Issues/98 (#99)

* edits based on lintr checks

* changes based on lintr

* suppressMessages when loading magrittr

* Don't print logs

R/correlation_threshold.R

@@ -16,7 +16,7 @@
 #' 
 #' @examples
 #' 
-#' library(magrittr)
+#' suppressMessages(suppressWarnings(library(magrittr)))
 #' sample <- tibble::data_frame(
 #'    x = rnorm(30),
 #'    y = rnorm(30)/1000

R/count_na_rows.R

@@ -29,5 +29,4 @@ count_na_rows <- function(population, variables) {
     dplyr::summarize_at(variables, dplyr::funs(sum)) %>%
     dplyr::collect() %>%
     data.frame()
-
 }

R/drop_na_columns.R

@@ -25,17 +25,17 @@
 #' @export
 drop_na_columns <- function(population, variables, cutoff = 0.05) {
   cutoff <- rlang::enquo(cutoff)
-  
+
   nrows <-
     population %>%
     dplyr::tally() %>%
     dplyr::collect() %>%
     magrittr::extract2("n")
 
   count <- rlang::sym("count")
-  
+
   feature <- rlang::sym("feature")
-  
+
   percent <- rlang::sym("percent")
 
   population %>%
@@ -44,6 +44,6 @@ drop_na_columns <- function(population, variables, cutoff = 0.05) {
     dplyr::collect() %>%
     tidyr::gather(!!feature, !!count, !!!variables) %>%
     dplyr::mutate(!!percent := (!!count) / nrows) %>%
-    dplyr::filter((!!percent) > (!!cutoff)) %>%
+    dplyr::filter( (!!percent) > (!!cutoff)) %>%
     magrittr::extract2("feature")
 }

R/drop_na_rows.R

@@ -25,9 +25,9 @@
 drop_na_rows <- function(population, variables) {
 
   key <- rlang::sym("key")
-  
+
   value <- rlang::sym("value")
-  
+
   rowname_temp <- rlang::sym("rowname_temp")
 
   if (is.data.frame(population)) {
@@ -37,13 +37,13 @@ drop_na_rows <- function(population, variables) {
       dplyr::filter(!is.na(!!value)) %>%
       tidyr::spread(!!key, !!value) %>%
       dplyr::select(-!!rowname_temp)
-    
-  } else { 
-    
+
+  } else {
+
     # Coalesce() must have at least 2 arguments.
-    if(length(variables) == 1)
+    if (length(variables) == 1)
       variables <- c(variables, variables)
-    
+
     population %>%
       dplyr::filter_(.dots =
           sprintf("!is.null(coalesce(%s))",

R/generalized_log.R

@@ -27,9 +27,9 @@ generalized_log <- function(population, variables, offset = 1) {
     x <- rlang::sym(variable)
 
     population %<>%
-      dplyr::mutate(!!x := log( ((!!x) + ((!!x) ^ 2 + (!!offset) ^ 2) ^ 0.5 ) / 2))
+      dplyr::mutate(!!x :=
+                      log( ( (!!x) + ( (!!x) ^ 2 + (!!offset) ^ 2) ^ 0.5 ) / 2))
   }
 
   population
 }
-

R/normalize.R

@@ -17,7 +17,7 @@
 #' @importFrom stats cor mad median sd setNames
 #'
 #' @examples
-#' library(magrittr)
+#' suppressMessages(suppressWarnings(library(magrittr)))
 #' population <- tibble::data_frame(
 #'    Metadata_group = c("control", "control", "control", "control",
 #'                       "experiment", "experiment", "experiment", "experiment"),
@@ -30,7 +30,8 @@
 #' cytominer::normalize(population, variables, strata, sample, operation = "standardize")
 #' 
 #' @export
-normalize <- function(population, variables, strata, sample, operation = "standardize", ...) {
+normalize <- function(population, variables, strata, sample,
+                      operation = "standardize", ...) {
   scale <- function(data, location, dispersion, variables) {
     if (is.data.frame(data)) {
       futile.logger::flog.debug(paste0("\t\tUsing base::scale (data is ",
@@ -59,7 +60,7 @@ normalize <- function(population, variables, strata, sample, operation = "standa
         s <- dispersion[[variable]]
 
         data %<>%
-          dplyr::mutate(!!x := ((!!x) - m) / s )
+          dplyr::mutate(!!x := ( (!!x) - m) / s )
 
       }
 

R/replicate_correlation.R

@@ -82,8 +82,7 @@ replicate_correlation <-
       strata <- c(strata, replicate_by)
     }
 
-    foreach::foreach(variable = variables, .combine = rbind) %dopar%
-    {
+    foreach::foreach(variable = variables, .combine = rbind) %dopar% {
 
       sample %>%
         split(.[split_by]) %>%
@@ -94,7 +93,8 @@ replicate_correlation <-
                 dplyr::arrange_(.dots = strata) %>%
                 dplyr::select_(.dots = c(strata, variable, replicate_by)) %>%
                 tidyr::spread_(replicate_by, variable) %>%
-                dplyr::select_(~-dplyr::one_of(setdiff(strata, replicate_by))) %>%
+                dplyr::select_(~-dplyr::one_of(setdiff(strata,
+                                                       replicate_by))) %>%
                 stats::cor()
               median(correlation_matrix[upper.tri(correlation_matrix)])
             }) %>%

R/transform.R

@@ -20,7 +20,8 @@
 #' @importFrom magrittr %>%
 #' @importFrom magrittr %<>%
 #' @export
-transform <- function(population, variables, operation = "generalized_log", ...) {
+transform <- function(population, variables,
+                      operation = "generalized_log", ...) {
   if (operation == "generalized_log") {
     generalized_log(population, variables, ...)
   } else {

R/variable_importance.R

@@ -45,17 +45,18 @@
 #' @importFrom magrittr %>%
 #' @importFrom magrittr %<>%
 #' @export
-variable_importance <- function(sample, variables, operation = "replicate_correlation", ...) {
+variable_importance <- function(sample, variables,
+                                operation = "replicate_correlation", ...) {
   if (operation == "replicate_correlation") {
    importance <- replicate_correlation(sample, variables, ...)
-   
+
   } else {
     error <- paste0("undefined operation `", operation, "'")
 
     futile.logger::flog.error(msg = error)
 
     stop(error)
-    
+
   }
 
   importance

R/variable_select.R

@@ -15,7 +15,7 @@
 #' # In this example, we use `correlation_threshold` as the operation for 
 #' # variable selection.
 #' 
-#' library(magrittr)
+#' suppressMessages(suppressWarnings(library(magrittr)))
 #' population <- tibble::data_frame(
 #'    x = rnorm(100),
 #'    y = rnorm(100)/1000
@@ -35,6 +35,8 @@
 #' 
 #' head(population)
 #' 
+#' futile.logger::flog.threshold(futile.logger::ERROR)
+#' 
 #' variable_select(population, variables, sample, operation) %>% head()
 #'
 #' @importFrom magrittr %>%

R/variance_threshold.R

@@ -35,7 +35,8 @@ variance_threshold <- function(variables, sample) {
 
     lunique <- apply(x, 2, function(data) length(unique(data[!is.na(data)])))
 
-    which((ratio > 19 & (100 * lunique / apply(x, 2, length)) <= 10) | (lunique == 1) | apply(x, 2, function(data) all(is.na(data))))
+    which( (ratio > 19 & (100 * lunique / apply(x, 2, length)) <= 10) |
+            (lunique == 1) | apply(x, 2, function(data) all(is.na(data))))
   }
 
   excluded_indexes <-

tests/testthat/test-aggregate.R

@@ -9,7 +9,7 @@ test_that("`aggregate` aggregates data", {
     )
 
   db <- dplyr::src_sqlite(":memory:", create = T)
-  
+
   data <- dplyr::copy_to(db, data)
 
   expect_equal(
@@ -31,7 +31,7 @@ test_that("`aggregate` aggregates data", {
       dplyr::collect(),
     data %>%
       dplyr::group_by(g) %>%
-      dplyr::summarise_at(.funs =dplyr::funs(median), .vars = c("x", "y"))
+      dplyr::summarise_at(.funs = dplyr::funs(median), .vars = c("x", "y"))
   )
 
   expect_equal(
@@ -42,6 +42,7 @@ test_that("`aggregate` aggregates data", {
       dplyr::collect(),
     data %>%
       dplyr::group_by(g) %>%
-      dplyr::summarise_at(.funs =c(dplyr::funs(mean), dplyr::funs(sd)), .vars = c("x", "y"))
+      dplyr::summarise_at(.funs = c(dplyr::funs(mean), dplyr::funs(sd)),
+                          .vars = c("x", "y"))
   )
 })

tests/testthat/test-correlation_threshold.R

@@ -1,14 +1,15 @@
 context("correlation_threshold")
 
-test_that("`correlation_threshold` selects variables that are not highly correlated", {
+test_that(
+  "`correlation_threshold` selects variables that are not highly correlated", {
 
   set.seed(123)
   data <- data.frame(x = rnorm(30))
   data$y <- rnorm(30) / 1000
   data$z <- data$x + rnorm(30) / 1000
 
   db <- DBI::dbConnect(RSQLite::SQLite(), ":memory:")
-  
+
   data <- dplyr::copy_to(db, data)
 
   expect_equal(

tests/testthat/test-count_na_rows.R

@@ -7,9 +7,9 @@ test_that("`count_na_rows` returns the frequency of NAs per variable", {
   data[c(2, 3, 5), "y"] <- NA
 
   db <- DBI::dbConnect(RSQLite::SQLite(), ":memory:")
-  
+
   data <- dplyr::copy_to(db, data)
-  
+
   expect_equal(
     count_na_rows(population = data,
                     variables = c("x", "y", "z")),