[R-pkg-devel] compression of vignettes

[Uwe Ligges]

On 25.11.2019 00:45, Spencer Graves wrote:
>        I encountered this or a similar problem with 
> "https://github.com/sbgraves237/Ecfun/blob/master/vignettes/time2nextNuclearWeaponState.Rmd". 
> I tried both "R CMD build Ecfun" and R CMD build Ecfun 
> --compact-vignettes".  With both, "R CMD check Ecfun_0.2-2.tar.gz" 
> included the following:
> 
> 
> * checking installed package size ... NOTE
>    installed size is  7.2Mb
>    sub-directories of 1Mb or more:
>      doc   6.6Mb


It is nice to have packages that are as small as possible, but we would 
not object to publish a package with the above size of (uncompressed) 
directories.

Best,
Uwe Ligges


> 
> 
>        Suggestions?
>        Thanks,
>        Spencer Graves
> 
> 
> On 2019-11-21 11:43, Dirk Eddelbuettel wrote:
>> On 21 November 2019 at 16:43, Joris Meys wrote:
>> | The workflow as described in the manual :
>> | 
>> https://cran.r-project.org/doc/manuals/R-exts.html#Checking-and-building-packages 
>>
>> |
>> | is as follows :
>> |
>> | R CMD build pkg --compact-vignettes
>> | R CMD check pkg_x.y.z.tar.gz --as-cran
>> |
>> | In a basic Travis script this can be used as:
>> |
>> | script:
>> | - R CMD build . --compact-vignettes=gs+qpdf
>> | - R CMD check *tar.gz --as-cran
>> |
>> | You find more information here : 
>> https://towardsdatascience.com/travis-ci-for-r-advanced-guide-719cb2d9e0e5 
>>
>>
>> You can also call the base R helper functions directly.
>>
>> I like having simple command-line wrapper, so I added one to `littler` 
>> which
>> I can call as `compactpdf.r`. All it does (besides argument checking) is
>> loop over either the given pdf files (or those found in the current
>> directory) to then call
>>
>> tools::compactPDF(f, gs_quality="ebook")
>>
>> for each of them. That way you can simply compare 'before' and 'after'.
>>
>> Similarly, I made compacting an option to the rendering script 
>> `render.r` I
>> often call directly from my editor when working on markdown files.
>>
>> Dirk
>>
> 
> ______________________________________________
> R-package-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-package-devel