differences
Differences Between TIBCO Enterprise Runtime for R and OpenSource R
Description
This file contains a compilation of the differences between TIBCO
Enterprise Runtime for R and OpenSource R.
General
TIBCO Enterprise Runtime for R issues a warning if a function is exported
more than once from a package NAMESPACE file. R provides no
such warning. (Duplicate exports in packages occur in a number of places
in the CRAN repository.)
abbreviate
In opensource R, the "use.classes" argument is silently ignored and treated as TRUE.
In some cases, this causes TIBCO Enterprise Runtime for R and opensource R to produce different results:
Example
abbreviate(c('a foobr','a foobar'))
# R: "a foobr" "afoobar"
# TERR: "afoobr" "afoobar"
Bessel
colMeans
 TIBCO Enterprise Runtime for R accepts vectors, but opensource R does not.
 In TIBCO Enterprise Runtime for R, rowMeans/colMeans return NaN if there are no values; opensource R returns NA:
rowMeans(matrix(NA), na.rm=T)
 colVars, colStdevs, rowVars, and rowStdevs are not supported by opensource R, but they are in TIBCO Enterprise Runtime for R.
 colSums, rowSums, colMeans and rowMeans are implemented both in opensource R and TIBCO Enterprise Runtime for R,
but there are more arguments in the TIBCO Enterprise Runtime for R implementation (for example, weights, freq and n. All of these might not be presented).
 colSums, rowSums, colMeans and rowMeans are NOT generic functions in opensource R.
connections
In opensource R, when the engine is run in
noninteractive mode with the script specified via f,
readLines(stdin()) reads lines from the script file.
In TIBCO Enterprise Runtime for R, in that case, readLines(stdin()) returns character(0).
data.matrix
The function numerical.matrix is not implemented in opensource R.
deparse
 TIBCO Enterprise Runtime for R has the extra control parameter noSplitLines.
 In TIBCO Enterprise Runtime for R, deparse currently supports only the control parameters 'keepInteger', 'keepNA', 'showAttributes', 'useSource', and (the TIBCO Enterprise Runtime for Rspecific) 'noSplitLines'.
difftime
Differences exist between TIBCO Enterprise Runtime for R and opensource R in the results for calculating difftime near DST;
the TIBCO Enterprise Runtime for R results appear more reasonable than those of opensource R.
Example
## The start of DST in USA 2013 is 20130310 2:00AM
> difftime("20130310 03:00:00", "20130310 02:00:00", tz = "America/New_York")
# Time difference of 2 hours (opensource R)
# Time difference of 0 secs (TERR)
> difftime("20130310 03:00:00", "20130310 02:59:00", tz = "America/New_York")
# Time difference of 2 hours (opensource R)
# Time difference of 59 mins (TERR)
The differences are caused by the differences in as.POSIXct, i.e.,
> as.POSIXct("20130310 02:00:00", tz = "America/New_York")
# [1] "20130310 EST" (opensource R)
# [1] "20130310 03:00:00 EDT" (TERR)
> as.POSIXct("20110313 02:59:59", tz = "America/New_York")
# [1] "20130310 EST" (opensource R)
# [1] "20130310 03:59:59 EDT" (TERR)
.deparseOpts
The "noSplitLines" option is available only in TIBCO Enterprise Runtime for R.
dput
See deparse for similar differences.
Encoding
General warning: TIBCO Enterprise Runtime for R might not
choose the same string encoding as opensource R in all cases. Indeed,
opensource R does not use the same string encoding consistently between
different platforms and locales. TIBCO Enterprise Runtime for R was
designed to be more consistent between different platforms.
 TIBCO Enterprise Runtime for R always uses the UTF8 character set for the "unknown"
encoding, whereas opensource R uses the native string encoding for
the particular OS where it is running, perhaps because it relies on
native OS routines that depend on that encoding.
 In opensource R, the Encoding< will not change a string's encoding in
some cases. For example, if a string consists totally of simple ASCII
characters, it seems to have its encoding stuck at "unknown". In
TIBCO Enterprise Runtime for R, Encoding< will always change the encoding.
 In opensource R, comparing a string with 'bytes' encoding to a string with
'latin1' or 'UTF8' encoding gives an error. Comparing xb (a
'bytes' string) with xu (a 'unknown' string) does not give an
error, but it acts weird: in some cases, xb==xu, xb<xu,
and xb>xu all return FALSE. TIBCO Enterprise Runtime for R is more consistant: any
'bytes' string is not equal and sorts as less than any non'bytes'
string.
 TIBCO Enterprise Runtime for R supports automatically converting string encodings on input and
output to the console. See getConsoleEncoding. Opensource R does not support this.
 TIBCO Enterprise Runtime for R provides the function getNativeEncoding for
reading the name of the "native" encoding. Opensource R does not support this.
 It is possible to construct strings with the "UTF8" encoding that are
not valid UTF8 byte sequences. In opensource R, attempting to manipulate these
strings can give errors such as "invalid multibyte string". TIBCO Enterprise Runtime for R
internally converts any string to a valid UTF8 string when doing
string manipulation, so this error should not occur in TIBCO Enterprise Runtime for R. Note
that TIBCO Enterprise Runtime for R does not actually change the bytes in an invalid UTF8
string (as viewed with charToRaw), but rather converts
the invalid UTF8 bytes to a valid UTF8 sequence when needed (see
getValidUtf8).
file.info
In TIBCO Enterprise Runtime for R, 'uid', 'gid', 'uname' and 'grname' fields are never included.
The 'exe' field is provided on all platforms, but says only "no" or "yes". In opensource R on Windows,
it says "no", "win64", or "win32".
format
Calling format.data.frame with zero columns returns a zerocolumn data.frame
with the same row names (hence number of rows) as the input data.frame. In opensource R,
this call returns an error.
getNativeSymbolInfo
In opensource R, getNativeSymbolInfo returns the address of the routine's entry point or its registration record
(if withRegistrationInfo is TRUE) in the address field. In TIBCO Enterprise Runtime for R, the field always
contains NA. The name of the third field of the return value is a package in opensource R, but it is a DLL in TIBCO Enterprise Runtime for R,
where it is compatible with information returned by getDLLRegisteredRoutines (where opensource R names the field DLL).
grep
grep calls the regexpr function in TIBCO Enterprise Runtime for R implementation.
However, both grep and grepl do not directly call the regexpr function in
opensource R implementation. Internal C functions other than regexpr are called respectively.
iconv
 In opensource R, iconv does not accept the string "native.enc" to
specify the native encoding like other functions.
 Like TIBCO Enterprise Runtime for R, opensource R ignores upper/lower case differences in encoding names,
but opensource R does not ignore characters other than letters and digits. Thus,
opensource R accepts "latin1" but not "latin1".
 In opensource R, the to argument must exactly match the strings
"latin1" or "UTF8" for these string encodings
to be marked when mark=TRUE. In TIBCO Enterprise Runtime for R, we also detect aliases
of these encodings.
 In opensource R, if an unconvertable character is found, and sub is not an
NA, the string sub is substituted once for every byte
in the original string character. TIBCO Enterprise Runtime for R substitutes it only once
for the entire unconvertable character.
 In TIBCO Enterprise Runtime for R, the sub string can have only a limited number of
characters (10), or an error is generated.
library.dynam
In opensource R, .dynLibs is used to set the internal list of "DLLInfo" objects,
which is an object with class "DLLInfoList", if new is specified.
Otherwise, it just returns this internal list if new is missing.
In TIBCO Enterprise Runtime for R, .dynLibs is equivalent to getLoadedDLLs, except the DLLInfoList
structure it returns carries no names. The new parameter is ingored.
list.files
Both TIBCO Enterprise Runtime for R and opensource R sort files depending on locale. In the "C" locale
(e.g., Sys.setlocale(locale='C')), characters are
sorted by their ascii codes. As a result, in TIBCO Enterprise Runtime for R, the list.files function lists
all files starting with upper case before listing files starting with lower case.
In opensource R, list.files lists files ignoring the case.
Logic
In opensource R, the negation (!) operator is not a group member of Logic.
Math.Date
The trunc.POSIXt(x, units = "secs") in opensource R seems to have problems
with dates after 20380118 19:14:06
For example:
R> pct < as.POSIXct(c("20380118 19:14:06.5", "20380118 19:14:07.5"))
R> trunc(pct, "secs")
[1] "20380118 19:14:06 PST" "20380118 19:14:08 PST"
TERR> pct < as.POSIXct(c("20380118 19:14:06.5", "20380118 19:14:07.5"))
TERR> trunc(pct, "secs")
{[1] "20380118 19:14:06 PST" "20380118 19:14:07 PST"}
nchar
In opensource R, nchar generates an error (or returns NA if
allowNA is TRUE) if the string contains an invalid
multibyte character sequence and the type argument is "chars" or "width".
norm
Opensource R does not handle complex matrices.
numeric_version
The TIBCO Enterprise Runtime for R functions S_system_version() and getSversion() are nonexistent in opensource R,
but they are similar to opensource R functions R_system_version() and getRversion().
path.expand
In opensource R, path.expand() appears to save the value used for "~" the first time it is called.
Even if the environment variables R_USER, and so on, are changed, further calls to path.expand do not use these environment variables
to determine the value for "~".
In contrast, TIBCO Enterprise Runtime for R reads the current values for R_USER, and so on, whenever path.expand is called.
qr
 In opensource R, LAPACK as.qr always returns an error.
 In opensource R, when x is a matrix of complex data, the QR factorization is always computed
with the LAPACK function ZGEQP3, regardless of the argument LAPACK.
When x is a matrix of noncomplex data, and LAPACK is FALSE,
the QR factorization is computed with a particular LINPACK subroutine
DQRDC2 (modified from LINPACK code DQRDC).
When LAPACK is TRUE, the returned qr contains a logical
attribute useLAPACK, set to TRUE.
 The implementation in TIBCO Enterprise Runtime for R has the linear model version of the QR for
complex matrices. Thus using LAPACK = TRUE assigns useLAPCK for
complex matrices, and qrcomplexmatrix may differ from qrcomplex, matrix, LAPACK = TRUE.
 Opensource R ignores the tolerance argument tol when LAPACK = TRUE and rank is
always equal to min(dim(X)) in this case. In TIBCO Enterprise Runtime for R, we have added a rank estimate
based on the tolerance and LAPACK routine dlaic1 (zlaic1).
qr.coef
 In TIBCO Enterprise Runtime for R, the argument k for qr.fitted is not supported and returns an error if used.
 In opensource R, qr.solve returns a vector of zeros when y contains NAs. In TIBCO Enterprise Runtime for R, an
error is returned.
 In opensource R, qr.coef computes the coefficients with the LAPACK
function qr_coef_cmplx for complex data, regardless if qr's
attribute useLAPACK is TRUE or FALSE. For noncomplex data, it computes
with the LAPACK function qr_coef_real when useLAPACK is TRUE
and the LINPACK function DQRCF when useLAPACK is FALSE.
 In opensource R, qr.fitted does not support LAPACK QR. It calls the LINPACK function DQRXB.
 In opensource R, qr.resid does not support LAPACK QR. It calls the LINPACK function DQRRSD.
 In opensource R, qr.qy computes the result with LAPACK
function qr_qy_cmplx for complex data, regardless if qr's
attribute useLAPACK is TRUE or FALSE. For noncomplex data, it computes
with the LAPACK function qr_qy_real when useLAPACK is TRUE
and the LINPACK function DQRQY when useLAPACK is FALSE.
 In opensource R, qr.qty computes the result with the LAPACK
function qr_qy_cmplx for complex data regardless if qr's
attribute useLAPACK is TRUE or FALSE. For noncomplex data, it computes
with the LAPACK function qr_qy_real when useLAPACK is TRUE
and the LINPACK function DQRQTY when useLAPACK is FALSE.
 In opensource R, qr.solve returns the result of qr.coef(a, b), if a is not an object of class qr;
the QR decomposition is computed first, then qr.coef is called.
readBin
 In opensource R, the useBytes argument to writeBin specifies
whether to write encoded strings in a form that can be printed. For
example, if R cannot display Unicode characters, writeBin("\u30A4",f,useBytes=FALSE) would write the bytes for
the string "a<U+30A4>;b", whereas
writeBin("\u30A4",f,useBytes=TRUE) would write the bytes
contained in the string as produced by charToRaw.
 TIBCO Enterprise Runtime for R ignores the useBytes argument, and always writes the
string bytes as produced by charToRaw.
reg.finalizer
The argument e can be any type of object in TIBCO Enterprise Runtime for R.
regexpr
NOTE Over time, we may modify TIBCO Enterprise Runtime for R so it is closer to opensource R's behavior, and remove items from this list.
In many cases, the TIBCO Enterprise Runtime for R matcher handles POSIXstandard forms that opensource R does not handle.
We might want to detect these cases and generate an error.
 In opensource R, the algorithms for POSIX and Perlstyle regular expression are different from TIBCO Enterprise Runtime for R.
When perl=FALSE, the TRE library(http://laurikari.net/tre/) is used for POSIX regular expression matching.
When perl = TRUE, the PCRE library(http://www.pcre.org) is used for Perl compatible regular expression matching.
 In opensource R, '.' does not match '\n' when perl=TRUE, but it does in TIBCO Enterprise Runtime for R.
Thus, regexpr('b.', 'ab\nc', perl=TRUE)
does not match in opensource R, but does in TIBCO Enterprise Runtime for R.
 In TIBCO Enterprise Runtime for R, the '[:print:]' character class matches several characters such as '\n' that opensource R does not.
Thus, regexpr('[[:print:]]', '\b\n\b')
does not match in opensource R, but does in TIBCO Enterprise Runtime for R.
 opensource R does not recognize the character classes
'[:d:]', '[:l:]', '[:s:]', '[:unicode:]', '[:u:]', '[:word:]' or '[:w:]',
and will generate an error if they are given in a pattern.
Thus, regexpr('[[:u::]]', 'aBc') matches in TIBCO Enterprise Runtime for R, but gives an error in opensource R.
 opensource R accepts a repeat qualifier of the form '{,max}', where the minimum value is taken to be zero.
TIBCO Enterprise Runtime for R generates an error if this form is used.
Actually, opensource R's implementation of this is a little strange.
If perl=TRUE, the form is accepted, but the expression never seems to match.
If perl=FALSE, the form is accepted, but is interpreted incorrectly:
'ab{,2}' acts like 'ab{0,3}', rather than 'ab{0,2}'.
Thus regexpr('ab{,2}', 'abc', perl=TRUE) does not match in opensource R,
and regexpr('ab{,2}', 'abbbc', perl=FALSE) matches the first four characters in the string.
Either one gives an error in TIBCO Enterprise Runtime for R.
 In opensource R with perl=TRUE,
the character class names ('[:alnum:]', '[:alpha:]', etc.)
do not match Unicode characters beyond the latin1 character set.
Thus, in opensource R regexpr('[[:alpha:]]', '1\u30A42', perl=TRUE) does not match,
but regexpr('[[:alpha:]]', '1\u30A42', perl=FALSE) does.
In TIBCO Enterprise Runtime for R, both of these will match.
 In general, there are many small differences between opensource R and TIBCO Enterprise Runtime for R
in exactly how the character class names ('[:alnum:]', '[:alpha:]', etc.)
classify particular Unicode characters beyond the latin1 character set.
 There are also differences between opensource R on Linux and Windows on
how the character class names ('[:alnum:]', '[:alpha:]', etc.)
are interpreted.
When comparing TIBCO Enterprise Runtime for R to opensource R, we usually compare it to opensource R running on Linux.
 opensource R with pel=FALSE accepts and ignores the construct '(*)', whereas TIBCO Enterprise Runtime for R generates an error.
Thus, regexpr('a(*)', 'abc', perl=FALSE) matches in opensource R, and produces an error in TIBCO Enterprise Runtime for R.
regexpr('a(*)', 'abc', perl=TRUE) produces an error in both opensource R and TIBCO Enterprise Runtime for R.
 opensource R does not recognize collating elements.
Therefore, regexpr('[[.^.]]', 'xxx^yyy') matches in TIBCO Enterprise Runtime for R, and generates an error in opensource R.
 opensource R does not recognize equivalence classes.
Therefore, regexpr('[[=a=]]', 'xxx\u00C4yyy') matches in TIBCO Enterprise Runtime for R, and generates an error in opensource R.
 opensource R does not recognize the following escape sequences:
'\v', '\cX', '\0ddd', '\Nname'.
 opensource R does not recognize the following abbreviations for character classes:
'\l', '\u', '\L', '\U'.
Opensource R also does not recognize the forms '\pX', '\pXXX', '\PX', '\PXXX'.
 opensource R does not accept the patterns '[[:<:]]' and '[[:>:]]' to match the start and end of a word.
Thus, regexpr('[[:<:]]a', 'bab aba') gives an error in opensource R, but matches in TIBCO Enterprise Runtime for R.
 opensource R does not accept '\<' and '\>' to match the start and end of a word when perl=TRUE.
Thus, in opensource R regexpr('\<a', 'bab aba', perl=FALSE) matches, but
regexpr('\<a', 'bab aba', perl=TRUE) does not match.
Both of these produce a match in TIBCO Enterprise Runtime for R.
strsplit opensource R does not accept '\`' or '\`' to match the beginning or end of the buffer.
opensource R only accepts '\A' or '\z' when perl=TRUE.
opensource R only accepts '\Z' when perl=TRUE, but interprets it the same as '\z'.
row.names
Unlike TIBCO Enterprise Runtime for R, opensource R has no function rowIds.
save
 The RFormat argument is TIBCO Enterprise Runtime for Rspecific.
 The arguments eval.promises and precheck are not supported in TIBCO Enterprise Runtime for R.
 The version argument supports only version 2 in TIBCO Enterprise Runtime for R.
seq.Date

If from is the last day of the month and by is either "months" or "years",
in TIBCO Enterprise Runtime for R the elements of the resulting sequence always adjust to the last day of month
and no month is skipped or duplicated.
Example
R> seq(as.Date("20000131"), by = "months", length = 12)
[1] "20000131" "20000302" "20000331" "20000501"
[5] "20000531" "20000701" "20000731" "20000831"
[9] "20001001" "20001031" "20001201" "20001231"
TERR> seq(as.Date("20000131"), by = "months", length = 12)
[1] "20000131" "20000229" "20000331" "20000430"
[5] "20000531" "20000630" "20000731" "20000831"
[9] "20000930" "20001031" "20001130" "20001231"
 Unlike opensource R, where exactly two of to, by, and length.out/along.with must be specified,
in TIBCO Enterprise Runtime for R, if only length.out / along.with is specified, by = 1 is assumed.
sequence
sequence(NULL) and sequence(integer(0)) both return integer(0) in TIBCO Enterprise Runtime for R, but return NULL in opensource R.
serialize
The RFormat argument is TIBCO Enterprise Runtime for Rspecific.
sink
In opensource R, the file argument must be an open connection
when type="message". In TIBCO Enterprise Runtime for R, it can also be a file name.
solve
 The LINPACK argument, which is available in opensource R, is ignored in TIBCO Enterprise Runtime for R.
 If b contains missing values (NAs), opensource R solve(qr(a), b) returns a vector of zeros; in TIBCO Enterprise Runtime for R an error is returned.
sort
 Opensource R handles only index.return=TRUE when na.last=NA. TIBCO Enterprise Runtime for R is more flexible:
sort(mydata, na.last = TRUE, index.return = TRUE) # Error: "index.return'"only for "na.last = NA"
 The sort algorithms are not the same. The result however, should be the same unless partial is used (in which case
the sort order is undefined for the ranges in between the specified partial indices).
Sys.time
Unlike opensource R, the return value of TIBCO Enterprise Runtime for R's Sys.timezone is platformindependent
and matches the value returned in the Windows version of opensource R.
traceback
If the max.lines argument is NULL or any other
nonnumeric value, opensource R prints all of the lines,
whereas TIBCO Enterprise Runtime for R prints only a maximum of 10 lines.
The print.context argument is in TIBCO Enterprise Runtime for R only; it is not in opensource R.
writeChar
 In R, if readChar is called with useBytes=FALSE,
the characters are interpreted according to the "current encoding".
In TIBCO Enterprise Runtime for R, readChar with useBytes=FALSE always
inteprets input characters using the UTF8 encoding.
 In R, readChar with useBytes=FALSE
reads all UTF8 byte sequences with two to five bytes.
In TIBCO Enterprise Runtime for R, reading a threebyte UTF8 sequence encoding a character greater than 0x10FFFF
or trying to read a four or fivebyte UTF8 sequence generates an error,
because these are not valid Unicode characters.
 In R, if writeChar is called with useBytes=FALSE,
it writes characters according to the "current encoding".
In TIBCO Enterprise Runtime for R, writeChar with useBytes=FALSE always
writes characters using the UTF8 encoding
(except for strings with the "bytes" encoding).
chull
The arguments: peel, maxpeel, onbdy and tol
are not available in Opensource R's chull function.
clusterApply
clusterSplit gives incorrect results in opensource R if
cl has one or fewer nodes, or seq has zero elements.
TIBCO Enterprise Runtime for R handles these cases better, returning a list with length equal
to the number of nodes in cl.
mclapply
TIBCO Enterprise Runtime for R defines mclapply for compatibility with opensource R,
but this function doesn't process list elements in separate parallel
processes.
mcparallel
TIBCO Enterprise Runtime for R defines mcparallel and mccollect for compatibility
with opensource R, but these functions don't evaluate expressions in
separate processes.
TIBCO Enterprise Runtime for R defines mc.reset.stream for compatibility with
opensource R, but this function doesn't reset or affect the random
number generators.
splitIndices
splitIndices gives incorrect results in opensource R if
nx or ncl is less than or equal to one. TIBCO Enterprise Runtime for R handles
these cases better, returning a list with ncl elements.
backSpline
In opensource R, backSpline uses a different algorithm, (usually) resulting in a slightly different splines.
bs
When the number of nonmissing values in x is less than 2, TIBCO Enterprise Runtime for R and Opensource R
currently give different results, neither of which is very useful.
ns
When the number of nonmissing values in x is less than 2, TIBCO Enterprise Runtime for R and Opensource R
currently give different results, neither of which is very useful.
Box.test
TIBCO Enterprise Runtime for R uses the truncated to integer value of lag,
as does Opensource R
but TIBCO Enterprise Runtime for R also uses the truncated integer value in computing the
degrees of freedom while R does uses the supplied lag value in
computing the degrees of freedom. This is only an issue if the value
specified for lag is not an integer.
TIBCO Enterprise Runtime for R will issue a warning if the combination of lag and fitdf
result in a negative value for the degrees of freedom,
Opensource R does not warn.
In this situation, both TIBCO Enterprise Runtime for R and Opensource R will report NA for
the test pvalue.
cor.test
When method="kendall", exact=TRUE, and length(x)>8,
the exact pvalues computed in TIBCO Enterprise Runtime for R differ from opensource R
by a small amount (usually less than 1e3).
This may be due to a different algorithm (that is, AS 71) is used in TIBCO Enterprise Runtime for R.
cutree
The group numbering might be different between TIBCO Enterprise Runtime for R and opensource R, but the
group memberships are the same.
In opensource R, when k is NULL, and if
the "height" component of tree is not sorted,
cutree is stopped. TIBCO Enterprise Runtime for R sorts the "height" component to
ensure "height" is ordered as increasing.
dist
In opensource R, the definition of the canberra method is to divide by
the absolute value of the sum of the variable values. This matters only when values are negative.
effects
In opensource R, when effects component of object is NULL,
effects is stopped with an error. In TIBCO Enterprise Runtime for R, it returns NULL.
factanal
The TIBCO Enterprise Runtime for R factanal uses a different algorithm for computing
the MLE estimates so the results can be slightly different.
The return value from TIBCO Enterprise Runtime for R does not include the rotmat
component when a rotation is performed.
The current TIBCO Enterprise Runtime for R version of factanal produces the same
results as factanal in SPLUS.
filter
NA handling for recursive filtering differs between TIBCO Enterprise Runtime for R and opensource R.
In TIBCO Enterprise Runtime for R, all values after an NA is encountered in x become NA.
hclust
Opensource R gives an error when the input d is a symmetric matrix,
while TIBCO Enterprise Runtime for R allows it.
The methods: "mcquitty", "ward", "median"
and "centroid"
which are available in R are not yet implemented in TIBCO Enterprise Runtime for R.
If one of the unimplemented methods is specified,
a warning is issued and the default method ("complete") is used.
Hypergeometric
 In opensource R, dhyper accepts only integer values. phyper, qhyper, and rhyper allow noninteger values for m, n, and k, but the values are rounded.
 In TIBCO Enterprise Runtime for R, for consistency the values for m, n, and k must be integers.
kmeans
The methods "Lloyd", "Forgy" and "MacQueen",
which are available in R, are not yet implemented in TIBCO Enterprise Runtime for R.
If one of the unimplemented methods is specified,
a warning is issued and the default method ("HartiganWong") is used.
ks.test
 When the alternative ="two.sided" and exact=FALSE, and for both onesample and twosample tests,
the pvalues computed in TIBCO Enterprise Runtime for R differ from opensource R by a small amount (usually less than 1e3).
This may be due to the algorithm used is not exactly the same in TIBCO Enterprise Runtime for R and opensource R.
 The argument y, if set to an a character string, specifies
the function that generates pvalues for the hypothesized distribution.
The argument distribution in TIBCO Enterprise Runtime for R is similar to that found in SPLUS
as an alternative way to specify the hypothesized distribution.
lag
In opensource R, the return value keeps the attribute of input object, whereas in TIBCO Enterprise Runtime for R, the return value is a "ts" object.
model.tables
The argument "type" for "feffects" and "residuals" are not implemented in opensource R.
optim
TIBCO Enterprise Runtime for R does not support the "Brent" optimization method.
optimize
 the argument "keep.xy" is not supported in opensource R.
 Only two components of list are returned in opensource R implementation:
"objective", "minimum" or "maximum".
ppoints
The ppoints function in opensource R does not have the weights argument.
princomp
 In TIBCO Enterprise Runtime for R, the returned component terms is
like that from SPLUS, which is not supported in opensource R.
 The returned value is different when the argument scores is numeric
or TRUE in TIBCO Enterprise Runtime for R, but they are same in opensource R.
shapiro.test
When you run the shapiro.test on a dataset
with missing values (NA),
TIBCO Enterprise Runtime for R reports a warning about NAs that is not shown in R.
(The test output is the same.)
In R, shapiro.test fails if the data includes Infs.
TIBCO Enterprise Runtime for R removes the nonfinite value(s), issues a warning about them,
and then runs the test on the remaining values.
smooth.spline
This function is based on the SPLUS implementation and has not been
completely converted to the R parameterization. The spar
parameters do not match: They are on the same scale, but differ
by a constant. Several arguments used in the R implementation are
included here but not used. This implementation might choose a slightly
rougher spline fit than the implementation in opensource R.
t.test
The argument "treatment" is not in opensource R.
terms
abb, neg.out and as.terms are not implemented in
opensource R. The object attribute  formula is removed in opensource R.
Uniform
 In opensource R, min = max is allowed for {p,q,r}unif but dunif(2, 2, 2) returns NaN and generates a warning.
 In TIBCO Enterprise Runtime for R, min = max is allowed for all {d,p,q,r}unif and dunif(2, 2, 2) returns Inf.
colMedians
colMedians, colQuantiles, and colProds are not in opensource R.
colMins
colMins, colMaxs, colRanges, rowMins, rowMaxs, and rowRanges are not in opensource R.
colVars
 TIBCO Enterprise Runtime for R accepts vectors but opensource R does not.
rowMeans and colMeans both return NaN if there are no values; opensource R returns NA:
Example
rowMeans(matrix(NA), na.rm=T)

All functions listed in this alias are generic functions;
methods currently exist for data.frame, resamp, and series.
 colVars, colStdevs, rowVars, rowStdevs are not supported by opensource R but they are
supported in TIBCO Enterprise Runtime for R.
 colSums, rowSums, colMeans and rowMeans are implemented both in opensource R and TIBCO Enterprise Runtime for R, but there are more arguments in TIBCO Enterprise Runtime for R implementation:
weights, freq and n. All of these may not be present.
 colSums, rowSums, colMeans and rowMeans are NOT generic functions in opensource R.
is.number
is.number is not implemented in opensource R.
ndotargs
This function does not exist in opensource R.
stdev
stdev is not in opensource R.
startDynamicHelp
 R does not maintain the server URL in tools:::httpdUrl.
 R uses R_DISABLE_HTTPD instead of TERR_DISABLE_RDHTTPD as the
name of the environment variable to disable starting the HTTP help server.
apropos
In opensource R, what must be a character vector.
browseURL
In opensource R, when browser is not NULL, the
function "browseURL" fails to run.
count.fields
Opensource R does not have the nmax argument.
find
In opensource R, what must be a character vector.
help.start
The update argument is not used by TIBCO Enterprise Runtime for R.
In opensource R, this argument is used to update the file
R_HOME/doc/html/packages.html. In TIBCO Enterprise Runtime for R, this information is maintained
internally to Rdhttpd and is generated dynamically on request.
ls.str
In opensource R, the name argument seems to be ignored.
packageDescription
In TIBCO Enterprise Runtime for R, the encoding argument is not implemented.
If this argument is given, a warning is generated.
remove.packages
 In R, the argument lib is given without the default NULL value.
 In R, this function does not display a
warning for the packages in use. No warning is given if the specified package is
not found in the specified library.
 In R, after the specified package(s) are removed, the HTML index of
packages is updated.
str
 The output from str() is roughly, but not exactly, the same.
 The arguments 'vec.len', 'nchar.max', 'envir', 'strict.width' and 'formatNum' are not yet supported (so they are ignored) in TIBCO Enterprise Runtime for R.
 The methods str.dendrogram and str.logLik are not implemented in TIBCO Enterprise Runtime for R.
R Version
R3.2.1