Hi Tal, In your example, note that x[[1]] is not a list; hence it must first be coerced to a list, and attributes are lost in this conversion. This is fairly standard in R code; eg x <- 1; attr(x, "animal") <- "cat"; x; as.list(x), so you can't blame Rcpp for being consistent with R here.
The onus is on you to be careful with your types when setting attributes. You should assume they are lost if you do any conversion. Indeed, the non-standard attributes are very brittle. That said, if you want the size of an RObject, there is always the C API to fall back on: Rf_length(x) gives you the length of the object, agnostic to its type. -Kevin On Tue, Aug 20, 2013 at 2:31 PM, Tal Galili <tal.gal...@gmail.com> wrote: > Dear Wush and others, > Just to followup and say I've used this idea here: > https://github.com/talgalili/dendextendRcpp/blob/master/src/cut_lower.cpp > > In general, it would have been better for me, where it possible to do some > (or all) of the things below: > 1) Get the length of a List > 2) Change the attribute of elements within a list > 3) Have this not happen: > > require(Rcpp) > cppFunction(' > SEXP temp(List x){ > return(x); > } > ') > x = list(structure(1L, animal = "cat")) > temp(x) > temp(x[[1]]) # I wish this didn't happen > > (I'm possibly missing out things, but I thought of listing it just in case) > > > Many thanks for the help! > Tal > > > > > > > ----------------Contact > Details:------------------------------------------------------- > Contact me: tal.gal...@gmail.com | > Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) | > www.r-statistics.com (English) > > ---------------------------------------------------------------------------------------------- > > > > On Tue, Aug 20, 2013 at 12:25 PM, Tal Galili <tal.gal...@gmail.com> wrote: > >> Hello Wush, >> Indeed, thank you! >> >> I will write more on this after I finish the code, since there are some >> bits which are a mystery to me. >> >> Tal >> >> >> ----------------Contact >> Details:------------------------------------------------------- >> Contact me: tal.gal...@gmail.com | >> Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) | >> www.r-statistics.com (English) >> >> ---------------------------------------------------------------------------------------------- >> >> >> >> On Tue, Aug 20, 2013 at 12:13 PM, Wush Wu <wush...@gmail.com> wrote: >> >>> Hi Tal, >>> >>> ``` >>> cppFunction(' >>> std::vector<RObject> temp2_fail(RObject x){ >>> std::vector<RObject> lower; >>> List x_list(wrap(x)); >>> int n = x_list.size(); >>> lower.push_back(x); >>> return(lower) ; >>> } >>> ') >>> ``` >>> >>> Is this what you want? >>> >>> >>> 2013/8/20 Tal Galili <tal.gal...@gmail.com> >>> >>>> Dear list members, >>>> >>>> I would like to be able to get an element pushed with its attributes, >>>> and I came across the following problem: >>>> >>>> cppFunction(' >>>> std::vector<List> temp(List x){ >>>> std::vector<List> lower; >>>> lower.push_back(x); >>>> return(lower) ; >>>> } >>>> ') >>>> x = list(structure(1L, animal = "cat")) >>>> temp(x) >>>> temp(x[1]) >>>> temp(x[[1]]) # is there a way to not make this output without >>>> attributes? >>>> >>>> >>>> >>>> If I use RObject, it will solve it, but then I would later not be able >>>> to iterate over the object, because I can't seem to get the .size() of an >>>> RObject. >>>> >>>> Here is the code with the RObject: >>>> >>>> >>>> cppFunction(' >>>> std::vector<RObject> temp2(RObject x){ >>>> std::vector<RObject> lower; >>>> lower.push_back(x); >>>> return(lower) ; >>>> } >>>> ') >>>> x = list(structure(1L, animal = "cat")) >>>> temp2(x) >>>> temp2(x[1]) >>>> temp2(x[[1]]) # Now it works as I want, but the following wouldn't work: >>>> >>>> >>>> >>>> >>>> cppFunction(' >>>> std::vector<RObject> temp2_fail(RObject x){ >>>> std::vector<RObject> lower; >>>> int n x.size(); >>>> lower.push_back(x); >>>> return(lower) ; >>>> } >>>> ') >>>> >>>> >>>> >>>> Any ideas? >>>> >>>> >>>> >>>> >>>> ----------------Contact >>>> Details:------------------------------------------------------- >>>> Contact me: tal.gal...@gmail.com | >>>> Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) >>>> | www.r-statistics.com (English) >>>> >>>> ---------------------------------------------------------------------------------------------- >>>> >>>> >>>> _______________________________________________ >>>> Rcpp-devel mailing list >>>> Rcpp-devel@lists.r-forge.r-project.org >>>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel >>>> >>> >>> >> > > _______________________________________________ > Rcpp-devel mailing list > Rcpp-devel@lists.r-forge.r-project.org > https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel >
_______________________________________________ Rcpp-devel mailing list Rcpp-devel@lists.r-forge.r-project.org https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel
