Skip to content

New versions of lattice and grid

6 messages · apjaworski@mmm.com, Peter Dalgaard, Deepayan Sarkar +1 more

apjaworski@mmm.com
# 
I just ran package update on my Win2000 R-1.6.1 and it installed new
versions of lattice and grid (0.6-7 and 0.7-3 respectively).  Then I ran
demo(lattice) and noticed that the last plot, which is supposed to
illustrate math expression usage, fails in the sense that all expression
are shown literally.  For example the title of the plot reads " pi*sum(x,
i=0, n)", the x-axis label is "sigma[i]", the subtitle reads
frac(demonstrating, expressions), etc.  This happed using windows device as
well as postscrip device (on Win2000 and Linux platforms).

I am almost sure that this worked before.  Could anybody confirm this?

Andy

__________________________________
Andy Jaworski
Engineering Systems Technology Center
3M Center, 518-1-01
St. Paul, MN 55144-1000
-----
E-mail: apjaworski at mmm.com
Tel:  (651) 733-6092
Fax:  (651) 736-3122
Deepayan Sarkar
#
On Thursday 19 December 2002 01:35 pm, apjaworski at mmm.com wrote:
The problem is definitely real. Fortunately I had a fairly recent (but not the 
most recent) version of grid that didn't have this problem, and comparing the 
two, the problem seems to be a small change in grid.text. The older version 
seems to work:

grid.text <- function(label, x=unit(0.5, "npc"), y=unit(0.5, "npc"),
                  just="centre", rot=0, check.overlap=FALSE,
                  default.units="npc", gp=gpar(), draw=TRUE, vp=NULL) {
  if (!is.unit(x))
    x <- unit(x, default.units)
  if (!is.unit(y))
    y <- unit(y, default.units)
  txt <- list(label=label, x=x, y=y, gp=gp,
              ## WAS label = as.character(label)
              just=just, rot=rot, check.overlap=check.overlap,
              vp=vp)
  cl <- "text"
  grid.grob(txt, cl, draw)
}

I have no idea whether this might break anything else. If Paul is able to look 
at this before he disappears for the holidays, great. Otherwise, we will 
probably have to wait to get this fixed.

Deepayan
Peter Dalgaard
# 
Deepayan Sarkar <deepayan at stat.wisc.edu> writes:
...for values of "WAS" meaning "Is currently", I presume.
It's pretty patently wrong and I think Paul will be gone until after
the 1.6.2 release, so it would be unfortunate to wait. Perhaps we
could program defensively using something like

   if (!is.language(label)) label <- as.character(label)
   txt <- list(label=label, x=x, y=y, gp=gp, ...etc...

??
Paul Murrell
# 
Hi
Peter Dalgaard BSA wrote:
Not gone yet (and not for a couple of hours it would seem).  This was a
brain-dead bug fix of mine.  I'm working on a better one.  Once I have
that and have checked that I will put up a grid_0.7-4 

Paul
Deepayan Sarkar
#
On Thursday 19 December 2002 04:30 pm, Peter Dalgaard BSA wrote:
Right.
That seems like a good idea. For what it's worth, both grid and lattice pass R 
CMD check, and the other changes don't look like they are related.

Deepayan
Paul Murrell
# 
Hi

grid_0.7-4 has been submitted to CRAN.
Apologies for the stuff up.
I may try to go on holiday now ... :)

Paul

technical note:  I used is.expression() rather than is.language()
because that is all that grid's C code checks for.  Yes, this should be
expanded to allow is.language() at some point.

even more technical note:  is.language() does not correspond to the
C-level function isLanguage ...
Paul Murrell wrote: