# ------  to be used to process raw occurrence data -----------------------
# ----  load required libraries -------------------------------------------
libs <- c('tidyverse','lubridate','purrr','here');lapply(libs, library,character.only=TRUE)


# ------- load raw occurrence data ----------------------------------------
all_csv_files <- list.files('data/raw/',pattern = '.csv')

# ------- process occurrence data -----------------------------------------
col_names_files <- purrr::map(here::here('data/raw',all_csv_files),
                              .f = function(csv_files = .x) read_csv(csv_files)%>%names())
cust_reorder_names <- function(fls,f_name){
  tmp_names <- c("Record_Type", "Platform", "ID_Number", "IDCR_code", "dates", 
                 "Year", "Month", "Day", "Latitude", "Longitude", "Summer_Winter")
  
  t_dat <- read_csv(fls)%>%
    dplyr::rename_at(.vars = vars(ends_with("_Typ")),
                     .funs = funs(str_replace_all(.,pattern = "_Typ",replacement = "_Type")))%>%
    dplyr::rename_at(.vars = vars(matches("SW")),
                     .funs = funs(str_replace_all(.,pattern = "SW",replacement = "Summer_Winter")))%>%
    dplyr::select(!!tmp_names)%>%mutate(f_name = str_replace_all(f_name,'.csv',''))
  t_dat
}

all_dat_sorted <-purrr::map2(.x = here::here('data/raw',all_csv_files),.y = all_csv_files,
                            .f = ~cust_reorder_names(fls = .x,f_name = .y))



lapply(all_dat_sorted,function(x) names(x))

# write processed data to a file ------------------------------------------
purrr::map2(.x = all_dat_sorted,.y = str_replace_all(all_csv_files,pattern = '.csv',replacement = '_processed.csv'),
            .f = function(dat = .x,fname = .y) {write_csv(path = here::here('data/tidy',fname),x = dat)})



# ------ combine all species data and write to a file ---------------------
all_data_combined <- do.call('rbind',all_dat_sorted)

all_data_combined <- all_data_combined%>%mutate(species = case_when(
  str_detect(string = f_name,pattern = "^BW") ~ "Bryde whale",
  str_detect(string = f_name,pattern = "^HBW") ~ "Humpback whale",
  str_detect(string = f_name,pattern = "^SRW") ~ "Southern right whale",
  str_detect(string = f_name,pattern = "^SW") ~ "Sperm whale"
),
season = case_when(
  str_detect(f_name,"summer") ~ "Summer",
  str_detect(f_name,"winter") ~ "Winter"
))

write_csv(x = all_data_combined,path = 'data/tidy/processed_combined_all_species_occurence.csv')


# ------- prepare data just for SSDM --------------------------------------
all_dat_ssdm <- all_data_combined%>%dplyr::select(species,season,Longitude,Latitude)%>%
  filter_if(any_vars(is.numeric(.)),any_vars(!is.na(.)))%>%
  unite('species_season',species:season,remove = FALSE,sep = '_')%>%
  dplyr::select(Longitude,Latitude,species_season)



#  ----- write processed data for ssdm ------------------------------------
write_csv(x = all_dat_ssdm,path = 'data/tidy/processed_combined_all_species_for_ssdm.csv')