---
title: "Combining zoning systems"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Combining zoning systems}
  %\VignetteEncoding{UTF-8}
  %\VignetteEngine{knitr::rmarkdown}
editor_options: 
  chunk_output_type: console
---
  
```{r, include = FALSE}
knitr::opts_chunk$set(
  eval = FALSE,
  collapse = TRUE,
  comment = "#>"
)
```

This vignettes demonstrates possibilities when zoning systems from different cities meet.
It raises the question: how should different systems be combined geographically?

You need the latest version of the package:
  
```{r, eval=FALSE}
remotes::install_github("zonebuilders/zonebuilder")
```


For this demo, we need the following libraries:


```{r setup}
library(zonebuilder) # for the zoning system
library(sf)          # for processing spatial data
library(dplyr)       # for processing general data
library(tmap)        # for visualizing spatial data
```


We will apply the zoning system to the main Dutch cities, and analyse commuting patterns between the zones. The data can be read as follows:


```{r}
NLD_cities = readRDS(url("https://github.com/zonebuilders/zonebuilder/releases/download/0.0.1/NLD_cities.Rds"))
NLD_wijk_od = readRDS(url("https://github.com/zonebuilders/zonebuilder/releases/download/0.0.1/NLD_wijk_od.Rds"))
NLD_wijk_centroids = readRDS(url("https://github.com/zonebuilders/zonebuilder/releases/download/0.0.1/NLD_wijk_centroids.Rds"))
```

Let's take a look at the `NLD_cities` data:

```{r}
NLD_cities %>% arrange(desc(population))
```

...and plot it on an interactive map:

```{r}
tmap_mode("view") # enable interactive mode in tmap
qtm(NLD_cities, symbols.size = "population")
```


The following code chunk generated zones for the Dutch cities:


```{r}
zbs = do.call(rbind, lapply(1:nrow(NLD_cities), function(i) {
  ci = NLD_cities[i, ]
  
  # Amsterdam 5, Eindhoven-Rotterdam 4, Roermond-Zeeland 2, others 3 
  nrings = ifelse(ci$population < 60000, 2,
           ifelse(ci$population < 220000, 3,
           ifelse(ci$population < 800000, 4, 5)))
  
  zb = zb_zone(x = ci, n_circles = nrings) %>% 
    mutate(name = ci$name,
           labelplus = paste(ci$name, label, sep = "_"))
  
  zb
}))
```


```{r}
tm_basemap("OpenStreetMap") +
tm_shape(zbs) +
  tm_polygons(col = "circle_id", id = "labelplus", style = "cat", palette = "YlOrBr", alpha = 0.7) + 
  tm_scale_bar()
```