Coming from openxlsx
you might know about
read.xlsx()
(two functions, one for files and one for
workbooks) and readWorkbook()
. Functions that do different
things, but mostly the same. In openxlsx2
we tried our best
to reduce the complexity under the hood and for the user as well. In
openxlsx2
they are replaced with read_xlsx()
,
wb_read()
and they share the same underlying function
wb_to_df()
.
For this example we will use example data provided by the package. You can locate it in our “inst/extdata” folder. The files are included with the package source and you can open them in any calculation software as well.
We begin with the openxlsx2_example.xlsx
file by telling
R where to find this file on our system
The object contains a path to the xlsx file and we pass this file to our function to read the workbook into R
# import workbook
library(openxlsx2)
wb_to_df(file)
#> Var1 Var2 NA Var3 Var4 Var5 Var6 Var7 Var8
#> 3 TRUE 1 NA 1 a 2023-05-29 3209324 This #DIV/0! 01:27:15
#> 4 TRUE NA NA #NUM! b 2023-05-23 <NA> 0 14:02:57
#> 5 TRUE 2 NA 1.34 c 2023-02-01 <NA> #VALUE! 23:01:02
#> 6 FALSE 2 NA <NA> #NUM! <NA> <NA> 2 17:24:53
#> 7 FALSE 3 NA 1.56 e <NA> <NA> <NA> <NA>
#> 8 FALSE 1 NA 1.7 f 2023-03-02 <NA> 2.7 08:45:58
#> 9 NA NA NA <NA> <NA> <NA> <NA> <NA> <NA>
#> 10 FALSE 2 NA 23 h 2023-12-24 <NA> 25 <NA>
#> 11 FALSE 3 NA 67.3 i 2023-12-25 <NA> 3 <NA>
#> 12 NA 1 NA 123 <NA> 2023-07-31 <NA> 122 <NA>
The output is created as a data frame and contains data types date,
logical, numeric and character. The function to import the file to R,
wb_to_df()
provides similar options as the
openxlsx
functions read.xlsx()
and
readWorkbook()
and a few new functions we will go through
the options. As you might have noticed, we return the column of the xlsx
file as the row name of the data frame returned. Per default the first
sheet in the workbook is imported. If you want to switch this, either
provide the sheet
parameter with the correct index or
provide the sheet name.
col_names
- first row as column nameIn the previous example the first imported row was used as column name for the data frame. This is the default behavior, but not always wanted or expected. Therefore this behavior can be disabled by the user.
# do not convert first row to column names
wb_to_df(file, col_names = FALSE)
#> B C D E F G H I J
#> 2 Var1 Var2 NA Var3 Var4 Var5 Var6 Var7 Var8
#> 3 TRUE 1 NA 1 a 2023-05-29 3209324 This #DIV/0! 01:27:15
#> 4 TRUE <NA> NA #NUM! b 2023-05-23 <NA> 0 14:02:57
#> 5 TRUE 2 NA 1.34 c 2023-02-01 <NA> #VALUE! 23:01:02
#> 6 FALSE 2 NA <NA> #NUM! <NA> <NA> 2 17:24:53
#> 7 FALSE 3 NA 1.56 e <NA> <NA> <NA> <NA>
#> 8 FALSE 1 NA 1.7 f 2023-03-02 <NA> 2.7 08:45:58
#> 9 <NA> <NA> NA <NA> <NA> <NA> <NA> <NA> <NA>
#> 10 FALSE 2 NA 23 h 2023-12-24 <NA> 25 <NA>
#> 11 FALSE 3 NA 67.3 i 2023-12-25 <NA> 3 <NA>
#> 12 <NA> 1 NA 123 <NA> 2023-07-31 <NA> 122 <NA>
detect_dates
- convert cells to R datesThe creators of the openxml standard are well known for mistakenly
treating something as a date and openxlsx2
has built in
ways to identify a cell as a date and will try to convert the value for
you, but unfortunately this is not always a trivial task and might fail.
In such a case we provide an option to disable the date conversion
entirely. In this case the underlying numerical value will be
returned.
# do not try to identify dates in the data
wb_to_df(file, detect_dates = FALSE)
#> Var1 Var2 NA Var3 Var4 Var5 Var6 Var7 Var8
#> 3 TRUE 1 NA 1 a 45075 3209324 This #DIV/0! 0.06059028
#> 4 TRUE NA NA #NUM! b 45069 <NA> 0 0.58538194
#> 5 TRUE 2 NA 1.34 c 44958 <NA> #VALUE! 0.95905093
#> 6 FALSE 2 NA <NA> #NUM! NA <NA> 2 0.72561343
#> 7 FALSE 3 NA 1.56 e NA <NA> <NA> NA
#> 8 FALSE 1 NA 1.7 f 44987 <NA> 2.7 0.36525463
#> 9 NA NA NA <NA> <NA> NA <NA> <NA> NA
#> 10 FALSE 2 NA 23 h 45284 <NA> 25 NA
#> 11 FALSE 3 NA 67.3 i 45285 <NA> 3 NA
#> 12 NA 1 NA 123 <NA> 45138 <NA> 122 NA
show_formula
- show formulas instead of resultsSometimes things might feel off. This can be because the openxml
files are not updating formula results in the sheets unless they are
opened in software that provides such functionality as certain tabular
calculation software. Therefore the user might be interested in the
underlying functions to see what is going on in the sheet. Using
show_formula
this is possible
# return the underlying Excel formula instead of their values
wb_to_df(file, show_formula = TRUE)
#> Var1 Var2 NA Var3 Var4 Var5 Var6 Var7 Var8
#> 3 TRUE 1 NA 1 a 2023-05-29 3209324 This E3/0 01:27:15
#> 4 TRUE NA NA #NUM! b 2023-05-23 <NA> C4 14:02:57
#> 5 TRUE 2 NA 1.34 c 2023-02-01 <NA> #VALUE! 23:01:02
#> 6 FALSE 2 NA <NA> #NUM! <NA> <NA> C6+E6 17:24:53
#> 7 FALSE 3 NA 1.56 e <NA> <NA> <NA> <NA>
#> 8 FALSE 1 NA 1.7 f 2023-03-02 <NA> C8+E8 08:45:58
#> 9 NA NA NA <NA> <NA> <NA> <NA> <NA> <NA>
#> 10 FALSE 2 NA 23 h 2023-12-24 <NA> SUM(C10,E10) <NA>
#> 11 FALSE 3 NA 67.3 i 2023-12-25 <NA> PRODUCT(C11,E3) <NA>
#> 12 NA 1 NA 123 <NA> 2023-07-31 <NA> E12-C12 <NA>
dims
- read specific dimensionSometimes the entire worksheet contains to much data, in such case we
provide functions to read only a selected dimension range. Such a range
consists of either a specific cell like “A1” or a cell range in the
notion used in the openxml
standard
# read dimension without column names
wb_to_df(file, dims = "A2:C5", col_names = FALSE)
#> A B C
#> 2 NA Var1 Var2
#> 3 NA TRUE 1
#> 4 NA TRUE <NA>
#> 5 NA TRUE 2
Alternatively, if you don’t know the Excel sheet’s address, you can
use wb_dims()
to specify the dimension. See below or
in?wb_dims
for more details.
cols
- read selected columnsIf you do not want to read a specific cell, but a cell range you can use the column attribute. This attribute takes a numeric vector as argument
rows
- read selected rowsThe same goes with rows. You can select them using numeric vectors
convert
- convert input to guessed typeIn xml exists no difference between value types. All values are per
default characters. To provide these as numerics, logicals or dates,
openxlsx2
and every other software dealing with xlsx files
has to make assumptions about the cell type. This is especially tricky
due to the notion of worksheets. Unlike in a data frame, a worksheet can
have a wild mix of all types of data. Even though the conversion process
from character to date or numeric is rather solid, sometimes the user
might want to see the data without any conversion applied. This might be
useful in cases where something unexpected happened or the import
created warnings. In such a case you can look at the raw input data. If
you want to disable date detection as well, please see the entry
above.
# convert characters to numerics and date (logical too?)
wb_to_df(file, convert = FALSE)
#> Var1 Var2 NA Var3 Var4 Var5 Var6 Var7 Var8
#> 3 TRUE 1 <NA> 1 a 2023-05-29 3209324 This #DIV/0! 01:27:15
#> 4 TRUE <NA> <NA> #NUM! b 2023-05-23 <NA> 0 14:02:57
#> 5 TRUE 2 <NA> 1.34 c 2023-02-01 <NA> #VALUE! 23:01:02
#> 6 FALSE 2 <NA> <NA> #NUM! <NA> <NA> 2 17:24:53
#> 7 FALSE 3 <NA> 1.56 e <NA> <NA> <NA> <NA>
#> 8 FALSE 1 <NA> 1.7 f 2023-03-02 <NA> 2.7 08:45:58
#> 9 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
#> 10 FALSE 2 <NA> 23 h 2023-12-24 <NA> 25 <NA>
#> 11 FALSE 3 <NA> 67.3 i 2023-12-25 <NA> 3 <NA>
#> 12 <NA> 1 <NA> 123 <NA> 2023-07-31 <NA> 122 <NA>
skip_empty_rows
- remove empty rowsEven though openxlsx2
imports everything as requested,
sometimes it might be helpful to remove empty lines from the data. These
might be either left empty intentional or empty because they are were
formatted, but the cell value was removed afterwards. This was added
mostly for backward comparability, but the default has been changed to
FALSE
. The behavior has changed a bit as well. Previously
empty cells were removed prior to the conversion to R data frames, now
they are removed after the conversion and are removed only if they are
completely empty
# erase empty rows from dataset
wb_to_df(file, sheet = 1, skip_empty_rows = TRUE) |> tail()
#> Var1 Var2 NA Var3 Var4 Var5 Var6 Var7 Var8
#> 6 FALSE 2 NA <NA> #NUM! <NA> <NA> 2 17:24:53
#> 7 FALSE 3 NA 1.56 e <NA> <NA> <NA> <NA>
#> 8 FALSE 1 NA 1.7 f 2023-03-02 <NA> 2.7 08:45:58
#> 10 FALSE 2 NA 23 h 2023-12-24 <NA> 25 <NA>
#> 11 FALSE 3 NA 67.3 i 2023-12-25 <NA> 3 <NA>
#> 12 NA 1 NA 123 <NA> 2023-07-31 <NA> 122 <NA>
skip_empty_cols
- remove empty columnsThe same for columns
# erase empty cols from dataset
wb_to_df(file, skip_empty_cols = TRUE)
#> Var1 Var2 Var3 Var4 Var5 Var6 Var7 Var8
#> 3 TRUE 1 1 a 2023-05-29 3209324 This #DIV/0! 01:27:15
#> 4 TRUE NA #NUM! b 2023-05-23 <NA> 0 14:02:57
#> 5 TRUE 2 1.34 c 2023-02-01 <NA> #VALUE! 23:01:02
#> 6 FALSE 2 <NA> #NUM! <NA> <NA> 2 17:24:53
#> 7 FALSE 3 1.56 e <NA> <NA> <NA> <NA>
#> 8 FALSE 1 1.7 f 2023-03-02 <NA> 2.7 08:45:58
#> 9 NA NA <NA> <NA> <NA> <NA> <NA> <NA>
#> 10 FALSE 2 23 h 2023-12-24 <NA> 25 <NA>
#> 11 FALSE 3 67.3 i 2023-12-25 <NA> 3 <NA>
#> 12 NA 1 123 <NA> 2023-07-31 <NA> 122 <NA>
row_names
- keep rownames from inputSometimes the data source might provide rownames as well. In such a
case you can openxlsx2
to treat the first column as
rowname
types
- convert column to specific typeIf the user know better than the software what type to expect in a
worksheet, this can be provided via types. This parameter takes a named
numeric. 0
is character, 1
is numeric and
2
is date
start_row
- where to beginOften the creator of the worksheet has used a lot of creativity and
the data does not begin in the first row, instead it begins somewhere
else. To define the row where to begin reading, define it via the
start_row
parameter
# start in row 5
wb_to_df(file, start_row = 5, col_names = FALSE)
#> B C D E F G H I J
#> 5 TRUE 2 NA 1.34 c 2023-02-01 NA #VALUE! 23:01:02
#> 6 FALSE 2 NA NA #NUM! <NA> NA 2 17:24:53
#> 7 FALSE 3 NA 1.56 e <NA> NA <NA> <NA>
#> 8 FALSE 1 NA 1.70 f 2023-03-02 NA 2.7 08:45:58
#> 9 NA NA NA NA <NA> <NA> NA <NA> <NA>
#> 10 FALSE 2 NA 23.00 h 2023-12-24 NA 25 <NA>
#> 11 FALSE 3 NA 67.30 i 2023-12-25 NA 3 <NA>
#> 12 NA 1 NA 123.00 <NA> 2023-07-31 NA 122 <NA>
na.strings
- define missing valuesThere is the “#N/A” string, but often the user will be faced with
custom missing values and other values we are not interested. Such
strings can be passed as character vector via
na.strings
# na strings
wb_to_df(file, na.strings = "")
#> Var1 Var2 NA Var3 Var4 Var5 Var6 Var7 Var8
#> 3 TRUE 1 NA 1 a 2023-05-29 3209324 This #DIV/0! 01:27:15
#> 4 TRUE NA NA #NUM! b 2023-05-23 <NA> 0 14:02:57
#> 5 TRUE 2 NA 1.34 c 2023-02-01 <NA> #VALUE! 23:01:02
#> 6 FALSE 2 NA <NA> #NUM! <NA> <NA> 2 17:24:53
#> 7 FALSE 3 NA 1.56 e <NA> <NA> <NA> <NA>
#> 8 FALSE 1 NA 1.7 f 2023-03-02 <NA> 2.7 08:45:58
#> 9 NA NA NA <NA> <NA> <NA> <NA> <NA> <NA>
#> 10 FALSE 2 NA 23 h 2023-12-24 <NA> 25 <NA>
#> 11 FALSE 3 NA 67.3 i 2023-12-25 <NA> 3 <NA>
#> 12 NA 1 NA 123 <NA> 2023-07-31 <NA> 122 <NA>