[R] max.print

Martin Maechler maechler at stat.math.ethz.ch
Mon Mar 12 14:10:41 CET 2012


>>>>> Peter Ehlers <ehlers at ucalgary.ca>
>>>>>     on Sat, 10 Mar 2012 11:04:06 -0800 writes:

    > On 2012-03-10 08:35, sybil kennelly wrote:
    >> Dear all.
    >> 
    >> I wanted to read in a 20,000 row X 60 column matrix
    >> (called "table") into R.
    >> 
    >> i did this:
    >> 
    >>> R table<- read.table("table", header=TRUE) table
    >> 
    >> it prints out the start of my table (~10000 rows by 7
    >> columns) and then this error:
    >> 
    >> 
    >> [ reached getOption("max.print") -- omitted 5465 rows ]]
    >> There were 50 or more warnings (use warnings() to see the
    >> first 50)
    >> 
    >> I have tried:
    >> 
    >> options(max.print = Inf)
    >> 
    >> and options(max.print = 9999999999999)
    >> 
    >> but i still get the same error. I have seen many people
    >> on R help have this problem. However the solution of
    >> options(max.print = Inf) does not seem to work for me.
    >> 
    >> 
    >> Any ideas?

    > Well, I don't know why you would want to do this to your
    > eyeballs, but View() would seem to be your friend and this
    > is probably somewhere in the archives.

well, I clearly agree with the first part...

    > You can't set max.print to anything that can't be coerced
    > to integer (see ?integer) and I think that setting it to
    > Inf is no longer legal (if ever it was).  [Perhaps
    > options() should generate a warning.]

Well, actually, it does .. a bit later... and every time you
print, from then on.
But you are right: R should alert the user more clearly!
I propose to implement that it will produce an *error*, if the
result is not a positive integer.

Another question:  Would anyone  (on R-core ?)
know why we print  two closing "]]" but only one initial "[" 
in the following ?

 [ reached getOption("max.print") -- omitted 19800 rows ]]

If nobody sees a good reason for the above asymmetry,
I'll change it to something symmetric, i.e., matching pair,
and would propose to use the more rare pair
 "{{" .. "}}"


Martin Maechler, ETH Zurich



More information about the R-help mailing list