def test_compare_counties(data_clean, gdf_processed):
neighbors = get_neighbors("Greene", gdf_processed)
compare_list = ["Greene"] + neighbors
data_clean_deaths = data_clean["deaths"]
clean_rules = DATA_INDEX["deaths"]["clean_rules"]
result = compare_counties(
data_clean_deaths,
clean_rules=clean_rules,
compare_field="moving_avg_per_capita",
counties=compare_list,
)
print("\n", result)
def test_process_neighbors_per_capita(data_clean, gdf_processed):
neighbors = get_neighbors("Dauphin", gdf_processed)
compare_list = ["Dauphin"] + neighbors
data_clean_cases = data_clean["cases"]
clean_rules = DATA_INDEX["cases"]["clean_rules"]
df = compare_counties(
data_clean_cases,
clean_rules=clean_rules,
compare_field="moving_avg_per_capita",
counties=compare_list,
)
df = df.set_index("date")
print(df)
def greene_region_deaths_moving_avg_per_cap(data_clean, gdf_processed) -> pd.DataFrame:
"""
A DataFrame representing a day-by-day comparison of moving avg number of new daily cases, per capita,
for Dauphin and neighboring counties.
"""
neighbors = get_neighbors("Greene", gdf_processed)
compare_list = ["Greene"] + neighbors
data_clean_deaths = data_clean["deaths"]
clean_rules = DATA_INDEX["deaths"]["clean_rules"]
return compare_counties(
data_clean_deaths,
clean_rules=clean_rules,
compare_field="moving_avg_per_capita",
counties=compare_list,
)
def test_process_neighbors_total(data_clean, gdf_processed):
neighbors = get_neighbors("Dauphin", gdf_processed)
compare_list = ["Dauphin"] + neighbors
data_clean_cases = data_clean["cases"]
clean_rules = DATA_INDEX["cases"]["clean_rules"]
df = compare_counties(
data_clean_cases,
clean_rules=clean_rules,
compare_field="total",
counties=compare_list,
)
df = df.set_index("date")
cases_lebanon_july_26 = df.at["2020-07-26", "lebanon"]
cases_lebanon_july_23 = df.at["2020-07-23", "cumberland"]
assert cases_lebanon_july_26 == 1544
assert cases_lebanon_july_23 == 1066
def gen_desc_neighbors(self, *, data_type: str) -> str:
neighbor_list = get_neighbors(self.county_name_clean, self.gdf)
neighbor_count = len(neighbor_list)
region_list = [self.county_name_clean] + neighbor_list
region_gdf = self.gdf[self.gdf["NAME"].isin(region_list)]
rank_from_top, rank_from_bottom, others_with_same_rank = self.gdf_get_ranking(
region_gdf, f"{data_type}_added_past_two_weeks_per_capita")
per_capita_rank_among_neighbors = rank_text(rank_from_top,
rank_from_bottom)
if others_with_same_rank > 0 and others_with_same_rank != neighbor_count:
sentence_frag = (
f"[b]{others_with_same_rank}[/b] other "
f"{p.singular_noun('counties', others_with_same_rank)} had the same per capita rate of"
f" {data_type}.")
else:
sentence_frag = ""
return (
f"Compared to its {p.number_to_words(neighbor_count)} neighboring counties, {self.county_name_clean} "
f"County had the [b]{per_capita_rank_among_neighbors}[/b] number of {data_type} per 100,000 people over "
f"the past two weeks. {sentence_frag} Here's how {self.county_name_clean}'s per capita 7-day moving average compares to "
f"its neighbors:")
def gen_chart(
county_name_clean: str,
data_type: str,
*,
data_index: Dict,
chart_dict: Dict,
data_clean: Dict,
county_data: Dict[str, pd.DataFrame],
gdf: geopandas.GeoDataFrame,
primary_color: str,
secondary_color: str,
aws_bucket: str,
aws_dir: str,
) -> Dict[str, Union[Union[str, None, List[Dict[str, str]]], Any]]:
"""
Creates a chart PNG using Altair and moves its to s3. Returns an URL to the image, a Dict representing a legend
for the chart
Args:
county_name_clean (str): Name of county without 'County' suffix. Eg. "Dauphin"
data_type (str): Type of data. Eg. "cases".
data_index (Dict): Config settings for data.
chart_dict (Dict): Config settings for chart.
data_clean (Dict[str, pd.DataFrame]): Dict of pandas dfs of cases, deaths, tests data for all Pa. counties
that has has some minimal cleaning.
county_data (Dict[str, pd.DataFrame]: Processed cases, deaths, tests, etc data for a specific county.
gdf (geopandas.GeoDataFrame): Pa geodataframe with cases, deaths, tests data merged on to it.
primary_color (str): Hex code for color theme.
secondary_color (str): Hex code for color theme.
aws_bucket (str): AWS bucket where charts will be uploaded to. Defaults to value stored in
definitions.py
aws_dir (str): Directory within AWS bucket where charts will be uploaded. Defaults to value stored in
definitions.py
Returns:
Dict[str, Union[Union[str, None, List[Dict[str, str]]], Any]]: Dict with keys relating to chart, legend,
and chart descriptive text.
"""
chart_type = chart_dict["type"]
custom_legend = None
fmt = "png"
content_type = "image/png"
gen_desc = GenStats(county_name_clean, gdf=gdf)
if "daily_and_avg" in chart_type:
chart = chart_bar_and_line(
data_type=data_type,
df=county_data[data_type],
line_color=primary_color,
bar_color=secondary_color,
)
custom_legend = chart_dict.get("custom_legend")
chart_desc = gen_desc.gen_desc_daily(data_type=data_type)
elif "choropleth" in chart_type:
chart = map_choropleth(
gdf,
color_field=chart_dict["color_field"],
highlight_polygon=county_name_clean,
min_color=secondary_color,
max_color=primary_color,
legend_title=chart_dict["legend_title"],
)
chart_desc = gen_desc.gen_desc_choro(data_type=data_type)
elif "neigbhors_per_capita" in chart_type:
compare_field = chart_dict["compare_field"]
neighbors = get_neighbors(county_name_clean, gdf)
neighbors = sort_counties_by_pop(neighbors)
compare_list = [county_name_clean] + neighbors
df_data_type = data_clean[data_type]
clean_rules = data_index[data_type]["clean_rules"]
df_multi_county = compare_counties(
df_data_type,
clean_rules=clean_rules,
compare_field=compare_field,
counties=compare_list,
)
county_cols = list(df_multi_county.columns)
county_cols.remove("date")
df_multi_county = stack_df(df_multi_county,
stack_cols=county_cols,
x_axis_col="date")
chart = chart_faceted(
df_multi_county,
category_col="category",
x_axis_col="date",
y_axis_col="value",
line_color=primary_color,
)
custom_legend = None
chart_desc = gen_desc.gen_desc_neighbors(data_type=data_type)
elif "stacked_area" in chart_type:
df = process_cumulative_tests(county_data["confirmed"],
county_data["tests"])
chart = chart_stacked_area(
df,
x_axis_col="date",
y_axis_col="count",
category_col="data_type",
domain=["positive", "negative"],
range_=[primary_color, secondary_color],
)
custom_legend = chart_dict.get("custom_legend")
chart_desc = gen_desc.gen_desc_area_tests()
else:
raise Exception(
"Chart type not found. Did you provide a valid chart type in chart_index?"
)
image_filename = f"{county_name_clean.lower()}_{data_type}_{chart_type}.{fmt}"
image_path = DIR_OUTPUT / image_filename
save(chart, str(image_path))
logging.info("...saved")
# Move to s3
copy_to_s3(image_path, aws_bucket, aws_dir, content_type=content_type)
return {
"title": chart_dict.get("title", "").upper(),
"custom_legend": custom_legend,
"image_path": f"https://{aws_bucket}/{aws_dir}/{image_filename}",
"description": chart_desc,
}
def test_get_neighbors(gdf_processed):
neighbors = get_neighbors("Dauphin", gdf_processed)
assert "Lebanon" in neighbors
assert "Allegheny" not in neighbors
print(neighbors)