R devel - Jun 2001 - expand.model.frame() fails when subset is specified (PR# 979)

> Thanks. This is also present in the current pre-1.3.0. Your patch
  > looks correct, but I wonder if the default for "enclos" 
  > should not be
  > environment(formula(model)) rather than parent.frame() as it is now?
  > 
  > (And wouldn't it be better named "envir"?)

Peter,

I was merely making an incremental improvement, your suggestions provide
additional improvements.   Go right ahead and make the changes.  [This seems
wierd, are you asking 
me, non-R-core person that I am, permission to make changes to the R source
code?]

Just for illustration, here's an example of why it is better to use  
environment(formula(model)) rather than parent.frame():

	> fun <- function() {
	+ data <- data.frame(x=1:10,y=1:10,z=1:10,m=1:10)
	+ model <- lm( y ~ x, data=data, subset=m<8)
	+ model
	}

	> test <- fun()
	> 

With my version:

	> expand.model.frame(test, ~ z )
	Error in as.data.frame.default(data) : can't coerce function into a
data.frame

With the additional changes:

	> test  <-  fun()
	> expand.model.frame(test, ~ z )
	  y x z
	1 1 1 1
	2 2 2 2
	3 3 3 3	
	4 4 4 4
	5 5 5 5
	6 6 6 6
	7 7 7 7

Pretty nifty.  

-Greg

Just in case its helpful, heres the patch:

-------BEGIN PATCH-------

diff -c R-1.2.3.orig/src/library/base/R/expand.model.frame.R
R-.2.3/src/library/base/R/expand.model.frame.R
*** R-1.2.3.orig/src/library/base/R/expand.model.frame.R	Thu Jun 14
17:29:43 2001
--- R-1.2.3/src/library/base/R/expand.model.frame.R	Thu Jun 14 17:29:43
2001
***************
*** 1,9 ****
! expand.model.frame <- function(model, extras,  enclos=parent.frame(),
                                 na.expand=FALSE)
  {
      ## don't use model$call$formula -- it might be a variable name
      f <- formula(model)
!     data <- eval(model$call$data, enclos)
  
      # new formula (there must be a better way...)
      ff <- foo ~ bar + baz
--- 1,10 ----
! expand.model.frame <- function(model, extras,
!                                envir=environment(formula(model)),
                                 na.expand=FALSE)
  {
      ## don't use model$call$formula -- it might be a variable name
      f <- formula(model)
!     data <- eval(model$call$data, envir)
  
      # new formula (there must be a better way...)
      ff <- foo ~ bar + baz
***************
*** 18,27 ****
      if (!na.expand){
          naa <- model$call$na.action
          subset <- model$call$subset
!         rval <- model.frame(ff, data=data, subset=subset, na.action=naa)
      } else {
          subset <- model$call$subset
!         rval <- model.frame(ff, data=data, subset=subset, na.action=I)
          oldmf <- model.frame(model)
          keep <- match(rownames(oldmf), rownames(rval))
          rval <- rval[keep, ]
--- 19,30 ----
      if (!na.expand){
          naa <- model$call$na.action
          subset <- model$call$subset
!         rval <- eval(call("model.frame",ff, data = data, subset =
subset,
!                       na.action = naa),envir )
      } else {
          subset <- model$call$subset
!         rval <- eval(call("model.frame",ff, data = data, subset =
subset,
!                           na.action = I), envir)
          oldmf <- model.frame(model)
          keep <- match(rownames(oldmf), rownames(rval))
          rval <- rval[keep, ]

--------END PATCH--------


LEGAL NOTICE
Unless expressly stated otherwise, this message is confidential and may be
privileged. It is intended for the addressee(s) only. Access to this E-mail by
anyone else is unauthorized. If you are not an addressee, any disclosure or
copying of the contents of this E-mail or any action taken (or not taken) in
reliance on it is unauthorized and may be unlawful. If you are not an addressee,
please inform the sender immediately.
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

"Warnes, Gregory R" <gregory_r_warnes@groton.pfizer.com> writes:
>   > Thanks. This is also present in the current pre-1.3.0. Your patch
>   > looks correct, but I wonder if the default for "enclos" 
>   > should not be
>   > environment(formula(model)) rather than parent.frame() as it is now?
>   > 
>   > (And wouldn't it be better named "envir"?)
> 
> Peter,
> 
> I was merely making an incremental improvement, your suggestions provide
> additional improvements.   Go right ahead and make the changes.  [This
seems
> wierd, are you asking 
> me, non-R-core person that I am, permission to make changes to the R source
> code?]
Sorry, I might have made that more clear. Followups on bug reports get
CC'ed into R-bugs and stored with the original report, as well as
circulated via R-devel. I was just making a note for whoever might have
a better informed opinion than myself, as well as for the person who'll
get around to putting in the fix eventually (which might well be
myself). 

-- 
   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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._