An NA in the rownames of a matrix (or dataframe) causes misalignment when the
matrix is printed:
R> x <- matrix(1:12, 3,4, dimnames=list(letters[1:3], LETTERS[1:4]))
R> rownames(x)[2] <- NA
R> x
A B C D
a 1 4 7 10
<NA> 2 5 8 11
c 3 6 9 12
The bug is in function Rstrlen, in src/main/printutils.c. MatrixRowLabel and
MatrixColumnLabel (same file) rely on Rstrlen to count the characters in each
label, but Rstrlen doesn't understand NA's. Here's a patch, to be
applied
between lines 242 ("int len;") and 243 ("len = 0;"):
if (s == CHAR(NA_STRING))
return quote ? R_print.na_width : R_print.na_width_noquote;
which solves the problem:
R-patched> x
A B C D
a 1 4 7 10
<NA> 2 5 8 11
c 3 6 9 12
-----------------------------------------------------------------------------
An NA in the colnames has similar misalignment problems:
R> x <- matrix(1:12, 3,4, dimnames=list(letters[1:3], LETTERS[1:4]))
R> colnames(x)[2] <- NA
R> x
A <NA> C D
a 1 4 7 10
b 2 5 8 11
c 3 6 9 12
But this time it's because lines 72, 139, 213, 299, and 375 of
src/main/printarray.c use "strlen", which doesn't know about
NA's:
clabw = strlen(CHAR(STRING_ELT(cl, j)));
My patch is to replace "strlen" with "Rstrlen" in each case:
clabw = Rstrlen(CHAR(STRING_ELT(cl, j)), 0);
and again this solves the problem:
R-patched> x
A <NA> C D
a 1 4 7 10
b 2 5 8 11
c 3 6 9 12
-----------------------------------------------------------------------------
References:
Don MacQueen reported a similar problem in the body of the matrix on 5/1/02:
http://www.r-project.org/nocvs/mail/r-help/2002/3455.html
but this was fixed in R-1.5.1.
Derek Yi <Derek.Yi@fmr.com> first noted the rownames problem on 8/1/02 in
http://www.r-project.org/nocvs/mail/r-help/2002/6366.html
and Brian D. Ripley <ripley@stats.ox.ac.uk> replied on 8/1/02 in
http://www.r-project.org/nocvs/mail/r-help/2002/6367.html
-----------------------------------------------------------------------------
Version:
platform = sparc-sun-solaris2.6
arch = sparc
os = solaris2.6
system = sparc, solaris2.6
status =
major = 1
minor = 5.1
year = 2002
month = 06
day = 17
language = R
Search Path:
.GlobalEnv, package:misc, package:io, package:arrays, package:ls1,
package:g.data, package:db, package:ts, package:ctest, Autoloads, package:base
--
-- David Brahm (brahm@alum.mit.edu)
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Have you looked at R-devel? Neither of these are currently incorrect. Rstrlen is an exported entry point, and I don't think we want to mess with it. It is used for other than printing. On Tue, 13 Aug 2002 brahm@alum.mit.edu wrote:> An NA in the rownames of a matrix (or dataframe) causes misalignment when the > matrix is printed: > > R> x <- matrix(1:12, 3,4, dimnames=list(letters[1:3], LETTERS[1:4])) > R> rownames(x)[2] <- NA > R> x > A B C D > a 1 4 7 10 > <NA> 2 5 8 11 > c 3 6 9 12 > > The bug is in function Rstrlen, in src/main/printutils.c. MatrixRowLabel and > MatrixColumnLabel (same file) rely on Rstrlen to count the characters in each > label, but Rstrlen doesn't understand NA's. Here's a patch, to be applied > between lines 242 ("int len;") and 243 ("len = 0;"): > > if (s == CHAR(NA_STRING)) > return quote ? R_print.na_width : R_print.na_width_noquote; > > which solves the problem: > > R-patched> x > A B C D > a 1 4 7 10 > <NA> 2 5 8 11 > c 3 6 9 12 > > ----------------------------------------------------------------------------- > > An NA in the colnames has similar misalignment problems: > > R> x <- matrix(1:12, 3,4, dimnames=list(letters[1:3], LETTERS[1:4])) > R> colnames(x)[2] <- NA > R> x > A <NA> C D > a 1 4 7 10 > b 2 5 8 11 > c 3 6 9 12 > > But this time it's because lines 72, 139, 213, 299, and 375 of > src/main/printarray.c use "strlen", which doesn't know about NA's: > > clabw = strlen(CHAR(STRING_ELT(cl, j))); > > My patch is to replace "strlen" with "Rstrlen" in each case: > > clabw = Rstrlen(CHAR(STRING_ELT(cl, j)), 0); > > and again this solves the problem: > > R-patched> x > A <NA> C D > a 1 4 7 10 > b 2 5 8 11 > c 3 6 9 12 > > ----------------------------------------------------------------------------- > > References: > > Don MacQueen reported a similar problem in the body of the matrix on 5/1/02: > http://www.r-project.org/nocvs/mail/r-help/2002/3455.html > but this was fixed in R-1.5.1. > > Derek Yi <Derek.Yi@fmr.com> first noted the rownames problem on 8/1/02 in > http://www.r-project.org/nocvs/mail/r-help/2002/6366.html > and Brian D. Ripley <ripley@stats.ox.ac.uk> replied on 8/1/02 in > http://www.r-project.org/nocvs/mail/r-help/2002/6367.html > > ----------------------------------------------------------------------------- > > Version: > platform = sparc-sun-solaris2.6 > arch = sparc > os = solaris2.6 > system = sparc, solaris2.6 > status > major = 1 > minor = 5.1 > year = 2002 > month = 06 > day = 17 > language = R > > Search Path: > .GlobalEnv, package:misc, package:io, package:arrays, package:ls1, package:g.data, package:db, package:ts, package:ctest, Autoloads, package:base > > -- > -- David Brahm (brahm@alum.mit.edu) > > -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- > 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 > _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._ >-- Brian D. Ripley, ripley@stats.ox.ac.uk Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/ University of Oxford, Tel: +44 1865 272861 (self) 1 South Parks Road, +44 1865 272860 (secr) Oxford OX1 3TG, UK Fax: +44 1865 272595 -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- 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 _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Brian D. Ripley <ripley@stats.ox.ac.uk> wrote:> Have you looked at R-devel? Neither of these are currently incorrect.I see you fixed these very issues on Aug 1 by modifying MatrixRowLabel and friends. Sorry for not checking there first! I will let Derek know. -- -- David Brahm (brahm@alum.mit.edu) -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- 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 _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Reasonably Related Threads
- Suppressing Scientific Notation
- Non-alignment of <NA> in rownames
- formatReal()-bug (or Is there anybody out there w/o IEEE754?)
- 111 FIXMEs in main/src
- Bug#454678: r-base-core: Crash when calling edit.matrix with edit.row.names = TRUE when there are no rownames (PR#10500)