r - Unwanted rows in data.frame -

i want write function checks number of files name id.csv , return number of rows no nas. here wrote:

        complete <- function(directory, id) {         setwd(directory)         my_list <- list.files(getwd(), pattern="*.csv", full.names=true)         my_id <- numeric()         my_output<-data.frame()         my_count<-numeric()         for(integer in id){             my_data <- read.csv(my_list[integer])             my_subset <- subset(my_data, sulfate !=0 & nitrate !=0)             my_count[integer]<-length(my_subset[[1]])             my_id[integer] <- integer         }         my_output<-cbind(my_id, my_count)         my_output     } complete("/home/jpasquier/téléchargements/specdata", c(1,3)) 

but here result:

> complete("/home/jpasquier/téléchargements/specdata", c(1,3))      my_id my_count [1,]     1      117 [2,]    na       na [3,]     3      243 

so don't understand why there unwanted row number2, after there no integer equal 2 in id argument. thanks.

you getting nas because assigning values 1st , 3rd value of variable, , nothing 2nd.

example:

v <- vector() v #logical(0) v[1] <- 111 v[3] <- 333  v #[1] 111  na 333 

edit:

here working example, no need create function:

#create dummy csv df1 <- data.frame(x=c(1,na,2,0,4, 4),                   y=c(1,3, 4,0,na,6)) df2 <- data.frame(x=c(1,0,2,0, 4,0),                   y=c(1,0,4,0,na,0))  write.csv(df1,"id_1.csv",row.names = false) write.csv(df1,"id_2.csv",row.names = false) write.csv(df2,"id_3.csv",row.names = false) write.csv(df2,"id_4.csv",row.names = false)  #choose csv id <- c(1,3)  #get non 0 counts res <- lapply(lapply(list.files(pattern="^id.*.csv")[id],read.csv),               function(i)nrow(subset(i,x!=0,y!=0)))  #update result names names(res) <- list.files(pattern="^id.*.csv")[id]  #output result counts res # $id_1.csv # [1] 4 #  # $id_3.csv # [1] 3