read.xls {gdata}R Documentation

Read Excel files


Read a Microsoft Excel file into a data frame


read.xls(xls, sheet=1, verbose=FALSE, pattern, na.strings=c("NA","#DIV/0!"),
         ..., method=c("csv","tsv","tab"), perl="perl")
xls2csv(xls, sheet=1, verbose=FALSE, blank.lines.skip=TRUE, ..., perl="perl")
xls2tab(xls, sheet=1, verbose=FALSE, blank.lines.skip=TRUE, ..., perl="perl")
xls2tsv(xls, sheet=1, verbose=FALSE, blank.lines.skip=TRUE, ..., perl="perl")
xls2sep(xls, sheet=1, verbose=FALSE, blank.lines.skip=TRUE, ...,
        method=c("csv","tsv","tab"), perl="perl")



path to the Microsoft Excel file. Supports "http://", "https://", and "ftp://" URLS.


name or number of the worksheet to read


logical flag indicating whether details should be printed as the file is processed.


if specified, them skip all lines before the first containing this string


name of the perl executable to be called.


intermediate file format, "csv" for comma-separated and "tab" for tab-separated


a character vector of strings which are to be interpreted as 'NA' values. See read.table for details.


logical flag indicating whether blank lines in the orginal file should be ignored.


additional arguments to read.table. The defaults for read.csv() are used.


This function works translating the named Microsoft Excel file into a temporary .csv or .tab file, using the xls2csv or xls2tab Perl script installed as part of this (gdata) package.

Caution: In the conversion to csv, strings will be quoted. This can be problem if you are trying to use the comment.char option of read.table since the first character of all lines (including comment lines) will be "\"" after conversion.

If you have quotes in your data which confuse the process you may wish to use read.xls(..., quote = ''). This will cause the quotes to be regarded as data and you will have to then handle the quotes yourself after reading the file in.

Caution: If you call "xls2csv" directly, is your responsibility to close and delete the file after using it.


"read.xls" returns a data frame.

"xls2sep" returns a temporary file in the specified format. "xls2csv" and "xls2tab" are simply wrappers for "xls2sep" specifying method as "csv" or "tab", respectively.


Excel file support is deprecated in gdata version 2.19.0, to be removed in the near future. At the user level, the functions installXLSXsupport, read.xls, sheetCount, sheetNames, xls2csv, xls2sep, xls2tab, xls2tsv, and xlsFormats now raise a warning about the functionality being deprecated.

Excel file support was first introduced in gregmisc/gdata 1.11.0 back in 2004, but today we have packages such as openxlsx, readxl, XLConnect, and xlsx offering dedicated Excel file support with more features. When Excel file support will be removed from the gdata package, it will result in lighter package maintenance and no more Perl warnings on the user side.


Either a working version of Perl must be present in the executable search path, or the exact path of the perl executable must be provided via the perl argument. See the examples below for an illustration.


Gregory R. Warnes, Jim Rogers, and Gabor Grothendiek

See Also



   ## Not run: 
   # iris.xls is included in the gregmisc package for use as an example
   xlsfile <- file.path(path.package('gdata'),'xls','iris.xls')

   iris <- read.xls(xlsfile) # defaults to csv format
   iris <- read.xls(xlsfile,method="csv") # specify csv format
   iris <- read.xls(xlsfile,method="tab") # specify tab format

   head(iris)  # look at the top few rows

   ## Example specifying exact Perl path for default MS-Windows install of
   ## ActiveState perl
   iris <- read.xls(xlsfile, perl="C:/perl/bin/perl.exe")

   ## Example specifying exact Perl path for Unix systems
   iris <- read.xls(xlsfile, perl="/usr/bin/perl")

   ## finding perl 
   ## (read.xls automatically calls findPerl so this is rarely needed)
   perl <- gdata:::findPerl("perl")
   iris <- read.xls(xlsfile, perl=perl)

   ## read example xls file from github
   xls.url <- paste0(
   xls <- read.xls(xls.url)

   ## read xls file ignoring all lines prior to first containing State
   ## (This URL is no longer valid.)
   crime.url <- ""
   crime <- read.xls(crime.url, pattern = "State")

   ## use of xls2csv - open con, print two lines, close con
   con <- xls2csv(crime.url)
   print(readLines(con, 2)) 

   ## Examples demonstrating selection of specific 'sheets'
   ## from the example XLS file 'ExampleExcelFile.xls'
   exampleFile <- file.path(path.package('gdata'),'xls',
   exampleFile2007 <- file.path(path.package('gdata'),'xls',

   ## see the number and names of sheets:
   if( 'XLSX' %in% xlsFormats() )  ## if XLSX is supported..

   ## show names of shets in the file
   if( 'XLSX' %in% xlsFormats() )  ## if XLSX is supported..

   data <- read.xls(exampleFile)          ## default is first worksheet
   data <- read.xls(exampleFile, sheet=2) ## second worksheet by number
   data <- read.xls(exampleFile, sheet="Sheet Second",v=TRUE) ## and by name

   ## load the third worksheet, skipping the first two non-data lines...
   if( 'XLSX' %in% xlsFormats() )  ## if XLSX is supported..
     data <- read.xls(exampleFile2007, sheet="Sheet with initial text", skip=2)

   ## load a file containing data and column names using latin-1
   ## characters
   latinFile <- file.path(path.package('gdata'),'xls','latin-1.xls')
   latin1 <- read.xls(latinFile, fileEncoding="latin1")
## End(Not run)

[Package gdata version 2.19.0 Index]