R/1_Manipulate_table_function.R

In CmlMagneville/SmaxNanalysis: What the Package Does (One Line, Title Case)

###########################################################################
##
## Script to code functions to manipulate df before computing maxN values
## and automoatize the computation of df for all/some species
##
## 1_Manipulate_table_function.R
##
## 11/2021
##
## Camille Magneville
##
###########################################################################



#' Create a dataframe with abundance data for different cameras and a given
#' species
#'
#' This function computes a dataframe containing the abundance of a given
#' species for a given period of time. The user gives as many dataframes as
#' there are cameras, a period of time and a species name and the function
#' checks on which cameras the species has been seen and gathers the dataframes
#' of these cameras.
#'
#' @param list_abund_df a list containing as many dataframes as there are
#' cameras. Each dataframe is a species*abundance dataframe and contains the
#' abundance of different species (columns) across continuous time (rows)
#'
#' @param species_nm a character string refering to the name of the species
#' for which dataframes must be gathered. \strong{Be sure that species names
#' is correctly spelled as in the vect_abund_df dataframes}.
#'
#' @param time_start a character string refering to the time when the dataframe
#' should begin.
#'
#' @param time_stop a character string refering to the time when the dataframe
#' should stop.
#'
#' @return the function returns a dataframe containing the abundance of a given
#' species for a given period of time. There are as many columns as there are
#' cameras and there are as many rows as there are timesteps (second) between
#' time_start and time_stop.
#'
#' @importFrom magrittr %>%
#'


gather.abund.df <- function(list_abund_df, species_nm,
                            time_start, time_stop) {


  ## 0 - check that the species names exists in at least one abundance df:

  # create a vector with the names of species present in all the df:
  sp_nm_df <- c()

  for (i in list_abund_df) {
    sp_nm_df <- append(colnames(i), sp_nm_df)
  }

  sp_nm_df <- unique(sp_nm_df)

  # check if studied species in it:
  if (! species_nm %in% sp_nm_df) {
    stop("Species in species_nm is not contained in the dataframes of list_abund_df.
         Please check, it may be mispelled.")
  }


  ## 1 - cut the df between time_start and time_stop:

  # create a vector that contains all seconds between time_start and time_stop:
  time_vect <- hms::as_hms(c(hms::as_hms(time_start):hms::as_hms(time_stop)))
  time_vect <- as.character(time_vect)

  # only keep rows in the list_abund_df that span in time_vect and keep the given sp:
  for (i in c(1:length(list_abund_df))) {
    list_abund_df[[i]] <- list_abund_df[[i]][which(rownames(list_abund_df[[i]]) %in% time_vect), ]
  }


  ## 2 - only keep df that present the studied sp:

  # create a vector that will contain the names of df containing the species
  cam_keep <- c()

  # fill the vector with df containing the studied species:
  for (i in c(1:length(list_abund_df))) {
    if (species_nm %in% colnames(list_abund_df[[i]])) {
      cam_keep <- append(list_abund_df[i], cam_keep)
    }
  }


  ## 3 - Combine df:

  # first add the 1st column:
  data <- data.frame(cam_keep[[1]][, species_nm])
  rownames(data) <- rownames(cam_keep[[1]])

  # then, if more than one camera (df in cam_keep) to keep:
  if (length(cam_keep) > 1) {
    for (i in c(2:length(cam_keep))) {
      df <- as.data.frame(cam_keep[[i]][, species_nm])
      rownames(df) <- rownames(cam_keep[[i]])
      data <- dplyr::bind_cols(data, df)
    }
  }

  # add columns names as camera names:
  colnames(data) <- names(cam_keep)


  ## 4 - Return
  return(list(abund_df = data,
              cam_nm = names(cam_keep)))

}



#' Automatise the computation of abundance df for each Pose for a given
#' set of species
#'
#' This function computes a list of abundance dataframe for the different
#' Poses for a given set of species or all species
#'
#' @param abund_allposes_list a list containing the three dataframes of the
#' three poses
#'
#' @param species_set a vector containing the name of species for which
#' dataframes must be retrieved
#'
#' @return the function returns a list for each species of list containing the
#' three df for the given species and the three poses
#'

automat.abund.df <- function(abund_allposes_list,
                             species_set) {



  # create a list that will contain the abundance df for all species for all poses:
  abund_allsp_list <- list()

  # loop on the species:
  for (j in species_set) {

    # create a list that will contain the df for the 3 poses of the given sp:
    sp_list <- list()

    # loop on the different poses:
    for (i in (1:length(abund_allposes_list))) {

      # get the list for the given pose:
      abund_list <- abund_allposes_list[[i]]

      # check that species is present during the given pose:
      if (j %in% colnames(abund_list[[1]])) {


        # if first Pose:
        if (i == 1) {
          abund_gather <- gather.abund.df(list_abund_df = abund_list, species_nm = j,
                                          time_start = "07:30:00", time_stop = "08:30:00")
          abund_df1 <- abund_gather$abund_df
        }

        # if second Pose:
        if (i == 2) {
          abund_gather <- gather.abund.df(list_abund_df = abund_list, species_nm = j,
                                          time_start = "11:30:00", time_stop = "12:30:00")
          abund_df2 <- abund_gather$abund_df
        }

        # if third Pose:
        if (i == 3) {
          abund_gather <- gather.abund.df(list_abund_df = abund_list, species_nm = j,
                                          time_start = "15:30:00", time_stop = "16:30:00")
          abund_df3 <- abund_gather$abund_df
        }

      }

    }

      # update the species list with the df of the 3 poses:
      sp_list <- list(abund_df1, abund_df2, abund_df3)

      # update the global list:
      abund_allsp_list[[j]] <-  sp_list
  }

  # return the global list of abundance df:
  return(abund_allsp_list)

}