Thanks, adding remotes to DESCRIPTION did what I was after.
For reference (including me), here is an outline of a possible workflow:
1. run revdep_check(), with many reverse dependencies something like
revdep_check(num_workers = 4)
speeds the things up dramatically.
2, if there are failures run revdep_add_broken()
3. bug-fix
4. run revdep_check() or revdep_check(num_workers = 4),
Due to 2. Above, this will run only the checks that failed the previous
time. Of course this assumes that the bug-fixes didn't break things that work.
5. repeat 2. - 4.
Georgi Boshnakov
-----Original Message-----
From: Gábor Csárdi <csardi.ga...@gmail.com>
Sent: 28 September 2020 18:46
To: Georgi Boshnakov <georgi.boshna...@manchester.ac.uk>
Cc: r-package-devel@r-project.org
Subject: Re: [R-pkg-devel] force revdepcheck to use local version of a
dependency
You can define the special dependency of the revdep-checked package in the
`Remotes` field. More info:
https://cran.rstudio.com/web/packages/remotes/vignettes/dependencies.html
E.g. if you are revdep-checking package PKG, then you can add something like
this in the DESCRIPTION file of PKG:
Remotes: local::/path/to/your/package
or if the package is on GitHub:
Remotes: user/repo
Then revdepcheck will automatically install the special version of the
dependency, for the dev version of the revdep-checked package. (But not for the
CRAN version of the package.)
Best,
Gabor
On Mon, Sep 28, 2020 at 6:17 PM Georgi Boshnakov
<georgi.boshna...@manchester.ac.uk> wrote:
>
> Hi,
>
> I am using package 'revdepcheck' to check reverse dependencies of a package
> and it works great. But now I have a scenario such that I wish it to use a
> local version of one of the dependencies of the checked package and am not
> able to find out how.
>
> Some context: Package A imports package B. Package A has many reverse
> dependencies. I wish to check if an update to package B will break some of
> the reverse dependencies of A and/or repair broken reverse dependencies of A.
> It is obvious to update B on CRAN but this may lead unnecessarily to a number
> of CRAN updates.
>
>
> Georgi Boshnakov
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-package-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-package-devel
______________________________________________
R-package-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-package-devel
