On 27 April 2017 at 22:43, Kirill Müller wrote: | For RSQLite, I wanted to double-check that I'm not inadvertently | truncating integer values. Unfortunately, enabling -Wconversion gives me | several false positives that originate from Rcpp headers. | | I was able to fix these locally by editing 11 header files (see list | below), mostly by replacing int with size_t or R_xlen_t, or by adding | casts. Happy to submit several pull requests, but wanted to discuss | scope and granularity first: | | - one PR each for Dimension.h, api/, date_datetime/, module/, sugar/, | vector/ | - one PR per affected file | - one PR per affected line of code?
Not sure :-/ James opened a relate issue #678 for remaining R_len_t -> R_xlen_t conversions but at least one of the SEXPs, namely CHARSXP, still uses R_len_t. So we can't do this "blindly" with search and replace. Maybe we can do them a file or group at a time. Ie the four Date/Datime ones could work as on. Maybe the sugar ones as another. Everything more internal I'd want to be pretty careful. Dirk | -Kirill | | | | M inst/include/Rcpp/Dimension.h | M inst/include/Rcpp/api/meat/module/Module.h | M inst/include/Rcpp/date_datetime/Date.h | M inst/include/Rcpp/date_datetime/Datetime.h | M inst/include/Rcpp/date_datetime/newDateVector.h | M inst/include/Rcpp/date_datetime/newDatetimeVector.h | M inst/include/Rcpp/module/Module.h | M inst/include/Rcpp/sugar/functions/rowSums.h | M inst/include/Rcpp/sugar/functions/sample.h | M inst/include/Rcpp/sugar/functions/strings/trimws.h | M inst/include/Rcpp/vector/no_init.h | | _______________________________________________ | 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 -- http://dirk.eddelbuettel.com | @eddelbuettel | e...@debian.org _______________________________________________ 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
