Skip to content

Cleaning up in the bugs repository

3 messages · Peter Dalgaard, Jens Oehlschlägel-Akiyoshi

Original
Peter Dalgaard
# 
In order not to have more items in the BUGS list than necessary, I've
gone through the list of open bugs and removed some of them for
reasons given below:

(Note that nothing really dies on R-bugs it just moves into trashcan,
wishlist, or Something-fixed, so if you complain loudly enough they
can be picked up again.)

*Graphics*

Remove three reports on panel.first() (2 + 1 answer with a different
number) because one really cannot expect to do what they were trying
to do. We may still want to use eval(panel.first) in plot.default,
though. 

*Low-level*

Jens O-A's:
"as.matrix.data.frame() modifies content of character columns"

was moved to Documentation-fixed. The behaviour is Splus compatible,
so hardly fixable.

*System-specific*

SuSE RPM (non-)issue got cleared up by Albrecht earlier today, I
believe. -->trashcan

*TooMuchAtOnce*

Two very old reports by Jim Lindsey and Rashid Nassar --> trashcan
(because, as the title implies, they are very long and most of the
items have been fixed or gone out of date). Please resubmit (in
smaller pieces!) if you still think there are bugs.
Jens Oehlschlägel-Akiyoshi
# 
Peter,

you wrote
this is shocking bad news, as it means, that we stay with a severe bug, be
it in S+ or not.
It means, we can't be shure, that the CONTENT of a character data remains
unchanged, if we rearange the data.

x1 <- '"'
x2 <- data.frame(I(x1))
x3 <- as.matrix(x2)
[1] "\""
[1] "\""
[1] "\\\""


I came across this, when I tried to write a function, which exports R data
reliably to flatfiles. Of course I used as.matrix.data.frame as it is a
basic function, but

IF this bug remains in R, there can just be one consequence: NEVER use
as.matrix, at least with character data AND REMEMBER not to use it.

Very unhappy about this




-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Jens Oehlschlägel-Akiyoshi
# 
- sorry the previous mail went out by accidentaly hitting a key, here is the
end -

.. please rethink, if there is any way to get that fixed. I mean, do we
really have to expect S+ code to break, because it relies on this bug?

Regards


Jens Oehlschlägel-Akiyoshi

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