On 5 March 2021 at 15:41, Duncan Murdoch wrote: | On 05/03/2021 2:40 p.m., Henrik Bengtsson wrote: | > Thank you. Glad to hear it's useful. | > | > This plain TeX/LaTeX vignette engine is implemented using base R. If | > someone is willing to drive the efforts, I think it's not too much | > work to refactor it and propose it for base R itself, where I think it | > belongs, | | What is your reasoning for this? As a former R Core member, I would say | the reasons for something to be added to base R are: | | - it's generally useful | - it doesn't have a lot of dependencies | - it doesn't work as well in a contributed package | | The first two conditions seem to be met, but what's wrong with leaving | it in R.rsp? Would something in base R be simplified by having this there?
It feels like this is about the third time in a decade+ I have seen you two discuss the same issue. Henrik advocates a fine technical solution from his r.rsp package, you point out that there is benefit to keeping base R smaller, and solutions in add-on packages. All the while we all carry on and just slap the four or five lines onto the file make it an Sweave vignette _without any executed code_ which works in base R today as a proxy for pure LaTeX, just as it has for 25 or so years. Or do as some of us do and follow Mark's trick [1] of including a pre-made pdf inside an even more minimal Rnw file for the Sweave driver. Works great with any pdf: pure LaTeX, LaTeX from (R)Markdown, ... Dirk [1] https://www.markvanderloo.eu/yaRb/2019/01/11/add-a-static-pdf-vignette-to-an-r-package/ -- https://dirk.eddelbuettel.com | @eddelbuettel | e...@debian.org ______________________________________________ R-package-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-package-devel