[Rd] vignette compilation times

[Friedrich Leisch]

>>>>> On Thu, 19 Feb 2009 17:47:53 -0500,
>>>>> Gabor Grothendieck (GG) wrote:

  > On Thu, Feb 19, 2009 at 5:36 PM, Friedrich Leisch
  > <Friedrich.Leisch at stat.uni-muenchen.de> wrote:
  >>>>>>> On Thu, 19 Feb 2009 11:47:12 +0000,
  >>>>>>> Robin Hankin (RH) wrote:
  >> 
  >> > thanks for this clarification Uwe
  >> > Could I include the r_env_cache/  directory in the package
  >> > and then assume that the CRAN checks use
  >> 
  >> > Sweave(.... , driver=weaver())
  >> 
  >> > in which case the process takes about 10 seconds?
  >> 
  >> That makes no sense, because then there are no checks done at all: if
  >> the code in your vignette does not change, weaver will not recompute
  >> anything, hence the cached results are used. But in that case you
  >> could as well include only the PDF (or the generated .tex if you like
  >> that better) ...
  >> 

  > Unless this has changed recently,I've tried including a PDF but it does not
  > appear in library(help = myPackage) nor on the CRAN site on
  > http://cran.r-project.org/package=myPackage
  > while Sweave'd PDFs do.

That was not the point of my email: If we process vignettes using the
weaver package during check, then in fact no checking is done at all
-> the whole purpose of vignettes is circumvented.

For vignettes you can be sure that all computations have been done
using the latest version of R and the package, i.e., the doc is up to
date. For a PDF this is not necessarily the case. There is a good
reason why we (more or less gently) try to push people writing docs as
Sweave files.  It is also unclear to me whether including a PDF
without sources in a GPLed package isn't a violation of the GPL (I
know people who very strongly think so). And source according to the
GPL means "the preferred form of the work for making modifications to
it." So for a PDF showing R output that would mean the text plus R
code plus data ... which boils down to XXXweave anyway.

But we really had this discussion several times ... perhaps somebody
volunteers to write an RFC for a more general system of including
manuals in R and volunteers to send patches implementing the mechanism
afterwards? Note that such a system needs to be compatible with the
package metadata system and work on all platforms.

Best,
Fritz