Skip to content

grid.grab(): strange behavior of "wrap" argument when working with lattice output

2 messages · John G. Bullock, Paul Murrell

John G. Bullock
# 
grid.grab() captures all of the current viewports and stores them as a 
grob. In theory, and often in practice, you can use grid.grab() to store 
an image now for drawing later on. But I have noticed some odd behavior 
of grid.grab() when working with lattice output. It is all related to 
the wrap argument:

1. When wrap = TRUE, the saved grob should produce a drawing faithful to 
the original viewports. But sometimes it doesn't.
 ??? 2. wrap = FALSE is supposed to produce less faithful output if it 
differs at all, but sometimes its output is more faithful.
 ??? 3. Even when using wrap = TRUE, grid.grab() warns me to use wrap = 
TRUE.

Here is a minimal example to illustrate all of these behaviors:

library(grid)
 ??? library(lattice)
 ??? xyplot(1:5 ~ 1:5)
trellis.focus()
 ??? panel.abline(h = 3)
 ??? panel.abline(h = 5)
myGrob <- grid.grab(wrap = TRUE)

At this point, I receive a warning --

 ?? ?? In grabDL(warn, wrap, ...) :
 ? ? ? ? one of more grobs overwritten (grab WILL not be faithful; try 
'wrap = TRUE')

-- which is strange, given that I specified wrap = TRUE.

I proceed by running grid.draw(myGrob). According to Paul Murrell (R 
Graphics, 2nd ed., page 239), using grid.grab(wrap = TRUE) is 
"guaranteed to replicate the original output." But it doesn't. 
Specifically, the top horizontal line (from panel.abline(h = 5)) isn't 
reproduced.

Matters get stranger when I consider the following code:

 ??? xyplot(1:5 ~ 1:5)
 ??? trellis.focus()
 ??? panel.abline(h = 3)
panel.abline(h = 5)
 ??? myGrob <- grid.grab(wrap = FALSE)
grid.draw(myGrob)

These commands are the same as in the previous example, except that this 
time, wrap = FALSE. But this time, grid.draw() does faithfully reproduce 
the captured viewports. So the behavior is the opposite of what I 
expect: when wrap = FALSE, grid.draw() faithfully reproduces the 
captured viewports, but when wrap = TRUE, it doesn't.

I've run these examples in R 3.3.2 and 3.5.1. (Session info from 3.3.2 
is appended.) The result is the same in both cases. Are these bugs, or 
am I overlooking something?

Thank you,
John Bullock


 > sessionInfo()
R version 3.3.2 (2016-10-31)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 17134)

locale:
[1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United 
States.1252??? LC_MONETARY=English_United States.1252 LC_NUMERIC=C
[5] LC_TIME=English_United States.1252

attached base packages:
[1] grid????? stats???? graphics? grDevices utils???? datasets methods base

other attached packages:
[1] lattice_0.20-34 RevoUtilsMath_10.0.0

loaded via a namespace (and not attached):
[1] RevoUtils_10.0.2
Paul Murrell
# 
Hi

Mostly these are bugs.

One source of confusion:  wrap=FALSE is NOT doing better than wrap=TRUE, 
which you can see if you start a new page before drawing the grabbed 
grob ...

xyplot(1:5 ~ 1:5)
trellis.focus()
panel.abline(h = 3)
panel.abline(h = 5)
myGrob <- grid.grab(wrap = FALSE)
## start new page, otherwise just drawing over the top
grid.newpage()
grid.draw(myGrob)

But you are correct that wrap=TRUE is underperforming.  The problem is 
that your code is producing two grobs with the same name (and 
grid.grab() gets upset because you cannot create a gTree with two grobs 
with the same name).

Here is a workaround (make sure that each grob has a different name) ...

xyplot(1:5 ~ 1:5)
trellis.focus()
panel.abline(h = 3)
panel.abline(h = 5, identifier="abline2")
myGrob <- grid.grab()

grid.grab(wrap=TRUE) does not fix the problem because it does not wrap 
grobs.  For a more permanent solution, I have added a 'wrap.grobs' 
argument to grid.grab() (in r-devel), so that you can now do this 
(wrap.grobs=TRUE implicitly sets wrap=TRUE) ...

xyplot(1:5 ~ 1:5)
trellis.focus()
panel.abline(h = 3)
panel.abline(h = 5)
myGrob <- grid.grab(wrap.grobs = TRUE)

I have also fixed grid.grab() so that it does not complain about grobs 
with the same name if you have set wrap.grobs=TRUE.

Thanks for the report.  I hope this fixes the problems for you.

Paul
On 31/10/18 7:34 AM, John G. Bullock wrote: