Just to update this thread, Tom added this FR(#1244). The result of merge is now keyed when all=FALSE, too. In 1.5.2. Matthew
On Thu, 2011-01-06 at 10:40 +0000, Matthew Dowle wrote: > If all is TRUE (or just all.x or all.y is TRUE) then result is keyed. > > Why the key is not set when all is FALSE (default) I'm not sure. Tom - can > you remember? Something to do with NAs in keys seems to ring a bell. > > Matthew > > > "Joseph Voelkel" <[email protected]> wrote in message > news:[email protected]... > > See code below. When two data tables are merged (by their key > > variable(s)), that key disappears in the resulting table. While this is > > probably not strictly a bug, it would seem that the user would most likely > > want to preserve the key in the output (especially if just enlarging the > > original data table, e.g. dt2<-merge(dt1,dt2).) > > > > > >> dt1<-data.table(a=1:6,b=11:16,key="a") > >> dt2<-data.table(a=1:6,c=21:26,key="a") > >> dt3<-merge(dt1,dt2) > >> > >> > > NAME NROW MB COLS KEY > > [1,] dt1 6 1 a,b a > > [2,] dt2 6 1 a,c a > > [3,] dt3 6 1 a,b,c > > Total: 3MB > > > > > > Joseph G. Voelkel, Ph.D. > > Professor, Center for Quality and Applied Statistics > > Kate Gleason College of Engineering > > Rochester Institute of Technology > > V 585-475-2231 > > F 585-475-5959 > > [email protected] > > > > _______________________________________________ > datatable-help mailing list > [email protected] > https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/datatable-help _______________________________________________ datatable-help mailing list [email protected] https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/datatable-help
