library(magrittr)
library(dplyr)
library(tidyr)
library(stringr)

library(readr)

# Import the race/hispanic-origin ACS data downloaded through American Fact-Finder ----
b03002_data <- read_csv(file = "Data/ACS/ZCTA/2013-2017/Raw/B03002/ACS_17_5YR_B03002_with_ann.csv",
                        skip = 1L,
                        col_types = cols(
                          .default = col_integer(),
                          Id = col_character(),
                          Id2 = col_character(),
                          Geography = col_character()
                        ))

# Tidy the data to a more user-friendly format
    b03002_data %<>%
      gather(key = "Measure_Name", value = "Value",
             -one_of(c("Race", "Id", "Id2", "Geography"))) %>%
      mutate(Measure_Type = case_when(str_detect(Measure_Name, "^Estimate") ~ "Estimate",
                                      str_detect(Measure_Name, "^Margin") ~ "Margin of Error"),
             Subpopulation = str_remove(Measure_Name, "(Estimate;)|(Margin of Error;)") %>%
               str_trim() %>% str_remove("(:|;)$")) %>%
      select(-Measure_Name) %>%
      spread(key = "Measure_Type", value = "Value") %>%
      rename_all(.funs = function(x) str_replace_all(x, "[[:space:]]+", "_"))

    b03002_data %<>%
      mutate(Hispanic_or_Latino_Status = case_when(str_detect(Subpopulation, "^Hispanic or Latino") ~ "Hispanic or Latino",
                                                   str_detect(Subpopulation, "^Not Hispanic or Latino") ~ "Not Hispanic or Latino",
                                                   str_detect(Subpopulation, "^Total") ~ "Total")) %>%
      mutate(Race = case_when(Subpopulation == "Hispanic or Latino" ~ "Total",
                              Subpopulation == "Not Hispanic or Latino" ~ "Total",
                              TRUE ~ str_remove(Subpopulation, "^(Not |)Hispanic or Latino: -") %>% str_trim)) %>%
      select(-Subpopulation)

    b03002_data %<>%
      filter(!Race %in% c("Two or more races: - Two races excluding Some other race, and three or more races",
                          "Two or more races: - Two races including Some other race"))

# Obtain easy-to-read table which separates all Hispanic/Latino persons
# into a distinct category
# (i.e. the 'Black or African-American' category excludes persons who are also Hispanic-Latino)

total_population_sizes_by_race_or_hisp_latino_status <- b03002_data %>%
  filter((Hispanic_or_Latino_Status == "Hispanic or Latino" & Race == "Total")
         |
           (Hispanic_or_Latino_Status == "Not Hispanic or Latino" & Race != "Total")) %>%
  mutate(Race_or_Hispanic_Latino_Status = case_when(Hispanic_or_Latino_Status == "Hispanic or Latino" ~ "Hispanic or Latino",
                                                    TRUE ~ Race)) %>%
  select(-Race, - Hispanic_or_Latino_Status) %>%
  select(Id, Id2, Geography, Race_or_Hispanic_Latino_Status,
         Estimate, Margin_of_Error)