R help - Jan 2004 - degree-min-sec data

Hello -

Have both astronomic and geodetic data sets with values in the form
"ddd:mm:ss.sssss", where dd is an integer between -180 and 180, mm is
an
integer between 0 and 60, and ss is a floating-point
number between 0 and 60.0.  In order to do anything useful with these
values they need to be turned into their "decimal degree" equivalent.

Assuming the data is a vector y, the following works, sort of:

z<-strsplit(as.character(y),split=":")
zz<-sapply(z,as.numeric)
x<- t(c(1,1/60,1/3600) %*% zz)

Trouble comes if there are any NA's in the data.  In this case, R refuses
to coerce zz into matrix form, but leaves it as a "list of lists". 
How can
the above routine be patched up to pass NAs through?
I can think of some inelegant solutions (e.g.  recoding NA as
"-200:-1:-1")
but surely there is a better way!
Or is this the wrong technique?

More generally, is there a way to "unzip" a character vector along
some
split marker, handling irregulatirites gracefully?  For example, from

y<-c("aaa#bbb", "ccc#ddd",
"eee","fff"),

strsplit(y,split="#") would produce

c(c("aaa","bbb"), c("ccc","ddd"),
c("eee","fff")).

Is there a way to produce instead

c(c("aaa","ccc","eee"),c("bbb","ddd","fff"))
?

Thanks for your assistance.

<>=<>=<>=<>=<>=<>=<>=<>=<>=<>=<>
George Heine, PhD
Mathematical Analyst
National IRM Center
U.S. Bureau of Land Management
voice   (303) 236-0099
fax       (303) 236-1974
pager   (303) 826-8182 or gheine at my2way.com
<>=<>=<>=<>=<>=<>=<>=<>=<>=<>=<>

George_Heine at blm.gov writes:
> Hello -
> 
> Have both astronomic and geodetic data sets with values in the form
> "ddd:mm:ss.sssss", where dd is an integer between -180 and 180,
mm is an
> integer between 0 and 60, and ss is a floating-point
> number between 0 and 60.0.  In order to do anything useful with these
> values they need to be turned into their "decimal degree"
equivalent.
> 
> Assuming the data is a vector y, the following works, sort of:
> 
> z<-strsplit(as.character(y),split=":")
> zz<-sapply(z,as.numeric)
> x<- t(c(1,1/60,1/3600) %*% zz)
> 
> Trouble comes if there are any NA's in the data.  In this case, R
refuses
> to coerce zz into matrix form, but leaves it as a "list of
lists".  How can
> the above routine be patched up to pass NAs through?
> I can think of some inelegant solutions (e.g.  recoding NA as
"-200:-1:-1")
> but surely there is a better way!
> Or is this the wrong technique?
Not necessarily. I might try something like

z <- strsplit(ifelse(is.na(y),y,"::NA"))
..etc..


There are also fancier possibilities:

z <-
scan(textConnection(y),sep=":",fill=TRUE,what=list(a=0,b=0,c=0))
x <- z$a + z$b/60 + z$c/3600

or even

x <- with(z, a + b/60 + c/3600)
> More generally, is there a way to "unzip" a character vector
along some
> split marker, handling irregulatirites gracefully?  For example, from
> 
> y<-c("aaa#bbb", "ccc#ddd",
"eee","fff"),
> 
> strsplit(y,split="#") would produce
> 
> c(c("aaa","bbb"), c("ccc","ddd"),
c("eee","fff")).
> 
> Is there a way to produce instead
> 
>
c(c("aaa","ccc","eee"),c("bbb","ddd","fff"))
?
?? Was that what you intended to write? 
> dput(strsplit(y,split="#"))
list(c("aaa", "bbb"), c("ccc", "ddd"),
"eee", "fff")


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