[Rd] Windows Source Install Without Rtools

[Gabor Grothendieck]

Does that mean that

   Rcmd INSTALL --binary myPackage/

is still ok or should one use:

   Rcmd INSTALL myPackage/

without --binary and is this advice the same with and without Rtools?

Also, as there is potential confusion here could the combinations of flags
that are not recommended result in a warning? In the latter case at least
one knows right off that its not the thing to do and can immediately abort
and try something else if they don't want to suffer the consequences of
bad help linking.

On Thu, Sep 17, 2009 at 5:57 PM, Duncan Murdoch <murdoch at stats.uwo.ca> wrote:
> On 17/09/2009 5:55 PM, Duncan Murdoch wrote:
>>
>> On 17/09/2009 4:40 PM, Gabor Grothendieck wrote:
>>>
>>> Maybe on other platforms but on Windows
>>> help does not get linked properly with all
>>> variations.
>>
>> I think you are thinking of building binaries, not installing packages.
>>  Generally the only variation we advise against is
>>
>> R CMD build --binary
>>
>> Everything else should work fine, as far as I know.
>
> And I forgot to add:  the way HTML help is handled in the upcoming 2.10.0 is
> different, so I suspect R CMD build --binary will work.  I haven't tested it
> though...
>
> Duncan Murdoch
>
>>
>> Duncan Murdoch
>>
>>> On Thu, Sep 17, 2009 at 4:27 PM, Simon Urbanek
>>> <simon.urbanek at r-project.org> wrote:
>>>>
>>>> On Sep 17, 2009, at 16:09 , Gabor Grothendieck wrote:
>>>>
>>>>> Note that that is not currently the recommended way.
>>>>>
>>>> Now that would be a surprise to me - to quote R-admin: 6.3.1: "R CMD
>>>> INSTALL
>>>> works in Windows to install source packages if you have the source-code
>>>> package files".
>>>>
>>>>
>>>>> Also, what comes after INSTALL? The source .tar.gz name?
>>>>> The pathname to the source directory?
>>>>>
>>>> R CMD INSTALL --help
>>>>
>>>> Usage: R CMD INSTALL [options] pkgs
>>>>
>>>> Install the add-on packages specified by pkgs.  The elements of pkgs can
>>>> be relative or absolute paths to directories with the package (bundle)
>>>> sources, or to gzipped package 'tar' archives.  The library tree
>>>> [...]
>>>>
>>>> Cheers,
>>>> S
>>>>
>>>>
>>>>> On Thu, Sep 17, 2009 at 3:58 PM, Simon Urbanek
>>>>> <simon.urbanek at r-project.org> wrote:
>>>>>>
>>>>>> On Sep 17, 2009, at 12:43 , Gabor Grothendieck wrote:
>>>>>>
>>>>>>> Thanks. Perhaps someone in the core group can
>>>>>>> still provide explicit information on how to install such
>>>>>>> a package.
>>>>>>>
>>>>>> R CMD INSTALL
>>>>>>
>>>>>> (see R --help and it has been *the* way to install packages for quite
>>>>>> a
>>>>>> while so I'm not sure how more explicit we have to make it...)
>>>>>>
>>>>>> Cheers,
>>>>>> Simon
>>>>>>
>>>>>>
>>>>>>
>>>>>>> On Thu, Sep 17, 2009 at 12:19 PM, Dirk Eddelbuettel <edd at debian.org>
>>>>>>> wrote:
>>>>>>>>
>>>>>>>> On 17 September 2009 at 11:32, Gabor Grothendieck wrote:
>>>>>>>> | Regarding this from this week:
>>>>>>>> |
>>>>>>>> | 2.10.0 PACKAGE INSTALLATION (Windows)
>>>>>>>> | It is possible to install source packages without Rtools provided
>>>>>>>> they
>>>>>>>> | have no compiled code and no configure.win or cleanup.win script.
>>>>>>>> |
>>>>>>>> | - what are the instructions to perform such a source install?
>>>>>>>> | - I assume one still needs MiKTeX -- is that right?
>>>>>>>> | - is the source install done from a tar.gz (in which case one
>>>>>>>> would
>>>>>>>> | need tar to create it) or directly from the source directory tree?
>>>>>>>>
>>>>>>>> You may want to stick this
>>>>>>>>
>>>>>>>>   http://developer.r-project.org/blosxom.cgi/R-devel
>>>>>>>>
>>>>>>>> into your RSS reader so that you get daily summaries of changes.
>>>>>>>>
>>>>>>>> I don't follow this too religiously (as I typically only built
>>>>>>>> r-devel
>>>>>>>> once we have
>>>>>>>> alpha/beta/rc candidates) but even I am aware that
>>>>>>>>
>>>>>>>>   a) dynamic .Rd conversion at runtime, so Miktex may no longer be
>>>>>>>> needed
>>>>>>>>
>>>>>>>>   b) untgz.exe on Windows to read (compressed) tarballs
>>>>>>>>
>>>>>>>> Dirk
>>>>>>>>
>>>>>>>> --
>>>>>>>> Three out of two people have difficulties with fractions.
>>>>>>>>
>>>>>>> ______________________________________________
>>>>>>> R-devel at r-project.org mailing list
>>>>>>> https://stat.ethz.ch/mailman/listinfo/r-devel
>>>>>>>
>>>>>>>
>>> ______________________________________________
>>> R-devel at r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-devel
>>
>>
>
>