Adding rows to different dataframe programatically - r

Question

Im creating a dataframe dynamically and Im using custom names to refer to those data frames.How ever, I can succesfully create the data frames dynamically and add information individually but manually when i try to add a record to it it will run the action but nothing happens. I can open the data frame and it shows as empty
#Extract unique machines on the system
machines <- unique(wo_raw$MACHINE)
for(machine in machines){
#Check if the machine is present on current data frames or has a record
if(exists(machine) && is.data.frame(get(machine))){
#Machine already exists on the system
cat(machine," is a dataframe","\n")
netlbs <- subset(wo_raw,((wo_raw$TYPE =="T" & wo_raw$TYPE2=="E") | (wo_raw$TYPE == "T" & is.na(wo_raw$TYPE2))) & wo_raw$WEEK<=curWeek & wo_raw$MACHINE == machine & wo_raw$YEAR == curYear,select = NET_LBS)
scraplbs<- subset(wo_raw,((wo_raw$TYPE =="T" & wo_raw$TYPE2=="E") | (wo_raw$TYPE == "T" & is.na(wo_raw$TYPE2))) & wo_raw$WEEK<=curWeek & wo_raw$MACHINE == machine & wo_raw$YEAR == curYear,select = SCRAP_LBS)
if(is.data.frame(netlbs) && nrow(netlbs)!=0){
totalNet<- sum(netlbs)
totalScrap<- sum(scraplbs)
scrapRate <- percent(totalScrap/(sum(totalNet,totalScrap)),accuracy = 2)
tempDf<-data.frame(curYear,curMonth,curDay,curWeek,totalNet,totalScrap,scrapRate)
names(tempDf)<-c("year","month","day","week","net_lbs","scrap_lbs","scrap_rate")
cat("Total Net lbs for ",machine,": ",totalNet,"\n")
cat("Total Scrap lbs for ",machine,": ",totalScrap,"\n")
cat("Total Scrap Rate for ",machine,": ",scrapRate,"\n")
#machine<-rbind(get(machine),tempDf)
#assign(machine,rbind(machine,tempDf))
add_row(get(machine),year=curYear,
month=curMonth,
day=curDay,
week=curWeek,
net_lbs=totalNet,
scrap_lbs=totalScrap,
scrap_rate=scrapRate)
cat("added row \n")
}
#info<-c(curYear,curMonth,curDay,curWeek,netlbs)
#cat("Total Net lbs: ",netlbs,"\n")
#netlbs <-NULL
}else{
cat("Creating machine dataframe: ",machine,"\n")
#Create a dataframe labeled with machine name contining
#date information, net lbs,scrap lbs and scrap rate
assign(paste0(machine,""),data.frame(year=integer(),
month=integer(),
day=integer(),
week=integer(),
net_lbs=double(),
scrap_lbs=double(),
scrap_rate=integer()
)
)
#machine$year<-curYear
}
#machine<-NULL
}
All the functions that I've tried are in commented lines from previous answers found on Stack Overflow. I did get working with a for but i dont think that would be really feasible since it will consume a lot of resources plus it doesn't work well when handling various data types . Does anybody have an idea of whats going on, I don't have an error to go by.

Answer 1

I think your code needs quite a lot of cleanup. Make sure you know for yourself at each step what exactly you are handling.
Some hints:
Try to make your code self-contained. If I run your code, I get an error right away, as I don't have wo_raw defined. I understand it's some kind of data.frame, but exactly what is in there? What do I need to do to try to run your code? Also with variables like curYear. I get that it needs to be 2019, but I need to type an awful lot to just get to the problem, I can't just copy-paste.
If you use any libraries, please also include a line for them. I don't know what add_row does or is supposed to do. So I also don't know if that's where your expectations are wrong?
Try to make your code minimal before posting it here. I like the comments and cats sprinkled throughout, but why a line such as netlbs <- subset(wo_raw,((wo_raw$TYPE =="T" & wo_raw$TYPE2=="E") | (wo_raw$TYPE == "T" & is.na(wo_raw$TYPE2))) & wo_raw$WEEK<=curWeek & wo_raw$MACHINE == machine & wo_raw$YEAR == curYear,select = NET_LBS)? For this problem, just something like subset(wo_raw, wo_raw$mach==machine, net) would suffice
I get that the code works, but try to work out where you are using what kind of objects. if (is.data.frame(netlbs)) {total=sum(netlbs)} may work, but summing a data.frame while you actually just need a column leads to confusion.
When using variables to store the names of other variables such as you are doing, be very aware of what you are actually refering to. For that reason, it's generally advisable to steer clear of these constructs, it's almost always easier to store your results in a list or something similar
Come to that: the variable machine is not a data.frame, it's a character. That character is the name of another variable, which is a data.frame. So (commented out) I see some instances of machine <- NULL and machine$year, those are wrong. As is rbind(machine, ...), as machine is not a data.frame
That being said, I think you got close with the assign-statement.
Does assign(machine,rbind(get(machine),tempDf)) work?