daily_update.R

In COHHIO/Rm_data: Data munging dependencies for Rminor & Rminor_elevated Shiny apps

# Optional: These are necessary when rapidly modifying these dependencies

# devtools::load_all("../../lookr")
# devtools::load_all("../hud.extract")

#1. must load RmData
# devtools::load_all()
pkgload::load_all()

#2. must set directories if using a directory structure differing from the default in clarity.looker:
# dirs <- clarity.looker::dirs
# Rm_env$gather_deps(dirs)

# 3. Extract the HUD Export zip

  # Use "get_bucket"
df_bucket_data <- aws.s3::get_bucket("hud.csv-daily", prefix = "HMIS",
                                     region = "us-east-2")


# Save the zip file
aws.s3::save_object(df_bucket_data$Contents$Key, bucket = "hud.csv-daily",
                    file = "hudx-111.zip", region = "us-east-2")

  # Unzip the file
clarity.looker::hud_export_extract(browser_dl_folder = "~/Documents/COHHIO/RmData")

  # Delete the zip from AWS
aws.s3::delete_object(df_bucket_data$Contents$Key, bucket = "hud.csv-daily",
                      region = "us-east-2")

# 4. Run update
daily_update(backup = FALSE, remote = FALSE)
beepr::beep(sound = 3)



# WHEN NEW GEOCODES ARE ISSUED BY HUD, some projects may fail to match with their county in `pe_add_regions`.  ggmap is used to determine the missing county for these projects. ggmap and its dependencies are currently ignored in the renv/settings.dcf See ?ggmap::register_google for details on setting up ggmap.
# 1. Set GGMAP_GOOGLE_API_KEY in .Renviron.
# 2. Set this option to bypass errors in ggmap
# options(ggmap = list(google = list(second_limit = 50L,
#                                    day_limit = 2500)))