[GRASS-stats] Some refactoring and fixing

Thu Jan 29 10:22:48 PST 2015

On Thu, 29 Jan 2015, Rainer M Krug wrote:

> Rainer M Krug <Rainer at krugs.de> writes:
>
>> Roger Bivand <Roger.Bivand at nhh.no> writes:
>>
>>> On Wed, 28 Jan 2015, Rainer M Krug wrote:
>>>
>>>> OK - the problem seems to be a name conflict. The build system status
>>>> report [1] states:
>>>>
>>>> ,----
>>>> | - Package name conflicts:
>>>> | ...
>>>> | [7] "rspatial: spgrass6"     "spgrass: spgrass6"
>>>> `----
>>>
>>> Right, spgrass6 is still hosted as part of rspatial (it was originally
>>> there). I'll try to resolve this by removing it from that project.
>>
>> Thanks Roger - spgrass6 is now available to be installed via
>>
>>
>> install.packages("spgrass6", repos="http://R-Forge.R-project.org")
>>
>>
>> for windows (source and compiled), linux (source) and Mac (source).
>>
>> Is there any possibility to remove the spgrass package (which has been
>> renamed to spgrass7) from the R Packages listing [1] to avoid confusion?
>>
>>> The file oChangeLog is the frozen ChangeLog from its original project,
>>> so when releasing, we need to run:
>>>
>>> cd spgrass6 # or 7
>>> svn up
>>> sh svn2cl.sh # must be on PATH
>>> # if formatted too narrowly
>>> # fmt < ChangeLog > tmp
>>> # mv tmp ChangeLog
>>> cat oChangeLog >> ChangeLog
>>> cp ChangeLog inst
>>> svn commit -m tidy
>>> svn up
>>> cd ..
>>>
>>> to build a ChangeLog including all past history.
>>
>> Concerning changes before release:
>>
>> 1) I would suggest to create for all functions which have a 6 at the end
>> (e.g. redRAST6) an alias without the 6 (e.g. readRAST). By using these
>> names without the 6 the switch between GRASS (and related spgrassX)
>> versions will require mainly a switch of the library used. For further
>> versions of spgrassX I would suggest to drop the numbers completely as,
>> even when spgrass6 and spgrass7 are loaded at the same time, specific
>> versions can be accessed by using the :: (i.e. spgrass6::readRAST).
>
> Committed (revision 18) - please check!

I've checked both spgrass6 (in GRASS 6.4.4) and spgrass7 (in GRASS 7.0) in 
today's R-devel as required. I've fixed what needed fixing to pass 
cleanly. When we're ready, I can submit them to CRAN.

Roger

>
>>
>> 2) I will go through the code and remove references (and version checks)
>> to GRASS GIS 7 to simplify the code. These checks are not needed anymore as
>> the check for the GRASS GIS version is done upon loading (only loads
>> when started outside grass session or inside grass 6.x session) and on
>> initGRASS() (aborts if gisBase points to grass 6.x installation)
>
> Committed (revision 17) - please check.
>
> The new spgrass6 concludes all examples without error messages and
> ,----
> | R CMD check --as-cran
> `----
> also without errors.
>
> Thomas - could you please do some testing on this version
> (revision 18) - the packages are building at the moment?
>
> Cheers,
>
> Rainer
>>
>> Otherwise I can't thinkl of any other points before release.
>>
>> Thanks,
>>
>> Rainer
>>
>>>
>>> Roger
>>>
>>>>
>>>> So the easiest to install at the moment seems to be to checkout the
>>>> sources via
>>>>
>>>> ,----
>>>> | svn checkout svn://r-forge.r-project.org/svnroot/spgrass/
>>>> `----
>>>>
>>>> and then install from the sources.
>>>>
>>>> Cheers,
>>>>
>>>> Rainer
>>>>
>>>> Rainer M Krug <Rainer at krugs.de> writes:
>>>>
>>>>> Hi Roger
>>>>>
>>>>> it seems that there is something not as it should be with the builds of
>>>>> spgrass6 and spgrass7 - there are at the moment GRASS, spgrass and
>>>>> spgrass7 but no spgrass6 - I don't have much experience with r-forge,
>>>>> could you please check what is going on?
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Rainer
>>>>>
>>>>> Rainer M Krug <r.m.krug at gmail.com> writes:
>>>>>
>>>>>> Envoyé de mon iPhone
>>>>>>
>>>>>>> Le 28 janv. 2015 à 15:55, Thomas Adams <tea3rd at gmail.com> a écrit :
>>>>>>>
>>>>>>> Rainer,
>>>>>>>
>>>>>>> Where can I get the new spgrass6 and spgrass7 to do some testing?
>>>>>>
>>>>>> The repository is on https://r-forge.r-project.org/projects/spgrass/   .
>>>>>>
>>>>>> It doesn't seem to be build yet so you have to install it from the svn
>>>>>> repo or wait a few hours - then you shoud be able to install it via
>>>>>>
>>>>>>  	R install command: install.packages("spgrass6", repos="http://R-Forge.R-project.org")
>>>>>>
>>>>>>  	Cheers,
>>>>>>
>>>>>> Rainer
>>>>>>> Thank you,
>>>>>>> Tom
>>>>>>>
>>>>>>>> On Wed, Jan 28, 2015 at 7:23 AM, Rainer M Krug <Rainer at krugs.de> wrote:
>>>>>>>> Rainer M Krug <Rainer at krugs.de> writes:
>>>>>>>>
>>>>>>>>> Roger Bivand <Roger.Bivand-aX1nC9UfZf8 at public.gmane.org> writes:
>>>>>>>>>
>>>>>>>>>> On Tue, 27 Jan 2015, Rainer M Krug wrote:
>>>>>>>>>>
>>>>>>>>>>> Hi
>>>>>>>>>>>
>>>>>>>>>>> I just committed a few of commits which refactor the code of readRAST,
>>>>>>>>>>> writeRAST, readVECT & writeVECT by creating internal fi=unctions for
>>>>>>>>>>> reading / writing for plugin / non-plugin and moving the default values
>>>>>>>>>>> (get...Options()) into the function definitions.
>>>>>>>>>>
>>>>>>>>>> Good, thanks; I've updated the default values in the help files. It's
>>>>>>>>>> easy to check for discrepancies in spgrass/pkg:
>>>>>>>>>>
>>>>>>>>>> R CMD build spgrass7
>>>>>>>>>> R CMD check spgrass7_0.1-0.tar.gz
>>>>>>>>>
>>>>>>>>> Thanks - haven't thought about about the check - will do the next time.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Maybe we could quieten the progress bar in the output from checking?
>>>>>>>>>> As in:
>>>>>>>>>>
>>>>>>>>>> pkg/spgrass7.Rcheck/spgrass7.Rout
>>>>>>>>>>
>>>>>>>>>> (local file, not added to repository, created by R CMD check if run in
>>>>>>>>>> spgrass/pkg).
>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> Also, I added tryCatch blocks to close open connections and to reset
>>>>>>>>>>> echoCmdOption. I ran the examples before each commit and I did not see
>>>>>>>>>>> any errors or changed behavior.
>>>>>>>>>>>
>>>>>>>>>>> Apologies for the typos in the commit messages - I only saw them later.
>>>>>>>>>>>
>>>>>>>>>>> To stick with consistency, I would recommend to rename the package to
>>>>>>>>>>> spgrass7 and to look into the possibility to create a meta-package which
>>>>>>>>>>> uses spgrass6 or spgrass7 (I don't know about the GRASS 5 interface -
>>>>>>>>>>> haven't used it) depending on the GRASS GIS version set / loaded.
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> I've renamed spgrass as spgrass7 - when you update, the change will
>>>>>>>>>> propagate.
>>>>>>>>>
>>>>>>>>> Thanks - got it.
>>>>>>>>>
>>>>>>>>>> I'm unsure about a metapackage, maybe block version 7 in
>>>>>>>>>> spgrass6 (following the parameter and flag name harmonisation).
>>>>>>>>>
>>>>>>>>> Yes - spgrass6 for 6.x, spgrass7 for 7.x.
>>>>>>>>>
>>>>>>>>> Will respnd in more detail tomorrow.
>>>>>>>>
>>>>>>>> OK - I just committed a blocking mechanism which for spgrass6 and
>>>>>>>> spgrass7 so that they
>>>>>>>>
>>>>>>>> a) when R is started in GRASS, the package can not be loaded if the
>>>>>>>> versions are not compatible and
>>>>>>>> b) initGRASS raises an error if the GRASS GIS version in gisBase is not
>>>>>>>> compatible with the package version
>>>>>>>>
>>>>>>>> If you think this is the right approach, the help files still need to be
>>>>>>>> amended to reflect this change.
>>>>>>>>
>>>>>>>> I was thinking if it would make sense to introduce an overwrite
>>>>>>>> mechanism for these checks to enable "non-standard" use or testing?
>>>>>>>>
>>>>>>>> I haven't looked at references for GRASS 7.0 in spgrass6 yet.
>>>>>>>>
>>>>>>>> Cheers,
>>>>>>>>
>>>>>>>> Rainer
>>>>>>>>
>>>>>>>>>
>>>>>>>>> Thanks,
>>>>>>>>>
>>>>>>>>> Rainer
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Best wishes,
>>>>>>>>>>
>>>>>>>>>> Roger
>>>>>>>>>>
>>>>>>>>>>> Cheers,
>>>>>>>>>>>
>>>>>>>>>>> Rainer
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>
>>>>>>>> --
>>>>>>>> Rainer M. Krug, PhD (Conservation Ecology, SUN), MSc (Conservation Biology, UCT), Dipl. Phys. (Germany)
>>>>>>>>
>>>>>>>> Centre of Excellence for Invasion Biology
>>>>>>>> Stellenbosch University
>>>>>>>> South Africa
>>>>>>>>
>>>>>>>> Tel :       +33 - (0)9 53 10 27 44
>>>>>>>> Cell:       +33 - (0)6 85 62 59 98
>>>>>>>> Fax :       +33 - (0)9 58 10 27 44
>>>>>>>>
>>>>>>>> Fax (D):    +49 - (0)3 21 21 25 22 44
>>>>>>>>
>>>>>>>> email:      Rainer at krugs.de
>>>>>>>>
>>>>>>>> Skype:      RMkrug
>>>>>>>>
>>>>>>>> PGP: 0x0F52F982
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> grass-stats mailing list
>>>>>>>> grass-stats at lists.osgeo.org
>>>>>>>> http://lists.osgeo.org/mailman/listinfo/grass-stats
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>
>>>>
>>>> Footnotes:
>>>> [1]  http://download.r-forge.r-project.org/STATUS
>>>>
>>>>
>>
>>
>> Footnotes:
>> [1]  https://r-forge.r-project.org/R/?group_id=2020
>
>

-- 
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; fax +47 55 95 91 00
e-mail: Roger.Bivand at nhh.no