diff --git a/app.py b/app.py index 1a388ed..5c9965f 100644 --- a/app.py +++ b/app.py @@ -1,11 +1,335 @@ -"""The Streamlit app. Run with `uv run streamlit run app.py`. +"""Mableton communication-needs frontend. Run: uv run streamlit run app.py. -It reads only files committed in data/, so it needs no keys or secrets to run. -Fetching and preparing data happens in scripts/, ahead of time. +Boundary polygons and 2020 population come from committed data/ GeoJSON. +Demographic rates are DEMONSTRATION ONLY until a validated CSV is supplied. """ +import json +from pathlib import Path + +import pandas as pd +import plotly.express as px +import pydeck as pdk import streamlit as st -st.set_page_config(page_title="Communication-needs Map", layout="wide") -st.title("Communication-needs Map") -st.write("Edit `app.py` and save; the page reloads by itself.") +BASE = Path(__file__).resolve().parent +BOUNDARIES = BASE / "data" / "mableton-council-districts.geojson" +METRICS_CSV = BASE / "data" / "district_metrics.csv" +METRICS = { + "Language other than English at home": "language_pct", + "Residents age 65+": "age65_pct", + "Households without internet": "no_internet_pct", +} +COLORS = { + "Language other than English at home": "#6366f1", + "Residents age 65+": "#0891b2", + "Households without internet": "#e07a39", +} + +st.set_page_config( + page_title="Mableton | Communication Needs", page_icon="🗺️", layout="wide" +) +st.markdown( + """""", + unsafe_allow_html=True, +) + + +@st.cache_data +def load_boundaries(): + with BOUNDARIES.open(encoding="utf-8") as f: + geo = json.load(f) + features = geo.get("features", []) + if len(features) != 6: + raise ValueError("Expected six council district polygons") + rows = [] + for feature in features: + props = feature["properties"] + number = int(props["district"]) + rows.append({"district": number, "population": int(props["population"])}) + return geo, pd.DataFrame(rows).sort_values("district").reset_index(drop=True) + + +def polygon_center(feature): + """Area-weighted polygon-ring centroid for 3D column positioning.""" + geometry = feature["geometry"] + polygon = ( + geometry["coordinates"][0] + if geometry["type"] == "Polygon" + else geometry["coordinates"][0][0] + ) + twice_area = 0.0 + sx = sy = 0.0 + for (x1, y1), (x2, y2) in zip(polygon, polygon[1:] + polygon[:1]): + cross = x1 * y2 - x2 * y1 + twice_area += cross + sx += (x1 + x2) * cross + sy += (y1 + y2) * cross + if abs(twice_area) > 1e-12: + return sx / (3 * twice_area), sy / (3 * twice_area) + return ( + sum(p[0] for p in polygon) / len(polygon), + sum(p[1] for p in polygon) / len(polygon), + ) + + +def load_metrics(): + if METRICS_CSV.exists(): + frame = pd.read_csv(METRICS_CSV) + required = ["district", *METRICS.values()] + missing = [name for name in required if name not in frame.columns] + if missing: + st.error("district_metrics.csv is missing columns: " + ", ".join(missing)) + st.stop() + frame = frame[required].copy() + for name in required: + frame[name] = pd.to_numeric(frame[name], errors="coerce") + if frame[required].isna().any().any() or frame["district"].duplicated().any(): + st.error("CSV has missing, nonnumeric, or duplicate district values.") + st.stop() + if not frame["district"].between(1, 6).all() or set(frame["district"]) != set( + range(1, 7) + ): + st.error("CSV must contain exactly one row for each district, 1 through 6.") + st.stop() + if not all(frame[c].between(0, 100).all() for c in METRICS.values()): + st.error("Demographic percentages must be between 0 and 100.") + st.stop() + return frame, False + # Fictional values ONLY to demonstrate the interface. Not Census estimates. + return pd.DataFrame( + { + "district": [1, 2, 3, 4, 5, 6], + "language_pct": [31, 22, 27, 18, 25, 34], + "age65_pct": [15, 19, 24, 21, 17, 13], + "no_internet_pct": [12, 16, 11, 20, 18, 14], + } + ), True + + +try: + geojson, population = load_boundaries() +except (OSError, ValueError, KeyError, json.JSONDecodeError) as exc: + st.error(f"Cannot read the official district boundary file: {exc}") + st.stop() + +rates, demo_mode = load_metrics() +df = population.merge(rates, on="district", how="left", validate="one_to_one") +if df[list(METRICS.values())].isna().any().any(): + st.error("Some districts are missing demographic data") + st.stop() + +st.sidebar.markdown("### Map controls") +district = st.sidebar.selectbox( + "Council district", df["district"].tolist(), format_func=lambda x: f"District {x}" +) +metric_label = st.sidebar.radio("Color / column height", list(METRICS)) +map_style = st.sidebar.radio( + "Map display", ["District boundaries", "3D columns"], horizontal=False +) +metric = METRICS[metric_label] +selected = df.loc[df["district"] == district].iloc[0] + +st.title("Mableton Communication-needs Map") +st.caption( + "Georgia 311 Challenge · Reach Every Resident · Six Mableton council districts" +) +if demo_mode: + st.warning( + "DEMO DATA: Language, age 65+, and internet percentages are fictional placeholders. The district shapes and 2020 populations are real. Add data/district_metrics.csv to use prepared statistics." + ) +else: + st.info( + "Showing the team's prepared district_metrics.csv. Verify data source, year, denominator, and district aggregation before presenting figures as official." + ) + +st.subheader(f"District {district} overview") +a, b, c, d = st.columns(4) +a.metric("2020 population", f"{int(selected['population']):,}") +b.metric("Other language at home", f"{selected['language_pct']:.1f}%") +c.metric("Age 65+", f"{selected['age65_pct']:.1f}%") +d.metric("No internet", f"{selected['no_internet_pct']:.1f}%") + +st.subheader("District map") +st.caption( + "Choose a district in the sidebar. Hover over a district or column to inspect its information." +) + +# Attach all three metrics to each official district polygon, enabling tooltips. +for feature in geojson["features"]: + props = feature["properties"] + num = int(props["district"]) + item = df.loc[df["district"] == num].iloc[0] + props.update( + { + "district_name": f"District {num}", + "language_pct": float(item["language_pct"]), + "age65_pct": float(item["age65_pct"]), + "no_internet_pct": float(item["no_internet_pct"]), + "metric_value": float(item[metric]), + "fill_color": [37, 99, 235, 185] + if num == district + else [99, 102, 241, 65 + int(item[metric] * 3)], + } + ) + +# Map uses official GeoJSON geometry rather than approximate district markers. +polygon_layer = pdk.Layer( + "GeoJsonLayer", + data=geojson, + pickable=True, + auto_highlight=True, + filled=True, + stroked=True, + get_fill_color="properties.fill_color", + get_line_color=[255, 255, 255, 220], + get_line_width=120, + line_width_min_pixels=2, +) + +layers = [polygon_layer] +if map_style == "3D columns": + column_rows = [] + for feature in geojson["features"]: + props = feature["properties"] + lng, lat = polygon_center(feature) + column_rows.append( + { + "district": int(props["district"]), + "lng": lng, + "lat": lat, + "population": int(props["population"]), + "language_pct": props["language_pct"], + "age65_pct": props["age65_pct"], + "no_internet_pct": props["no_internet_pct"], + "metric_value": props["metric_value"], + "fill_color": [37, 99, 235, 230] + if props["district"] == district + else [224, 122, 57, 190], + } + ) + layers.append( + pdk.Layer( + "ColumnLayer", + data=pd.DataFrame(column_rows), + get_position="[lng, lat]", + get_elevation="metric_value", + elevation_scale=130, + radius=430, + get_fill_color="fill_color", + pickable=True, + auto_highlight=True, + extruded=True, + ) + ) + +view = pdk.ViewState( + latitude=33.82, + longitude=-84.57, + zoom=10.8, + pitch=52 if map_style == "3D columns" else 0, + bearing=20 if map_style == "3D columns" else 0, +) +tooltip = { + "html": "District {district}
Other language: {language_pct}%
Age 65+: {age65_pct}%
No internet: {no_internet_pct}%
2020 population: {population}", + "style": {"backgroundColor": "#172033", "color": "white"}, +} +st.pydeck_chart( + pdk.Deck( + layers=layers, + initial_view_state=view, + tooltip=tooltip, + map_style="https://basemaps.cartocdn.com/gl/positron-gl-style/style.json", + ), + width="stretch", + height=490, +) + +left, right = st.columns([1.1, 1]) +with left: + st.subheader("Compare all districts") + chart = px.bar( + df, + x="district", + y=metric, + text_auto=".1f", + labels={"district": "Council district", metric: "Percent (%)"}, + color_discrete_sequence=[COLORS[metric_label]], + ) + chart.update_layout( + yaxis_range=[0, 100], showlegend=False, margin={"l": 0, "r": 0, "t": 10, "b": 0} + ) + st.plotly_chart(chart, width="stretch") +with right: + st.subheader(f"District {district}: three measures") + breakdown = pd.DataFrame( + { + "Measure": list(METRICS), + "Percent": [float(selected[col]) for col in METRICS.values()], + } + ) + fig = px.bar( + breakdown, + x="Percent", + y="Measure", + orientation="h", + text_auto=".1f", + color="Measure", + color_discrete_map=COLORS, + ) + fig.update_layout( + xaxis_range=[0, 100], showlegend=False, margin={"l": 0, "r": 0, "t": 10, "b": 0} + ) + st.plotly_chart(fig, width="stretch") + +st.subheader("Possible outreach considerations") +notes = [] +if selected["language_pct"] >= 30: + notes.append( + "Explore multilingual notices and partnerships with trusted community organizations." + ) +if selected["age65_pct"] >= 20: + notes.append( + "Consider accessible print materials and outreach through senior/community centers." + ) +if selected["no_internet_pct"] >= 18: + notes.append( + "Include offline options such as mailed notices, flyers, and in-person announcements." + ) +if not notes: + notes.append( + "Consider combining digital updates with community partnerships and accessible print notices." + ) +for note in notes: + st.write("• " + note) +st.caption( + "These are illustrative suggestions, not measured outreach preferences. Thresholds are design rules, not research findings." +) + +st.subheader("District data") +st.dataframe( + df.rename( + columns={ + "district": "District", + "population": "2020 population", + "language_pct": "Other language (%)", + "age65_pct": "Age 65+ (%)", + "no_internet_pct": "No internet (%)", + } + ), + hide_index=True, + width="stretch", +) +st.download_button( + "Download displayed data (CSV)", + data=df.to_csv(index=False), + file_name="mableton_district_dashboard_data.csv", + mime="text/csv", +) +st.caption( + "Boundary and 2020 population source: Mableton council district GeoJSON committed in the team's data/ directory." +) diff --git a/pyproject.toml b/pyproject.toml index cbf14ac..92b8b32 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -6,6 +6,7 @@ requires-python = ">=3.12" dependencies = [ "pandas>=3.0.6", "plotly>=7.1.0", + "pydeck>=0.9.3", "streamlit>=1.66.0", ] diff --git a/requirements.txt b/requirements.txt index 196f336..e402c1f 100644 --- a/requirements.txt +++ b/requirements.txt @@ -63,7 +63,9 @@ protobuf==7.36.2 pyarrow==26.0.0 # via streamlit pydeck==0.9.3 - # via streamlit + # via + # communication-needs-map + # streamlit python-dateutil==2.9.0.post0 # via pandas python-multipart==0.0.32 diff --git a/tests/test_app.py b/tests/test_app.py index 4de47da..2112943 100644 --- a/tests/test_app.py +++ b/tests/test_app.py @@ -6,5 +6,5 @@ def test_app_runs(): - app = AppTest.from_file(APP).run() + app = AppTest.from_file(APP, default_timeout=30).run() assert not app.exception diff --git a/uv.lock b/uv.lock index f63c908..054a3f3 100644 --- a/uv.lock +++ b/uv.lock @@ -1,5 +1,5 @@ version = 1 -revision = 3 +revision = 5 requires-python = ">=3.12" resolution-markers = [ "python_full_version >= '3.14' and sys_platform == 'win32'", @@ -307,6 +307,7 @@ source = { virtual = "." } dependencies = [ { name = "pandas" }, { name = "plotly" }, + { name = "pydeck" }, { name = "streamlit" }, ] @@ -324,6 +325,7 @@ dev = [ requires-dist = [ { name = "pandas", specifier = ">=3.0.6" }, { name = "plotly", specifier = ">=7.1.0" }, + { name = "pydeck", specifier = ">=0.9.3" }, { name = "streamlit", specifier = ">=1.66.0" }, ]