# our very first code ----

## Downloading the data ----
download.file(
  url = "https://kamapu.gitlab.io/multivar/data/course-data.zip",
  destfile = "course-data.zip", method = "curl")
unzip("course-data.zip", overwrite = TRUE)
unlink("course-data.zip")

# Load the data
getwd()

wetlands <- readRDS("wetlands.rds")

names(wetlands)
head(wetlands$env)
tail(wetlands$env)

head(wetlands$spp)
head(wetlands$obs)

head(wetlands$cross)

dim(wetlands$cross)
325 * 335
dim(wetlands$obs)

## Reshaping data ----
library(tidyr)

cross_table <- pivot_wider(
  data = wetlands$obs[c("plot_id", "taxon_code",
                        "cover_perc")],
      names_from = "taxon_code",
      values_from = "cover_perc",
      values_fill = 0,
      values_fn = mean)
head(cross_table)

cross_table <- as.data.frame(cross_table)
head(cross_table)
rownames(cross_table) <- cross_table$plot_id
head(cross_table)

cross_table <- cross_table[names(cross_table) !=
                             "plot_id"]
cross_table <- as.matrix(cross_table)
dim(cross_table)
dim(wetlands$cross)

## using vegtable ----
library(vegtable)

cross_table2 <- crosstable(
  formula = cover_perc ~ plot_id + taxon_code,
  FUN = mean,
  data = wetlands$obs,
  na_to_zero = TRUE,
  as_matrix = TRUE)

is.matrix(cross_table2)

# Common issues
spp_list <- c("Ranunculus repens", "ranunculus repens")
unique(spp_list)
wetlands$spp$taxon_name

spp_check <- unique(wetlands$spp$taxon_name)
spp_check <- spp_check[order(spp_check)]
spp_check

spp_list <- c("Ranunculus repens", "Ranunculus repens ",
              " Ranunculus repens", "Ranunculus   repens ")
unique(spp_list)

spp_list <- clean_strings(spp_list)
unique(spp_list)

# Consistency
all(wetlands$obs$taxon_code %in% wetlands$spp$taxon_code)
any(is.na(wetlands$obs$taxon_code))

any(duplicated(wetlands$spp$taxon_code))