XLGetRange: Import Data Directly From Excel

Description

The package RDCOMClient is used to open an Excel workbook and return the content (value) of one (or several) given range(s) in a specified sheet. This is helpful, whenever pathologically scattered data on an Excel sheet, which can't simply be saved as CSV-file, has to be imported in R.

XLGetWorkbook() does the same for all the sheets in an Excel workbook.

Usage

XLGetRange(file = NULL, sheet = NULL, range = NULL, as.data.frame = TRUE,
           header = FALSE, stringsAsFactors = FALSE, echo = FALSE,
           na.strings = NULL, skip = 0)
XLGetWorkbook(file, compactareas = TRUE)
XLCurrReg(cell)
XLNamedReg(x)

Value

If as.data.frame is set to TRUE, a single data.frame or a list of data.frames will be returned. If set to FALSE a list of the cell values in the specified Excel range, resp. a list of lists will be returned.

XLGetWorkbook() returns a list of lists of the values in the given workbook.

Arguments

file: the fully specified path and filename of the workbook. If it is left as NULL, the function will look for a running Excel-Application and use its current sheet. The parameter sheet will be ignored in this case.
sheet: the name of the sheet containing the range(s) of interest.
range: a scalar or a vector with the address(es) of the range(s) to be returned (characters). Use "A1"-address mode to specify the ranges, for example "A1:F10".
If set to NULL (which is the default), the function will look for a selection that contains more than one cell. If found, the function will use this selection. If there is no selection then the current region of the selected cell will be used. Use XLCurrReg() if the current region of a cell, which is currently not selected, should be used. Range names can be provided with XLNamedReg("name").
as.data.frame: logical. Determines if the cellranges should be coerced into data.frames. Defaults to TRUE, as this is probably the common use of this function.
header: a logical value indicating whether the range contains the names of the variables as its first line. Default is FALSE. header is ignored if as.data.frame has been set to FALSE.
stringsAsFactors: logical. Should character columns be coerced to factors? The default is FALSE, which will return character vectors.
echo: logical. If set to TRUE, the function will print the full command used, such that it can be copied into the R-script for future use.
na.strings: a character vector of strings which are to be interpreted as NA values. Blank fields are always considered to be missing values. Default is NULL, meaning none.
compactareas: logical, defining if areas should be returned by XLGetWorkbook as list or as matrix (latter is default).
cell: range of the left uppe cell, when current region should be used.
x: the name or the index of the XL-name to be used.
skip: the number of lines of the data file to skip before beginning to read data.

Author

Andri Signorell <andri@signorell.net>

Details

The result consists of a list of lists, if as.data.frame is set to FALSE. Be then prepared to encounter NULL values. Those will prevent from easily being able to coerce the square data structure to a data.frame.

The following code will replace the NULL values by NA and coerce the data to a data.frame.


  # get the range D1:J69 from an excel file
  xlrng <- XLGetRange(file="myfile.xlsx", sheet="Tabelle1",
                      range="D1:J69", as.data.frame=FALSE)
  # replace NULL values by NA
  xlrng[unlist(lapply(xlrng, is.null))] <- NA
  # coerce the square data structure to a data.frame
  d.lka <- data.frame(lapply(data.frame(xlrng), unlist))

This of course can be avoided by setting as.data.frame = TRUE.

The function will return dates as integers, because MS-Excel stores them internally as integers. Such a date can subsequently be converted with the (unusual) origin of as.Date(myDate, origin="1899-12-30"). See also XLDateToPOSIXct, which does the job. The conversion can directly be performed by XLGetRange() if datecols is used and contains the date columns in the sheet data.

Examples

Run this code

if (FALSE)  # Windows-specific example

XLGetRange(file="C:/My Documents/data.xls",
           sheet="Sheet1",
           range=c("A2:B5","M6:X23","C4:D40"))


# if the current region has to be read (incl. a header), place the cursor in the interesting region
# and run:
d.set <- XLGetRange(header=TRUE)

# Get XL nameslist
nm <- xl$ActiveWorkbook()$names()

lst <- list()
for(i in 1:nm$count())
  lst[[i]] <- c(name=nm[[i]]$name(), 
                address=nm[[i]]$refersToRange()$Address())
  
# the defined names
as.data.frame(do.call(rbind, lst), stringsAsFactors = FALSE)

Run the code above in your browser using DataLab