[Bioc-devel] Help finding helpful diagnostic info in the build results report

Mike Smith gr|mbough @end|ng |rom gm@||@com
Sat Sep 26 11:09:43 CEST 2020


Hi Lillian,

For the most part skipping happens because there is a hierarchy to the
package testing.  If you can't BUILD a package, then you automatically
can't CHECK or build a binary version (BUILD BIN), so the build system will
simply skip trying those steps if it fails at the BUILD stage.  (The
exception in the BUILD BIN step on Linux, which is skipped because for the
most part it doesn't make sense to build binary versions Linux).

So you don't need to worry about the skipped steps for now, but then the
question is "why is your package failing to BUILD on the Windows
machine?".  My gut instinct is that it's down to the use of left- and
right-quotation marks (“/”) in your Title field, rather than just the
regular quotation mark (").

I don't know why it would be OK on Linux, but not Windows, but character
encoding is fiddly and different across operating systems. Also the
get_dcf_info error actually looks like it comes from the python code that
runs the build system - so it's pretty hard for you to debug directly, and
might behave differently from similarly named functions in R.  I'd try
changing out the quotes and then giving it another try.

Best,
Mike

On Fri, 25 Sep 2020 at 23:17, Thistlethwaite, Lillian Rosa <
Lillian.Thistlethwaite using bcm.edu> wrote:

> The newest build report for the checking of a package I submitted to
> Bioconductor skipped both the Mac and Windows builds. The Mac build was
> skipped because of a known issue with another R package dependency, gmp,
> and this can be ignored.
>
> However, the Windows build being skipped is a mystery to me. I did see the
> "get_dcf_info failed" message in the build tokay1 (Windows) Summary, but
> the read.dcf() function call from devtools and the system build works just
> fine on a Windows machine I tested the package on. There isn't much else
> diagnostic information as to why tokay1 is skipping the checks which I can
> use besides this. How are these undiagnostic SKIPS typically debugged?
>
> For reference, please view the most recent build results report here:
> http://bioconductor.org/spb_reports/CTD_buildreport_20200917232759.html
>
>
>
>
> Warmest wishes,
>
> Lillian Thistlethwaite
>
>
> Baylor College of Medicine
>
> Quantitative & Computational Biosciences Program
>
> Houston, TX
>
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> Bioc-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>

	[[alternative HTML version deleted]]



More information about the Bioc-devel mailing list