[R-pkg-devel] PDF Vignette

2019-12-06 Thread Charith Karunarathna
Dear All, I am trying to add a PDF file (perfectphyloR.pdf) for vignette of my R package (perfectphyloR). When I check my package using win-builder, I am getting the following warning message: * checking re-building of vignette outputs ... [2s] WARNING Error in re-building vignettes: ... Wa

Re: [R-pkg-devel] Error while checking if Pandoc is available

2019-12-06 Thread Henrik Bengtsson
Hugh is correct. To clarify, update if (!rmarkdown::pandoc_available() || rmarkdown::pandoc_version() < "1.12.3") { stop("'brinton' requires Pandoc v < 1.12.3 (https://pandoc.org/)") } The reason is that when using 'a || b', expression 'b' will NOT be evaluated if 'a' is not TRUE. In contrast,

Re: [R-pkg-devel] Including in my vignette a figure drawn outside R

2019-12-06 Thread Michael Dewey
Thank you Dirk for the last point which has helped my grasp the implications of something in WRE which will help me go forward. Michael On 03/12/2019 16:17, Dirk Eddelbuettel wrote: On 3 December 2019 at 15:36, Michael Dewey wrote: | I am planning to include a flow-chart in the vignette of on

Re: [R-pkg-devel] Error while checking if Pandoc is available

2019-12-06 Thread Hugh Parsonage
Change | to || On Fri, 6 Dec 2019 at 11:08 pm, Pere Millan Martinez wrote: > Yesterday the publication of the new brinton package war archived > (https://cran.r-project.org/web/packages/brinton/index.html) because the > following line produces the following error: > > if(rmarkdown::pandoc_availa

[R-pkg-devel] Error while checking if Pandoc is available

2019-12-06 Thread Pere Millan Martinez
Yesterday the publication of the new brinton package war archived (https://cran.r-project.org/web/packages/brinton/index.html) because the following line produces the following error: if(rmarkdown::pandoc_available() == FALSE | rmarkdown::pandoc_version() < "1.12.3") {stop("'brinton' requires P

Re: [R-pkg-devel] CRAN submission with "possibly unsafe calls" or alternative approach?

2019-12-06 Thread Rainer M Krug
Thanks for the clarification - I was not aware of that. I will put the variables in an environment and call them via a function before submission to CRAN.. Thanks, Rainer > On 6 Dec 2019, at 12:03, peter dalgaard wrote: > > I think you may want to rethink the mechanism. > > Locked binding

Re: [R-pkg-devel] CRAN submission with "possibly unsafe calls" or alternative approach?

2019-12-06 Thread peter dalgaard
I think you may want to rethink the mechanism. Locked bindings are generally there to allow the compiler to make assumptions about the type, etc., of objects (or rather: not make assumptions because it will know what the type is). Unlocking invalidates this and may trigger recompilation or int

Re: [R-pkg-devel] Rcpp specific Linux warning related to sprint function

2019-12-06 Thread Duncan Murdoch
On 06/12/2019 5:48 a.m., Sebastian SOSA wrote: Dear all, I am trying to submit my R package on CRAN however I have a specific Linux warning message related to a Rcpp function I found online and included in my package: https://gallery.rcpp.org/articles/faster-data-frame-creation/ More specific

[R-pkg-devel] Rcpp specific Linux warning related to sprint function

2019-12-06 Thread Sebastian SOSA
Dear all, I am trying to submit my R package on CRAN however I have a specific Linux warning message related to a Rcpp function I found online and included in my package: https://gallery.rcpp.org/articles/faster-data-frame-creation/ More specifically, the issue is related to sprint (lines 7 and

Re: [R-pkg-devel] Package write access for storing new files

2019-12-06 Thread Dr Gregory Jefferis
Possibly you can use the rappdirs location *if it exists* and tempdir() otherwise. You could make the rappdirs location when running the shiny app. Since the user has to do something to run the app, this is probably ok. I think the problem is that loading the package itself cannot result in a d

Re: [R-pkg-devel] Package write access for storing new files

2019-12-06 Thread Rainer M Krug
Thanks for the tip with rappdirs. I am just not sure, how I should handle the questions, as it should also run in a shiny application? > On 6 Dec 2019, at 10:12, Gregory Jefferis wrote: > > I use rappdirs to construct a platform appropriate consistent package > download directory. But for C

Re: [R-pkg-devel] Package write access for storing new files

2019-12-06 Thread Gregory Jefferis
I use rappdirs to construct a platform appropriate consistent package download directory. But for CRAN I believe you will need to obtain some kind of user confirmation at least once to allow this behaviour. Best, Greg. Sent from my iPhone > On 6 Dec 2019, at 08:43, Rainer M Krug wrote: > > H

[R-pkg-devel] CRAN submission with "possibly unsafe calls" or alternative approach?

2019-12-06 Thread Rainer M Krug
Hi In my package `dmdScheme` I define three variables. Depending on a state in the package (a selected metadata schemes) these are set as followed from within the package: ``` unlockBinding("dmdScheme_example", as.environment("package:dmdScheme")) assign("dmdScheme_example", scheme_example,

[R-pkg-devel] Package write access for storing new files

2019-12-06 Thread Rainer M Krug
Hi I am writing a package which downloads files which become part of the package (only useful for functions in the package and independent of the working directory) into system.file("installedSchemes", package = "dmdScheme") (See https://github.com/Exp-Micro-Ecol-Hub/dmdScheme/blob/b9