Hmmm - I just wrote a set of automated tests and it all looks ok (f7ba8cd). Could you please try the latest development version and see if you still have problems?

from roxygen.

On 10/24/2011 8:42 PM, hadley wickham wrote:

Hmmm - I just wrote a set of automated tests and it all looks ok (f7ba8cd). Could you please try the latest development version and see if you still have problems?

I tried with v2.1 roxygen2 which was at link and it did not work. See
attached .r and the created .rd file. If I take away quotes it works.

--jeff
\name{EnvironCovariates}
\alias{EnvironCovariates}
\title{Extracts environmental covariate data for SMI Zc analysis
: by extracting data from ACCESS database and creating a series of
anomaly summaries for sea surface temperature(SST), upwelling index (UWI) and
multivariate ENSO index (MEI).}
\arguments{
\item{average.years}{years to use for average for anomaly
creationccc}

\item{fdir}{directory containing environmental.data.mdb}

\item{sites}{SST sites to be used in SST averaging (from
create.SST.anomalies) 1) ESB, 2)WSB, 3)PtArg, 4)PtSM,
5)PtSL, 6)CSM, 7)MB, 8)PtReyes}
}
\value{
dataframe with rows as years from 1987 to last year in
data and the columns are:
time,OcttoJuneSSTAnomalies,ApriltoJuneSSTAnomalies,JulytoJuneSSTAnomalies,
OcttoJuneUWI33Anomalies, ApriltoJuneUWI33Anomalies,
JulytoJuneUWI33Anomalies, OcttoJuneUWI36Anomalies,
ApriltoJuneUWI36Anomalies, JulytoJuneUWI36Anomalies,
OcttoJuneMEI, ApriltoJuneMEI, JulytoJuneMEI
}
\description{
Extracts environmental covariate data for SMI Zc analysis
: by extracting data from ACCESS database and creating a
series of anomaly summaries for sea surface
temperature(SST), upwelling index (UWI) and multivariate
ENSO index (MEI).
}
\author{
Jeff Laake
}

#' Extracts environmental covariate data for SMI Zc analysis
#' : by extracting data from ACCESS database and creating a series of
#' anomaly summaries for sea surface temperature(SST), upwelling index (UWI) and
#' multivariate ENSO index (MEI).
#'
#' @export
#' @import CIPinnipedAnalysis
#' @param average.years years to use for average for anomaly creationccc
#' @param fdir directory containing environmental.data.mdb
#' @param sites SST sites to be used in SST averaging (from create.SST.anomalies)
#' 1) ESB, 2)WSB, 3)PtArg, 4)PtSM, 5)PtSL, 6)CSM, 7)MB, 8)PtReyes
#' @return dataframe with rows as years from 1987 to last year in data and the columns are:
#' time,OcttoJuneSSTAnomalies,ApriltoJuneSSTAnomalies,JulytoJuneSSTAnomalies,
#' OcttoJuneUWI33Anomalies, ApriltoJuneUWI33Anomalies, JulytoJuneUWI33Anomalies,
#' OcttoJuneUWI36Anomalies, ApriltoJuneUWI36Anomalies, JulytoJuneUWI36Anomalies,
#' OcttoJuneMEI, ApriltoJuneMEI, JulytoJuneMEI
#' @author Jeff Laake
"EnvironCovariates"<-function(average.years=c(1994:1996,1998:2008),fdir="",sites=1:5)
{

SST Anomalies

if(fdir=="")fdir=system.file(package = "CIPinnipedAnalysis")
anomalies=create.SST.anomalies(average.years,fdir=fdir,store=FALSE)
fpath=file.path(fdir,"environmental.data.mdb")
connection=odbcConnectAccess(fpath)
SSTAnomalies=t(apply(anomalies[,,sites],c(2,1),mean,na.rm=TRUE))
SSTAnomalies[is.nan(SSTAnomalies)]=NA
nyears=nrow(SSTAnomalies)
OcttoJuneSSTAnomalies=rowMeans(cbind(SSTAnomalies[1:(nyears-1),c("Oct","Nov","Dec")],SSTAnomalies[2:nyears,c("Jan","Feb","Mar","Apr","May","June")]),na.rm=TRUE)
ApriltoJuneSSTAnomalies=rowMeans(SSTAnomalies[,c("Apr","May","June")],na.rm=TRUE)
names(ApriltoJuneSSTAnomalies)=as.character(as.numeric(names(ApriltoJuneSSTAnomalies))-1)
JulytoJuneSSTAnomalies=rowMeans(cbind(SSTAnomalies[1:(nyears-1),c("July","Aug","Sept","Oct","Nov","Dec")],SSTAnomalies[2:nyears,c("Jan","Feb","Mar","Apr","May","June")]),na.rm=TRUE)

UpwellingIndex for 33N & 36N

UWI=sqlFetch(connection,"UWIAnomaly")
UWI=UWI[order(UWI$Year,UWI$Month),]
UWI=tapply(UWI$UWI,list(UWI$Year,UWI$Month,UWI$Location),unique)
minyear=min(as.numeric(dimnames(UWI)[[1]]))
maxyear=max(as.numeric(dimnames(UWI)[[1]]))
nyears=maxyear-minyear+1
OcttoJuneUWI33Anomalies=rowMeans(cbind(UWI[1:(nyears-1),as.character(10:12),1],UWI[2:nyears,as.character(1:6),1]),na.rm=TRUE)
ApriltoJuneUWI33Anomalies=rowMeans(UWI[,as.character(4:6),1],na.rm=TRUE)[1:nyears]
names(ApriltoJuneUWI33Anomalies)=as.character(as.numeric(names(ApriltoJuneUWI33Anomalies))-1)
JulytoJuneUWI33Anomalies=rowMeans(cbind(UWI[1:(nyears-1),as.character(7:12),1],UWI[2:nyears,as.character(1:6),1]),na.rm=TRUE)
OcttoJuneUWI36Anomalies=rowMeans(cbind(UWI[1:(nyears-1),as.character(10:12),2],UWI[2:nyears,as.character(1:6),2]),na.rm=TRUE)
ApriltoJuneUWI36Anomalies=rowMeans(UWI[,as.character(4:6),2],na.rm=TRUE)
names(ApriltoJuneUWI36Anomalies)=as.character(as.numeric(names(ApriltoJuneUWI36Anomalies))-1)
JulytoJuneUWI36Anomalies=rowMeans(cbind(UWI[1:(nyears-1),as.character(7:12),2],UWI[2:nyears,as.character(1:6),2]),na.rm=TRUE)

Multivariate ENSO Index - lagged by 3 months

MEI=sqlFetch(connection,"MEI")
minyear=min(MEI$Year)
maxyear=max(MEI$Year)
nyears=maxyear-minyear+1
MEI=tapply(MEI$MEI,list(MEI$Year,MEI$Month),unique)
OcttoJuneMEI=rowMeans(cbind(MEI[1:(nyears-1),as.character(7:12)],MEI[2:nyears,as.character(1:3)]),na.rm=TRUE)
ApriltoJuneMEI=rowMeans(MEI[,as.character(1:3)],na.rm=TRUE)
names(ApriltoJuneMEI)=as.character(as.numeric(names(ApriltoJuneMEI))-1)
JulytoJuneMEI=rowMeans(cbind(MEI[1:(nyears-1),as.character(4:12)],MEI[2:nyears,as.character(1:3)]),na.rm=TRUE)

Create dataframe with values from 1987:2009

maxyear=max(as.numeric(names(JulytoJuneMEI)))

time is meant to match the beginning year of a survival interval from time to time+1; that is why the

April-to-June is set as the previous year to model survival of pup cohort born in the previous year

envcovdf=data.frame(time=1987:maxyear,
        OcttoJuneSSTAnomalies=OcttoJuneSSTAnomalies[as.character(1987:maxyear)],
        ApriltoJuneSSTAnomalies=ApriltoJuneSSTAnomalies[as.character(1987:maxyear)],
        JulytoJuneSSTAnomalies=JulytoJuneSSTAnomalies[as.character(1987:maxyear)],
        OcttoJuneUWI33Anomalies=OcttoJuneUWI33Anomalies[as.character(1987:maxyear)],
        ApriltoJuneUWI33Anomalies=ApriltoJuneUWI33Anomalies[as.character(1987:maxyear)],
        JulytoJuneUWI33Anomalies=JulytoJuneUWI33Anomalies[as.character(1987:maxyear)],
        OcttoJuneUWI36Anomalies=OcttoJuneUWI36Anomalies[as.character(1987:maxyear)],
        ApriltoJuneUWI36Anomalies=ApriltoJuneUWI36Anomalies[as.character(1987:maxyear)],
        JulytoJuneUWI36Anomalies=JulytoJuneUWI36Anomalies[as.character(1987:maxyear)],
        OcttoJuneMEI=OcttoJuneMEI[as.character(1987:maxyear)],
        ApriltoJuneMEI=ApriltoJuneMEI[as.character(1987:maxyear)],
        JulytoJuneMEI=JulytoJuneMEI[as.character(1987:maxyear)]
        )
odbcClose(connection)
return(envcovdf)

}

from roxygen.

Ok - should definitely be fixed now.

from roxygen.

On 10/25/2011 12:40 PM, hadley wickham wrote:

Ok - should definitely be fixed now.

That worked but now it is throwing in an odd character into the usage
regardless of whether I use quotes. See attached. --jeff

#' Extracts environmental covariate data for SMI Zc analysis
#' : by extracting data from ACCESS database and creating a series of
#' anomaly summaries for sea surface temperature(SST), upwelling index (UWI) and
#' multivariate ENSO index (MEI).
#'
#' @export
#' @import CIPinnipedAnalysis
#' @param average.years years to use for average for anomaly creation
#' @param fdir directory containing environmental.data.mdb
#' @param sites SST sites to be used in SST averaging (from create.SST.anomalies)
#' 1) ESB, 2)WSB, 3)PtArg, 4)PtSM, 5)PtSL, 6)CSM, 7)MB, 8)PtReyes
#' @return dataframe with rows as years from 1987 to last year in data and the columns are:
#' time,OcttoJuneSSTAnomalies,ApriltoJuneSSTAnomalies,JulytoJuneSSTAnomalies,
#' OcttoJuneUWI33Anomalies, ApriltoJuneUWI33Anomalies, JulytoJuneUWI33Anomalies,
#' OcttoJuneUWI36Anomalies, ApriltoJuneUWI36Anomalies, JulytoJuneUWI36Anomalies,
#' OcttoJuneMEI, ApriltoJuneMEI, JulytoJuneMEI
#' @author Jeff Laake
EnvironCovariates<-function(average.years=c(1994:1996,1998:2008),fdir="",sites=1:5)
{

SST Anomalies

if(fdir=="")fdir=system.file(package = "CIPinnipedAnalysis")
anomalies=create.SST.anomalies(average.years,fdir=fdir,store=FALSE)
fpath=file.path(fdir,"environmental.data.mdb")
connection=odbcConnectAccess(fpath)
SSTAnomalies=t(apply(anomalies[,,sites],c(2,1),mean,na.rm=TRUE))
SSTAnomalies[is.nan(SSTAnomalies)]=NA
nyears=nrow(SSTAnomalies)
OcttoJuneSSTAnomalies=rowMeans(cbind(SSTAnomalies[1:(nyears-1),c("Oct","Nov","Dec")],SSTAnomalies[2:nyears,c("Jan","Feb","Mar","Apr","May","June")]),na.rm=TRUE)
ApriltoJuneSSTAnomalies=rowMeans(SSTAnomalies[,c("Apr","May","June")],na.rm=TRUE)
names(ApriltoJuneSSTAnomalies)=as.character(as.numeric(names(ApriltoJuneSSTAnomalies))-1)
JulytoJuneSSTAnomalies=rowMeans(cbind(SSTAnomalies[1:(nyears-1),c("July","Aug","Sept","Oct","Nov","Dec")],SSTAnomalies[2:nyears,c("Jan","Feb","Mar","Apr","May","June")]),na.rm=TRUE)

UpwellingIndex for 33N & 36N

UWI=sqlFetch(connection,"UWIAnomaly")
UWI=UWI[order(UWI$Year,UWI$Month),]
UWI=tapply(UWI$UWI,list(UWI$Year,UWI$Month,UWI$Location),unique)
minyear=min(as.numeric(dimnames(UWI)[[1]]))
maxyear=max(as.numeric(dimnames(UWI)[[1]]))
nyears=maxyear-minyear+1
OcttoJuneUWI33Anomalies=rowMeans(cbind(UWI[1:(nyears-1),as.character(10:12),1],UWI[2:nyears,as.character(1:6),1]),na.rm=TRUE)
ApriltoJuneUWI33Anomalies=rowMeans(UWI[,as.character(4:6),1],na.rm=TRUE)[1:nyears]
names(ApriltoJuneUWI33Anomalies)=as.character(as.numeric(names(ApriltoJuneUWI33Anomalies))-1)
JulytoJuneUWI33Anomalies=rowMeans(cbind(UWI[1:(nyears-1),as.character(7:12),1],UWI[2:nyears,as.character(1:6),1]),na.rm=TRUE)
OcttoJuneUWI36Anomalies=rowMeans(cbind(UWI[1:(nyears-1),as.character(10:12),2],UWI[2:nyears,as.character(1:6),2]),na.rm=TRUE)
ApriltoJuneUWI36Anomalies=rowMeans(UWI[,as.character(4:6),2],na.rm=TRUE)
names(ApriltoJuneUWI36Anomalies)=as.character(as.numeric(names(ApriltoJuneUWI36Anomalies))-1)
JulytoJuneUWI36Anomalies=rowMeans(cbind(UWI[1:(nyears-1),as.character(7:12),2],UWI[2:nyears,as.character(1:6),2]),na.rm=TRUE)

Multivariate ENSO Index - lagged by 3 months

MEI=sqlFetch(connection,"MEI")
minyear=min(MEI$Year)
maxyear=max(MEI$Year)
nyears=maxyear-minyear+1
MEI=tapply(MEI$MEI,list(MEI$Year,MEI$Month),unique)
OcttoJuneMEI=rowMeans(cbind(MEI[1:(nyears-1),as.character(7:12)],MEI[2:nyears,as.character(1:3)]),na.rm=TRUE)
ApriltoJuneMEI=rowMeans(MEI[,as.character(1:3)],na.rm=TRUE)
names(ApriltoJuneMEI)=as.character(as.numeric(names(ApriltoJuneMEI))-1)
JulytoJuneMEI=rowMeans(cbind(MEI[1:(nyears-1),as.character(4:12)],MEI[2:nyears,as.character(1:3)]),na.rm=TRUE)

Create dataframe with values from 1987:2009

maxyear=max(as.numeric(names(JulytoJuneMEI)))

time is meant to match the beginning year of a survival interval from time to time+1; that is why the

April-to-June is set as the previous year to model survival of pup cohort born in the previous year

envcovdf=data.frame(time=1987:maxyear,
        OcttoJuneSSTAnomalies=OcttoJuneSSTAnomalies[as.character(1987:maxyear)],
        ApriltoJuneSSTAnomalies=ApriltoJuneSSTAnomalies[as.character(1987:maxyear)],
        JulytoJuneSSTAnomalies=JulytoJuneSSTAnomalies[as.character(1987:maxyear)],
        OcttoJuneUWI33Anomalies=OcttoJuneUWI33Anomalies[as.character(1987:maxyear)],
        ApriltoJuneUWI33Anomalies=ApriltoJuneUWI33Anomalies[as.character(1987:maxyear)],
        JulytoJuneUWI33Anomalies=JulytoJuneUWI33Anomalies[as.character(1987:maxyear)],
        OcttoJuneUWI36Anomalies=OcttoJuneUWI36Anomalies[as.character(1987:maxyear)],
        ApriltoJuneUWI36Anomalies=ApriltoJuneUWI36Anomalies[as.character(1987:maxyear)],
        JulytoJuneUWI36Anomalies=JulytoJuneUWI36Anomalies[as.character(1987:maxyear)],
        OcttoJuneMEI=OcttoJuneMEI[as.character(1987:maxyear)],
        ApriltoJuneMEI=ApriltoJuneMEI[as.character(1987:maxyear)],
        JulytoJuneMEI=JulytoJuneMEI[as.character(1987:maxyear)]
        )
odbcClose(connection)
return(envcovdf)

}

\name{EnvironCovariates}
\alias{EnvironCovariates}
\title{Extracts environmental covariate data for SMI Zc analysis
: by extracting data from ACCESS database and creating a series of
anomaly summaries for sea surface temperature(SST), upwelling index (UWI) and
multivariate ENSO index (MEI).}
\usage{
EnvironCovariates(average.years = c(1994:1996, 1998:2008),
fdir = "", sites = 1:5)
}
\arguments{
\item{average.years}{years to use for average for anomaly
creation}

\item{fdir}{directory containing environmental.data.mdb}

\item{sites}{SST sites to be used in SST averaging (from
create.SST.anomalies) 1) ESB, 2)WSB, 3)PtArg, 4)PtSM,
5)PtSL, 6)CSM, 7)MB, 8)PtReyes}
}
\value{
dataframe with rows as years from 1987 to last year in
data and the columns are:
time,OcttoJuneSSTAnomalies,ApriltoJuneSSTAnomalies,JulytoJuneSSTAnomalies,
OcttoJuneUWI33Anomalies, ApriltoJuneUWI33Anomalies,
JulytoJuneUWI33Anomalies, OcttoJuneUWI36Anomalies,
ApriltoJuneUWI36Anomalies, JulytoJuneUWI36Anomalies,
OcttoJuneMEI, ApriltoJuneMEI, JulytoJuneMEI
}
\description{
Extracts environmental covariate data for SMI Zc analysis
: by extracting data from ACCESS database and creating a
series of anomaly summaries for sea surface
temperature(SST), upwelling index (UWI) and multivariate
ENSO index (MEI).
}
\author{
Jeff Laake
}

from roxygen.

What's the odd character?

from roxygen.

On 10/25/2011 2:04 PM, hadley wickham wrote:

What's the odd character?

A capital A with a hat. You should be able to see it in the file I
attached to last message. I retyped the line in the function and it is
still there.

from roxygen.

Where is it? I wonder if it's an encoding problem, or if I'm just missing it.

from roxygen.

On 10/25/2011 2:15 PM, hadley wickham wrote:

Where is it? I wonder if it's an encoding problem, or if I'm just missing it.

average.years = c(1994:1996, 1998:2008),

from roxygen.

On 10/25/2011 2:15 PM, hadley wickham wrote:

Where is it? I wonder if it's an encoding problem, or if I'm just missing it.

It has something to do with the comma in c(1994:1996, 1998:2008). When I
take out the comma and text before or after, it works fine. It is not a
weird character in that part because if I retype it it is still there.

--jeff

from roxygen.

I think it's probably a non-breaking space that shouldn't be there - I've been adding them to get the usage wrapping correctly, but somehow I must be failing to remove it. I'll take a look in the next couple of days.

from roxygen.

On 10/26/2011 2:02 PM, hadley wickham wrote:

I think it's probably a non-breaking space that shouldn't be there - I've been adding them to get the usage wrapping correctly, but somehow I must be failing to remove it. I'll take a look in the next couple of days.

No hurry. Just thought I'd try a few things to see if it helped you
debug. --jeff

from roxygen.