R，R编程，循环目录

Question

Hi I am trying to loop through a directory of excel files for analysis.

My Variable is named FileToGrab which obtains the name of the excel file.

Where I have FileToGrab in bold is what I want the data frame to be named not the actual FileToGrab data frame.

Example FileToGrab = 2013ExcelSheet23

I want my Data Frame to be named 2013ExcelSheet23 and not FileToGrab.

FileToGrab = 2013ExcelSheet24

I want my Data Frame to be named 2013ExcelSheet24 and not FileToGrab.

FileToGrab = 2013ExcelSheet25

I want my Data Frame to be named 2013ExcelSheet25 and not FileToGrab.

..... and so on.

New to R sorry if this does not make sense. Thanks

x <- 1:50
for(i in seq(along=x))
{


FileToGrab  = gsub("(^ +)|( +$)", "",listofFile[i])
FileToGrab  = str_replace_all(string=FileToGrab, pattern=" ", repl="")

DirFileName = paste("C:\\Users\\w47593\\Desktop\\RProjects\\CallCenterProjectJuly2013\\Files\\",FileToGrab)
DirFileName = str_replace_all(string=DirFileName, pattern=" ", repl="")

file.name <- DirFileName 
sheet.name <- "Detail"
FileToGrab = str_replace_all(string=FileToGrab, pattern=".xls", repl="")


## Connect to Excel File Pull and Format Data
excel.connect <- odbcConnectExcel(DirFileName)
**FileToGrab**  <- sqlFetch(excel.connect, sheet.name, na.strings=c("","-"))
odbcClose(excel.connect)


}

Answer 1

You probably don't want to name your objects starting with numbers as you would have to quote them each time you used them

> 11Foo <- 1
Error: unexpected symbol in "11Foo"
> `11Foo` <- 1
> 11Foo
Error: unexpected symbol in "11Foo"
> `11Foo`
[1] 1

Like wise, I doubt you want 25+ objects clogging up your workspace. A far better solution is often to import the data into a list and work with those objects. You have similar issues with accessing the names

> ll <- list(`1` = 1, `2` = 2)
> ll$1
Error: unexpected numeric constant in "ll$1"
> ll$`1`
[1] 1

but then you don't need to refer to them by name necessarily, and you benefit by being able to iterate over the list using lapply etc.

I would use something like

fs <- list.file("dir/to/excel/files", glob2rx("*.xls"))
ll <- vector(mode = "list", length = length(fs))

for (i in seq_along(ll)) {
  excel.connect <- odbcConnectExcel(fs[i])
  ll[[i]] <- sqlFetch(excel.connect, sheet.name, na.strings=c("","-"))
  odbcClose(excel.connect)
}

names(ll) <- sub("\\.xls", "", fs)

You would still have to extract via

ll$"2013ExcelSheet25"

but you can also use

ll[["2013ExcelSheet25"]]

or better

ll[[1]]

or even

ll[[which(names(ll) == "2013ExcelSheet25")]]

But because these are all in a single list the are contained, and you can operate on them via lapply and co.

Answer 2

Why not use

files = list.files(DirFileName)

and then iterate through that to load them into R?

Assignment to object using filenames:

objects = list()
objects[[files[1]]] = ...