exportDataToSBDF
Export Data to an SBDF File
Description
Exports data to Spotfire's proprietary binary data format (SBDF).
Usage
exportDataToSBDF(data, file, rowNames = FALSE)
Arguments
| data |
a data.frame or list object to export.
|
| file |
a character string specifying the name of the Spotfire Binary Data File
(SBDF) to create.
|
| rowNames |
a logical flag.
If rowNames=TRUE, the row names are exported to the data file.
|
Details
The following table shows the supported data types
and how they are converted when you export from an R engine into SBDF.
|
| R data type | | SBDF data type |
|
| logical | | Boolean |
|
| integer | | Integer (32 bit) |
|
| numeric | | Real |
|
| character | | String |
|
| factor | | String |
|
| POSIXct or POSIXlt | | DateTime |
|
| difftime | | TimeSpan |
|
| raw | | Binary |
|
|
Most of the supported R data types are stored as data.frame
columns containing atomic vectors. One exception is the raw data
type, which is stored in a data.frame column as a list of raw
objects and NULL values, such as
data.frame(x=I(list(as.raw(1:3),as.raw(4:90),NULL))). The raw
objects in such a list can have different lengths. Any elements of such
a list that are not raw objects are exported as NULL.
Note: TIBCO Enterprise Runtime for R does not support the Spotfire data type Currency
natively. Any Currency value contained in an SBDF file is imported to
TIBCO Enterprise Runtime for R as numeric. If this numeric is then exported to an SDBF, it is
converted to the Spotfire data type Real. This can result in a loss of
precision for very large values.
SBDF does not support factors (categorical data types), so these are
exported as simple strings.
It is possible to specify which SBDF data type to use for exporting each
column (including types "Date" and "Time"), by giving
data as a
data.frame with an attribute "SBDFExportTypes". This attribute
should be a named string vector associating column names with the
desired exported SBDF data type for the column. This can be most easily
set using the function
setSBDFExportTypes. If a column
name is not associated with an SBDF data type in "SBDFExportTypes", it
is exported using the default types listed above.
The SBDF file format can represent a set of "data table properties"
associated with the whole table, as well as a set of "column properties"
values associated with each table column. Data table properties are
represented in the data argument by adding a
"SpotfireTableMetaData" attribute to the whole object. Column
properties for an individual column are represented by adding a
"SpotfireColumnMetaData" attribute to the column element of
data. The value of a "SpotfireTableMetaData" or
"SpotfireColumnMetaData" attribute should be a list of named
elements, where each element can be one of the supported R data types
that are listed above. Each list element must have length 1, except for
raw data vectors, which can have any length.
Value
returns NULL.
See Also
Examples
df <- data.frame(x=1:5, y=sin(1:5), z=LETTERS[1:5],
aa=I(lapply(1:5, function(i) as.raw(1:i))),
stringsAsFactors=FALSE)
## add a data table property
attr(df, "SpotfireTableMetaData") <- list(TMDInt=123L, TMDString="abc")
## add column properties to the 'x' column
attr(df[["x"]], "SpotfireColumnMetaData") <- list(CMDInt=345L, CMDString="def")
## Export the data to a temporary file:
tfile <- tempfile(fileext = ".sbdf")
exportDataToSBDF(df, tfile)
## Read the file in with importDataFromSBDF:
df2 <- importDataFromSBDF(tfile)
all.equal(df, df2) ## two data objects are the same
## Clean up:
unlink(tfile)