Created
August 25, 2021 14:10
-
-
Save mossmatters/dbc4020f5c63f78e0e5b68fa6a0129eb to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
library(jsonlite) | |
library(readxl) | |
library(dplyr) | |
library(ggplot2) | |
#define the source data location and get a JSON with the assetIDs | |
url = 'https://beta.healthdata.gov/api/views/gqxm-d9w9' | |
document <- fromJSON(txt=url) | |
files = document[["metadata"]]$attachments | |
excels = files[endsWith(files$filename,".xlsx"),] | |
#Function for downloading data, but only if it's not already there | |
dataget = function(json_data){ | |
base_url = "https://beta.healthdata.gov/api/views/gqxm-d9w9/files/%s" | |
dest = sprintf("~/Dropbox/lbbcovid/data/%s",json_data[3]) | |
if (!file.exists(dest)){ | |
download.file(sprintf(base_url,json_data[2]),destfile = dest) | |
} | |
} | |
#download the data with the function (only if it's not already there) | |
apply(excels,1,dataget) | |
excel.files = list.files("data",pattern = "\\.xlsx$") | |
lubbock.list = vector(mode="list",length = length(excel.files)) | |
for (f in 1:length(excel.files)){ | |
date.data = read_excel(sprintf("data/%s",excel.files[f]),sheet="Counties",skip=1) | |
my.date = as.Date(strsplit(excel.files[f],'_')[[1]][4],"%Y%m%d") | |
lubbock.date = date.data %>% filter(.[1] == "Lubbock County, TX") | |
lubbock.date$Date = my.date | |
lubbock.list[[f]] = lubbock.date | |
} | |
names(lubbock.list[[f]]) | |
column_title = "% staffed adult ICU beds occupied by COVID-19 patient" | |
multiplier = 100 | |
lubbock.icu = data.frame(matrix(NA,nrow=length(excel.files),ncol=2)) | |
for (f in 1:length(excel.files)){ | |
if(column_title %in% colnames(lubbock.list[[f]])){ | |
pct.icu.occupied = lubbock.list[[f]][column_title] | |
} else{ | |
pct.icu.occupied = "NA" | |
} | |
lubbock.icu[f,1]=lubbock.list[[f]]$Date | |
lubbock.icu[f,2]=pct.icu.occupied | |
} | |
names(lubbock.icu) = c("Date","Value") | |
class(lubbock.icu$Date) = "Date" | |
class(lubbock.icu$Value) = "numeric" | |
lubbock.icu$Value = lubbock.icu$Value*multiplier | |
g = ggplot(lubbock.icu,aes(x=Date,y=Value)) + geom_line() + | |
scale_x_date(date_labels="%b %y",date_breaks ="1 month") + | |
ylab(column_title)+ | |
labs (title = "Lubbock County, Texas", | |
caption= "data: beta.health.gov" | |
)+ | |
theme_bw()+ | |
theme(axis.title.x = element_blank(), | |
axis.title.y = element_text(size=15), | |
plot.title = element_text(size=15), | |
plot.caption = element_text(size=15)) | |
g |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment