R help - Nov 2010 - Aggregate on identical link attributes

Hello R community, 
   I need to do some aggregation based on the test data below.  The below
code works ok, but when its applied to my real data which includes over
9,000 records the process runs for over an hour.  I know there is a more
efficient way of doing this.  I want to Sum the below data's volumes where
FNODE and TNODE match, 

i.e. FNODE=1 and TNODE =20 -> Volume=100 ,
      FNODE=20 and TNODE =1 -> Volume=100 

These should be aggregated toTotalVolume= 200.  
      
Also, there are some link without partner links(see record 21 in test data). 
This record should not be summed with another link since there isnt a
compliment to sum with.

Thanks in advance.
-JR


#Create test data
TNode<-c(1:20,21)
FNode<-c(rev(1:20),22)
Volume<-c(rep(100,20),200)
ClassCode=c(rep("Local",20),rep("Freeway",1))
#Create data frame with test data
EmmeData..<-data.frame(TNode=TNode,FNode=FNode,Volume=Volume,ClassCode=ClassCode)

#Create list to store results
 LinkSum_<-list()
#Create vecotrs to establish loops
  TNodes<-unique(EmmeData..$TNode)
  FNodes<-unique(EmmeData..$FNode)

  for(tn in TNodes){
      for(fn in FNodes){
          TF<-EmmeData..[EmmeData..$TNode==tn & EmmeData..$FNode==fn,]
          FT<-EmmeData..[EmmeData..$TNode==fn & EmmeData..$FNode==tn,]
          if(length(TF$TNode)>0){
              
LinkSum_[[tn]]<-list(ID=list(c(TF$TNode,TF$FNode)),ClassCode=TF$ClassCode,TotalVolume=TF$Volume+FT$Volume)
          }
         
      }
  }    

-- 
View this message in context:
http://r.789695.n4.nabble.com/Aggregate-on-identical-link-attributes-tp3044009p3044009.html
Sent from the R help mailing list archive at Nabble.com.

Okay here is a solution that works in less than 60 minutes but i feel likes
its messy, if anyone has an alternative solution i would very much
appreciate your insights.  

#Create test data
TNode<-c(1:20,21)
FNode<-c(rev(1:20),22)
Volume<-c(rep(100,20),200)
ClassCode=c(rep("Local",20),rep("Freeway",1))
#Create data frame with test data
EmmeData..<-data.frame(TNode=TNode,FNode=FNode,Volume=Volume,ClassCode=ClassCode)
 
#Create vectors to establish loops
 TNodes<-unique(ED..$TNode)
 FNodes<-unique(ED..$FNode)
   
EmmeData..$TF<-paste(EmmeData..$TNode,EmmeData..$FNode,sep="-")
EmmeData..$FT<-paste(EmmeData..$FNode,EmmeData..$TNode,sep="-")
 
#Split string into 2 elements to paste back together in reverse order as a
matching ID
EmmeData..$F<-unlist(lapply(strsplit(EmmeData..$FT,split="-"),
"[", 1))
EmmeData..$T<-unlist(lapply(strsplit(EmmeData..$FT,split="-"),
"[", 2))
EmmeData..$Match<-paste(EmmeData..$F,EmmeData..$T,sep="-")
z<-list()
Nd<-EmmeData..$TF
for(nd in Nd){
    x<-EmmeData..[EmmeData..$TF==nd,]
    y<-EmmeData..[EmmeData..$Match==x$TF,]
    if(length(y$Volume)==0){
        z[[nd]]<-list(ID=x$TF,Volume=x$Volume,Class=x$ClassCode) 
    }
    if(length(y$Volume)==1){
        z[[nd]]<-list(ID=x$TF,Volume=x$Volume+y$Volume,Class=x$ClassCode) 
    }
    
}
 
CalcVolumes.. <- do.call('rbind', z) 


-- 
View this message in context:
http://r.789695.n4.nabble.com/Aggregate-on-identical-link-attributes-tp3044009p3045949.html
Sent from the R help mailing list archive at Nabble.com.