sandbox/test_joins.R

# load libraries
library(npstools) # devtools::load_all()
library(tidyverse)
library(here)
library(glue)
library(fs)
library(dplyr)
library(lubridate)
here = here::here

# load your own configuration, which could be based off package
nps_config_yaml <- system.file(package="npstools", "nps_config.yaml")
cfg <- get_nps_config(nps_config_yaml)

# big memory hog table wrangle
for (park in c("CABR", "SAMO")){
  get_pct_cover_tbl(cfg, park, 2015) %>%
    write_csv(here(glue("sandbox/pct_cover_{park}-2015_post-fix.csv")))
}

# compare pre and post
for (park in c("CABR", "SAMO")){ # park = "CABR" # park = "SAMO"
  pre  <- here(glue("sandbox/pct_cover_{park}-2015_pre-fix.csv")) %>%
    read_csv()
  post <- here(glue("sandbox/pct_cover_{park}-2015_post-fix.csv")) %>%
    read_csv()
  cat(park, "\n")
  if (all(dim(pre) == dim(post))){
    pre == post
  } else {
    cat(glue("dim (rows x cols) PRE ({paste(dim(pre), collapse=' x ')}) != POST ({paste(dim(post), collapse=' x ')})"), "\n")
  }
}

# get_spp_park_tbl(cfg, "CABR")
# get_spp_park_tbl(cfg, "CHIS")
#
# # specify park and year of interest
# sz <- function(o){ format(object.size(o), units = "auto") }
# get_total_eventpoints_tbl(cfg, "CABR") %>% sz()  #  18.2 Kb
# get_total_eventpoints_tbl(cfg, "SAMO") %>% sz()  #  40.5 Kb
# get_total_eventpoints_tbl(cfg, "CHIS") %>% sz()  # 300.6 Kb
#
# pct_cover_tbl %>%
#   select(-Query_type) %>%
#   DT::datatable() %>%
#   DT::formatRound(columns=c("Average", "StdDev"), digits=3)
ecoquants/npstools documentation built on May 10, 2019, 9:50 a.m.