R devel - Feb 2000 - "a.matrix[a.char.matrix]<- " crashes R (PR#447)

To wit:
---
> junk <-
array(3,c(10,3),dimnames=list(as.character(1:10),as.character(1:3)))
> junk[cbind(as.character(1:5),as.character(c(1,2,3,2,3)))] <- 20
> 
> junk
Process R segmentation violation (core dumped) at Wed Feb 16 15:57:49
2000
---

I know that the use of a ***character*** matrix is not documented for
this context. Still, it should not crash R.

In addition to 0.90.1, this also happens on rw0991b


--please do not edit the information below--

Version:
 platform = sparc-sun-solaris2.7
 arch = sparc
 os = solaris2.7
 system = sparc, solaris2.7
 status = 
 major = 0
 minor = 90.1
 year = 1999
 month = December
 day = 15
 language = R

Search Path:
 .GlobalEnv, Autoloads, package:base

-- 
Charles C. Berry                        (858) 534-2098 
                                         Dept of Family/Preventive
Medicine
E mailto:cberry@tajo.ucsd.edu	         UC San Diego
http://hacuna.ucsd.edu/members/ccb.html  La Jolla, San Diego 92093-0645

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

Peter Dalgaard BSA <p.dalgaard@biostat.ku.dk> writes:
> Still crashes for me!
[Directed back to R-bugs, and please keep it there!]

OK, there are two issues here:

(a) matrix subscripting doesn't work with character type. Plain and
simple. A matrix subscript just get flattened into a character vector
and then used for subscripting. We might want to give an error instead
or simply implement the expected behaviour, but not this close to a
major release, I think. (Splus cant do this either)

(b) There's a bug in string indexing:

  x<-1:2
  x[c("2","2")]<-2
  x

Program received signal SIGSEGV, Segmentation fault. <etc>

It seems to happen only when there are repeated indices. The bug is
that the names component of x which is originally empty isn't filled
with "" for the first elements as it usually is but has NULL (0x0)
pointers there.

Will dig deeper...

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)             FAX: (+45) 35327907

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

> From: p.dalgaard@biostat.ku.dk
> Date: Fri, 18 Feb 2000 17:05:50 +0100 (MET)
> 
> Peter Dalgaard BSA <p.dalgaard@biostat.ku.dk> writes:
> 
> > Still crashes for me!
> 
> [Directed back to R-bugs, and please keep it there!]
> 
> OK, there are two issues here:
> 
> (a) matrix subscripting doesn't work with character type. Plain and
> simple. A matrix subscript just get flattened into a character vector
> and then used for subscripting. We might want to give an error instead
> or simply implement the expected behaviour, but not this close to a
> major release, I think. (Splus cant do this either)
I think that is the correct behaviour. (What is the expected
behaviour?  It is what I expected.) A k-index (aka matrix or array
subset) is supposed to give the indices of the elements to be
extracted/replaced. I suppose character values could be matched to
dimnames, but that would be a new idea.  We might want to *warn*.

On the other hand, S-PLUS's print routine seems to ignore the added
elements, and R loses the dim attribute.
> (b) There's a bug in string indexing:
> 
>   x<-1:2
>   x[c("2","2")]<-2
>   x
> 
> Program received signal SIGSEGV, Segmentation fault. <etc>
> 
> It seems to happen only when there are repeated indices. The bug is
repeated *new *indices.

The hackery in StringSubscript was copying the wrong string as a name
for duplicate string indices. Fixed.

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