Dear all, We'd like to submit an update of the "swmmr" package to CRAN, which interfaces the rainfall runoff model SWMM from the US environmental protection agency. The package also includes three vignettes which demonstrate how to run the external model from the R command line. Within the vignettes, the locally installed SWMM model is executed. Of course, the model is not available on CRAN. My question therefore is: How can we pass the incoming checks although processing the vignettes would fail? Thanks in advance! Best, Dominik Dominik Leutnant M.Sc. Muenster University of Applied Sciences Department of Civil Engineering Institute for Infrastucture?Water?Resources?Environment (IWARU) WG Urban Hydrology and Water Management Corrensstr. 25 FRG-48149 M?nster, Germany Tel.: +49 (0) 251/83-65274 Fax: +49 (0) 251/83-65915 Mail: leutnant at fh-muenster.de<mailto:leutnant at fh-muenster.de> Web: https://www.fh-muenster.de/
[R-pkg-devel] vignette with third party software
3 messages · Dominik Leutnant, Iñaki Ucar
2018-02-26 15:03 GMT+01:00 Dominik Leutnant <leutnant at fh-muenster.de>:
Dear all, We'd like to submit an update of the "swmmr" package to CRAN, which interfaces the rainfall runoff model SWMM from the US environmental protection agency. The package also includes three vignettes which demonstrate how to run the external model from the R command line. Within the vignettes, the locally installed SWMM model is executed. Of course, the model is not available on CRAN. My question therefore is: How can we pass the incoming checks although processing the vignettes would fail?
If the size of the Example1.inp file is reasonable, you may include it
under the inst directory and retrieve it as follows in your vignettes:
inp_file <- system.file("Example1.inp", package="swmmr")
Otherwise, you may set knitr::opts_chunk$set(eval=FALSE) in the first
chunk for each vignette, so that your code is not executed and thus it
won't fail.
Thanks in advance! Best, Dominik Dominik Leutnant M.Sc.
I?aki
2018-02-26 16:16 GMT+01:00 I?aki ?car <i.ucar86 at gmail.com>:
2018-02-26 15:03 GMT+01:00 Dominik Leutnant <leutnant at fh-muenster.de>:
Dear all, We'd like to submit an update of the "swmmr" package to CRAN, which interfaces the rainfall runoff model SWMM from the US environmental protection agency. The package also includes three vignettes which demonstrate how to run the external model from the R command line. Within the vignettes, the locally installed SWMM model is executed. Of course, the model is not available on CRAN. My question therefore is: How can we pass the incoming checks although processing the vignettes would fail?
If the size of the Example1.inp file is reasonable, you may include it
under the inst directory and retrieve it as follows in your vignettes:
inp_file <- system.file("Example1.inp", package="swmmr")
Let me add a couple more comments about this. In case you can include the data file, this has the benefit that you can use it in your tests too (I've seen that you simply skip_on_cran and skip_on_travis). Also, I've seen that you write files in one vignette. Note that you should use the tempdir() for that.
Otherwise, you may set knitr::opts_chunk$set(eval=FALSE) in the first chunk for each vignette, so that your code is not executed and thus it won't fail.
Thanks in advance! Best, Dominik Dominik Leutnant M.Sc.
I?aki
I?aki