[R-SIG-Mac] CRAN OSX rgdal binaries 1.3-3 and 1.3-4

[Roger Bivand]

On Sun, 5 Aug 2018, David Winsemius wrote:

>
>> On Aug 5, 2018, at 6:28 AM, Roger Bivand <Roger.bivand using nhh.no> wrote:
>>
>> The CRAN OSX rgdal binaries 1.3-3 and 1.3-4 do not include PROJ datum metadata files. To check whether your OSX binary rgdal has the PROJ datum metadata files, try:
>>
>> library(rgdal)
>> projNAD()
>>
>> If they are not present, consider running this snippet:
>>
>> pkg <- "rgdal"
>> exdir <- file.path(system.file(package=pkg), "proj")
>> list.files(exdir)
>> tf <- tempfile()
>> fl <- "http://download.osgeo.org/proj/proj-datumgrid-1.7.zip"
>> download.file(fl, tf)
>> unzip(tf, exdir=exdir)
>> list.files(exdir)
>>
>> Hope this helps,
>
> I think a lot of us Mac useRs install the KyngChaos super-package (GDAL, GEOS, PROJ, SQLite and UnixImageIO framework described and linked to at:
>
> https://www.kyngchaos.com/software/frameworks
>
> I see that another version was released earlier this year which I had not installed, so I only had GDAL 2.1. I got this when loading `rgdal` version 1.3-1 (and I see that I'm a couple of minor versions behind. so I'll update) I did wonder that the message about linking to `sp` ver 1.2-7 didn't match my current version of `sp` which is 1.3-1
>
> library(rgdal)
> Loading required package: sp
> rgdal: version: 1.3-1, (SVN revision 747)
> Geospatial Data Abstraction Library extensions to R successfully loaded
> Loaded GDAL runtime: GDAL 2.1.3, released 2017/20/01
> Path to GDAL shared files: /Library/Frameworks/R.framework/Versions/3.5/Resources/library/rgdal/gdal
> GDAL binary built with GEOS: FALSE
> Loaded PROJ.4 runtime: Rel. 4.9.3, 15 August 2016, [PJ_VERSION: 493]
> Path to PROJ.4 shared files: /Library/Frameworks/R.framework/Versions/3.5/Resources/library/rgdal/proj
> Linking to sp version: 1.2-7
>
> # When I updated rgdal the startup message from rgdal now says:
>
> Linking to sp version: 1.3-1
>
> # ... despite the fact that I installed 1.3-4

Thanks for reporting back. The line you refer to states the version of sp 
that rgdal links to, and sp_1.3-1 is the current CRAN version. Sometimes 
we forget to touch the header file in sp, so that it may be out of sync, 
but here it is OK. Possibly rgdal was updated before sp?

>
> That's never cause a problem that I've noticed, but I'm not even a 
> moderate user of GIS so am not stressing the interlocking programs very 
> hard.
>
> I then mounted the newer super-package 2.2 disk image from KyngChaos and 
> ran the pkg-installer to see if I could break anything. Nothing seems to 
> be broken, but the rgdal package still reports loading GDAL 2.1.3, 
> despite GDAL 2.2 being the more recent version and the one to which the 
> 'Current' folder-symlink is pointing. The PROJ framework now has a new 
> 4B directory but the startup message for the PROJ.4 runtime is also 
> unchanged.

Because rgdal does not try to re-order the load list for 
shared/dynamically loaded objects, it takes the first one offered. If this 
(for GDAL) is not the same as the version used at install time, it should 
complain, but if install and runtime versions are the same, it shows the 
runtime version (getGDALVersionInfo() and getGDALCheckVersion()). 
getPROJ4VersionInfo() shows the PROJ version, using two sources, one a 
call to a runtime function, the other the PROJ header file version value 
from install time. These may differ, but the runtime simply reports the 
same header information, so this isn't foolproof (and versions of PROJ 
years ago were released without the header files getting the correct 
version numbers.

Provided the examples and tests run, things should be OK, but code 
coverage wrt. external software (and revisions of the EPSG database) is 
not like the textbook definition.

Roger

>
>>
>
>> Roger
>>
>> --
>> Roger Bivand
>> Department of Economics, Norwegian School of Economics,
>> Helleveien 30, N-5045 Bergen, Norway.
>> voice: +47 55 95 93 55; e-mail: Roger.Bivand using nhh.no
>> http://orcid.org/0000-0003-2392-6140
>> https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
>>
>> _______________________________________________
>> R-SIG-Mac mailing list
>> R-SIG-Mac using r-project.org
>> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
>
> David Winsemius
> Alameda, CA, USA
>
> 'Any technology distinguishable from magic is insufficiently advanced.'   -Gehm's Corollary to Clarke's Third Law
>
>
>
>
>
>

-- 
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; e-mail: Roger.Bivand using nhh.no
http://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en