[R] Alternative to cxy in R

[Martin Maechler]

>>>>> "BDR" == Prof Brian D Ripley <ripley at stats.ox.ac.uk> writes:

    BDR> On Wed, 5 Jan 2000, [iso-8859-1] Jesús María Frías Celayeta wrote:
    >> good morning to the R-help list!,
    >> 
    >> I would like to ask if there is an alternative to the S ploting
    >> parameter "cxy" in R. I am trying to port some S code to R that makes use
    >> of this "cxy" parameter to scale the graphics, and I don't find this
    >> parameter in help(par).

    BDR> R has cin, which gives this in inches, and pin and usr will enable
    BDR> you to convert these to user coordinates (I think those are the
    BDR> right ones).

    BDR> You can usually make use of csi, for example yinch(par("csi")) is a
    BDR> reasonable replacement for par("cxy")[2], and by use of cra you
    BDR> can rescale to the width.

Last time this was a topic on R-help, your answer was a bit more
elaborated: 

>> From: Prof Brian D Ripley <ripley at stats.ox.ac.uk>
>> Sender: owner-r-help at stat.math.ethz.ch
>> To: John Maindonald <john.maindonald at anu.edu.au>
>> cc: r-help at stat.math.ethz.ch
>> Subject: Re: [R] Character height & width in user co-ordinates
>> Date: Mon, 17 May 1999 07:49:29 +0100 (BST)
>> 
>> On Mon, 17 May 1999, John Maindonald wrote:
>> 
>> > Is there an equivalent to the S-PLUS par()$cxy in R?
>> 
>> That is specified as giving:
>> 
>> cxy=c(x1,y1)    character size (width and height) in usr coordinates.
>> 
>> When I needed this (V&R2, p.74) I used for the height yinch(par("csi")).
>> The width is a lot less clearcut, since R (and indeed S-PLUS) allows
>> variable-width fonts. I would use strwidth() to find the actual width.
>> There is also strheight(), but I think that computes the actual
>> height rather than the nominal height given by par("cxy").
>> 
>> I think that  (0.57*xinch(par("csi")), yinch(par("csi"))) is a reasonable
>> substitute, 0.57 coming from some experiments I did when I first
>> encountered this.

In the mean time, the font width/height settings have undergone a few
patches; but I don't know how much your proposal above would be affected.

But anyway: I think   par.Rd (i.e.  help(par) )
should contain something like the above information.
Alternatively, we could introduce  par("cxy") (as "read_only") into R,
which would solve porting problems, 
but as you said we should tell users, rather to use strwidth() instead.

By the way, the following script is somewhat revealing:

  plot(1)##-- using x11()  {default fonts, ...}
  chs <- c(strsplit(" .,;:!'`|#@$%^&*()-_=+/\\<>",NULL)[[1]],
	   letters,LETTERS, paste(0:9))
  chs <- chs[chs != ""]# fix bug in R 0.99 (unreleased)
  wh <- t(sapply(chs, function(s)c(w = strwidth(s),h = strheight(s))))

  apply(round(10000 * wh,2), 2, table)
  ##- $w
  ##-  60.02  80.03 100.04 120.04 140.05 160.06 180.07 200.07 220.08 240.09 
  ##-     13      7      1      5     34      9     13      2      3      1 

  ##- $h
  ##- 201.45 
  ##-     88 

  summary(10000 * wh[,"w"])
  ##-    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  ##-   60.02  120.00  140.10  135.10  160.10  240.10 

  round(10000* wh[,"w"])
  ##-       .   ,   ;   :   !   '   `   |   #   @   $   %   ^   &   *   ( 
  ##-  80  60  80  60  60  60  60  60  60 140 240 140 220 120 180 100  80 
  ##-   )   -   _   =   +   /  \\   <   >   a   b   c   d   e   f   g   h 
  ##-  80 160 140 140 140  80  80 140 140 140 140 140 140 140  60 140 140 
  ##-   i   j   k   l   m   n   o   p   q   r   s   t   u   v   w   x   y 
  ##-  60  60 120  60 180 140 140 140 140  80 120  60 140 140 180 120 140 
  ##-   z   A   B   C   D   E   F   G   H   I   J   K   L   M   N   O   P 
  ##- 120 180 160 180 180 160 160 180 180  60 140 160 140 220 180 200 160 
  ##-   Q   R   S   T   U   V   W   X   Y   Z   0   1   2   3   4   5   6 
  ##- 200 160 160 140 160 180 220 180 180 180 140 140 140 140 140 140 140 
  ##-   7   8   9 
  ##- 140 140 140 


  10000* xyinch(par("csi")) #---- CONSIDERABLY larger (also for y !)
  ## 300.6317 335.7513


Martin Maechler <maechler at stat.math.ethz.ch>	http://stat.ethz.ch/~maechler/
Seminar fuer Statistik, ETH-Zentrum  LEO D10	Leonhardstr. 27
ETH (Federal Inst. Technology)	8092 Zurich	SWITZERLAND
phone: x-41-1-632-3408		fax: ...-1228			<><
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help 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-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._