R help - Oct 2013 - behaviour of read.xls (gdata package) when worksheet usesuser-defined cells formats

(I'm afraid this post didn't reach the list on last Wednesday, here it
is again )

hi R-list,

And sorry for my frenglish !

I am running R < Good Sport > release ( i386-w64-mingw32/i386 (32-bit) ) )
under Windows 7 Professional, Service Pack 1.
My perl executable is ActivePerl build 817 [257965] (i.e. version 5.8.8.817).
Usually it is working fine.

Using the gdata ::read.xls function I am used to, I am now facing a stupid,
probably trivial problem I never encountered before... and can't fix it by
myself.

I have received Excel files (saved as Excel 97-2003 files)  in which the values
(say 11.6185410334347) are displayed in a user-defined format in the cells (here
the format is defined as "0.00", giving a displayed value of 0.12) but
fully and correctly displayed in the formula bar.
In these files, the cells in the 4 first lines are used for header information
(strings).

I read any of the files using :

foo.data<-
read.xls(xls=my.path.to.xls.file,sheet=sheet.number,perl=my.perl.path,as.is=TRUE,pattern="[0123456789]",head=FALSE)

The value I get in the resulting foo.data$the.variable is the displayed value
(0.12), not the "true", "underlying" value of 11.6185......
and it is neither a factor, nor a string :
> is.what(foo.data$the.variable)
[1] "is.atomic"   "is.double"   "is.numeric" 
"is.standard" "is.unsorted" "is.vector"
> foo.data$the.variable [2]*1e10
[1] 1.2e+09

Is this normal behavior ? Is this related to how I chose the read.xls arguments
? How should I specify the arguments to recover the full, "true"
values directly from the files (without changing the format manually to
"standard", of course !) ? Or is this related to my perl executable ?

... Or do I miss a point ?

Thanks for helping, all the best , Olivier


	[[alternative HTML version deleted]]

Hello
I had draw the results of PCA (Principal Components Analysis) (1)
Is it possible to put on this graphic the 75% ellipse confidence of each 
sex calculated by dataEllipse (library(car)) ?

My code is
1) PCA data frame with 169 rows and 2 columns
library(ggplot2)
p <- ggplot(PCA, aes(x=F1, y=F2 ))
p + geom_point(aes(colour = Sexe, shape = Sexe), size=3) +
geom_hline(yintercept = 0) +
geom_vline(xintercept = 0) +
labs(title = "ACP",
x = "Facteur 1", y = "Facteur 2",fill = "Sexe") +
theme(legend.position = c(0.95,0.95),
legend.background = element_rect(colour = "black"))

2) library(car)
XH <- dataEllipse(
X1[ACP$Sexe=="H"],
X2[ACP$Sexe=="H"],
levels=0.75, lty=2, add=TRUE, plot.points=FALSE, center.cex=0, col=4)

XF <- dataEllipse(
X1[ACP$Sexe=="F"],
X2[ACP$Sexe=="F"],
levels=0.75, lty=2, add=TRUE, plot.points=FALSE, center.cex=0, col=2)

XH and XF are two matrix with 2 columns and 52 rows
Thank you for your help

-- 
Michel ARNAUD
Charg? de mission aupr?s du DRH
DGDRD-Drh - TA 174/04
Av Agropolis 34398 Montpellier cedex 5
tel : 04.67.61.75.38
fax : 04.67.61.57.87
port: 06.47.43.55.31