Closed AngusMcLure closed 3 years ago
Thanks for letting us know about this and sharing the fix. Glad you were able to workaround it. The edit of reordering the first two lines seems fine to me, so I guess we should just go ahead and do that in rstantools. @bgoodri Is that fine by you too?
No problem. Thanks for all your work in building, developing and maintaining these tools.
Yeah, we can put the # Generated by rstantools ... thing on the second line to avoid the need for the checkbashisms script.
On Tue, Dec 1, 2020 at 6:41 PM Jonah Gabry notifications@github.com wrote:
Thanks for letting us know about this and sharing the fix. Glad you were able to workaround it. The edit of reordering the first two lines seems fine to me, so I guess we should just go ahead and do that in rstantools. @bgoodri https://github.com/bgoodri Is that fine by you too?
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/stan-dev/rstantools/issues/82#issuecomment-736891225, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAZ2XKSH3QIN4XQSQB2EQ43SSV5K5ANCNFSM4UJV2ZJA .
Minor issue, with an easy fix.
rstantools version: 2.1.1 R version: 4.0.3 Platform: OS X
When checking a package that I set up using
rstan_create_package
, I got a warning (new to R 4.0?) prompting me to add thecheckbashisms
script to my $PATH. I did this, resolving the warning, but producing a new NOTE:I eliminated the NOTE (both on my local OS X build and on win-builder) by making small changes to
configure
andconigure.win
automatically generated byrstan_create_package
. The defaultconfigure
is:Noting, but disobeying the imperative not to edit, I reordered the first two lines so that the
#!
lines is at the top. i.e.:I did the same for
configure.win
The package now passes
devtools::check()
anddevtools::check_win_release()