[R] Odp: nice way to find or not a value (problem with numeric(0))

Petr PIKAL petr.pikal at precheza.cz
Fri Mar 6 10:06:50 CET 2009 

Hi

r-help-bounces at r-project.org napsal dne 05.03.2009 15:21:22:

> 
> Hello Petr,
> 
> In fact spec is data.frame with a column called code (containing 
numerical
> values) and some other columns called data1, data2, ... containing data 
for
> each equipment (that is for each code).
> 
> But I don't have the data for all my devices. It means that some 'code' 
are
> not in spec$Code.
> In that case I want to assign a default value to the data.
> 
> for example :
> spec$Code     spec$data1
> 4                           12.5
> 8                            20.2
> 
> Then, with code=4 
> spec$data1[spec$Code==code]*(code %in%specmodules$Code) + 1*(!code 
> %in% specmodules$Code) 
> gives 12.5
> 
> But with code=654,
> I get numeric(0) instead of 1
> because 
> this value is not in spec$Code and returns numeric(0).
> 
> I hope it is clearer (is it ?) and that you could find a nice way to 
write
> my test (the if-test works but is not very elegant).

Not much. Guess what I get using your code? If you think it is 12.5 you 
are mistaken.

> spec
  code data1
1    4  12.5
2    8  20.2
> code=4
> spec$data1[spec$code==code]*(code %in%specmodules$Code) + 1*(!code %in% 
specmodules$Code)
Error in inherits(x, "factor") : object "specmodules" not found
>

This is why you shall provide reproducible code. Now I still need to only 
**think** what you want to do.

Maybe merge is what you want

Having spec like above and

test<-data.frame(code = c(4,5,8,12), value= c(10, 20, 30,40))

then

merge(test, spec, by="code", all.x=T)
  code value data1
1    4    10  12.5
2    5    20    NA
3    8    30  20.2
4   12    40    NA

Gives you new data frame which you can then filter by is.na and which to 
replace NA values with standard ones.

Regards
Petr

> 
> Thanks in adance for you help, 
> Ptit Bleu.
> 
> 
> 
> Petr Pikal wrote:
> > 
> > Hi
> > 
> > r-help-bounces at r-project.org napsal dne 04.03.2009 09:11:06:
> > 
> >> 
> >> Hello,
> >> 
> >> I have a data.frame called spec containing data about samples. But I 
> > don't
> >> have these data for all my samples.
> >> So if I have data (that is code of the sample is in spec$Code), I 
would 
> > like
> >> to assign data1 to the variable m.
> >> If I don't have this data, I would like to assign 1 to m.
> >> 
> >> I tried this : 
> >> m<-spec$data1[spec$Code==code]*(code %in%specmodules$Code) + 1*(!code 

> > %in%
> >> specmodules$Code) 
> >> 
> >> It works when I have the data but if it is not the case I get 
numeric(0)
> >> instead of 1.
> >> 
> >> I finally use the following command. It works but I'm sure there is a 

> > more
> >> elegant way.
> >> if (code %in%spec$Code) m<-spec$data1[spec$Code==code] else m<-1
> > 
> > It is a bit cryptic what do you want. Above version shall not work as 
it 
> > takes only one logical value but you probably have vector of values. 
(We 
> > do not know code, spec$Code or any other data you have).
> > 
> > when I try your first construction with some values I have I get 
sensible 
> > results so without trying to find out how your data really look like I 

> > suggest you to inspect it more closely and/or provide some working 
example 
> > demonstrating what you did, what is the result and how the result 
shall 
> > look like.
> > 
> > zdrz$sklon*zdrz$otac %in% c(.6,1.2,2)+1*!(zdrz$otac %in% c(.6,1.2,2))
> >  [1] 110  80  50  50  10   1 120  80  50  20
> > zdrz$otac[5]<-NA
> > zdrz$sklon*zdrz$otac %in% c(.6,1.2,2)+1*!(zdrz$otac %in% c(.6,1.2,2))
> >  [1] 110  80  50  50   1   1 120  80  50  20
> > zdrz$sklon[4]<-Inf
> > zdrz$sklon*zdrz$otac %in% c(.6,1.2,2)+1*!(zdrz$otac %in% c(.6,1.2,2))
> >  [1] 110  80  50 Inf   1   1 120  80  50  20
> > zdrz$sklon[4]<-NA
> > zdrz$sklon*zdrz$otac %in% c(.6,1.2,2)+1*!(zdrz$otac %in% c(.6,1.2,2))
> >  [1] 110  80  50  NA   1   1 120  80  50  20
> > 
> > Regards
> > Petr
> > 
> >> 
> >> Is there a way to avoid an if-test ?
> >> 
> >> Thanks for your help,
> >> Have a good day,
> >> Ptit Bleu.
> >> 
> >> -- 
> >> View this message in context: 
> > http://www.nabble.com/nice-way-to-find-or-not-a-
> >> value-%28problem-with-numeric%280%29%29-tp22325406p22325406.html
> >> Sent from the R help mailing list archive at Nabble.com.
> >> 
> >> ______________________________________________
> >> R-help at r-project.org mailing list
> >> https://stat.ethz.ch/mailman/listinfo/r-help
> >> PLEASE do read the posting guide 
> > http://www.R-project.org/posting-guide.html
> >> and provide commented, minimal, self-contained, reproducible code.
> > 
> > ______________________________________________
> > R-help at r-project.org mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide
> > http://www.R-project.org/posting-guide.html
> > and provide commented, minimal, self-contained, reproducible code.
> > 
> > 
> 
> -- 
> View this message in context: 
http://www.nabble.com/nice-way-to-find-or-not-a-
> value-%28problem-with-numeric%280%29%29-tp22325406p22352529.html
> Sent from the R help mailing list archive at Nabble.com.
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide 
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

More information about the R-help mailing list