ivabrunec · July 18, 2022 22:06
diff --git a/experiential_map.R b/experiential_map.R
 # Create a map of traces using Google Location History #

 # get your google maps data:
 # https://www.howtogeek.com/725241/how-to-download-your-google-maps-data/

 library(dplyr) # data wrangling
 library(lubridate) # date parsing
 library(ggplot2) # plotting
 library(jsonlite) # reading in json file
 library(sf) # spatial
 library(osmdata) # osm api
 library(sfnetworks) # spatial
 library(showtext) # fonts

 # switch to current wd
 setwd(dirname(rstudioapi::getActiveDocumentContext()$path))

 # add font
 font_add_google(name = "Abril Fatface", family = "abril")
 showtext_auto()

 # read in file
 myHistory <- jsonlite::fromJSON("Takeout/Location History/Records.json", flatten=TRUE)

 # data cleaning & transformations
 # based on https://medium.com/geekculture/explore-your-activity-on-google-with-r-how-to-analyze-and-visualize-your-location-history-2ea8edabe733
 # grab just locations
 myData <- myHistory$locations

 myData <- myData %>% 
  filter(activity!="NULL")

 # clean up dates and lat/long
 myData <- myData %>% 
  mutate(time  = as_datetime(myData$timestamp),
         date = date(time),
         hour  = paste(hour(time),minute(time),sep=":"),
         week = isoweek(time),
         month = paste(month(time)),
         year = isoyear(time),
         latitude = latitudeE7/1e7,
         longitude= longitudeE7/1e7) %>%
  select(-timestamp,-latitudeE7,-longitudeE7,-time)

 ### !! EDIT THESE DEPENDING ON YOUR LOCATION AND DATES !! ###
 # grab everything before the end of April
 philly_data <- filter(myData, date < '2022-04-28')

 # add underlying osm map
 bbx = c(-75.20, 39.92,
        -75.13, 39.98)

 # all roads for dataviz
 roads_all <- bbx %>%
  opq() %>%
  add_osm_feature(key = 'highway') %>%
  osmdata_sf()

 # larger roads only to snap to network
 # takes a second to run
 highways <- bbx %>%
  opq() %>%
  add_osm_feature(
    key = "highway",
    value = c(
      'road',
      "motorway",
      "trunk",
      "primary",
      "secondary",
      "tertiary",
      "motorway_link",
      "trunk_link",
      "primary_link",
      "secondary_link",
      "tertiary_link",
      'unclassified'
    )
  ) %>%
  osmdata_sf()

 # now, we want to link the points to the nearest osm segment
 # build sf network
 my_roads <- st_geometry(highways$osm_lines)
 net = as_sfnetwork(my_roads, directed = FALSE)

 # grab coordinates
 philly_coords <- myData %>%
  st_as_sf(coords = c('longitude', 'latitude'))
 st_crs(philly_coords) = 4326

 # snap points to their nearest node
 nearest_nodes = st_nearest_feature(philly_coords, net)

 snapped_pois = philly_coords %>%
  st_set_geometry(st_geometry(net)[nearest_nodes])

 # okay, now we want to filter the network by the points
 filtered = net %>%
  activate("edges") %>%
  st_filter(snapped_pois$geometry)

 # split geometry into lon and lat
 snapped_pois <- snapped_pois %>%
  dplyr::mutate(lon = sf::st_coordinates(.)[,1],
                lat = sf::st_coordinates(.)[,2])

 # plot
 ggplot()+
  geom_sf(data = roads_all$osm_lines, col = 'grey40', size = .4) +
  geom_sf(data = st_as_sf(filtered, "edges"), col = "#f88379") +
  #geom_point(data = snapped_pois, aes(x = lon, y = lat), size = .5, col = '#f88379') +
  theme_void() +
  theme(plot.background=element_rect(fill = 'grey20', color=NA),
        panel.background = element_rect(fill = 'grey20', color=NA),
        plot.margin = unit(c(t=.2,r=.2,b=1,l=.2), "cm"),
        plot.caption = element_text(color = "#f88379", size = 70, 
                                    hjust = .65, vjust = .5, family = "abril"))+
  labs(caption = 'Philly, 2022')

 # save
 ggsave('philly_map.png', dpi = 300, height = 7, width = 5)
	# Create a map of traces using Google Location History #

	# get your google maps data:
	# https://www.howtogeek.com/725241/how-to-download-your-google-maps-data/

	library(dplyr) # data wrangling
	library(lubridate) # date parsing
	library(ggplot2) # plotting
	library(jsonlite) # reading in json file
	library(sf) # spatial
	library(osmdata) # osm api
	library(sfnetworks) # spatial
	library(showtext) # fonts

	# switch to current wd
	setwd(dirname(rstudioapi::getActiveDocumentContext()$path))

	# add font
	font_add_google(name = "Abril Fatface", family = "abril")
	showtext_auto()

	# read in file
	myHistory <- jsonlite::fromJSON("Takeout/Location History/Records.json", flatten=TRUE)

	# data cleaning & transformations
	# based on https://medium.com/geekculture/explore-your-activity-on-google-with-r-how-to-analyze-and-visualize-your-location-history-2ea8edabe733
	# grab just locations
	myData <- myHistory$locations

	myData <- myData %>%
	filter(activity!="NULL")

	# clean up dates and lat/long
	myData <- myData %>%
	mutate(time = as_datetime(myData$timestamp),
	date = date(time),
	hour = paste(hour(time),minute(time),sep=":"),
	week = isoweek(time),
	month = paste(month(time)),
	year = isoyear(time),
	latitude = latitudeE7/1e7,
	longitude= longitudeE7/1e7) %>%
	select(-timestamp,-latitudeE7,-longitudeE7,-time)

	### !! EDIT THESE DEPENDING ON YOUR LOCATION AND DATES !! ###
	# grab everything before the end of April
	philly_data <- filter(myData, date < '2022-04-28')

	# add underlying osm map
	bbx = c(-75.20, 39.92,
	-75.13, 39.98)

	# all roads for dataviz
	roads_all <- bbx %>%
	opq() %>%
	add_osm_feature(key = 'highway') %>%
	osmdata_sf()

	# larger roads only to snap to network
	# takes a second to run
	highways <- bbx %>%
	opq() %>%
	add_osm_feature(
	key = "highway",
	value = c(
	'road',
	"motorway",
	"trunk",
	"primary",
	"secondary",
	"tertiary",
	"motorway_link",
	"trunk_link",
	"primary_link",
	"secondary_link",
	"tertiary_link",
	'unclassified'
	)
	) %>%
	osmdata_sf()

	# now, we want to link the points to the nearest osm segment
	# build sf network
	my_roads <- st_geometry(highways$osm_lines)
	net = as_sfnetwork(my_roads, directed = FALSE)

	# grab coordinates
	philly_coords <- myData %>%
	st_as_sf(coords = c('longitude', 'latitude'))
	st_crs(philly_coords) = 4326

	# snap points to their nearest node
	nearest_nodes = st_nearest_feature(philly_coords, net)

	snapped_pois = philly_coords %>%
	st_set_geometry(st_geometry(net)[nearest_nodes])

	# okay, now we want to filter the network by the points
	filtered = net %>%
	activate("edges") %>%
	st_filter(snapped_pois$geometry)

	# split geometry into lon and lat
	snapped_pois <- snapped_pois %>%
	dplyr::mutate(lon = sf::st_coordinates(.)[,1],
	lat = sf::st_coordinates(.)[,2])

	# plot
	ggplot()+
	geom_sf(data = roads_all$osm_lines, col = 'grey40', size = .4) +
	geom_sf(data = st_as_sf(filtered, "edges"), col = "#f88379") +
	#geom_point(data = snapped_pois, aes(x = lon, y = lat), size = .5, col = '#f88379') +
	theme_void() +
	theme(plot.background=element_rect(fill = 'grey20', color=NA),
	panel.background = element_rect(fill = 'grey20', color=NA),
	plot.margin = unit(c(t=.2,r=.2,b=1,l=.2), "cm"),
	plot.caption = element_text(color = "#f88379", size = 70,
	hjust = .65, vjust = .5, family = "abril"))+
	labs(caption = 'Philly, 2022')

	# save
	ggsave('philly_map.png', dpi = 300, height = 7, width = 5)