[Rd] Bugs and comments. (PR#410)

[plindsey@alpha.luc.ac.be]

Hi,
Here are a few errors I found as well as a few comments.

1) In the man page of par:
          lty: The line type.  Line types can either be specified
               as an integer (0=blank, 1=solid, 2=dashed, 3=dot-
               ted, 4=dotdash, 5=longdash, 6=twodash) or as one
               of the character strings `"blank"', `"solid"',
               `"dashed"', `"dotted"' `"dotdash"', `"longdash"',
               or `"twodash"', where `"blank"' uses `invisible
               lines' (i.e. doesn't draw them).

               Alternatively, a string of up to 8 characters
               (from code{c(0:9, "A":"F")}) may be given, giving
  -------------------^^^^^---------------^
               the length (typically in points/pixels) of line
               segments which are alternatively drawn and
               skipped.  For example, `"44"' is dashed and `"13"'
               is dotted.

2) I was trying to get the warnings to look as they use to do but even
  setting the option warn to 1 does not work. For example what kind is
  generating the following warning when a library is loaded???

  > options(warn=1)
  > library(eda)
  Warning: Explicit local dynamic loading not supported on this platform. Using default.
  > 

  Problem can also occurs with error messages:

  > a <- function(a,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p,q,r,s,t,u,v,w,x,y,z)
      stop("This is an very long error message which will be difficult to see especially in a Microsoft windows version of R.")
  > a(1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26)
  Error in a(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17,  : This is an very long error message which will be difficult to see especially in a Microsoft windows version of R.

  Basically, in Microsoft windows you can only see about as much as "...  : This$"
  and then the rest is of the screen. As their is no horizontal scroll
  bar... the students are unable to see the error message!

  I personally, prefer warnings to be printed out as they occur, that way
  I can see what is going on. By storing the warnings in a an object rather
  than printing them out if there were more than 10 seems to me as the
  opposite thing to do. Indeed, if lost of warnings occurred it means that
  there is more than certainly a problem. This seems to be getting like SAS
  where people will soon just run something get a result and not even notice
  that a model has not converged (if there were 9 warnings of NA/Inf replaced
  by maximum value and then the no convergence message... they will all be
  stored and the user will not even noticed that the model has not even
  converged!!). This is especially annoying when teaching with students as
  you can imagine.

  As well, I personally think that too much is printed out about the
  function which generated the error. I think that its name is enough
  and that its arguments are of no relevance to where the error
  occurred and just make things harder to read.

3) I got the following error message from the plot function. 

  Error in xy.coords(x, y, xlabel, ylabel, log) : x and y lengths differ

  Now how is this possible as the arguments of xy.coord are the following??

  > args(xy.coords)
  function (x, y, xlab = NULL, ylab = NULL, log = NULL, recycle = FALSE) 
  NULL

  As well if you do:

  >  xy.coords(x=response,y=time,xlabel="time",ylabel="weight")
  Error in xy.coords(x = response, y = time, xlabel = "time", ylabel = "weight") : unused argument to function

  It is pretty hard to figure out what argument is unused... it might
  be a good idea to print out which argument was not found.

Patrick.


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