Merge pull request #1994 from Gilead-BioStats/fix-1976

Closes #1976 wrap all our stop-esque calls into the log4r setup
Gilead-BioStats · Dec 18, 2024 · 5cc2056 · 5cc2056
2 parents 9332851 + 4045ddf
commit 5cc2056
Show file tree

Hide file tree

Showing 29 changed files with 170 additions and 214 deletions.
diff --git a/NAMESPACE b/NAMESPACE
@@ -131,6 +131,7 @@ importFrom(rlang,":=")
 importFrom(rlang,.data)
 importFrom(rlang,as_label)
 importFrom(rlang,as_name)
+importFrom(rlang,caller_env)
 importFrom(rlang,check_installed)
 importFrom(rlang,enquo)
 importFrom(rlang,enquos)

diff --git a/R/Analyze_Fisher.R b/R/Analyze_Fisher.R
@@ -44,13 +44,11 @@ Analyze_Fisher <- function(
   dfTransformed,
   strOutcome = "Numerator"
 ) {
-  stopifnot(
-    "dfTransformed is not a data.frame" = is.data.frame(dfTransformed),
-    "GroupID or the value in strOutcome not found in dfTransformed" = all(c("GroupID", strOutcome) %in% names(dfTransformed)),
-    "NA value(s) found in GroupID" = all(!is.na(dfTransformed[["GroupID"]])),
-    "strOutcome must be length 1" = length(strOutcome) == 1,
-    "strOutcome is not character" = is.character(strOutcome)
-  )
+  stop_if(cnd = !is.data.frame(dfTransformed), message = "dfTransformed is not a data.frame")
+  stop_if(cnd = !all(c("GroupID", strOutcome) %in% names(dfTransformed)), message = "GroupID or the value in strOutcome not found in dfTransformed")
+  stop_if(cnd = !all(!is.na(dfTransformed[["GroupID"]])), message = "NA value(s) found in GroupID")
+  stop_if(cnd = length(strOutcome) != 1, message = "strOutcome must be length 1")
+  stop_if(cnd = !is.character(strOutcome), message = "strOutcome is not character")
 
   fisher_model <- function(site) {
     SiteTable <- dfTransformed %>%

diff --git a/R/Analyze_Identity.R b/R/Analyze_Identity.R
@@ -21,10 +21,8 @@
 #' @export
 
 Analyze_Identity <- function(dfTransformed, strValueCol = "Metric") {
-  stopifnot(
-    "dfTransformed is not a data.frame" = is.data.frame(dfTransformed),
-    "strValueCol not found in dfTransformed" = strValueCol %in% names(dfTransformed)
-  )
+  stop_if(!is.data.frame(dfTransformed), message = "dfTransformed is not a data.frame")
+  stop_if(!(strValueCol %in% names(dfTransformed)), message = "strValueCol not found in dfTransformed")
 
   dfAnalyzed <- dfTransformed %>%
     mutate(

diff --git a/R/Analyze_NormalApprox.R b/R/Analyze_NormalApprox.R
@@ -36,13 +36,14 @@ Analyze_NormalApprox <- function(
   dfTransformed,
   strType = "binary"
 ) {
-  stopifnot(
-    "dfTransformed is not a data.frame" = is.data.frame(dfTransformed),
-    "One or more of these columns not found: GroupID, GroupLevel, Denominator, Numerator, Metric" =
-      all(c("GroupID", "GroupLevel", "Denominator", "Numerator", "Metric") %in% names(dfTransformed)),
-    "NA value(s) found in GroupID" = all(!is.na(dfTransformed[["GroupID"]])),
-    "strType is not 'binary' or 'rate'" = strType %in% c("binary", "rate")
+
+  stop_if(cnd = !is.data.frame(dfTransformed), message ="dfTransformed is not a data.frame")
+  stop_if(
+    cnd = !all(c("GroupID", "GroupLevel", "Denominator", "Numerator", "Metric") %in% names(dfTransformed)),
+    message = "One or more of these columns not found: GroupID, GroupLevel, Denominator, Numerator, Metric"
   )
+  stop_if(cnd = !all(!is.na(dfTransformed[["GroupID"]])), message = "NA value(s) found in GroupID")
+  stop_if(cnd = !strType %in% c("binary", "rate"), message = "strType is not 'binary' or 'rate'")
 
   # Caclulate Z-score with overdispersion --------------------------------------
   if (strType == "binary") {

diff --git a/R/Analyze_Poisson.R b/R/Analyze_Poisson.R
@@ -27,12 +27,12 @@
 #' @export
 
 Analyze_Poisson <- function(dfTransformed) {
-  stopifnot(
-    "dfTransformed is not a data.frame" = is.data.frame(dfTransformed),
-    "One or more of these columns not found: GroupID, GroupLevel, Denominator, Numerator, Metric" =
-      all(c("GroupID", "GroupLevel", "Denominator", "Numerator", "Metric") %in% names(dfTransformed)),
-    "NA value(s) found in GroupID" = all(!is.na(dfTransformed[["GroupID"]]))
+  stop_if(cnd = !is.data.frame(dfTransformed), message = "dfTransformed is not a data.frame")
+  stop_if(
+    cnd = !all(c("GroupID", "GroupLevel", "Denominator", "Numerator", "Metric") %in% names(dfTransformed)),
+    message = "One or more of these columns not found: GroupID, GroupLevel, Denominator, Numerator, Metric"
   )
+  stop_if(cnd = !all(!is.na(dfTransformed[["GroupID"]])), message = "NA value(s) found in GroupID")
 
   dfModel <- dfTransformed %>%
     mutate(LogDenominator = log(.data$Denominator))

diff --git a/R/Flag.R b/R/Flag.R
@@ -56,22 +56,17 @@ Flag <- function(
   vThreshold = NULL,
   strValueColumn = NULL
 ) {
-  stopifnot(
-    "dfAnalyzed is not a data frame" = is.data.frame(dfAnalyzed),
-    "strColumn is not character" = is.character(strColumn),
-    "vThreshold is not numeric" = is.numeric(vThreshold),
-    "vThreshold must be length of 2" = length(vThreshold) == 2,
-    "vThreshold cannot be NULL" = !is.null(vThreshold),
-    "strColumn must be length of 1" = length(strColumn) == 1,
-    "strColumn not found in dfAnalyzed" = strColumn %in% names(dfAnalyzed),
-    "strValueColumn not found in dfAnalyzed" = strValueColumn %in% names(dfAnalyzed),
-    "GroupID not found in dfAnalyzed" = "GroupID" %in% names(dfAnalyzed)
-  )
+  stop_if(cnd = !is.data.frame(dfAnalyzed), message = "dfAnalyzed is not a data frame")
+  stop_if(cnd = !is.character(strColumn), message = "strColumn is not character")
+  stop_if(cnd = !is.numeric(vThreshold), message = "vThreshold is not numeric")
+  stop_if(cnd = length(vThreshold) != 2, message = "vThreshold must be length of 2")
+  stop_if(cnd = is.null(vThreshold), message = "vThreshold cannot be NULL")
+  stop_if(cnd = length(strColumn) != 1, message = "strColumn must be length of 1")
+  stop_if(cnd = !(strColumn %in% names(dfAnalyzed)), message = "strColumn not found in dfAnalyzed")
+  stop_if(cnd = !("GroupID" %in% names(dfAnalyzed)), message = "GroupID not found in dfAnalyzed")
 
   if (all(!is.na(vThreshold))) {
-    stopifnot(
-      "vThreshold must contain a minimum and maximum value (i.e., vThreshold = c(1, 2))" = vThreshold[2] > vThreshold[1]
-    )
+    stop_if(cnd = vThreshold[2] <= vThreshold[1], "vThreshold must contain a minimum and maximum value (i.e., vThreshold = c(1, 2))")
   }
 
   # Flag values outside the specified threshold.
@@ -86,6 +81,8 @@ Flag <- function(
 
   # If strValueColumn is supplied, it can only affect sign of Flag (1 or -1).
   if (!is.null(strValueColumn)) {
+    stop_if(cnd = !(strValueColumn %in% names(dfAnalyzed)), message = "strValueColumn not found in dfAnalyzed")
+
     nMedian <- dfFlagged %>%
       pull(strValueColumn) %>%
       stats::median(na.rm = TRUE)

diff --git a/R/Flag_Fisher.R b/R/Flag_Fisher.R
@@ -36,18 +36,14 @@ Flag_Fisher <- function(
   dfAnalyzed,
   vThreshold = NULL
 ) {
-  stopifnot(
-    "dfAnalyzed is not a data frame" = is.data.frame(dfAnalyzed),
-    "vThreshold is not numeric" = is.numeric(vThreshold),
-    "vThreshold must be length of 2" = length(vThreshold) == 2,
-    "vThreshold cannot be NULL" = !is.null(vThreshold),
-    "GroupID not found in dfAnalyzed" = "GroupID" %in% names(dfAnalyzed)
-  )
+  stop_if(cnd = !is.data.frame(dfAnalyzed), message = "dfAnalyzed is not a data frame")
+  stop_if(cnd = !is.numeric(vThreshold), message = "vThreshold is not numeric")
+  stop_if(cnd = length(vThreshold) != 2, message = "vThreshold must be length of 2")
+  stop_if(cnd = is.null(vThreshold), message = "vThreshold cannot be NULL")
+  stop_if(cnd = !("GroupID" %in% names(dfAnalyzed)), message = "GroupID not found in dfAnalyzed")
 
   if (all(!is.na(vThreshold))) {
-    stopifnot(
-      "vThreshold must contain a minimum and maximum value (i.e., vThreshold = c(1, 2))" = vThreshold[2] > vThreshold[1]
-    )
+    stop_if(cnd = vThreshold[2] <= vThreshold[1], "vThreshold must contain a minimum and maximum value (i.e., vThreshold = c(1, 2))")
   }
 
   dfFlagged <- dfAnalyzed %>%

diff --git a/R/Flag_NormalApprox.R b/R/Flag_NormalApprox.R
@@ -43,22 +43,24 @@ Flag_NormalApprox <- function(
   dfAnalyzed,
   vThreshold = NULL
 ) {
-  stopifnot(
-    "dfAnalyzed is not a data frame" = is.data.frame(dfAnalyzed),
-    "vThreshold is not numeric" = is.numeric(vThreshold),
-    "vThreshold must be length of 4" = length(vThreshold) == 4,
-    "vThreshold cannot be NULL" = !is.null(vThreshold)
-  )
+  stop_if(cnd = !is.data.frame(dfAnalyzed), message = "dfAnalyzed is not a data frame")
+  stop_if(cnd = !is.numeric(vThreshold), message = "vThreshold is not numeric")
+  stop_if(cnd = !(length(vThreshold) == 4), message = "vThreshold must be length of 4")
+  stop_if(cnd = is.null(vThreshold), message = "vThreshold cannot be NULL")
 
 
   if (all(!is.na(vThreshold))) {
-    stopifnot(
-      "vThreshold must contain cutoff for moderate/high risks in two directions (i.e., vThreshold = c(-3, -2, 2, 3))" =
-        vThreshold[1] < vThreshold[2],
-      "vThreshold must contain cutoff for moderate/high risks in two directions (i.e., vThreshold = c(-3, -2, 2, 3))" =
-        vThreshold[2] < vThreshold[3],
-      "vThreshold must contain cutoff for moderate/high risks in two directions (i.e., vThreshold = c(-3, -2, 2, 3))" =
-        vThreshold[3] < vThreshold[4]
+    stop_if(
+      cnd = vThreshold[2] <= vThreshold[1],
+      message = "vThreshold must contain cutoff for moderate/high risks in two directions (i.e., vThreshold = c(-3, -2, 2, 3))"
+    )
+    stop_if(
+      cnd = vThreshold[3] <= vThreshold[2],
+      message = "vThreshold must contain cutoff for moderate/high risks in two directions (i.e., vThreshold = c(-3, -2, 2, 3))"
+    )
+    stop_if(
+      cnd = vThreshold[4] <= vThreshold[3],
+      message = "vThreshold must contain cutoff for moderate/high risks in two directions (i.e., vThreshold = c(-3, -2, 2, 3))"
     )
   }
 

diff --git a/R/Flag_Poisson.R b/R/Flag_Poisson.R
@@ -34,26 +34,27 @@ Flag_Poisson <- function(
   dfAnalyzed,
   vThreshold = NULL
 ) {
-  stopifnot(
-    "dfAnalyzed is not a data frame" = is.data.frame(dfAnalyzed),
-    "vThreshold is not numeric" = is.numeric(vThreshold),
-    "vThreshold must be length of 4" = length(vThreshold) == 4,
-    "vThreshold cannot be NULL" = !is.null(vThreshold)
-  )
+  stop_if(cnd = !is.data.frame(dfAnalyzed), message = "dfAnalyzed is not a data frame")
+  stop_if(cnd = !is.numeric(vThreshold), message = "vThreshold is not numeric")
+  stop_if(cnd = !(length(vThreshold) == 4), message = "vThreshold must be length of 4")
+  stop_if(cnd = is.null(vThreshold), message = "vThreshold cannot be NULL")
 
 
   if (all(!is.na(vThreshold))) {
-    stopifnot(
-      "vThreshold must contain cutoff for moderate/high risks in two directions (i.e., vThreshold = c(-7, -5, 5, 7))" =
-        vThreshold[1] < vThreshold[2],
-      "vThreshold must contain cutoff for moderate/high risks in two directions (i.e., vThreshold = c(-7, -5, 5, 7))" =
-        vThreshold[2] < vThreshold[3],
-      "vThreshold must contain cutoff for moderate/high risks in two directions (i.e., vThreshold = c(-7, -5, 5, 7))" =
-        vThreshold[3] < vThreshold[4]
+    stop_if(
+      cnd = vThreshold[2] <= vThreshold[1],
+      message = "vThreshold must contain cutoff for moderate/high risks in two directions (i.e., vThreshold = c(-7, -5, 5, 7))"
+    )
+    stop_if(
+      cnd = vThreshold[3] <= vThreshold[2],
+      message = "vThreshold must contain cutoff for moderate/high risks in two directions (i.e., vThreshold = = c(-7, -5, 5, 7))"
+    )
+    stop_if(
+      cnd = vThreshold[4] <= vThreshold[3],
+      message = "vThreshold must contain cutoff for moderate/high risks in two directions (i.e., vThreshold = = c(-7, -5, 5, 7))"
     )
   }
 
-
   # Flag values outside the specified threshold.
   dfFlagged <- dfAnalyzed %>%
     mutate(

diff --git a/R/Input_Rate.R b/R/Input_Rate.R
@@ -74,49 +74,33 @@ Input_Rate <- function(
   strDenominatorCol = NULL
 ) {
   # Check if data frames are NULL
-  if (is.null(dfSubjects)) {
-    stop("dfSubjects must be provided")
-  }
-  if (is.null(dfDenominator)) {
-    stop("dfDenominator, must be provided")
-  }
-  if (is.null(dfNumerator)) {
-    stop("dfNumerator, must be provided")
-  }
+  stop_if(cnd = is.null(dfSubjects), message = "dfSubjects must be provided")
+  stop_if(cnd = is.null(dfDenominator), message = "dfDenominator, must be provided")
+  stop_if(cnd = is.null(dfNumerator), message = "dfNumerator, must be provided")
 
   # must be eit
   strNumeratorMethod <- match.arg(strNumeratorMethod)
   strDenominatorMethod <- match.arg(strDenominatorMethod)
 
   # Check if strNumeratorCol is Null when strNumeratorMethod is 'Sum'
   if (strNumeratorMethod == "Sum") {
-    if (is.null(strNumeratorCol)) {
-      stop("strNumeratorCol must be provided when strNumeratorMethod is 'Sum'")
-    }
-    if (!is.numeric(dfNumerator[[strNumeratorCol]])) {
-      stop("strNumeratorCol must be numeric when strNumeratorMethod is `Sum`")
-    }
+    stop_if(cnd = is.null(strNumeratorCol), message = "strNumeratorCol must be provided when strNumeratorMethod is 'Sum'")
+    stop_if(cnd = !is.numeric(dfNumerator[[strNumeratorCol]]), message = "strNumeratorCol must be numeric when strNumeratorMethod is `Sum`")
   }
 
   # Check if strDenominatorCol is Null when strDenominatorMethod is 'Sum'
   if (strDenominatorMethod == "Sum") {
-    if (is.null(strDenominatorCol)) {
-      stop("strDenominatorCol must be provided when strDenominatorMethod is 'Sum'")
-    }
-    if (!is.numeric(dfDenominator[[strDenominatorCol]])) {
-      stop("strDenominatorCol must be numeric when strDenominatorMethod is `Sum`")
-    }
+    stop_if(cnd = is.null(strDenominatorCol), message = "strDenominatorCol must be provided when strDenominatorMethod is 'Sum'")
+    stop_if(cnd = !is.numeric(dfDenominator[[strDenominatorCol]]), message = "strDenominatorCol must be numeric when strDenominatorMethod is `Sum`")
   }
 
   # check that "strSubjectCol" is in all dfs
-  stopifnot(
-    strSubjectCol %in% colnames(dfSubjects),
-    strSubjectCol %in% colnames(dfNumerator),
-    strSubjectCol %in% colnames(dfDenominator)
-  )
+  stop_if(cnd = !(strSubjectCol %in% colnames(dfSubjects)), message = "`strSubjectCol` must be a column name of `dfSubjects`")
+  stop_if(cnd = !(strSubjectCol %in% colnames(dfNumerator)), message = "`strSubjectCol` must be a column name of `dfNumerator`")
+  stop_if(cnd = !(strSubjectCol %in% colnames(dfDenominator)), message = "`strSubjectCol` must be a column name of `dfDenominator`")
 
   # check that "strGroupCol" is in dfSubjects
-  stopifnot(strGroupCol %in% colnames(dfSubjects))
+  stop_if(cnd = !(strGroupCol %in% colnames(dfSubjects)), message = "`strGroupcol` must be a column name of `dfSubjects`")
 
   # if `strGroupLevel` is null, use `strGroupCol`
   if (is.null(strGroupLevel)) {
@@ -147,12 +131,7 @@ Input_Rate <- function(
     group_by(.data$SubjectID) %>%
     summarise(Denominator = sum(.data$Denominator, na.rm = TRUE))
 
-  if (all(dfDenominator_subj$Denominator == 0)) {
-    LogMessage(
-      level = "fatal",
-      message = "All denominator values are 0, please check `dfDenominator`"
-    )
-  }
+  stop_if(cnd = all(dfDenominator_subj$Denominator == 0), message = "All denominator values are 0, please check `dfDenominator`")
 
   # Merge Numerator and Denominator with Subject Data. Keep all data in Subject. Fill in missing numerator/denominators with 0
   dfInput <- dfSubjects %>%

diff --git a/R/RunQuery.R b/R/RunQuery.R
@@ -29,18 +29,17 @@
 #'
 #' @export
 RunQuery <- function(strQuery, df, bUseSchema = FALSE, lColumnMapping = NULL) {
-  stopifnot(is.character(strQuery))
+  stop_if(cnd = !is.character(strQuery), message = "strQuery must be a query")
 
   # Check that strQuery contains "FROM df"
-  if (!stringr::str_detect(strQuery, "FROM df")) {
-    stop("strQuery must contain 'FROM df'")
-  }
+  stop_if(cnd = !stringr::str_detect(strQuery, "FROM df"), message = "strQuery must contain 'FROM df'")
 
   # Check that columnMapping exists if use_schema == TRUE
 
-  if (bUseSchema && is.null(lColumnMapping)) {
-    stop("if use_schema = TRUE, you must provide lColumnMapping spec")
-  }
+  stop_if(
+    cnd = (bUseSchema && is.null(lColumnMapping)),
+    message = "if use_schema = TRUE, you must provide lColumnMapping spec"
+  )
 
   # Enforce data structure of schema.
   if (bUseSchema) {

diff --git a/R/Summarize.R b/R/Summarize.R
@@ -38,10 +38,9 @@ Summarize <- function(
   dfFlagged,
   nMinDenominator = NULL
 ) {
-  stopifnot(
-    "dfFlagged is not a data frame" = is.data.frame(dfFlagged),
-    "One or more of these columns: GroupID, GroupLevel, Flag, Score, not found in dfFlagged" = all(c("GroupID", "GroupLevel", "Flag", "Score") %in% names(dfFlagged))
-  )
+  stop_if(cnd = !is.data.frame(dfFlagged), message = "dfFlagged is not a data frame")
+  stop_if(cnd = !all(c("GroupID", "GroupLevel", "Flag", "Score") %in% names(dfFlagged)),
+          message = "One or more of these columns: GroupID, GroupLevel, Flag, Score, not found in dfFlagged" )
 
   if (!("Numerator" %in% colnames(dfFlagged))) {
     dfFlagged$Numerator <- NA

diff --git a/R/Transform_Count.R b/R/Transform_Count.R
@@ -36,12 +36,10 @@ Transform_Count <- function(
   strCountCol,
   strGroupCol = "GroupID"
 ) {
-  stopifnot(
-    "dfInput is not a data frame" = is.data.frame(dfInput),
-    "strCountCol not found in input data" = strCountCol %in% names(dfInput),
-    "strCountCol is not numeric or logical" = is.numeric(dfInput[[strCountCol]]) | is.logical(dfInput[[strCountCol]]),
-    "NA's found in strCountCol" = !anyNA(dfInput[[strCountCol]])
-  )
+  stop_if(cnd = !is.data.frame(dfInput), message = "dfInput is not a data frame")
+  stop_if(cnd = !(strCountCol %in% names(dfInput)), message = "strCountCol not found in input data")
+  stop_if(cnd = !(is.numeric(dfInput[[strCountCol]]) | is.logical(dfInput[[strCountCol]])), message = "strCountCol is not numeric or logical")
+  stop_if(cnd = anyNA(dfInput[[strCountCol]]), message = "NA's found in strCountCol")
 
   dfTransformed <- dfInput %>%
     group_by(GroupID = .data[[strGroupCol]]) %>%

diff --git a/R/Transform_Rate.R b/R/Transform_Rate.R
@@ -38,14 +38,13 @@ Transform_Rate <- function(
   strNumeratorCol = "Numerator",
   strDenominatorCol = "Denominator"
 ) {
-  stopifnot(
-    "dfInput is not a data frame" = is.data.frame(dfInput),
-    "strNumeratorColumn is not numeric" = is.numeric(dfInput[[strNumeratorCol]]),
-    "strDenominatorColumn is not numeric" = is.numeric(dfInput[[strDenominatorCol]]),
-    "NA's found in numerator" = !anyNA(dfInput[[strNumeratorCol]]),
-    "NA's found in denominator" = !anyNA(dfInput[[strDenominatorCol]]),
-    "Required columns not found in input data" = c(strNumeratorCol, strDenominatorCol, "GroupID", "GroupLevel") %in% names(dfInput)
-  )
+  stop_if(cnd = !is.data.frame(dfInput), message = "dfInput is not a data frame")
+  stop_if(cnd = !is.numeric(dfInput[[strNumeratorCol]]), message = "strNumeratorColumn is not numeric")
+  stop_if(cnd = !is.numeric(dfInput[[strDenominatorCol]]), message = "strDenominatorColumn is not numeric")
+  stop_if(cnd = anyNA(dfInput[[strNumeratorCol]]), message = "NA's found in numerator")
+  stop_if(cnd = anyNA(dfInput[[strDenominatorCol]]), message = "NA's found in denominator")
+  stop_if(cnd = !(all(c(strNumeratorCol, strDenominatorCol, "GroupID", "GroupLevel") %in% names(dfInput))), message = "Required columns not found in input data")
+
 
   dfTransformed <- dfInput %>%
     group_by(.data$GroupID, .data$GroupLevel) %>%

diff --git a/R/Visualize_Score.R b/R/Visualize_Score.R
@@ -39,13 +39,11 @@ Visualize_Score <- function(
   bFlagFilter = FALSE,
   strTitle = ""
 ) {
-  stopifnot(
-    "strTitle must be character" = is.character(strTitle),
-    "bFlagFilter must be logical" = is.logical(bFlagFilter),
-    "dfResults must be a data.frame" = is.data.frame(dfResults),
-    "strType must be 'Metric' or 'Score'" = strType %in% c("Metric", "Score"),
-    "strType must be length 1" = length(strType) == 1
-  )
+  stop_if(cnd = !is.character(strTitle), message = "strTitle must be character" )
+  stop_if(cnd = !is.logical(bFlagFilter), message = "bFlagFilter must be logical")
+  stop_if(cnd = !is.data.frame(dfResults), message = "dfResults must be a data.frame")
+  stop_if(cnd = !(length(strType) == 1), message = "strType must be length 1" )
+  stop_if(cnd = !(strType %in% c("Metric", "Score")), message = "strType must be 'Metric' or 'Score'" )
 
   dfResults$FlagAbs <- abs(dfResults$Flag)
   flagBreaks <- as.character(unique(sort(dfResults$FlagAbs)))