update fails after specific sequence of steps (PR#474)

Strange things happen on Unix too.

source('fox.R')
Moore<-read.table('Moore.txt')
attach(Moore)
mod.full<-lm(conform~status*fcategory)
summary(mod.full)

...

mod.m<-update(mod.full, ~.~-status:fcategory)

Error in model.frame.default(formula = ~conform ~ 1, drop.unused.levels = TRUE) : 
	object is not a matrix

mod.m<-update(mod.full, .~.-status:fcategory) # works following summary and error
mod.m

Call:
lm(formula = conform ~ status + fcategory)

Coefficients:
    (Intercept)        statuslow     fcategorylow  fcategorymedium  
       14.72356         -4.60667          0.08089         -1.09511  


#### Well you *thought* it worked, but watch: ####

mod.full<-lm(conform~status*fcategory)
summary(mod.full)

Call:
lm(formula = conform ~ L×€@ * fcategory)
                       ^^^^^^^

...and the attached dataframe is corrupted. Must investigate further.

ls("Moore")

[1] "conform"   "fcategory" "fscore"    "L×€@"      "subject"  
                                         ^^^^^^^

O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)             FAX: (+45) 35327907
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._