- Visualize geographic data as a choropleth
geo <- get_geo_data("brazil", get_munis = TRUE, scale = 10)
geo$muni$state_code <- geo$muni$state_code2
geo$muni$state_code2 <- NULL
country_data <- data.frame(
year = 2001:2010,
country_code = "BRA",
dat = runif(10),
stringsAsFactors = FALSE
)
state_data <- data.frame(
year = rep(2001:2010, nrow(geo$state)),
country_code = "BRA",
state_code = rep(geo$state$state_code, each = 10),
dat = runif(10 * nrow(geo$state)),
stringsAsFactors = FALSE
)
muni_data <- data.frame(
year = rep(2001:2010, nrow(geo$muni)),
country_code = "BRA",
state_code = rep(geo$muni$state_code, each = 10),
muni_code = rep(geo$muni$muni_code, each = 10),
dat = runif(10 * nrow(geo$muni)),
stringsAsFactors = FALSE
)
geovis(geo, path = "~/Desktop/geowidget3",
name = "Brazil SINASC Explorer",
view_level = "country",
view_country_code = "BRA",
default_var = "dat",
country_data = country_data,
state_data = state_data,
muni_data = muni_data
)
- Support geo grid morhping to view all time series simultaneously while preserving some notion of geographic location (as is done here). This will not always be effective, especially when the number of geographic entities is very large.
- Make setting height / width work (currently always full page)
- Broader testing with multiple countries, etc.
- More robust checking of config to ensure a bad spec isn't sent to the JS library