R/LoadRefreshDataFrame.r
In BlueSkyStatistics/BlueSky: BlueSky Collection of Functions for Data Access, Analysis, and Formatting.
Defines functions
BSkySetMaxFactor
BSkyGetMaxFactor
BSkyRemoveRefreshDataframe
BSkyLoadRefreshDataframe
BSkyLoadRefreshDataframe2
BSky.RemoveRefresh.Dataframe
BSky.LoadRefresh.Dataframe
#For loading and refreshing in memory dataframe. (not a disk file dataframe)
# dframe is the dataframe name and is a string.
BSky.LoadRefresh.Dataframe <- function(dframe)
{
#print(environment())
#print( parent.frame())
#print(globalenv())
#print("Summ Exists")
#print(exists("summary", envir = parent.frame(), inherits = FALSE))
#print(class(summary))
BSkyFunctionInit()
BSkySetCurrentDatasetName(dframe)
success <- NULL
logflag=FALSE
#print("BSky Load Refresh starts..")
#print(eval( parse(text=paste('class(',dframe,')',sep=''))))#.GlobalEnv$
#BSkyStoreApplicationWarnErrMsg(BSkyWarnMsg, BSkyErrMsg)
curidx <- UAgetIndexOfDataSet(dframe) # current index of dataset if its already loaded ( before )
filetype = "RDATA" #default
#print("BSky Load Refresh..")
#cat("\nCur Index :",curidx)
############
# 20-Jan-2021 Sanjay and I (Anil) Concluded that following code is unnecessary because it is not executing the 'if' condition
# anyway and that is why the old code (#30Dec2020) happens to work fine.
# The new code (#20Jan2021) appear like a logical approach bu then it convertes dframe from string to actual dataframe which
# in turn makes rest of the calls to fail (becuase all that code expects dframe to be a string.
# Here is an example that we tried
#
#df = data.frame(A=c(1,2,3), B=c(4,5,6), C=c(11,22,33))
# tbdf = as_tibble(df)
# class(tbdf)
# class(tbdf) = c("tbl_df") #make tbdf a list
# df2 = as.data.frame(tbdf)
# class(df2)
# dframe2= "tbdf"
# dfstr= "df"
# dfclass = eval( parse(text=paste('class(.GlobalEnv$',dfstr,')',sep='')))
# eval(parse(text=paste('hascolumns <- ncol(.GlobalEnv$',dframe2,') > 0')))
# ## [Aggregate to data set]
# require(dplyr);
# #Aggregate the dataset
# test<-mtcars %>%
# dplyr::group_by(transmission,region) %>%
# dplyr::summarize(mean_mpg=mean(mpg,na.rm =TRUE))
# class(test)
# is.data.frame(test)
# dframe='test'
# dfclass = class(dframe)
# tbl_df_index = which(dfclass == 'tbl_df')
# length(tbl_df_index) > 0
# eval( parse(text=paste('is.data.frame(',dframe,')',sep='')))
# eval(parse(text=paste('hascolumns <- ncol(.GlobalEnv$',dframe,') > 0')))
# dframe = as.data.frame(dframe)
#
#We may have to come back and fix the following commented code if our dframe class is only tbl_df and does not contain data.frame class.
# # #if dframe class is tbl_df then we convert it to data.frame type
# # dfclass = class(dframe) #30Dec2020. This does not look right. dframe is supposed to be string. Fixed, line below
# # #20Jan2021 dfclass = eval( parse(text=paste('class(.GlobalEnv$',dframe,')',sep='')))
# # tbl_df_index = which(dfclass == 'tbl_df')
# # if( length(tbl_df_index) > 0)#length is empty if tbl_df is not found
# # {
# # dframe = as.data.frame(dframe)#30Dec2020. This does not look right. dframe is supposed to be string. Fixed, line below
# # # #20Jan2021 dframe = eval( parse(text=paste('as.data.frame(.GlobalEnv$',dframe,')',sep='')))
# # }
############
#### CHANGE THIS TO ls()
#if(eval( parse(text=paste('exists("dframe")',sep=''))))#exists("dframe"))
# # # if(eval( parse(text=paste('exists("',dframe,'", where=globalenv())',sep=''))))#exists(dframe)) in any env
# # # {
# # # if(eval( parse(text=paste('is.data.frame(',dframe,')',sep='')))) ##
# # # {
# # # ####29Jan2021 we do not need if check because we care making a local copy of the dataset
# # # # if(!eval( parse(text=paste('exists("',dframe,'", where=globalenv(), inherits=FALSE)',sep=''))))
# # # # {
# # # eval( parse(text=paste('.GlobalEnv$',dframe,' <- as.data.frame(',dframe,')',sep=''))) # making a copy of the dataset that may be in some package. This is will also take care of "tibble"
# # # # }
# # # }
# # # }
### 01Feb2021 Whole if block above can be replaced by following commented line.
BSkyCheckIfTibble(dataframeName=dframe) # converts to data.frame only if dataframe has tbl_df class
#print("Step5")
if(eval( parse(text=paste('exists("',dframe,'", where=globalenv(), inherits=FALSE)',sep=''))))#exists(dframe))
{
#cat("\nDataframe exists\n Is it a DataFrame : ")
#print(ls())
varenv <- eval( parse(text=paste('getEnv(dframe)',sep='')))
#print(varenv)
#print(eval( parse(text=paste('is.data.frame(.GlobalEnv$',dframe,')',sep='')))) #.GlobalEnv$
#cat("\n")
# if(eval( parse(text=paste('is.data.frame(',dframe,')',sep='')),envir=globalenv()))
if(eval( parse(text=paste('is.data.frame(.GlobalEnv$',dframe,')',sep='')))) #.GlobalEnv$
{
#print("HasCol and Has row")
eval(parse(text=paste('hascolumns <- ncol(.GlobalEnv$',dframe,') > 0'))) #20Oct2016 data.frame has cols
#print("A")
#print(hascolumns)
eval(parse(text=paste('hasrows <- nrow(.GlobalEnv$',dframe,') > 0'))) #20Oct2016 data.frame has rows
#print("B")
#print(hasrows)
if(hascolumns && hasrows)#20Oct2016 'if' condition added for the code below. And a full 'else' block is also added for this 'if'
{
#at this point we need to assign some column name(Var1, Var2) to the column those do not have any col-names.
GenerateUniqueColName(paste('.GlobalEnv$',dframe,sep='') )#its string so no need to enclose in single quotes
#Now replace special chars with underscore and prefix 'X' if there is digit in the first index of colum name.
colcount <- eval(parse(text=paste('length(names(.GlobalEnv$',dframe,'))' ))) #colcount = length(names(datasetname))
#cat('Count=')
#cat(colcount)
#ptm= proc.time()
for( i in 1:colcount)
{
if(logflag)
{
cat('\n-----------------------------------------------------------------')
cat("\nSpl Chr:OldName = ")
print(eval(parse(text=paste('names(.GlobalEnv$',dframe,')[',i,']',sep=''))))
}
eval(parse(text=paste('names(.GlobalEnv$',dframe,')[',i,'] <- make.names(names(.GlobalEnv$',dframe,')[',i,'])', sep='')))
#names(datasetname)[i] = ReplaceSplChrsAndPrefixXForDigitInBegining(names(datasetname)[i])
if(logflag)
{
cat("\nSpl Chr removed ColName = ")
print(eval(parse(text=paste('names(.GlobalEnv$',dframe,')[',i,']',sep=''))))
cat('\n-----------------------------------------------------------------')
}
}
#cat("\nIts Dataframe\n")
#eval( parse(text=paste(dframe,' <<- dframe',sep='')))
if(curidx == -1) # if not exists in list
{
#cat("\n adding to uadatasets$name:")
uadatasets$name <- c(uadatasets$name, dframe)
#cat("\n added to uadatasets$name:")
}
else
{
#cat("\nCreating filetype attribute\n")
filetype <- eval(parse(text=paste('attr(.GlobalEnv$',dframe,',"filetype")',sep=''))) ## this may not work. because dframe is new object. may not have attributes.
#bskymaxfactors <- eval(parse(text=paste('attr(',datasetname,',"maxfactor")',sep='')))
}
#cat("\n Creating Attrs for DS:")
#eval(parse(text=paste('attributes(',dframe,') <<- c(attributes(',dframe,'), list(split = c(FALSE), splitcolumnnames = c(), splitcolumnindex = c(), filetype = c("CSV"), slice = c(FALSE), align = c("Left"), maxfactor= c(bskymaxfactors)))', sep='')))
if(is.null(filetype) || nchar(filetype)==0)
filetype = "RDATA"
#cat("File Type:", filetype)
#eval(parse(text=paste('uadatasets$lst$',dframe,' <- .GlobalEnv$',dframe,sep='')))
#cat("\n added to uadatasets$lst$dfname:")
UAcreateExtraAttributes(dframe, filetype)
#cat("\n Finished Creating Attrs for DS:\n")
success = "SUCCESS"
#print(eval(parse(text=paste('colnames(',dframe,')'))))
replacePeriodWithUnderscore(dframe)
#print(eval(parse(text=paste('colnames(.GlobalEnv$',dframe,')'))))
#cat("\nFixing UTC")
#MakeAllDateColUTC(dframe)
#cat("\nUTC Fixed")
}
else
{
cat("Dataframe object does not have row(s) or column(s)!\n")
BSkyErrMsg =paste("BSky.LoadRefresh.Dataframe: Dataset has no row(s) or column(s)"," Dataset Name:", dframe)
warning("BSky.LoadRefresh.Dataframe: Dataset has no row(s) or column(s) ")
}
}
else
{
cat("Object is not data frame object");
warning("BSky.LoadRefresh.Dataframe: Object is not data frame object ")
}
}
else
{
cat("Dataframe object does not exists!")
BSkyErrMsg =paste("BSky.LoadRefresh.Dataframe: Dataset with the same name already on the global list"," Dataset Name:", dframe)
warning("BSky.LoadRefresh.Dataframe: Dataset with the same name already on the global list ")
}
BSkyFunctionWrapUp()
invisible(success)
}
#Closes a Dataframe, Makes it NULL
BSky.RemoveRefresh.Dataframe <- function(dframe)
{
BSkyFunctionInit()
BSkySetCurrentDatasetName(dframe)
#BSkyStoreApplicationWarnErrMsg(BSkyWarnMsg, BSkyErrMsg)
#curidx <- UAgetIndexOfDataSet(dframe) # current index of dataset if its already loaded ( before )
#cat("\nCur Index of CSV:",curidx)
if(eval( parse(text=paste('exists("dframe")',sep=''))))#exists(dframe))
{
if(eval( parse(text=paste('is.null("dframe")',sep='')))) #Already closed from elsewhere
{
#Just remove external attributes
}
else if(eval( parse(text=paste('is.data.frame(',dframe,')',sep='')))) # not closed. Close it from here
{
UAcloseDataset(dframe)
#remove external attributes
}
else
{
cat("Object is not data frame object");
}
}
else
{
cat("Dataframe object does not exists!")
BSkyErrMsg =paste("BSky.RemoveRefresh.Dataframe: Dataset with the same name already on the global list"," Dataset Name:", dframe)
warning("BSky.RemoveRefresh.Dataframe: Dataset with the same name already on the global list ")
}
BSkyFunctionWrapUp()
}
BSkyLoadRefreshDataframe2 <- function(dframe)
{
BSkyFunctionInit()
BSkySetCurrentDatasetName(dframe)
success <- NULL
curidx <- UAgetIndexOfDataSet(dframe) # current index of dataset if its already loaded ( before )
filetype = "RDATA" #default
#### CHANGE THIS TO ls()
if(eval( parse(text=paste('exists("',dframe,'", where=globalenv())',sep=''))))#exists(dframe))
{
varenv <- eval( parse(text=paste('getEnv(dframe)',sep='')))
#print(varenv)
#print(eval( parse(text=paste('is.data.frame(.GlobalEnv$',dframe,')',sep='')))) #.GlobalEnv$
#cat("\n")
# if(eval( parse(text=paste('is.data.frame(',dframe,')',sep='')),envir=globalenv()))
if(eval( parse(text=paste('is.data.frame(.GlobalEnv$',dframe,')',sep='')))) #.GlobalEnv$
{
eval(parse(text=paste('uadatasets$lst$',dframe,' <- .GlobalEnv$',dframe,sep='')))
#cat("\n added to uadatasets$lst$dfname:")
#UAcreateExtraAttributes(dframe, filetype)
#cat("\n Finished Creating Attrs for DS:\n")
success = "SUCCESS"
#print(eval(parse(text=paste('colnames(',dframe,')'))))
#replacePeriodWithUnderscore(dframe)
#print(eval(parse(text=paste('colnames(.GlobalEnv$',dframe,')'))))
}
else
{
cat("Object is not data frame object");
warning("BSky.LoadRefresh.Dataframe: Object is not data frame object ")
}
}
else
{
cat("Dataframe object does not exists!")
BSkyErrMsg =paste("BSky.LoadRefresh.Dataframe: Dataset with the same name already on the global list"," Dataset Name:", dframe)
warning("BSky.LoadRefresh.Dataframe: Dataset with the same name already on the global list ")
}
BSkyFunctionWrapUp()
invisible(success)
}
# dframe must be a name (a string) rather than an actual dataframe object
##Refreshes values in C1Datagrid, if dataset/dataframe was modified from syntax.
#Basically loads/refreshes the grid from memory(dataset/dataframe). Not by reading disk file from disk.
# BSkyLoadRefreshDataframe loads the specified dataframe in the BlueSky Statistics UI datagrid.
# If dataframe is already loaded in the UI datagrid then it refresh it for any changes done.
#
# To use BSkyLoadRefreshDataframe you must run it from within the BlueSky Statistics application.
#
# @name BSkyLoadRefreshDataframe
# @param dframe is the dataframe object(do not use quotes around the object)
# @param load.dataframe is provided so that you can selectively refresh or not refresh the UI datagrid.
# This parameter is defined in the dialog to make the decision, whether to refresh the UI datagrid or not.
# Default value is TRUE.
# @export
# @examples
# df <-data.frame(A=c(1,2,3), B=c(4,5,6), C=c(6,7,8))
# BSkyLoadRefreshDataframe(df)
BSkyLoadRefreshDataframe <- function(dframe, load.dataframe = TRUE)
{
#if(load.dataframe)
#{
#invisible(dframe)
#}
#else
#{
#invisible(NULL)
#}
BSkyLoadRefresh(dframe, load.dataframe);
}
### title should fit on one line, be written in sentence case, but not end in a full stop
### to print @ in the documentation, escape with one more @ (e.g. @@ prints @)
#' @title Loading a dataframe in the BlueSky Statistics UI datagrid
#'
#' @description loads the specified dataframe in the BlueSky Statistics UI datagrid.
#'
#' @param bskyDatasetName is the dataframe name (use quotes around the name)
#' @param load.dataframe is provided so that you can selectively refresh or not refresh the UI datagrid.
#' This parameter is defined in the dialog to make the decision, whether to refresh the UI datagrid or not. Default value is TRUE.
#'
#'
#' @examples
#' df <-data.frame(A=c(1,2,3), B=c(4,5,6), C=c(6,7,8))
#' BSkyLoadRefresh('df')
BSkyLoadRefresh <- function (bskyDatasetName, load.dataframe = TRUE, load.UIgrid = TRUE, isRmarkdownOutputOn = BSkyIsRmarkdownOutputOn(), maxFactor=BSkyGetMaxFactor(), createAttr=TRUE)## change this to a string parameter from a dataset object
{
isdataframe=FALSE
isdesign=FALSE
ists=FALSE
isPkgLoaded = FALSE
isexists = FALSE
error = FALSE
pkgname = c()
dsname = c()
pkgEnv = c()
ischar = is.character(bskyDatasetName)
originalDatasetname = bskyDatasetName
#for FALSE we can just return from the function.
if(!load.dataframe)
{
return(invisible())
}
#print(originalDatasetname)
#print(class(originalDatasetname))
if(ischar)## is argument a character. If not we print error in the sink and exit
{
hasPkgname = grepl("::", bskyDatasetName)
if(hasPkgname)#package name is passed
{
arr = strsplit(bskyDatasetName, "::")
pkgname = arr[[1]][1]
dsname = arr[[1]][2]
isPkgLoaded = pkgname %in% (.packages())
if(!isPkgLoaded)# if packge is not loaded
{
cat("\n") # forcing a new line in case someone created a cat() without a trailing new line
cat("Package ")
cat(pkgname)
cat(" not loaded. Use library(packageName) to load the package.")
error=TRUE
invisible()
}
pkgEnv = eval(parse(text=paste('as.environment("package:',pkgname,'")', sep='')))
isexists = eval(parse(text=paste('exists("',dsname,'", where=pkgEnv, inherits=FALSE)', sep='' )))# DS exists in the pkg.
if(isexists)
{
# dsclass = eval(parse(text=paste('"class(',pkgname,'::',dsname,')',sep='')))
# if( length(dsclass) > 1)
# {
#check if data.frame
# isdataframe = "data.frame" %in% dsclass
isdataframe = eval(parse(text=paste('"data.frame" %in% c(class(',pkgname,'::',dsname,'))',sep='')))
# if(isdataframe)
# {
# eval(parse(text=paste('"class(',pkgname,'::',dsname,')',sep='')))
# }
# }
isdesign = eval(parse(text=paste('"design" %in% c(class(',pkgname,'::',dsname,'))',sep='')))#for DoE
if(isdataframe && isdesign)#DoE
{
eval( parse(text=paste('.GlobalEnv$',dsname,' <- (',pkgname,'::',dsname,')',sep='')))##make a copy in globalEnv
#dsname = 'mtcars'
bskyDatasetName=dsname #overwrite bskyDatasetName as it may still have 'datasets::mtcars'
}
else if(isdataframe)
{
eval( parse(text=paste('.GlobalEnv$',dsname,' <- as.data.frame(',pkgname,'::',dsname,')',sep='')))##make a copy in globalEnv
#dsname = 'mtcars'
bskyDatasetName=dsname #overwrite bskyDatasetName as it may still have 'datasets::mtcars'
}
else
{
#lets check if it is a "ts" object, if it is, we try to convert it to data.frame object.
ists = eval(parse(text=paste('"ts" %in% c(class(',pkgname,'::',dsname,'))',sep='')))
if(ists)
{
##make it a data.frame and also make a copy in globalEnv
eval( parse(text=paste('.GlobalEnv$',dsname,' <- as.data.frame(',pkgname,'::',dsname,')',sep='')))
#dsname = 'mtcars'
bskyDatasetName=dsname #overwrite bskyDatasetName as it may still have 'datasets::mtcars'
}
}
}
else
{
cat("\n") # forcing a new line in case someone created a cat() without a trailing new line
cat("Dataset ")
cat(dsname)
cat(" is not found in the package ")
cat(pkgname)
error= TRUE
invisible()
}
}
else#package name is not passed so, whatever is passed should be considered as a datasetname in global env.
{
inGlobalEnv = eval( parse(text=paste('exists("',bskyDatasetName,'", where=globalenv(), inherits=FALSE)',sep='')))
# if dataset is not found in the Global environment then we should not go searching it
# because same named datasets can be found in multiple packages. And user may get confused if we load
# the first one we found. So we simply error out. User is supposed to pass package name to avoid ambiguity
if(inGlobalEnv)
{
isdataframe = eval(parse(text=paste('"data.frame" %in% c(class(',bskyDatasetName,'))',sep='')))
isdesign = eval(parse(text=paste('"design" %in% c(class(',bskyDatasetName,'))',sep=''))) # for DoE
#check if it is a "ts" object
ists = eval(parse(text=paste('"ts" %in% c(class(',bskyDatasetName,'))',sep='')))
if(isdesign && isdataframe)
{
##make it a data.frame and also make a copy in globalEnv
eval( parse(text=paste('.GlobalEnv$',bskyDatasetName,' <- (',bskyDatasetName,')',sep='')))
}
else if(ists || isdataframe)#DoE
{
##make it a data.frame and also make a copy in globalEnv
eval( parse(text=paste('.GlobalEnv$',bskyDatasetName,' <- as.data.frame(',bskyDatasetName,')',sep='')))
}
if ( !isdataframe && !ists ) #not a data.frame (and ts) and the object exists
{
cat("\n") # forcing a new line in case someone created a cat() without a trailing new line
cat("Syntax error (BSkyLoadRefresh expects the name of the data.frame (or ts) object as a character string) or the object referenced is not an object of class data.frame (or ts)")
cat("\nCorrect syntax is:")
cat("\n\tBSkyLoadRefresh('dataframe-name') #load an object of class data.frame (or ts)")
cat("\n\tBSkyLoadRefresh('packageName::dataframe-name') #load an object of class data.frame (or ts) from a specific R package")
error= TRUE
invisible()
}
}
else
{
isexists = eval(parse(text=paste('exists("' ,bskyDatasetName,'")', sep='' )))# DS exists in the pkg.
if (isexists)
{
eval( parse(text=paste('.GlobalEnv$',bskyDatasetName,' <- ',bskyDatasetName,sep='')))##make a copy in globalEnv
#dsname = 'mtcars'
isdataframe = eval(parse(text=paste('"data.frame" %in% c(class(',bskyDatasetName,'))',sep='')))
isdesign = eval(parse(text=paste('"design" %in% c(class(',bskyDatasetName,'))',sep='')))
#check if it is a "ts" object
ists = eval(parse(text=paste('"ts" %in% c(class(',bskyDatasetName,'))',sep='')))
if(isdesign && isdataframe)#DoE
{
eval( parse(text=paste('.GlobalEnv$',bskyDatasetName,' <- (',bskyDatasetName,')',sep='')))
}
else if(ists)
{
##make it a data.frame and also make a copy in globalEnv
eval( parse(text=paste('.GlobalEnv$',bskyDatasetName,' <- as.data.frame(',bskyDatasetName,')',sep='')))
}
if ( !isdataframe && !ists ) #not a data.frame or ts and the object exists
{
cat("\n") # forcing a new line in case someone created a cat() without a trailing new line
cat("Syntax error (BSkyLoadRefresh expects the name of the data.frame (or ts) object as a character string) or the object referenced is not an object of class data.frame (or ts)")
cat("\nCorrect syntax is:")
cat("\n\tBSkyLoadRefresh('dataframe-name') #load an object of class data.frame (or ts)")
cat("\n\tBSkyLoadRefresh('packageName::dataframe-name') #load an object of class data.frame (or ts) from a specific R package")
error= TRUE
invisible()
}
}
else
{
cat("\n") # forcing a new line in case someone created a cat() without a trailing new line
cat("Dataset ")
cat(bskyDatasetName)
cat(" cannot be found. Please check the spelling of the dataset name or load the package containing the dataset")
error= TRUE
invisible()
}
}
}
}
hasrows=FALSE
##Check if number of rows is greater than zero. Can't rely on column count because col count was greater than zero in subset-ed dataset that had no rows.
rowcount = eval( parse(text=paste('nrow(.GlobalEnv$',bskyDatasetName,')',sep='')))
if(rowcount>0)
{
hasrows = TRUE
}
else
{
# The data frame 'bskyDatasetName' has zero rows, hence it will not be loaded in the datagrid
cat("\nERROR: The data frame '") # forcing a new line in case someone created a cat() without a trailing new line
cat(bskyDatasetName)
cat("' has zero rows, hence it will not be loaded in the datagrid")
error= TRUE
invisible()
}
if(load.dataframe && !error)
{
# first check the whether the dataset exsits in the uadatasets list - if so no processing of the dataset is needed - Anil needs to verify this
# if the dataset name does not exist in the uadatasets, then do all prep (attrbutes, variables, etc) the dataset and add to the uadatasets
#Following block is added by Anil. 28Nov2020
#curidx <- UAgetIndexOfDataSet(bskyDatasetName) #strip out .GlobalEnv$ from the bskyDatasetName and then looks in to uadatasets$name
#if(curidx == -1)
#{
#BSky.LoadRefresh.Dataframe(bskyDatasetName)#This is what we call from C# behind the scenes if the memory dataframe is a new one.
#}
##get dataframe object from the dataframename
#dframeobj = eval( parse(text=paste('as.data.frame(.GlobalEnv$',bskyDatasetName,')',sep='')))
##we can put dframeobj object=c(dframeobj)in the Sanjay's code below or we can just put the name object=c(bskyDatasetName)
##if we put object then python gets the object directly.
##if we put name then python code must know that its a name and get object of that name from R memory and then process it.
# if((exists("uadatasets.sk") && exists("BSkyKableFormatting", env=uadatasets.sk) && uadatasets.sk$BSkyKableFormatting == FALSE))
# {
# doKableFormatting = FALSE
# }
# else
# {
# doKableFormatting = TRUE
# if((exists("uadatasets.sk") && exists("BSkyRmarkdownFormatting", env=uadatasets.sk) && uadatasets.sk$BSkyRmarkdownFormatting == FALSE))
# {
# doRmarkdownFormatting = FALSE
# }
# else
# {
# doRmarkdownFormatting = TRUE
# }
# isRmarkdownOutputOn = doRmarkdownFormatting
# }
# after the above processing do the following
# if(isRmarkdownOutputOn == FALSE && trimws(originalDatasetname)!= "bsky_piped_temp_dataset")
if(isRmarkdownOutputOn == FALSE && load.UIgrid == TRUE)
{
cat("\n") # forcing a new line in case someone created a cat() without a trailing new line
#print("Entered")
#print(originalDatasetname)
print("BSkyDataGridRefresh")
}
###################################################################################
#11/24/20 - The following code is for creating an in memory queue instead of a sync file to signal the new (Python, etc) application tier
####################################################################################
#if(trimws(originalDatasetname)!= "bsky_piped_temp_dataset")
if(load.UIgrid == TRUE)
{
if(!exists("holdBSkyFormatObjectNew", env=uadatasets.sk) || is.null(uadatasets.sk$holdBSkyFormatObjectNew))
{
uadatasets.sk$holdBSkyFormatObjectNew = list(list(type=c("BSkyDataGridRefresh"), object=c(bskyDatasetName)))
}
else
{
uadatasets.sk$holdBSkyFormatObjectNew = c(uadatasets.sk$holdBSkyFormatObjectNew, list(list(type=c("BSkyDataGridRefresh"), object=c(bskyDatasetName))))
}
}
###### 23 Jan 2021 ### Add dataset in uadataset$name if it is not in the list ######
datasetname <- BSkyValidateDataset(bskyDatasetName)
if(is.null(datasetname))
{
uadatasets$name <- c(uadatasets$name, bskyDatasetName)
}
if(createAttr)
{
UAcreateExtraAttributes(bskyDatasetName, "RDATA")
}
# if maxFactor = -1 then we do not convert factor col to character
# if maxFactor is a positive integer and factor columns has levels more than maxFactor we convert this col to character.
## 09Oct2022
## we do not put maaxFactor limit to RDATA and RDA files while opening the file.
## Here we are mostly dealing with memory dataset.
## Say I loaded a RDATA file with a col having factor levels more the the maxFactor setting
## because we do not restrict RData we load that col as factor (and not make it char)
## now user made a copy of this dataset (in R Editor) and loadRefresh' this memory dataset
## will load and the factor col turns to character because of the following code. User may get
## confused why the factor to character conversion happened.
## So we should comment the code below (or make that IF FALSE)
if(FALSE) ## (maxFactor > 0 ) ##
{
# columns having more than 30 factors are converted back to character
colcount = eval(parse(text=paste('ncol(',bskyDatasetName,')')))
for(i in 1:colcount)
{
coluname = eval(parse(text=paste('colnames(.GlobalEnv$',bskyDatasetName,')[',i,']',sep='')))
colclass = eval(parse(text=paste('class(.GlobalEnv$',bskyDatasetName,'$',coluname,')',sep='')))
if("factor" %in% colclass)
{
lvlcount = eval(parse(text=paste('length(levels(.GlobalEnv$',bskyDatasetName,'$',coluname,'))',sep='')))
if(lvlcount > maxFactor)
{
eval(parse(text=paste('.GlobalEnv$',bskyDatasetName,'$',coluname,'<- as.character(.GlobalEnv$',bskyDatasetName,'$',coluname,')',sep='' )))
}
}
}
}
}
}
BSkyRemoveRefreshDataframe <- function(dframe)
{
BSky.RemoveRefresh.Dataframe (dframe)
}
##10Mar2022
BSkyGetMaxFactor <- function()
{
if(exists("maxFactor", env=uadatasets.sk))
{
return(invisible(uadatasets.sk$maxFactor))
}
else
{
# -1 here means no restriction on maximum factors in a column
return(invisible(-1))#return(invisible(30))
}
}
##10Mar2022
BSkySetMaxFactor <- function(maxFactor = -1)
{
uadatasets.sk$maxFactor = maxFactor
return(invisible(uadatasets.sk$maxFactor))
}
BlueSkyStatistics/BlueSky documentation built on Jan. 29, 2025, 4:15 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Defines functions BSkySetMaxFactor BSkyGetMaxFactor BSkyRemoveRefreshDataframe BSkyLoadRefreshDataframe BSkyLoadRefreshDataframe2 BSky.RemoveRefresh.Dataframe BSky.LoadRefresh.Dataframe
#For loading and refreshing in memory dataframe. (not a disk file dataframe)
# dframe is the dataframe name and is a string.
BSky.LoadRefresh.Dataframe <- function(dframe)
{
#print(environment())
#print( parent.frame())
#print(globalenv())
#print("Summ Exists")
#print(exists("summary", envir = parent.frame(), inherits = FALSE))
#print(class(summary))
BSkyFunctionInit()
BSkySetCurrentDatasetName(dframe)
success <- NULL
logflag=FALSE
#print("BSky Load Refresh starts..")
#print(eval( parse(text=paste('class(',dframe,')',sep=''))))#.GlobalEnv$
#BSkyStoreApplicationWarnErrMsg(BSkyWarnMsg, BSkyErrMsg)
curidx <- UAgetIndexOfDataSet(dframe) # current index of dataset if its already loaded ( before )
filetype = "RDATA" #default
#print("BSky Load Refresh..")
#cat("\nCur Index :",curidx)
############
# 20-Jan-2021 Sanjay and I (Anil) Concluded that following code is unnecessary because it is not executing the 'if' condition
# anyway and that is why the old code (#30Dec2020) happens to work fine.
# The new code (#20Jan2021) appear like a logical approach bu then it convertes dframe from string to actual dataframe which
# in turn makes rest of the calls to fail (becuase all that code expects dframe to be a string.
# Here is an example that we tried
#
#df = data.frame(A=c(1,2,3), B=c(4,5,6), C=c(11,22,33))
# tbdf = as_tibble(df)
# class(tbdf)
# class(tbdf) = c("tbl_df") #make tbdf a list
# df2 = as.data.frame(tbdf)
# class(df2)
# dframe2= "tbdf"
# dfstr= "df"
# dfclass = eval( parse(text=paste('class(.GlobalEnv$',dfstr,')',sep='')))
# eval(parse(text=paste('hascolumns <- ncol(.GlobalEnv$',dframe2,') > 0')))
# ## [Aggregate to data set]
# require(dplyr);
# #Aggregate the dataset
# test<-mtcars %>%
# dplyr::group_by(transmission,region) %>%
# dplyr::summarize(mean_mpg=mean(mpg,na.rm =TRUE))
# class(test)
# is.data.frame(test)
# dframe='test'
# dfclass = class(dframe)
# tbl_df_index = which(dfclass == 'tbl_df')
# length(tbl_df_index) > 0
# eval( parse(text=paste('is.data.frame(',dframe,')',sep='')))
# eval(parse(text=paste('hascolumns <- ncol(.GlobalEnv$',dframe,') > 0')))
# dframe = as.data.frame(dframe)
#
#We may have to come back and fix the following commented code if our dframe class is only tbl_df and does not contain data.frame class.
# # #if dframe class is tbl_df then we convert it to data.frame type
# # dfclass = class(dframe) #30Dec2020. This does not look right. dframe is supposed to be string. Fixed, line below
# # #20Jan2021 dfclass = eval( parse(text=paste('class(.GlobalEnv$',dframe,')',sep='')))
# # tbl_df_index = which(dfclass == 'tbl_df')
# # if( length(tbl_df_index) > 0)#length is empty if tbl_df is not found
# # {
# # dframe = as.data.frame(dframe)#30Dec2020. This does not look right. dframe is supposed to be string. Fixed, line below
# # # #20Jan2021 dframe = eval( parse(text=paste('as.data.frame(.GlobalEnv$',dframe,')',sep='')))
# # }
############
#### CHANGE THIS TO ls()
#if(eval( parse(text=paste('exists("dframe")',sep=''))))#exists("dframe"))
# # # if(eval( parse(text=paste('exists("',dframe,'", where=globalenv())',sep=''))))#exists(dframe)) in any env
# # # {
# # # if(eval( parse(text=paste('is.data.frame(',dframe,')',sep='')))) ##
# # # {
# # # ####29Jan2021 we do not need if check because we care making a local copy of the dataset
# # # # if(!eval( parse(text=paste('exists("',dframe,'", where=globalenv(), inherits=FALSE)',sep=''))))
# # # # {
# # # eval( parse(text=paste('.GlobalEnv$',dframe,' <- as.data.frame(',dframe,')',sep=''))) # making a copy of the dataset that may be in some package. This is will also take care of "tibble"
# # # # }
# # # }
# # # }
### 01Feb2021 Whole if block above can be replaced by following commented line.
BSkyCheckIfTibble(dataframeName=dframe) # converts to data.frame only if dataframe has tbl_df class
#print("Step5")
if(eval( parse(text=paste('exists("',dframe,'", where=globalenv(), inherits=FALSE)',sep=''))))#exists(dframe))
{
#cat("\nDataframe exists\n Is it a DataFrame : ")
#print(ls())
varenv <- eval( parse(text=paste('getEnv(dframe)',sep='')))
#print(varenv)
#print(eval( parse(text=paste('is.data.frame(.GlobalEnv$',dframe,')',sep='')))) #.GlobalEnv$
#cat("\n")
# if(eval( parse(text=paste('is.data.frame(',dframe,')',sep='')),envir=globalenv()))
if(eval( parse(text=paste('is.data.frame(.GlobalEnv$',dframe,')',sep='')))) #.GlobalEnv$
{
#print("HasCol and Has row")
eval(parse(text=paste('hascolumns <- ncol(.GlobalEnv$',dframe,') > 0'))) #20Oct2016 data.frame has cols
#print("A")
#print(hascolumns)
eval(parse(text=paste('hasrows <- nrow(.GlobalEnv$',dframe,') > 0'))) #20Oct2016 data.frame has rows
#print("B")
#print(hasrows)
if(hascolumns && hasrows)#20Oct2016 'if' condition added for the code below. And a full 'else' block is also added for this 'if'
{
#at this point we need to assign some column name(Var1, Var2) to the column those do not have any col-names.
GenerateUniqueColName(paste('.GlobalEnv$',dframe,sep='') )#its string so no need to enclose in single quotes
#Now replace special chars with underscore and prefix 'X' if there is digit in the first index of colum name.
colcount <- eval(parse(text=paste('length(names(.GlobalEnv$',dframe,'))' ))) #colcount = length(names(datasetname))
#cat('Count=')
#cat(colcount)
#ptm= proc.time()
for( i in 1:colcount)
{
if(logflag)
{
cat('\n-----------------------------------------------------------------')
cat("\nSpl Chr:OldName = ")
print(eval(parse(text=paste('names(.GlobalEnv$',dframe,')[',i,']',sep=''))))
}
eval(parse(text=paste('names(.GlobalEnv$',dframe,')[',i,'] <- make.names(names(.GlobalEnv$',dframe,')[',i,'])', sep='')))
#names(datasetname)[i] = ReplaceSplChrsAndPrefixXForDigitInBegining(names(datasetname)[i])
if(logflag)
{
cat("\nSpl Chr removed ColName = ")
print(eval(parse(text=paste('names(.GlobalEnv$',dframe,')[',i,']',sep=''))))
cat('\n-----------------------------------------------------------------')
}
}
#cat("\nIts Dataframe\n")
#eval( parse(text=paste(dframe,' <<- dframe',sep='')))
if(curidx == -1) # if not exists in list
{
#cat("\n adding to uadatasets$name:")
uadatasets$name <- c(uadatasets$name, dframe)
#cat("\n added to uadatasets$name:")
}
else
{
#cat("\nCreating filetype attribute\n")
filetype <- eval(parse(text=paste('attr(.GlobalEnv$',dframe,',"filetype")',sep=''))) ## this may not work. because dframe is new object. may not have attributes.
#bskymaxfactors <- eval(parse(text=paste('attr(',datasetname,',"maxfactor")',sep='')))
}
#cat("\n Creating Attrs for DS:")
#eval(parse(text=paste('attributes(',dframe,') <<- c(attributes(',dframe,'), list(split = c(FALSE), splitcolumnnames = c(), splitcolumnindex = c(), filetype = c("CSV"), slice = c(FALSE), align = c("Left"), maxfactor= c(bskymaxfactors)))', sep='')))
if(is.null(filetype) || nchar(filetype)==0)
filetype = "RDATA"
#cat("File Type:", filetype)
#eval(parse(text=paste('uadatasets$lst$',dframe,' <- .GlobalEnv$',dframe,sep='')))
#cat("\n added to uadatasets$lst$dfname:")
UAcreateExtraAttributes(dframe, filetype)
#cat("\n Finished Creating Attrs for DS:\n")
success = "SUCCESS"
#print(eval(parse(text=paste('colnames(',dframe,')'))))
replacePeriodWithUnderscore(dframe)
#print(eval(parse(text=paste('colnames(.GlobalEnv$',dframe,')'))))
#cat("\nFixing UTC")
#MakeAllDateColUTC(dframe)
#cat("\nUTC Fixed")
}
else
{
cat("Dataframe object does not have row(s) or column(s)!\n")
BSkyErrMsg =paste("BSky.LoadRefresh.Dataframe: Dataset has no row(s) or column(s)"," Dataset Name:", dframe)
warning("BSky.LoadRefresh.Dataframe: Dataset has no row(s) or column(s) ")
}
}
else
{
cat("Object is not data frame object");
warning("BSky.LoadRefresh.Dataframe: Object is not data frame object ")
}
}
else
{
cat("Dataframe object does not exists!")
BSkyErrMsg =paste("BSky.LoadRefresh.Dataframe: Dataset with the same name already on the global list"," Dataset Name:", dframe)
warning("BSky.LoadRefresh.Dataframe: Dataset with the same name already on the global list ")
}
BSkyFunctionWrapUp()
invisible(success)
}
#Closes a Dataframe, Makes it NULL
BSky.RemoveRefresh.Dataframe <- function(dframe)
{
BSkyFunctionInit()
BSkySetCurrentDatasetName(dframe)
#BSkyStoreApplicationWarnErrMsg(BSkyWarnMsg, BSkyErrMsg)
#curidx <- UAgetIndexOfDataSet(dframe) # current index of dataset if its already loaded ( before )
#cat("\nCur Index of CSV:",curidx)
if(eval( parse(text=paste('exists("dframe")',sep=''))))#exists(dframe))
{
if(eval( parse(text=paste('is.null("dframe")',sep='')))) #Already closed from elsewhere
{
#Just remove external attributes
}
else if(eval( parse(text=paste('is.data.frame(',dframe,')',sep='')))) # not closed. Close it from here
{
UAcloseDataset(dframe)
#remove external attributes
}
else
{
cat("Object is not data frame object");
}
}
else
{
cat("Dataframe object does not exists!")
BSkyErrMsg =paste("BSky.RemoveRefresh.Dataframe: Dataset with the same name already on the global list"," Dataset Name:", dframe)
warning("BSky.RemoveRefresh.Dataframe: Dataset with the same name already on the global list ")
}
BSkyFunctionWrapUp()
}
BSkyLoadRefreshDataframe2 <- function(dframe)
{
BSkyFunctionInit()
BSkySetCurrentDatasetName(dframe)
success <- NULL
curidx <- UAgetIndexOfDataSet(dframe) # current index of dataset if its already loaded ( before )
filetype = "RDATA" #default
#### CHANGE THIS TO ls()
if(eval( parse(text=paste('exists("',dframe,'", where=globalenv())',sep=''))))#exists(dframe))
{
varenv <- eval( parse(text=paste('getEnv(dframe)',sep='')))
#print(varenv)
#print(eval( parse(text=paste('is.data.frame(.GlobalEnv$',dframe,')',sep='')))) #.GlobalEnv$
#cat("\n")
# if(eval( parse(text=paste('is.data.frame(',dframe,')',sep='')),envir=globalenv()))
if(eval( parse(text=paste('is.data.frame(.GlobalEnv$',dframe,')',sep='')))) #.GlobalEnv$
{
eval(parse(text=paste('uadatasets$lst$',dframe,' <- .GlobalEnv$',dframe,sep='')))
#cat("\n added to uadatasets$lst$dfname:")
#UAcreateExtraAttributes(dframe, filetype)
#cat("\n Finished Creating Attrs for DS:\n")
success = "SUCCESS"
#print(eval(parse(text=paste('colnames(',dframe,')'))))
#replacePeriodWithUnderscore(dframe)
#print(eval(parse(text=paste('colnames(.GlobalEnv$',dframe,')'))))
}
else
{
cat("Object is not data frame object");
warning("BSky.LoadRefresh.Dataframe: Object is not data frame object ")
}
}
else
{
cat("Dataframe object does not exists!")
BSkyErrMsg =paste("BSky.LoadRefresh.Dataframe: Dataset with the same name already on the global list"," Dataset Name:", dframe)
warning("BSky.LoadRefresh.Dataframe: Dataset with the same name already on the global list ")
}
BSkyFunctionWrapUp()
invisible(success)
}
# dframe must be a name (a string) rather than an actual dataframe object
##Refreshes values in C1Datagrid, if dataset/dataframe was modified from syntax.
#Basically loads/refreshes the grid from memory(dataset/dataframe). Not by reading disk file from disk.
# BSkyLoadRefreshDataframe loads the specified dataframe in the BlueSky Statistics UI datagrid.
# If dataframe is already loaded in the UI datagrid then it refresh it for any changes done.
#
# To use BSkyLoadRefreshDataframe you must run it from within the BlueSky Statistics application.
#
# @name BSkyLoadRefreshDataframe
# @param dframe is the dataframe object(do not use quotes around the object)
# @param load.dataframe is provided so that you can selectively refresh or not refresh the UI datagrid.
# This parameter is defined in the dialog to make the decision, whether to refresh the UI datagrid or not.
# Default value is TRUE.
# @export
# @examples
# df <-data.frame(A=c(1,2,3), B=c(4,5,6), C=c(6,7,8))
# BSkyLoadRefreshDataframe(df)
BSkyLoadRefreshDataframe <- function(dframe, load.dataframe = TRUE)
{
#if(load.dataframe)
#{
#invisible(dframe)
#}
#else
#{
#invisible(NULL)
#}
BSkyLoadRefresh(dframe, load.dataframe);
}
### title should fit on one line, be written in sentence case, but not end in a full stop
### to print @ in the documentation, escape with one more @ (e.g. @@ prints @)
#' @title Loading a dataframe in the BlueSky Statistics UI datagrid
#'
#' @description loads the specified dataframe in the BlueSky Statistics UI datagrid.
#'
#' @param bskyDatasetName is the dataframe name (use quotes around the name)
#' @param load.dataframe is provided so that you can selectively refresh or not refresh the UI datagrid.
#' This parameter is defined in the dialog to make the decision, whether to refresh the UI datagrid or not. Default value is TRUE.
#'
#'
#' @examples
#' df <-data.frame(A=c(1,2,3), B=c(4,5,6), C=c(6,7,8))
#' BSkyLoadRefresh('df')
BSkyLoadRefresh <- function (bskyDatasetName, load.dataframe = TRUE, load.UIgrid = TRUE, isRmarkdownOutputOn = BSkyIsRmarkdownOutputOn(), maxFactor=BSkyGetMaxFactor(), createAttr=TRUE)## change this to a string parameter from a dataset object
{
isdataframe=FALSE
isdesign=FALSE
ists=FALSE
isPkgLoaded = FALSE
isexists = FALSE
error = FALSE
pkgname = c()
dsname = c()
pkgEnv = c()
ischar = is.character(bskyDatasetName)
originalDatasetname = bskyDatasetName
#for FALSE we can just return from the function.
if(!load.dataframe)
{
return(invisible())
}
#print(originalDatasetname)
#print(class(originalDatasetname))
if(ischar)## is argument a character. If not we print error in the sink and exit
{
hasPkgname = grepl("::", bskyDatasetName)
if(hasPkgname)#package name is passed
{
arr = strsplit(bskyDatasetName, "::")
pkgname = arr[[1]][1]
dsname = arr[[1]][2]
isPkgLoaded = pkgname %in% (.packages())
if(!isPkgLoaded)# if packge is not loaded
{
cat("\n") # forcing a new line in case someone created a cat() without a trailing new line
cat("Package ")
cat(pkgname)
cat(" not loaded. Use library(packageName) to load the package.")
error=TRUE
invisible()
}
pkgEnv = eval(parse(text=paste('as.environment("package:',pkgname,'")', sep='')))
isexists = eval(parse(text=paste('exists("',dsname,'", where=pkgEnv, inherits=FALSE)', sep='' )))# DS exists in the pkg.
if(isexists)
{
# dsclass = eval(parse(text=paste('"class(',pkgname,'::',dsname,')',sep='')))
# if( length(dsclass) > 1)
# {
#check if data.frame
# isdataframe = "data.frame" %in% dsclass
isdataframe = eval(parse(text=paste('"data.frame" %in% c(class(',pkgname,'::',dsname,'))',sep='')))
# if(isdataframe)
# {
# eval(parse(text=paste('"class(',pkgname,'::',dsname,')',sep='')))
# }
# }
isdesign = eval(parse(text=paste('"design" %in% c(class(',pkgname,'::',dsname,'))',sep='')))#for DoE
if(isdataframe && isdesign)#DoE
{
eval( parse(text=paste('.GlobalEnv$',dsname,' <- (',pkgname,'::',dsname,')',sep='')))##make a copy in globalEnv
#dsname = 'mtcars'
bskyDatasetName=dsname #overwrite bskyDatasetName as it may still have 'datasets::mtcars'
}
else if(isdataframe)
{
eval( parse(text=paste('.GlobalEnv$',dsname,' <- as.data.frame(',pkgname,'::',dsname,')',sep='')))##make a copy in globalEnv
#dsname = 'mtcars'
bskyDatasetName=dsname #overwrite bskyDatasetName as it may still have 'datasets::mtcars'
}
else
{
#lets check if it is a "ts" object, if it is, we try to convert it to data.frame object.
ists = eval(parse(text=paste('"ts" %in% c(class(',pkgname,'::',dsname,'))',sep='')))
if(ists)
{
##make it a data.frame and also make a copy in globalEnv
eval( parse(text=paste('.GlobalEnv$',dsname,' <- as.data.frame(',pkgname,'::',dsname,')',sep='')))
#dsname = 'mtcars'
bskyDatasetName=dsname #overwrite bskyDatasetName as it may still have 'datasets::mtcars'
}
}
}
else
{
cat("\n") # forcing a new line in case someone created a cat() without a trailing new line
cat("Dataset ")
cat(dsname)
cat(" is not found in the package ")
cat(pkgname)
error= TRUE
invisible()
}
}
else#package name is not passed so, whatever is passed should be considered as a datasetname in global env.
{
inGlobalEnv = eval( parse(text=paste('exists("',bskyDatasetName,'", where=globalenv(), inherits=FALSE)',sep='')))
# if dataset is not found in the Global environment then we should not go searching it
# because same named datasets can be found in multiple packages. And user may get confused if we load
# the first one we found. So we simply error out. User is supposed to pass package name to avoid ambiguity
if(inGlobalEnv)
{
isdataframe = eval(parse(text=paste('"data.frame" %in% c(class(',bskyDatasetName,'))',sep='')))
isdesign = eval(parse(text=paste('"design" %in% c(class(',bskyDatasetName,'))',sep=''))) # for DoE
#check if it is a "ts" object
ists = eval(parse(text=paste('"ts" %in% c(class(',bskyDatasetName,'))',sep='')))
if(isdesign && isdataframe)
{
##make it a data.frame and also make a copy in globalEnv
eval( parse(text=paste('.GlobalEnv$',bskyDatasetName,' <- (',bskyDatasetName,')',sep='')))
}
else if(ists || isdataframe)#DoE
{
##make it a data.frame and also make a copy in globalEnv
eval( parse(text=paste('.GlobalEnv$',bskyDatasetName,' <- as.data.frame(',bskyDatasetName,')',sep='')))
}
if ( !isdataframe && !ists ) #not a data.frame (and ts) and the object exists
{
cat("\n") # forcing a new line in case someone created a cat() without a trailing new line
cat("Syntax error (BSkyLoadRefresh expects the name of the data.frame (or ts) object as a character string) or the object referenced is not an object of class data.frame (or ts)")
cat("\nCorrect syntax is:")
cat("\n\tBSkyLoadRefresh('dataframe-name') #load an object of class data.frame (or ts)")
cat("\n\tBSkyLoadRefresh('packageName::dataframe-name') #load an object of class data.frame (or ts) from a specific R package")
error= TRUE
invisible()
}
}
else
{
isexists = eval(parse(text=paste('exists("' ,bskyDatasetName,'")', sep='' )))# DS exists in the pkg.
if (isexists)
{
eval( parse(text=paste('.GlobalEnv$',bskyDatasetName,' <- ',bskyDatasetName,sep='')))##make a copy in globalEnv
#dsname = 'mtcars'
isdataframe = eval(parse(text=paste('"data.frame" %in% c(class(',bskyDatasetName,'))',sep='')))
isdesign = eval(parse(text=paste('"design" %in% c(class(',bskyDatasetName,'))',sep='')))
#check if it is a "ts" object
ists = eval(parse(text=paste('"ts" %in% c(class(',bskyDatasetName,'))',sep='')))
if(isdesign && isdataframe)#DoE
{
eval( parse(text=paste('.GlobalEnv$',bskyDatasetName,' <- (',bskyDatasetName,')',sep='')))
}
else if(ists)
{
##make it a data.frame and also make a copy in globalEnv
eval( parse(text=paste('.GlobalEnv$',bskyDatasetName,' <- as.data.frame(',bskyDatasetName,')',sep='')))
}
if ( !isdataframe && !ists ) #not a data.frame or ts and the object exists
{
cat("\n") # forcing a new line in case someone created a cat() without a trailing new line
cat("Syntax error (BSkyLoadRefresh expects the name of the data.frame (or ts) object as a character string) or the object referenced is not an object of class data.frame (or ts)")
cat("\nCorrect syntax is:")
cat("\n\tBSkyLoadRefresh('dataframe-name') #load an object of class data.frame (or ts)")
cat("\n\tBSkyLoadRefresh('packageName::dataframe-name') #load an object of class data.frame (or ts) from a specific R package")
error= TRUE
invisible()
}
}
else
{
cat("\n") # forcing a new line in case someone created a cat() without a trailing new line
cat("Dataset ")
cat(bskyDatasetName)
cat(" cannot be found. Please check the spelling of the dataset name or load the package containing the dataset")
error= TRUE
invisible()
}
}
}
}
hasrows=FALSE
##Check if number of rows is greater than zero. Can't rely on column count because col count was greater than zero in subset-ed dataset that had no rows.
rowcount = eval( parse(text=paste('nrow(.GlobalEnv$',bskyDatasetName,')',sep='')))
if(rowcount>0)
{
hasrows = TRUE
}
else
{
# The data frame 'bskyDatasetName' has zero rows, hence it will not be loaded in the datagrid
cat("\nERROR: The data frame '") # forcing a new line in case someone created a cat() without a trailing new line
cat(bskyDatasetName)
cat("' has zero rows, hence it will not be loaded in the datagrid")
error= TRUE
invisible()
}
if(load.dataframe && !error)
{
# first check the whether the dataset exsits in the uadatasets list - if so no processing of the dataset is needed - Anil needs to verify this
# if the dataset name does not exist in the uadatasets, then do all prep (attrbutes, variables, etc) the dataset and add to the uadatasets
#Following block is added by Anil. 28Nov2020
#curidx <- UAgetIndexOfDataSet(bskyDatasetName) #strip out .GlobalEnv$ from the bskyDatasetName and then looks in to uadatasets$name
#if(curidx == -1)
#{
#BSky.LoadRefresh.Dataframe(bskyDatasetName)#This is what we call from C# behind the scenes if the memory dataframe is a new one.
#}
##get dataframe object from the dataframename
#dframeobj = eval( parse(text=paste('as.data.frame(.GlobalEnv$',bskyDatasetName,')',sep='')))
##we can put dframeobj object=c(dframeobj)in the Sanjay's code below or we can just put the name object=c(bskyDatasetName)
##if we put object then python gets the object directly.
##if we put name then python code must know that its a name and get object of that name from R memory and then process it.
# if((exists("uadatasets.sk") && exists("BSkyKableFormatting", env=uadatasets.sk) && uadatasets.sk$BSkyKableFormatting == FALSE))
# {
# doKableFormatting = FALSE
# }
# else
# {
# doKableFormatting = TRUE
# if((exists("uadatasets.sk") && exists("BSkyRmarkdownFormatting", env=uadatasets.sk) && uadatasets.sk$BSkyRmarkdownFormatting == FALSE))
# {
# doRmarkdownFormatting = FALSE
# }
# else
# {
# doRmarkdownFormatting = TRUE
# }
# isRmarkdownOutputOn = doRmarkdownFormatting
# }
# after the above processing do the following
# if(isRmarkdownOutputOn == FALSE && trimws(originalDatasetname)!= "bsky_piped_temp_dataset")
if(isRmarkdownOutputOn == FALSE && load.UIgrid == TRUE)
{
cat("\n") # forcing a new line in case someone created a cat() without a trailing new line
#print("Entered")
#print(originalDatasetname)
print("BSkyDataGridRefresh")
}
###################################################################################
#11/24/20 - The following code is for creating an in memory queue instead of a sync file to signal the new (Python, etc) application tier
####################################################################################
#if(trimws(originalDatasetname)!= "bsky_piped_temp_dataset")
if(load.UIgrid == TRUE)
{
if(!exists("holdBSkyFormatObjectNew", env=uadatasets.sk) || is.null(uadatasets.sk$holdBSkyFormatObjectNew))
{
uadatasets.sk$holdBSkyFormatObjectNew = list(list(type=c("BSkyDataGridRefresh"), object=c(bskyDatasetName)))
}
else
{
uadatasets.sk$holdBSkyFormatObjectNew = c(uadatasets.sk$holdBSkyFormatObjectNew, list(list(type=c("BSkyDataGridRefresh"), object=c(bskyDatasetName))))
}
}
###### 23 Jan 2021 ### Add dataset in uadataset$name if it is not in the list ######
datasetname <- BSkyValidateDataset(bskyDatasetName)
if(is.null(datasetname))
{
uadatasets$name <- c(uadatasets$name, bskyDatasetName)
}
if(createAttr)
{
UAcreateExtraAttributes(bskyDatasetName, "RDATA")
}
# if maxFactor = -1 then we do not convert factor col to character
# if maxFactor is a positive integer and factor columns has levels more than maxFactor we convert this col to character.
## 09Oct2022
## we do not put maaxFactor limit to RDATA and RDA files while opening the file.
## Here we are mostly dealing with memory dataset.
## Say I loaded a RDATA file with a col having factor levels more the the maxFactor setting
## because we do not restrict RData we load that col as factor (and not make it char)
## now user made a copy of this dataset (in R Editor) and loadRefresh' this memory dataset
## will load and the factor col turns to character because of the following code. User may get
## confused why the factor to character conversion happened.
## So we should comment the code below (or make that IF FALSE)
if(FALSE) ## (maxFactor > 0 ) ##
{
# columns having more than 30 factors are converted back to character
colcount = eval(parse(text=paste('ncol(',bskyDatasetName,')')))
for(i in 1:colcount)
{
coluname = eval(parse(text=paste('colnames(.GlobalEnv$',bskyDatasetName,')[',i,']',sep='')))
colclass = eval(parse(text=paste('class(.GlobalEnv$',bskyDatasetName,'$',coluname,')',sep='')))
if("factor" %in% colclass)
{
lvlcount = eval(parse(text=paste('length(levels(.GlobalEnv$',bskyDatasetName,'$',coluname,'))',sep='')))
if(lvlcount > maxFactor)
{
eval(parse(text=paste('.GlobalEnv$',bskyDatasetName,'$',coluname,'<- as.character(.GlobalEnv$',bskyDatasetName,'$',coluname,')',sep='' )))
}
}
}
}
}
}
BSkyRemoveRefreshDataframe <- function(dframe)
{
BSky.RemoveRefresh.Dataframe (dframe)
}
##10Mar2022
BSkyGetMaxFactor <- function()
{
if(exists("maxFactor", env=uadatasets.sk))
{
return(invisible(uadatasets.sk$maxFactor))
}
else
{
# -1 here means no restriction on maximum factors in a column
return(invisible(-1))#return(invisible(30))
}
}
##10Mar2022
BSkySetMaxFactor <- function(maxFactor = -1)
{
uadatasets.sk$maxFactor = maxFactor
return(invisible(uadatasets.sk$maxFactor))
}
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.