Hi Ramon,
We try to not be adventurous with what we put on the build machines so
we'd like to stick to whatever Rtools version is officially the current
stable one.
Also we don't have a mechanism for switching to a different package
code base or branch on some platforms. This wouldn't be a good idea
anyway because then Windows users trying to install the source package
would get something different than what they get when installing the
binary. This would maybe be acceptable if the source and binaries had
different versions but I can't think of any versioning scheme that would
both allow you to handle this and play nice with Bioconductor versioning
scheme.
So I'm afraid the 2 only solutions for now are:
1) Let the package fail on the Windows builder (we can mark it
as unsupported on Windows).
2) Use conditionals in your code (e.g. "if (platform is Windows)
then do this else do that") to make the new functionality available
on non Windows platforms. With this solution, Windows users
will be able to use everything in the package except this new
functionality.
HTH,
H.
On 6/4/19 04:13, Ramon Diaz-Uriarte wrote:
>
> Dear All,
>
> (Follow up to
> https://urldefense.proofpoint.com/v2/url?u=https-3A__stat.ethz.ch_pipermail_bioc-2Ddevel_2019-2DMay_015151.html&d=DwIFaQ&c=eRAMFD45gAfqt84VtBcfhQ&r=BK7q3XeAvimeWdGbWY_wJYbW0WYiZvSXAJJKaaPhzWA&m=_6xTfTTU7Nf_ytIv4RV3f0dN5rUUFrfwJu3b0_3lqsY&s=PtKX7VQrfC7KQbl6O-yXaEKys9IGQuUBSMVKtJ1EnO0&e=
> )
>
> The newest version of a package we develop, OncoSimulR[1] cannot be built
> with Rtools35, the current toolchain[2], because of a known problem where
> compilation and linking fail[3].
>
>
> However, we can build the package with Rtools40, and that produces a
> binary zip file that works with current R-3.6.0, R-3.6.0 patched, and
> R-devel.
>
>
> Is there any way to make the new functionality available through BioC in
> all the platforms? (I understand the answer is probably "no", since testing
> under Windows would not be possible with the current Rtools35 toolchain).
>
> Alternatively, is it possible to have some platforms use one code base or
> branch in a repo, and another platform (Windows) use a different code
> base/branch (i.e., use a code base without the new functionality so that it
> can build and check OK with Rtools 35)?
>
> Of course, any other suggestions welcome.
>
>
>
> Thanks,
>
> R.
>
>
> [1]
> https://urldefense.proofpoint.com/v2/url?u=https-3A__bioconductor.org_packages_release_bioc_html_OncoSimulR.html&d=DwIFaQ&c=eRAMFD45gAfqt84VtBcfhQ&r=BK7q3XeAvimeWdGbWY_wJYbW0WYiZvSXAJJKaaPhzWA&m=_6xTfTTU7Nf_ytIv4RV3f0dN5rUUFrfwJu3b0_3lqsY&s=MJq-nWf5OPsjWdfA_AO_5FQlFgHJDfSHzOpL3ldX4js&e=
> https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_rdiaz02_OncoSimul_tree_freq-2Ddep-2Dfitness&d=DwIFaQ&c=eRAMFD45gAfqt84VtBcfhQ&r=BK7q3XeAvimeWdGbWY_wJYbW0WYiZvSXAJJKaaPhzWA&m=_6xTfTTU7Nf_ytIv4RV3f0dN5rUUFrfwJu3b0_3lqsY&s=uWPj2OmSHXduU9GTa2HuDAfSjT77Q643cHo7-dgLSw0&e=
>
>
> [2]
> https://urldefense.proofpoint.com/v2/url?u=https-3A__cran.r-2Dproject.org_bin_windows_Rtools_&d=DwIFaQ&c=eRAMFD45gAfqt84VtBcfhQ&r=BK7q3XeAvimeWdGbWY_wJYbW0WYiZvSXAJJKaaPhzWA&m=_6xTfTTU7Nf_ytIv4RV3f0dN5rUUFrfwJu3b0_3lqsY&s=AcfE1hiD6aC6pTBm2dLCoPLanwoDo0DabfZagD3k4jk&e=
>
> [3] The "too many sections" error or the linker never finishing. E.g.,
> https://urldefense.proofpoint.com/v2/url?u=https-3A__sourceforge.net_p_mingw-2Dw64_discussion_723797_thread_c6b70624_&d=DwIFaQ&c=eRAMFD45gAfqt84VtBcfhQ&r=BK7q3XeAvimeWdGbWY_wJYbW0WYiZvSXAJJKaaPhzWA&m=_6xTfTTU7Nf_ytIv4RV3f0dN5rUUFrfwJu3b0_3lqsY&s=VyV58euSzIx7Bw678kuf3VQ-StYVUZT1r-gYuUA0qiU&e=
>
>
> --
> Ramon Diaz-Uriarte
> Department of Biochemistry, Lab B-25
> Facultad de Medicina
> Universidad Autónoma de Madrid
> Arzobispo Morcillo, 4
> 28029 Madrid
> Spain
>
> Phone: +34-91-497-2412
>
> Email: rdia...@gmail.com
> ramon.d...@iib.uam.es
>
> https://urldefense.proofpoint.com/v2/url?u=http-3A__ligarto.org_rdiaz&d=DwIFaQ&c=eRAMFD45gAfqt84VtBcfhQ&r=BK7q3XeAvimeWdGbWY_wJYbW0WYiZvSXAJJKaaPhzWA&m=_6xTfTTU7Nf_ytIv4RV3f0dN5rUUFrfwJu3b0_3lqsY&s=r6FEs_1DV66f_CPzlq__0vNjOtoYXaTZCdjSATihXmY&e=
>
> _______________________________________________
> Bioc-devel@r-project.org mailing list
> https://urldefense.proofpoint.com/v2/url?u=https-3A__stat.ethz.ch_mailman_listinfo_bioc-2Ddevel&d=DwIFaQ&c=eRAMFD45gAfqt84VtBcfhQ&r=BK7q3XeAvimeWdGbWY_wJYbW0WYiZvSXAJJKaaPhzWA&m=_6xTfTTU7Nf_ytIv4RV3f0dN5rUUFrfwJu3b0_3lqsY&s=5oVrYps_p71fghBetRc49b8su8Q5Ufdze9XyUlo8ZKI&e=
>
--
Hervé Pagès
Program in Computational Biology
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N, M1-B514
P.O. Box 19024
Seattle, WA 98109-1024
E-mail: hpa...@fredhutch.org
Phone: (206) 667-5791
Fax: (206) 667-1319
_______________________________________________
Bioc-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel
