In ReseauBiodiversiteQuebec/mapselector: Map selector module for Shiny

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  eval = FALSE
)

library(mapselector)

library(tidyverse)
rcoleo_sites <- rcoleo::get_sites()

rcoleo_sites[[1]]$body[[1]] %>% glimpse
 # from Claire's code

leaflet() %>%
  addTiles() %>% # Affichage du fond de carte
  addCircleMarkers(lng = obs_an()$long_site, # Positionnement des sites avec les coordonnées long/lat
                   lat = obs_an()$lat_site,
                   radius = 8, # taille du cercle
                   popup = obs_an()$popup_info, # Ajout de fenêtres pop-up
                   color = obs_an()$col,
                   layerId = obs_an()$site_code)

# lon and lat come from the geom coordinates

# sf point object??

ppopup_info = paste0("<b> Region</b> ",
                   Region,
                   "<br/>","<b> id_cellule</b> ",
                   cell_id,
                   "<br/>",
                   "<b> code_cellule</b> ",
                   cell.cell_code,
                   "<br/>",
                   "<b> nom_cellule</b> ",
                   cell.name,
                   "<br/>",
                   "<b> code_site</b> ",
                   site_code,
                   "<br/>",
                   "<b> type_echantillonnage</b> ",
                   type,
                   "<br/>",
                   "<b> annee_creation_site</b> ",
                   open_year)

library(sf)
# st_as_sf(rcoleo_sites[[1]]$body[[1]], coords = "geom.coordinates", dim = "XY")

rcoleo_sites[[1]]$body[[1]]$geom.coordinates %>% lapply(st_point)


site_info <- rcoleo_sites[[1]]$body[[1]]

site_info$lng <- map_dbl(site_info$geom.coordinates, 1)

site_info$lat <- map_dbl(site_info$geom.coordinates, 2)

library(leaflet)

leaflet(site_info) %>%
  addTiles() %>% 
  addCircleMarkers()


rcoleo_sites <- rcoleo::sf_sites()

Doesn't work. here is a better way:

library(tidyverse)
library(sf)
library(leaflet)
# get data
rcoleo_sites <- rcoleo::get_sites()

# get the real info out
site_info_df <- rcoleo_sites[[1]]$body[[1]]

# convert to sf
site_info_df$geom.coordinates <- lapply(site_info_df$geom.coordinates, sf::st_point)

site_info_sf <-  sf::st_as_sf(site_info_df)

site_info_sf %>% 
  leaflet %>% 
  addTiles() %>% 
  addCircleMarkers()

TODO: it might be interesting to apply the same procedure to cells (assuming that this even works! ) See other example from my attempt with ggiraph

adding awesome markers

from Guillaume, via Steve in original app:

    "lac": markerAwsomeMarker({
        icon: "fish",
        markerColor: "darkblue",
        layer: 'Aquatique',
    }),
    "rivière": markerAwsomeMarker({
        icon: "fish",
        markerColor: "blue",
        layer: 'Aquatique',
    }),
    "toundrique": markerAwsomeMarker({
        icon: "leaf",
        markerColor: "orange",
        layer: 'Toundra',
    }),
    "marais": markerAwsomeMarker({
        icon: "frog",
        markerColor: "darkgreen",
        layer: 'Marais',
    }),
    "marais côtier": markerAwsomeMarker({
        icon: "water",
        markerColor: "cadetblue",
        layer: 'Marais côtier',
    }),
    "forestier": markerAwsomeMarker({
        icon: "tree",
        markerColor: "green",
        layer: 'Forestier',
    }),
    "tourbière": markerAwsomeMarker({
        icon: "seedling",
        markerColor: "darkred",
        layer: 'Tourbière',
    }),

and based on the leaflet docs

Seems like the first step is to create an awesomeIconList

# is this too fancy

library(fontawesome)

fa_col <- function(txt) fontawesome::fa(txt, fill = "white")

icon_options <- list(
  lac             = list(text = fa_col("fish"),     markerColor = "darkblue"),
  rivière         = list(text = fa_col("fish"),     markerColor = "blue"),
  toundrique      = list(text = fa_col("leaf"),     markerColor = "orange"),
  marais          = list(text = fa_col("frog"),     markerColor = "darkgreen"),
  "marais côtier" = list(text = fa_col("water"),    markerColor = "cadetblue"),
  forestier       = list(text = fa_col("tree"),     markerColor = "green"),
  tourbière       = list(text = fa_col("seedling"), markerColor = "darkred")
)

awesome_icon_list <- lapply(icon_options, do.call, what = makeAwesomeIcon)

site_icons <- do.call(awesomeIconList, awesome_icon_list)

plot using the dataset

site_info_sf %>% 
  leaflet %>% 
  addTiles() %>% 
  addAwesomeMarkers(icon = leaflet::makeAwesomeIcon(
    text = mapselector::fa_tbl[mapselector::fa_tbl$name == "fish", "svg"],
    iconColor = "black",
    markerColor = "red"
))

site_info_sf %>%  
  leaflet %>% 
  addTiles() %>% 
  addAwesomeMarkers(icon = ~site_icons[type])

You can filter out a habitat type

site_info_sf %>%  
  filter(type != "tourbière") %>% 
  leaflet %>% 
  addTiles() %>% 
  addAwesomeMarkers(icon = ~site_icons[type])

based on the issues in this thread here

group according to site?? is that possible?

set groups and use layer controls

# make awesomeicon functions
site_icons_adders <- lapply(site_icons, function(ic) function(...) addAwesomeMarkers(..., icon = ic))

blank_map <- site_info_sf %>%  
  leaflet %>% 
  addTiles()

runfun <- function(x, fun){
  fun(x)
}

runfun(c(2,3,6), mean)


Reduce(f = runfun, site_icons_adders, init = blank_map)

Ok that gives you all things mapped to the last icon. You need to split up the data first

# filter the data

# split the data

site_info_sf_split <- split(site_info_sf, site_info_sf[["type"]])

# filter the icons -- only what is in data

filter_icons <- site_icons[names(site_info_sf_split)]

# lengths should be same
stopifnot(length(site_info_sf_split) == length(filter_icons))

add_partial_awesome <- function(ic, dat, grp, pt_id) function(map) addAwesomeMarkers(map = map, icon = ic, data = dat, group = grp, layerId = pt_id, popup = pt_id, label = pt_id)

markers <- mapply(add_partial_awesome,
                  filter_icons, 
                  site_info_sf_split, 
                  names(site_info_sf_split),
                  lapply(site_info_sf_split, `[[`, "site_code"))


p <- Reduce(f = runfun, markers, init = blank_map)

addLayersControl(p ,
    overlayGroups = names(site_info_sf_split),
    options = layersControlOptions(collapsed = FALSE)
  )

getting obs off of site

rc <- mapselector::get_rcoleo_sites_sf()

library(dplyr)
camps <- rc$campaigns[[8]][c("type", "opened_at", "closed_at", "site_id")]


rcoleo:::endpoints()$campaigns


camp_list <- purrr::transpose(camps) %>% 
  purrr::map(rcoleo::get_gen, endpoint = rcoleo:::endpoints()$campaigns)


camp_df <- camp_list %>% purrr::map_dfr("body")

camp_df

camp_df$efforts



# obs

as.list(camp_df$id)
forty_two <- rcoleo::get_gen(endpoint = rcoleo:::endpoints()$observations, query = list(campaign_id = 42))

df42 <- forty_two$body %>% do.call(rbind, .)

View(df42)


# note that it goes totally wild and gives back EVERYTHING if you accidenitally name the quiery campaign and not campaign_id