoom = relayout.get("mapbox.zoom")
+    cache_key = (
+        loaded_dataset,
+        flux_source,
+        color_by,
+        int(color_period) if color_period is not None else None,
+        tuple(int(p) for p in (periods or [])),
+        float(zoom) if zoom is not None else None,
+        int(_update or 0),
+    )
+    cached_fig = _MAP_FIG_CACHE.get(cache_key)
+    if cached_fig is not None:
+        fig = go.Figure(cached_fig)
+        _apply_selection(fig, gdf, selected_ids)
+        return fig
     wel = data["wel"]
     rch = data["rch"]
     periods = periods or []
@@ -1455,21 +1540,19 @@ def update_map(
             f"[flux] period={period_label} cells={count} nonzero={nz_count} "
             f"min={vmin:.6g} max={vmax:.6g}"
         )
-    zoom = None
-    if isinstance(relayout, dict):
-        zoom = relayout.get("map.zoom")
-        if zoom is None:
-            zoom = relayout.get("mapbox.zoom")
-    return _build_map_figure(
+    fig = _build_map_figure(
         gdf,
         active_cells,
         label,
         color_by,
         force_linear,
-        selected_ids or [],
+        [],
         zoom=zoom,
         show_grid=False,
     )
+    _MAP_FIG_CACHE[cache_key] = fig
+    _apply_selection(fig, gdf, selected_ids)
+    return fig
 
 
 if __name__ == "__main__":