similar to: weird warning in [.data.frame with 2.10

I've just installed R2.10.0 v50082, and found the problem described below the first few times I launched it. Did not occur with v49613 (the previous most recent version I have). The problem has now gone away, and is therefore non-reproducible, but there seems to be something up with parameter processing or environment variable checking or loading of .RData files immediately after installation.

I'm encountering problems when making lazy-loadable databases of the output from 'parse_Rd'. The lazy-load database is of seemingly limitless size when I try to reload it... Admittedly I am using functions that I'm not really supposed to use, which is why this isn't a bug report, but there does seem to be something strange going on; my code is very similar to code that lives

Under R2.10.0 beta (2009-10-14 r50082) my .First does not run reliably. I reported this last week, but not reproducibly, and it seemed to go away, so I thought it might be installation-related. But it has now recurred reproducibly, at least on my machine. The website ftp://ftp.csiro.au/MarkBravington contains two .RData files each containing (only) a .First. One of them does what it should, and

Looks like there is a buglet in 'installed.packages', around line 17: for (lib in lib.loc) { dest <- file.path(tempdir(), paste("libloc_", URLencode(lib,=20 TRUE), paste(fields, collapse =3D ","), ".rds", sep =3D "")) if (!noCache && file.exists(dest) && file.info(dest)$mtime >=20

In R1.7 and above (including R 1.9 alpha), 'update.formula' forgets to copy any offset(...) term in the original '.' formula: test> df <- data.frame( x=1:4, y=sqrt( 1:4), z=c(2:4,1)) test> fit1 <- glm( y~offset(x)+z, data=df) test> fit1$call glm(formula = y ~ offset(x) + z, data = df) test> fit1u <- update( fit1, ~.) test> fit1u$call glm(formula = y ~ z,

You might want to have a look at 'source.mvb' & friends in the 'mvbutils' package. It's designed to allow control of nested sourcing, and to allow interspersed data and commands in a single self-contained file. Unlike 'source', 'source.mvb' reads each statement and immediately executes it, before proceeding to the next; hence it has do the parsing to figure

Problem below with PCRE grep in R-devel; works fine in R-patched. (Unless there's been an absolutely massive change in rules for updated PCRE version 8.13; jeez I hope not) > grep( '[.][.]', '', perl=TRUE) Error in grep("[.][.]", "", perl = TRUE) : invalid regular expression '[.][.]' In addition: Warning message: In grep("[.][.]",

There's a problem new to R2.4.0 when rbinding an auto-row-named data.frame to a list: > rbind( data.frame( x=1), list( x=2)) Error in attr(value, "row.names") <- rlabs : row names must be 'character' or 'integer', not 'double' Works OK with 2 data.frames or 2 lists. Mark Bravington CSIRO Mathematical & Information Sciences Marine

There is a bug in `predict' whereby the order of variables sometimes gets re-arranged compared to the original fit, and then disaster results. Specifically, the 'variables' and 'predvars' attributes of a 'terms' object get out of synch. This only happens when the terms in the original formula get re-ordered during fitting: test> scrunge.data_ data.frame(

# r-bugs@r-project.org `termplot' bombs if the original dataset contains NAs in the predictors, even if these were ignored during fitting: test> bug.data_ data.frame( x=c(1:3, NA), y=runif( 4)) test> lm.tpbug_ lm( y~x, data=bug.data) test> termplot( lm.tpbug) Error in xy.coords(x, y, xlabel, ylabel, log) : subscript out of bounds A simple fix seems to be add the

This is not so much a bug as an infelicity in the code that can easily be fixed. The initialize expression in the quasi family function is, (uniformly for all links and all variance functions): initialize <- expression({ n <- rep.int(1, nobs) mustart <- y + 0.1 * (y == 0) }) This is inappropriate (and often fails) for variance function "mu(1-mu)".

I'm encountering excruciatingly slow load times for character vectors in R 2.6.0-- up to 30sec for a 15K file that contains a no-attributes character vector of length ~1e4 and object size ~0.5MB. In R 2.5.1, repeated loads of the same set of files are near-instantaneous. The problem is proving tricky to reproduce consistently from scratch, so I have attached the 3 files used in the examples

> I think we should get rid of source attributes completely, > since they are no longer needed, but your comment still > applies to source references. We should strip them when code > gets modified. > > Duncan Murdoch I would be very concerned about losing source attributes-- it would break a lot of my code :(! It's very useful to have a single portable R object that

# r-bugs@r-project.org `predict' complains about new factor levels, even if the "new" levels are merely levels in the original that didn't occur in the original fit and were sensibly dropped, and that don't occur in the prediction data either. (At least if `drop.unused.levels' was set to TRUE, which the default.) test> scrunge.data.2_ data.frame( y=runif( 3),

Is this the last bug of the year? Well, it's the last one from me, anyway... The "parse" function seems to give erratic behaviour when used in conjunction with "pushBack" on an open connection (R1.6.1, Windows 2000). Try this: > { cat( c( '1', 'a+b', '2'), file='r123.r', sep='\n'); tcon_ file( 'r123.r'); open( tcon);

The function 'getAnywhere' crashes if given a non-existent name containing a period: > getAnywhere( 'nomethod.noclassforme') Error in get(x, envir, mode, inherits) : variable "nomethod" was not found However, 'getAnywhere' behaves gracefully if the non-existent name lacks a period: > getAnywhere( 'nomethod') no object named `nomethod' was

When replacing all elements of a column in a data.frame, a corrupt data.frame will be generated if the replacer is "short": test> bbb <- data.frame( x=1:2) test> bbb[,'x'] <- 3 # should fill the column test> bbb x 1 3 2 <NA> Warning message: corrupt data frame: columns will be truncated or padded with NAs in: format.data.frame(x, digits = digits)

In R-1.7.1, I used to be able to append a character vector to a 'tklistbox' with e.g. listio <- tklistbox( tktoplevel(), font='Courier', height=20, width=20, setgrid=TRUE) tkinsert( listio, 'end', letters[1:3]) tkpack( listio,side='left', expand=TRUE, fill='both') and three items would be added to 'listio'. This doesn't work in R-devel-- it

Thanks, Mark - I'm taking up your invitation to forward your message to the list just because it gives us some valuable data on (1) how long the behavior has been around, and (2) how many other people (don't) understand the behavior, and (3) how we might fix or work around it. I notice some other people also seem to be diffident about posting on R-devel; perhaps I should conclude that

Dear R-devel When 'install.packages' runs, it updates all html files in all packages. Mostly, there seems to be no actual change to the html file contents, but the date/time does change. This has causing been me a bit of trouble, because I keep synchronized versions of R on several different machines, and whenever I install a package, many MB of file transfers are required; my slow upload