[R] grid(Base): How to avoid "Figure region too small and/or viewport too large" by specifying 'relative' units?

[Marius Hofert]

Dear Paul,

Thank you very much for helping, works perfectly fine. 

I now see the scaling when looking at grid.show.layout(), but I would have never
come up with looking there in the first place :-)

Cheers,

Marius


Paul Murrell <paul at stat.auckland.ac.nz> writes:

> Hi
>
> Apologies for the slow response.
>
> This part of the problem is I think just a bit of a misunderstanding. The
> diagram drawn by grid.show.viewport() is drawn within a *subset* of the current
> page (or viewport), to allow room for labelling, which is why your subsequent
> "real" viewports do not align with the diagram.
> If you change your grid.show.layout() call to the following (which "removes" the
> normal margin used by grid.show.layout()) ...
>
> grid.show.layout(gl, vp=viewport(width=1.25, height=1.25))
>
> ... then you should find your viewports line up with the diagram properly.
>
> Paul
>
> On 20/10/12 19:10, Marius Hofert wrote:
>> In the meanwhile, I found a more minimal example which shows the problem (just
>> change 'inch' to TRUE to see the difference):
>>
>>
>> require(grid)
>>
>> inch <- FALSE # TRUE
>>
>> d <- if(inch) 5 else 1
>> pspc <- d*c(0.3, 0.3) # width, height of panels
>> spc <- d*c(0.05, 0.05) # width, height of space
>> axlabspc <- d*c(0.1, 0.1) # width y label, height x label
>> labspc <- d*c(0.05, 0.05) # width label boxes, height label boxes
>>
>> par. <- par(no.readonly=TRUE)
>> gl <- grid.layout(5, 5, default.units=if(inch) "inches" else "npc",
>>                    widths=c(axlabspc[1], pspc[1], spc[1], pspc[1], labspc[1]),
>>                    heights=c(labspc[2], pspc[2], spc[2], pspc[2], axlabspc[2]))
>> grid.show.layout(gl)
>> pushViewport(viewport(layout=gl))
>> for(i in 1:2) {
>>      for(j in 1:2) {
>>          pushViewport(viewport(layout.pos.row=2*i, layout.pos.col=2*j, name="foo"))
>>          grid.rect()
>>          upViewport()
>>      }
>> }
>> par(par.)
>>
>> ______________________________________________
>> R-help at r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>> and provide commented, minimal, self-contained, reproducible code.
>>