From 4bf0e6258558c801e2152ed99a4bb09f6225b95a Mon Sep 17 00:00:00 2001
From: Mattia Greco <mattia_greco@outlook.com>
Date: Mon, 8 Jan 2024 13:51:18 +0100
Subject: [PATCH] modify filter functions

---
 R/filter.R                   | 89 +++++++++++++++++++++++++-----------
 man/filter_by_coordinates.Rd |  2 +-
 man/filter_by_month.Rd       |  2 +-
 man/filter_by_species.Rd     |  4 +-
 man/filter_by_year.Rd        |  2 +-
 5 files changed, 68 insertions(+), 31 deletions(-)

diff --git a/R/filter.R b/R/filter.R
index ee4e579..9a30e56 100644
--- a/R/filter.R
+++ b/R/filter.R
@@ -12,12 +12,12 @@
 reshape_forcis <- function(data){
   
   if (get_data_type(data) %in% c("CPR North")) {
-    stop(paste0("This function is not designed to work with 'CPR North' data"), call. = FALSE) 
+    stop("This function is not designed to work with 'CPR North' data", call. = FALSE) 
   }
   
   taxa_cols <- get_species_names(data) 
   metadat_cols <- get_required_columns()
-  dat_reshaped <- dat_reshaped %>% 
+  dat_reshaped <- data %>% 
     select(all_of(taxa_cols),metadat_cols) %>% 
     pivot_longer(all_of(taxa_cols), 
                  names_to  = 'taxa', 
@@ -29,7 +29,7 @@ reshape_forcis <- function(data){
 
 #' Filter forcis data by year of sampling
 #'
-#' @param data forcis data in long format
+#' @param data forcis data
 #' @param years numeric vector of selected years
 #'
 #' @return A `data.frame`.
@@ -40,21 +40,32 @@ reshape_forcis <- function(data){
 
 filter_by_year <- function(data,years){
   year_vector <- as.numeric(years)
-  filtered_dat <- data %>% 
-    filter(! is.na(.data$profile_date_time)) %>% 
-    mutate(new_profile_date_time = dmy(.data$profile_date_time)) %>% 
-             mutate(year=year(.data$new_profile_date_time)) %>% 
-             filter(.data$year %in% year_vector) %>% 
-             select(-c(.data$year,.data$new_profile_date_time))
-           
-           return(filtered_dat)
-           
+  
+  if (get_data_type(data)=="Sediment trap"){
+    filtered_dat <- data %>%
+      filter(! is.na(.data$sample_date_time_start)) %>%
+      mutate(new_sample_date_start=gsub(' .*','',sample_date_time_start)) %>% 
+      mutate(new_sample_date_start = dmy(.data$new_sample_date_start)) %>%
+      mutate(year=year(.data$new_sample_date_start)) %>% 
+      filter(.data$year %in% year_vector) %>%
+      select(-c(.data$year,.data$new_sample_date_start))
+    return(filtered_dat)
+    
+  } else {
+    filtered_dat <- data %>% 
+      filter(! is.na(.data$profile_date_time)) %>% 
+      mutate(new_profile_date_time = dmy(.data$profile_date_time)) %>% 
+      mutate(year=year(.data$new_profile_date_time)) %>% 
+      filter(.data$year %in% year_vector) %>% 
+      select(-c(.data$year,.data$new_profile_date_time))
+    return(filtered_dat)
+  }
 }
 
 
 #' Filter forcis data by month of sampling
 #'
-#' @param data forcis data in long format
+#' @param data forcis data 
 #' @param months  numeric vector of selected months
 #'
 #' @return A `data.frame`.
@@ -66,22 +77,32 @@ filter_by_year <- function(data,years){
 filter_by_month <- function(data,months){
   
   month_vector <- as.numeric(months)
-  
-  filtered_dat <- data %>% 
-    filter(! is.na(.data$profile_date_time)) %>% 
-    mutate(new_profile_date_time =dmy(.data$profile_date_time)) %>% 
-    mutate(month=month(.data$new_profile_date_time)) %>% 
-    filter(.data$month %in% month_vector)%>% 
-    select(-c(.data$month,.data$new_profile_date_time))
-  
-  return(filtered_dat)
-  
+  if (get_data_type(data)=="Sediment trap"){
+    filtered_dat <- data %>%
+      filter(! is.na(.data$sample_date_time_start)) %>%
+      mutate(new_sample_date_start=gsub(' .*','',sample_date_time_start)) %>% 
+      mutate(new_sample_date_start = dmy(.data$new_sample_date_start)) %>%
+      mutate(month=month(.data$new_sample_date_start)) %>% 
+      filter(.data$month %in% month_vector) %>%
+      select(-c(.data$month,.data$new_sample_date_start))
+    return(filtered_dat)
+    
+  } else {
+    filtered_dat <- data %>% 
+      filter(! is.na(.data$profile_date_time)) %>% 
+      mutate(new_profile_date_time =dmy(.data$profile_date_time)) %>% 
+      mutate(month=month(.data$new_profile_date_time)) %>% 
+      filter(.data$month %in% month_vector)%>% 
+      select(-c(.data$month,.data$new_profile_date_time))
+    
+    return(filtered_dat)
+  } 
 }
 
 
 #' Filter forcis data by coordinate square 
 #'
-#' @param data forcis data in long format
+#' @param data forcis data 
 #' @param coord_square a numeric vector containing in this order minimum latitute,
 #' minimum longitude, maximum latitude, maximum longitude
 #'
@@ -113,19 +134,33 @@ filter_by_coordinates <- function(data, coord_square){
 #'
 #' @param data forcis data in long format, except for CPR North data
 #' @param species a character vector listing species of interest
-#'
+#' @param remove_NAs logical, If FALSE, retains all taxa including those with NA counts
 #' @return A `data.frame`
 #' @export
 #'
 #' @examples
 #' ## ADD EXAMPLE ----
 
-filter_by_species <- function (data,species ){
+filter_by_species <- function (data,species, remove_NAs=TRUE ){
   my_species <- as.character(species)
   
+  taxa_cols <- get_species_names(data) 
+  
+  if (length(taxa_cols)>0) {
+    stop("This function requires data in long format", call. = FALSE) 
+  }
+  
+  if (get_data_type(data) %in% c("CPR North")) {
+    stop("This function is not designed to work with 'CPR North' data", call. = FALSE) 
+  }
   filtered_dat <- data %>% 
     filter(.data$taxa %in% my_species)
-  # filter(! is.na(counts))
+  
+  if(remove_NAs) {
+    
+    filtered_dat <- filtered_dat %>% 
+      filter(! is.na(.data$counts))
+  }
   
   return(filtered_dat)
 }
diff --git a/man/filter_by_coordinates.Rd b/man/filter_by_coordinates.Rd
index 9939af1..1353e2b 100644
--- a/man/filter_by_coordinates.Rd
+++ b/man/filter_by_coordinates.Rd
@@ -7,7 +7,7 @@
 filter_by_coordinates(data, coord_square)
 }
 \arguments{
-\item{data}{forcis data in long format}
+\item{data}{forcis data}
 
 \item{coord_square}{a numeric vector containing in this order minimum latitute,
 minimum longitude, maximum latitude, maximum longitude}
diff --git a/man/filter_by_month.Rd b/man/filter_by_month.Rd
index b28736f..ee112fa 100644
--- a/man/filter_by_month.Rd
+++ b/man/filter_by_month.Rd
@@ -7,7 +7,7 @@
 filter_by_month(data, months)
 }
 \arguments{
-\item{data}{forcis data in long format}
+\item{data}{forcis data}
 
 \item{months}{numeric vector of selected months}
 }
diff --git a/man/filter_by_species.Rd b/man/filter_by_species.Rd
index 347b9b5..2f6fe1d 100644
--- a/man/filter_by_species.Rd
+++ b/man/filter_by_species.Rd
@@ -4,12 +4,14 @@
 \alias{filter_by_species}
 \title{Filter forcis data by species}
 \usage{
-filter_by_species(data, species)
+filter_by_species(data, species, remove_NAs = TRUE)
 }
 \arguments{
 \item{data}{forcis data in long format, except for CPR North data}
 
 \item{species}{a character vector listing species of interest}
+
+\item{remove_NAs}{logical, If FALSE, retains all taxa including those with NA counts}
 }
 \value{
 A \code{data.frame}
diff --git a/man/filter_by_year.Rd b/man/filter_by_year.Rd
index b038dd8..eae2dbd 100644
--- a/man/filter_by_year.Rd
+++ b/man/filter_by_year.Rd
@@ -7,7 +7,7 @@
 filter_by_year(data, years)
 }
 \arguments{
-\item{data}{forcis data in long format}
+\item{data}{forcis data}
 
 \item{years}{numeric vector of selected years}
 }