how create coordinates that include c(lng,lat) in a column? about deckgl

[crazycapivara]

Hi,

data must NOT be JSON. Sorry, I have to update the docs showing how to use R data objects.

You just need to pass the column names for lat and lng via get_position R function, e. g.

scatterplot:

deckgl(latitude = -24.655, longitude = 176.9, zoom = 4) %>%
   add_scatterplot_layer(
      data = quakes,
      getPosition = get_position("lat", "long"),
      getRadius = JS("d => d.mag * 100"),
      radiusScale = 60,
      getColor = JS("d => [Math.sqrt(d.depth) * 20, 130, 10]")
   )

icon:

deckgl(latitude = -24.655, longitude = 176.9, zoom = 4) %>%
   add_icon_layer(
      data = quakes,
      getPosition = get_position("lat", "long"),
      getSize = get_property("mag"),
      getColor = JS("d => [Math.sqrt(d.depth) * 20, 130, 10]")
   )

[crazycapivara]

Another example using data from the deckgl package

sample_data <- deckgl::bart_stations

sample_data$radius <- sqrt(sample_data$exits)

properties <- list(
  getPosition = get_position("lat", "lng"),
  getRadius = get_property("radius"),
  radiusScale = 6,
  getColor = c(255, 140, 20)
)

deck <- deckgl(zoom = 10.5, pitch = 35) %>%
  add_scatterplot_layer(data = sample_data, properties = properties) %>%
  add_mapbox_basemap()

if (interactive()) deck

[crazycapivara]

see also road-safety-example

It really helped
Many thanks @crazycapivara

Another example using data from the deckgl package

sample_data <- deckgl::bart_stations

sample_data$radius <- sqrt(sample_data$exits)

properties <- list(
  getPosition = get_position("lat", "lng"),
  getRadius = get_property("radius"),
  radiusScale = 6,
  getColor = c(255, 140, 20)
)

deck <- deckgl(zoom = 10.5, pitch = 35) %>%
  add_scatterplot_layer(data = sample_data, properties = properties) %>%
  add_mapbox_basemap()

if (interactive()) deck

hi @crazycapivara
Some other questions：

head(a)
lng lat
1 116.71148 39.434895
2 116.71148 39.434895
3 116.71148 39.434895
4 116.71144 39.435100
5 116.71144 39.435060
6 116.71140 39.435090

How to get path, Is get_position still valid?

when I try that ,I get nothing
code as follow:
properties <- list(
pickable = TRUE,
widthScale = 20,
widthMinPixels = 2,
getPosition = get_position("lat", "lng"),
getColor = c(255, 140, 20),
getWidth = 5

,getTooltip = get_property("pd")

)

deck <- deckgl(latitude = 38.38687967,longitude = 114.42064498,pitch = 25, zoom = 5.5) %>%
add_path_layer(data = a, properties = properties) %>%
add_mapbox_basemap()

if (interactive()) deck

properties <- list(
pickable = TRUE,
widthScale = 20,
widthMinPixels = 2,
getPosition = get_position("lat", "lng"),
getColor = c(255, 140, 20),
getWidth = 5
)

deck <- deckgl(latitude = 38.38687967,longitude = 114.42064498,pitch = 25, zoom = 5.5) %>%
add_path_layer(data = a, properties = properties) %>%
add_mapbox_basemap()

if (interactive()) deck

[crazycapivara]

Hi @liygzting,
you need to use the right data format as specified in deckgl-path-layer-api-reference:

// Data format:
[
      {
        path: [[-122.4, 37.7], [-122.5, 37.8], [-122.6, 37.85]],
        name: 'Richmond - Millbrae',
        color: [255, 0, 0]
      },
      ...
]

Furthermore, you need to use the getPath property as shown in the docs.

So, you need a list column in your data.frame containing the coords of the path(es).

Here is an example using the bart_stations data set:

coords <- deckgl::bart_stations[, c("lng", "lat")] %>%
  as.matrix()

data <- tibble::tibble(
  tooltip = "Bart Stations",
  path = list(coords)
)

# Check, if your data format is correct
jsonlite::toJSON(data) %>% print()

properties <- list(
  pickable = TRUE,
  widthScale = 20,
  widthMinPixels = 2,
  getPath = get_property("path"),
  getColor = c(250, 140, 20),
  getWidth = 5,
  getTooltip = get_property("tooltip")
)

deckgl(zoom = 9, pickingRadius = 20) %>%
  add_path_layer(data = data, properties = properties) %>%
  add_icon_layer(
    data = bart_stations,
    getPosition = get_position("lat", "lng"),
    getColor = c(10, 140, 50),
    getTooltip = get_property("name")
  )

[crazycapivara]

Alternatively, you can create the data using a list of lists:

coords <- deckgl::bart_stations[, c("lng", "lat")] %>%
  as.matrix()

data <- list(
  list(
    tooltip = "Bart Stations",
    path = coords
  )
)

thanks @crazycapivara ,
for path , scatter,icon or grid etc, I change data to tibble not dataframe so that i can plot most of layer.
Does that mean you can just use this？
Or is there anything else to notice？

[crazycapivara]

Nope, you can use a data.frame aswell. A tibble is just a variant and often more comfortable to handle. That's why I prefer it, e. g. when using list columns. Here is how to create a data.frame with list column which works fine:

# coords same as above
df <- data.frame(
  tooltip = "Bart Stations",
  path = I(list(coords))
)

For all layers expecting point data it's fine to have a data.frame with lat and lng columns.

It is planned to support sf objects, see #81.

thanks @crazycapivara
It really makes sense
Since you told me this good method, I decided to start using tibble too.