[R] Problem in loading all packages all at once
Prof Brian Ripley
ripley at stats.ox.ac.uk
Mon Mar 26 15:36:59 CEST 2007
I really don't know what is going on here, beyond that 'utils' has not
been loaded.
But
1) You are not supposed to edit system files like library/base/R/Rprofile.
If you do, don't ask for help!
2) I don't see how the output shown came from the input shown: almost
surely it did not.
3) It is extremely bad practice to load packages you do not need.
This both slows R down unnecessarily and creates conflicts where packages
override default behaviour.
4) I believe there is still a limit to the number of DLLs you can have
loaded in a single R session, and I believe it to be 100. So you probably
cannot load all 751 packages, and at some point package loading will fail.
Now, loading library/base/R/Rprofile has no error-checking (as it is
supposed to be unchanged, src/main/main.c:818) and so you may get no
indication of this.
So my guess is that there was an attempt to load a few hundred packages
and only some of them got loaded, not as far down the list as 'utils'.
If you want to load some packages in every session, do so from your
.Rprofile file: that is run as user code and you should get full
error-checking (and messages).
As for Charilaos' points:
a) .First.sys is defined in the system profile, library/base/R/Rprofile.
It is run from the base environment.
b) The environment variable R_DEFAULT_PACKAGES justs sets the default for
the option 'defaultPackages', and the format is described in the surely
obvious place, the documentation for the option. There are examples in
the 'check' script.
On Mon, 26 Mar 2007, Charilaos Skiadas wrote:
> A couple of things:
>
> 1) It would have been a lot easier to help you if you had created a
> much smaller example. perhaps trying to load 2-3 packages instead of
> 1000. Now we have to wade through a lot of stuff to get to the point.
>
> 2) Perhaps your call to Sys.putenv doesn't do quite what you expect
> it to do. Example:
> > TEMP <- c("a","b")
> > Sys.getenv("TEST")
> TEST
> ""
> > Sys.putenv(TEST=TEMP)
> > Sys.getenv("TEST")
> TEST
> "b"
>
> 3) I couldn't really see where in your code all these packages are
> loaded. Reading the help for .First.sys tells me that that's the call
> that actually loads the default packages, unless I've misunderstood
> something. If that's the case, then the following line you have at
> the end would prevent this loading:
>
> .First.sys <- function()
>
>
> I must say that from the ?Startup help there are two things that are
> not very clear:
> a) At what point in the load process is R_DEFAULT_PACKAGES
> checked? .First.sys doesn't seem to know about it.
> Ok, a bit of search provided an answer to this: Reading the help for
> "options", I see there, in the section about defaultPackages, what is
> probably the most helpful bit (for me) in all this:
>
> defaultPackages:
> the packages that are attached by default when R starts up.
> Initially set from value of the environment variable
> R_DEFAULT_PACKAGES, or if that is unset to c("datasets", "utils",
> "grDevices", "graphics", "stats", "methods"). (Set R_DEFAULT_PACKAGES
> to NULL or a comma-separated list of package names.) A call to
> options should be in your ?.Rprofile? file to ensure that the change
> takes effect before the base package is initialized (see Startup).
>
>
> b) What is the format that the R_DEFAULT_PACKAGES variable must have
> (I guess now I know, since it is mentioned in ?options, though that's
> not really the place I would have expected it to be explained) ? The
> only examples I could find on the web were setting it to NULL.
>
>
>
> Hope all this helps in some way.
>
> Haris Skiadas
> Department of Mathematics and Computer Science
> Hanover College
>
>
> On Mar 26, 2007, at 6:01 AM, gyadav at ccilindia.co.in wrote:
>
>>
>> Hi All
>>
>> Please see the Rprofile file which i have modified as follows and
>> after
>> that when I start R then I see that R says to me "TRUE" for all the
>> packages implying that all loaded at once.
>> But when i try to use commands as simple as help("lm"), it doesnt
>> work nor
>> any of the menu "Packages" is not working.
>> Although the regression using lm ( Y ~ X ) is working even summary and
>> rnorm is working fine.
>> Please tell me why menu and help command is not working
>>
>> Rprofile is in between +++++++++++++++++++++++++++ lines
>> R initialization is in between @@@@@@@@@@@@@@ lines
>> Commands are in between $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ lines
>>
>> Please also suggest me should i send this to r-devel list or not
>>
>> ### This is the system Rprofile file. It is always run on startup.
>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>> ++++
>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>> ++++
>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>> ++++
>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>> ++++
>> options(scipen = 0)
>> options(max.print = 10000)# not yet exercised in 'core R'
>> options(add.smooth = TRUE)# currently only used in 'plot.lm'
>>
>> ######################################################################
>> ########################
>> ######################################################################
>> ########################
>> #############code added by gaurav yadav to install many other
>> packages at
>> startup#############
>> print ("START OF CODE ADDED BY GAURAV YADAV")
>> TEMP <- Sys.getenv("R_DEFAULT_PACKAGES")
>> print ("R_DEFAULT_PACKAGES")
>> print (TEMP)
>> print ("=======")
>> TEMP <-
>> c
>> (TEMP,"aplpack","approximator","apTreeshape","ArDec","arules","ash","a
>> space","assist","aster","asypow","aws","base","bayesm","bayesmix","bay
>> esSurv","BayesTree","BayesValidate","betareg","Bhat","BHH2","bicreduc"
>> ,"biglm","bim","bindata","Biodem","biopara","bitops","bivpois","blight
>> y","blockrand","BMA","boa","Bolstad","boolean","boost","boot","bootstr
>> ap","bqtl","BradleyTerry","brlr","BRugs","BSDA","BsMD","butler","calib
>> rate","calibrator","caMassClass","car","cat","caTools","catspec","cba"
>> ,"cclust","CDNmoney","cfa","CGIwithR","changeLOS","chplot","chron","Ci
>> rcStats","circular","clac","class","classInt","classPP","clim.pact","c
>> limatol","clines","clue","cluster","clusterRepro","clustvarsel","cmprs
>> k","cobs","CoCo","CoCoCg","CoCoCore","CoCoGraph","CoCoObjects","CoCoOl
>> dData","CoCoRaw","cocorresp","coda","coin","colorspace","combinat","co
>> mpositions","concor","concord","cond","conf.design","connectedness","c
>> opula","corpcor","corpora","covRobust","coxrobust","cramer","crossdes"
>> ,"crq","c
>> sampling","cslogistic","CTFS","ctv")
>>
> [snip]
>> #TEMP <- c(TEMP,"runfirst","base","datasets", "utils", "grDevices",
>> "graphics","stats","methods","lmtest")
>> print ("MODIFIED R_DEFAULT_PACKAGES+++++++++++")
>> print (TEMP)
>> print ("++++++++++++++++")
>> print(Sys.putenv(R_DEFAULT_PACKAGES=TEMP))
>> print ("END OF CODE ADDED BY GAURAV YADAV")
>>
>> ######################################################################
>> ########################
>> ######################################################################
>> ########################
>>
>>
>> local({dp <- as.vector(Sys.getenv("R_DEFAULT_PACKAGES"))
>> if(identical(dp, "")) # marginally faster to do methods last
>> dp <- c("datasets", "utils", "grDevices", "graphics",
>> "stats", "methods")
>> else if(identical(dp, "NULL")) dp <- character(0)
>> else dp <- strsplit(dp, ",")[[1]]
>> dp <- sub("[[:blank:]]*([[:alnum:]]+)", "\\1", dp) # strip
>> whitespace
>> options(defaultPackages = dp)
>> })
>>
>> .First.sys <- function()
>>
>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>> ++++
>>
>> THE R SHOWS ME THIS GIVEN HEREIN BELOW WHEN I START R
>> @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
>> [1] "START OF CODE ADDED BY GAURAV YADAV"
>> [1] "R_DEFAULT_PACKAGES+++++++++++"
>> R_DEFAULT_PACKAGES
>> ""
>> [1] "++++++++++++++++"
>> [1] "MODIFIED R_DEFAULT_PACKAGES+++++++++++"
>> R_DEFAULT_PACKAGES
>> "" "aplpack" "approximator"
>>
>>
> [snip tons of packages]
>> [1] "++++++++++++++++"
>> [1] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
>> TRUE TRUE
>> [751] TRUE TRUE TRUE TRUE TRUE TRUE TRUE
>> [1] "END OF CODE ADDED BY GAURAV YADAV"
>>
>> R : Copyright 2006, The R Foundation for Statistical Computing
>> Version 2.3.0 (2006-04-24)
>> ISBN 3-900051-07-0
>>
>> R is free software and comes with ABSOLUTELY NO WARRANTY.
>> You are welcome to redistribute it under certain conditions.
>> Type 'license()' or 'licence()' for distribution details.
>>
>> Natural language support but running in an English locale
>>
>> R is a collaborative project with many contributors.
>> Type 'contributors()' for more information and
>> 'citation()' on how to cite R or R packages in publications.
>>
>> Type 'demo()' for some demos, 'help()' for on-line help, or
>> 'help.start()' for an HTML browser interface to help.
>> Type 'q()' to quit R.
>>
>> [Previously saved workspace restored]
>>
>>>
>>
>> @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
>>
>> $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
>>
>> Type 'demo()' for some demos, 'help()' for on-line help, or
>> 'help.start()' for an HTML browser interface to help.
>> Type 'q()' to quit R.
>>
>> [Previously saved workspace restored]
>>
>>> help("lm")
>> Error: could not find function "help"
>>> help.search("regression")
>> Error: could not find function "help.search"
>>> local({pkg <- select.list(sort(.packages(all.available = TRUE)))
>> + if(nchar(pkg)) library(pkg, character.only=TRUE)})
>> Error in eval(expr, envir, enclos) : could not find function
>> "select.list"
>>> chooseCRANmirror()
>> Error: could not find function "chooseCRANmirror"
>>> setRepositories()
>> Error: could not find function "setRepositories"
>>> utils:::menuInstallPkgs()
>> --- Please select a CRAN mirror for use in this session ---
>> Error in open.connection(file, "r") : unable to open connection
>> In addition: Warning message:
>> unable to connect to 'cran.r-project.org' on port 80.
>> Error in contrib.url(repos, type) : trying to use CRAN without
>> setting a
>> mirror
>>> update.packages(ask='graphics')
>> Error: could not find function "update.packages"
>>> utils:::menuInstallLocal()
>> Error in install.packages(choose.files("", filters = Filters[c
>> ("zip", :
>> no packages were specified
>>> X<-rnorm(1000)
>>> Y<-rnorm(1000)
>>> result<-lm(Y ~ X)
>>> summary(result)
>>
>> Call:
>> lm(formula = Y ~ X)
>>
>> Residuals:
>> Min 1Q Median 3Q Max
>> -3.0047 -0.7205 0.0203 0.7232 3.1052
>>
>> Coefficients:
>> Estimate Std. Error t value Pr(>|t|)
>> (Intercept) 0.0347667 0.0322268 1.079 0.281
>> X -0.0009727 0.0320626 -0.030 0.976
>>
>> Residual standard error: 1.018 on 998 degrees of freedom
>> Multiple R-Squared: 9.223e-07, Adjusted R-squared: -0.001001
>> F-statistic: 0.0009204 on 1 and 998 DF, p-value: 0.9758
>>
>>>
>>
>> $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
--
Brian D. Ripley, ripley at stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595
More information about the R-help
mailing list