# Setup ----

#' What do we always do first? Besides the tidyverse, 
#' we need the rwolrdmap package (which you have to install).


# Country level map with rworldmap ----

eurost <- read_csv2("data/eurostat_data.csv")

eurost <- eurost %>%
  filter(time == 2014)

mpdta <- map_data("world")

eurost2 <-
  inner_join(x = eurost,
             y = mpdta,
             by = c("geo_name" = "region"))


#' The basic ggplot can be done with `geom_polygon`. 
#' As we said, you supply longitude as x and latitude as y data, 
#' group them according to the group indicator and fill in your 
#' variable of interest (youth unemployment.)
#' That's it.

P <- ggplot() +
  geom_polygon(
    data = eurost2,
    aes(
      x = long,
      y = lat,
      group = group,
      fill = unemp_youth_t
    ),
    colour = "white",
    size = 0.1
  )
P
 


#' Mapping is a normal ggplot operation like anything else. 
#' You can change the plot in the normal way.
P2 <- P +
  theme(
    panel.grid.minor = element_line(colour = NA),
    panel.background = element_rect(fill = NA, colour = NA),
    axis.text.x = element_blank(),
    axis.text.y = element_blank(),
    axis.ticks.x = element_blank(),
    axis.ticks.y = element_blank(),
    axis.title = element_blank()
  ) +
  scale_fill_gradient(
    name = "Youth Unemployment",
    low = rgb(207, 233, 238, maxColorValue = 255),
    high = rgb(79, 129, 189, maxColorValue = 255)
  ) +
  labs(title = "Map of youth unemployment per country in Europe (2014)",
       subtitle = "notice the different way of changing the colors",
       caption = "Source: Eurostat")


P2

# NUTS2 level plots with eurostat -----

#' The overall procedure is the same as above. 
#' Once we have data on long, lat, and group, we merge 
#' it with our data of interest. 

library("eurostat")

df <- get_eurostat_geospatial(output_class = "df", 
                              # data in a data.frame in order to be able to work with ggplot in a tidy way
                              
                              resolution = "20",   
                              # the higher the resolution, the more processing power needed
                              
                              nuts_level = "2"
                              # which NUTS level do you want
                              )   

data_b2 <-
  get_eurostat(id = "yth_empl_110",
               time_format = "num")

data_b2 <- data_b2 %>%
  filter(age == "Y15-24",
         sex == "T",
         time == 2014)  %>%
  subset(nchar(as.character(geo)) > 3)

df2 <- inner_join(data_b2, df, by = "geo")


#' And the plot in the exact same way.
new <- ggplot() +
  geom_polygon(
    data = df2,
    aes(
      x = long,
      y = lat,
      group = group,
      fill = values
    ),
    colour = "white",
    size = 0.1
  )

new


#' 
#' Again, we simply modify this chart. 
#' To zoom into the plot, we use the function `coord_map` 
#' which builds upon the mapproj package (which you have 
#' to install first).

library(mapproj)
new2 <- new +
  theme(
    panel.grid.minor = element_line(colour = NA),
    panel.background = element_rect(fill = NA, colour = NA),
    axis.text.x = element_blank(),
    axis.text.y = element_blank(),
    axis.ticks.x = element_blank(),
    axis.ticks.y = element_blank(),
    axis.title = element_blank()
  ) +
  scale_fill_gradient(
    name = "Youth Unemployment",
    low = rgb(207, 233, 238, maxColorValue = 255),
    high = rgb(79, 129, 189, maxColorValue = 255)
  ) +
  coord_map(xlim = c(-30, 45), ylim = c(30, 75)) +
  labs(title = "Map of youth unemployment per NUTS2 region in Europe (2014)",
       subtitle = "We zoom into main continental Europe with `coord_map()` for better readability",
       caption = "Source: Eurostat")
new2