Testing out the mechanics for a shiny web app/game with maps and street names

street_map_game_mechanics.qmd

---
output: html_document
editor_options: 
  chunk_output_type: console
---

## Load Packages

```{r}
setwd(here::here('streep_map_game/'))
library(tidyverse)
library(sf)
library(osmdata)
library(ggmap)
```

## Query street data from OpenStreetMap

```{r}
# Set bounding box.
# Bounding box adjusted from getbb('Ulm')
bbx <- matrix(
  data = c(9.93, 10.04283, 48.37, 48.44),
  nrow = 2,
  ncol = 2,
  byrow = T,
  dimnames = list(c('x', 'y'), c('min', 'max'))
)

# Build query for streets
# Features https://wiki.openstreetmap.org/wiki/Map_features#Roads
query <- bbx %>% 
  opq() %>% 
  add_osm_feature(
    key = 'highway',  
    value = c('motorway', 'trunk', 'primary', 'secondary', 
              'tertiary', 'unclassified', 'residential',
              'motorway_link', 'trunk_link', 'primary_link', 
              'secondary_link')
  )

# Get data from Openstreetmap
sf_data <- osmdata_sf(query)
# Object of class 'osmdata' with:
#                  $bbox : 48.37,9.93,48.44,10.04283
#         $overpass_call : The call submitted to the overpass API
#                  $meta : metadata including timestamp and version numbers
#            $osm_points : 'sf' Simple Features Collection with 21143 points
#             $osm_lines : 'sf' Simple Features Collection with 3927 linestrings
#          $osm_polygons : 'sf' Simple Features Collection with 13 polygons
#        $osm_multilines : NULL
#     $osm_multipolygons : NULL
```

## Make sure that we only use streets from Ulm and not the adjacent Neu-Ulm (New Ulm)

```{r}
# New query for administrative boundaries
ulm_boundary_query <- getbb('Ulm') %>% 
  opq() %>% 
  add_osm_feature(
    key = 'boundary',  
    value = 'administrative'
  )
ulm_boundary <- osmdata_sf(ulm_boundary_query)

# Find Ulm's boundaries via filter()
boundary <- ulm_boundary$osm_multipolygons %>% 
  filter(name == 'Ulm') %>% 
  pull(geometry)

# Find streets (lines) in results from previous query that are within Ulm's boundaries
# Key function st_contains() from {sf}
indices_in_ulm <- st_contains(boundary, sf_data$osm_lines) %>% unlist()
street_lines_ulm <- sf_data$osm_lines %>% slice(indices_in_ulm)
```

## Draw a map with a highlighted street

```{r}
# Filter specific street
street <- street_lines_ulm %>% 
  filter(str_detect(name, 'Zinglerstr')) %>% 
  pull(geometry) %>% 
  st_union() 
  # Streets can be represented by multiple lines so throw them together

#our background map
background_map <- get_map(bbx, maptype = "toner-background")

# inherit.aes must be false
ggmap(background_map) +
  geom_sf(data = street_lines_ulm, inherit.aes = FALSE) +
  geom_sf(
    data = street, 
    inherit.aes = F, 
    col = 'red', 
    size = 1.5
  ) 
```

## Imitate a point guess and compute distance to highlighted street

```{r}
# Create point with st_point() from {sf}
# Make sure every geometry uses same coordinate reference system (CRS)
point_guess <- st_point(c(9.96, 48.405)) %>% st_sfc(crs = 4326) 
street_crs_corrected <- st_transform(street, crs = 4326)
# Compute distance with st_distance from {sf}
distance <-  st_distance(point_guess, street_crs_corrected)
distance
# Units: [m]
#          [,1]
# [1,] 1966.924
```

## Create circle with radius of previously calculated distance
## Plot everything into one image.

```{r}
circle <- point_guess %>% st_buffer(distance)

ggmap(background_map) +
  geom_sf(data = street_lines_ulm, inherit.aes = FALSE) +
  geom_sf(
    data = street, 
    inherit.aes = F, 
    col = 'red', 
    size = 1.5
  )  +
  geom_sf(
    data = circle, 
    inherit.aes = F, 
    col = 'blue', 
    size = 1.5, 
    fill = NA
  ) +
  geom_sf(
    data = point_guess, 
    inherit.aes = F, 
    col = 'blue', 
    size = 2, 
    fill = NA
  ) +
  annotate(
    'label',
    x = 9.96,
    y = 48.405 + 0.0025,
    label = glue::glue('Distance: {round(distance,2)}m'),
  )
```