add county comparison script

ad-hoc/county_comparisons.R

@@ -0,0 +1,75 @@
+rm(list=ls())
+suppressPackageStartupMessages(library(tidyverse))
+suppressPackageStartupMessages(library(behindbarstools))
+suppressPackageStartupMessages(library(lubridate))
+suppressPackageStartupMessages(library(argparse))
+
+# get carceral data for the state of Illinois
+all_df <- read_scrape_data(TRUE)
+
+parser <- ArgumentParser()
+
+# specify our desired options
+# by default ArgumentParser will add an help option
+parser$add_argument(
+    "-id", "--facid",
+    help="The facility id to make for the plot")
+parser$add_argument(
+    "-alt", "--alttext",
+    default = NULL,
+    help="The name of the geography to use in the plot.")
+parser$add_argument(
+    "-st", "--start", default = "2021-10-01",
+    help="Date to start the active case analysis should be YYYY-MM-DD format")
+parser$add_argument(
+    "-en", "--end",
+    default = as.character(round_date(Sys.Date(), unit = "day")),
+    help="Date to start the active case analysis should be YYYY-MM-DD format")
+
+args <- parser$parse_args()
+
+make_county_plot <- function(fac_id, adf = all_df, sub_name = NULL){
+    sub_df <- adf %>%
+        filter(Facility.ID == fac_id)
+
+    cname <- ifelse(
+        is.null(sub_name),
+        str_c(str_to_title(first(sub_df$County), " County")), sub_name)
+    cid <- as.numeric(first(sub_df$County.FIPS))
+    gen_df <- get_genpop_covid(cid)
+
+    final_plot <- sub_df %>%
+        # trim the data down
+        select(
+            Date, Active = Residents.Active, Pop = Population.Feb20) %>%
+        mutate(Group = str_c(cname, "\nJail Population")) %>%
+        # pull in the gen population data
+        bind_rows(
+            gen_df %>%
+                arrange(Date) %>%
+                mutate(Active = diff_roll_sum(General.Confirmed, Date)) %>%
+                select(Date, Active, Pop = General.Population) %>%
+                mutate(Group = str_c(cname, "\nOverall Population"))
+        ) %>%
+        # remove missing values
+        arrange(Group, Date) %>%
+        na.omit() %>%
+        # only look at recent dates
+        filter(Date >= ymd(args$start)) %>%
+        filter(Date <= ymd(args$end)) %>%
+        # calculate rate
+        mutate(Rate = Active/Pop * 100000) %>%
+        # make a purrty plot
+        ggplot(aes(x = Date, y = Rate, color = Group)) +
+        geom_line(size = 2) +
+        theme_behindbars() +
+        scale_color_bbdiscrete() +
+        labs(color = "", y = "Active Cases\nPer 100,000")
+
+    ggsave(
+        str_c(cname, "_compare_plot.png"),
+        final_plot,
+        width = 10, height = 6)
+}
+
+make_county_plot(as.integer(args$facid), sub_name = args$alttext)