Oh dear, I seem to have suffered a case of reversed arguments. This explains my surprise why R didn't have this in a function already - as it does! I was following the pattern of search.vector %in% pattern, but match() arguments are opposite this.
Thanks to both Davids. Michael -----Original Message----- From: David Winsemius [mailto:dwinsem...@comcast.net] Sent: July-09-14 2:01 PM To: Folkes, Michael Cc: David L Carlson; r-help@r-project.org Subject: Re: [R] using match to obtain non-sorted index values from non-sortedvector On Jul 9, 2014, at 1:13 PM, Folkes, Michael wrote: > So nice! > Apply wins again. I doubt that `sapply( ..., which(,) )` would win a foot race with `match`: > match(Tset, pop.df$pop) [1] 5 4 2 -- David. > Thanks David. > Michael > > -----Original Message----- > From: David L Carlson [mailto:dcarl...@tamu.edu] > Sent: July-09-14 1:11 PM > To: Folkes, Michael; r-help@r-project.org > Subject: RE: using match to obtain non-sorted index values from > non-sortedvector > > There may be a faster way, but > >> sapply(Tset, function(x) which(pop.df$pop==x)) > [1] 5 4 2 > > ------------------------------------- > David L Carlson > Department of Anthropology > Texas A&M University > College Station, TX 77840-4352 > > -----Original Message----- > From: r-help-boun...@r-project.org > [mailto:r-help-boun...@r-project.org] > On Behalf Of Folkes, Michael > Sent: Wednesday, July 9, 2014 2:58 PM > To: r-help@r-project.org > Subject: [R] using match to obtain non-sorted index values from > non-sorted vector > > Hello all, > > I've been struggling with the best way to find index values from a > large vector with elements that will match elements of a subset vector > [the table argument in match()]. > > BUT the index values can't come out sorted (as we'd get in which(X > %in% > Y) ). > > My 'population' vector can't be sorted. > > pop.df <- data.frame(pop=c(1,6,4,3,10)) > > The subset: Tset = c(10,3,6) > > > > So I'd like to get these index values (from pop.df) , in this order: > 5,4,2 > > > > If it could be sorted I could use: > > which(sort(pop.df$pop) %in% sort(Tset)) > > > > But sorting will cause more grief later, so best not mess with it. > > Here is my hopefully adequate MWE of a solution. I'm keen to see if > anybody has a better suggestion. > > Thanks! > > _____________________ > > ###BEGIN R > > #pop is the full set of values, it has no info on their ranking > > # I don't want to sort these data. They need to remain in this order. > > pop.df <- data.frame(pop=c(1,6,4,3,10)) > > > > #rank.df is my dataframe that tells me the top three rankings (derived > elsewhere) > > rank.df <- data.frame(rank=1:3, Tset = c(10,3,6)) # Target set > > > > #match.df will be my source of row index based on rank > > match.df <- data.frame(match.vec= match(pop.df$pop, > table=rank.df$Tset), > index.vec=1:nrow(pop.df)) > > > #rank.df will now include the index location in the pop.df where I can > find the top three ranks. > > rank.df <- merge(rank.df, match.df, by.x='rank', by.y='match.vec') > > rank.df > > > ####END > > > > _______________________________________________________ > > Michael Folkes > > Salmon Stock Assessment > David Winsemius Alameda, CA, USA ______________________________________________ R-help@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.
