Pablo Alvarez-2 wrote:
Hello,
We (two mac users) have been attempting to install rgdal from
"http://www.kyngchaos.com/software:frameworks";, given that it is not
available as a binary on the CRAN (binaries) of the "Package
Installer".
The GDAL library contains an impressive collection of utilities and
also
leverages several other libraries such as HDF, netCDF and PROJ4,
just to
name a few. Building and distributing a binary R package that
contains the
GDAL library bloats the size of the package and causes redundancies
if a
user already has GDAL installed somewhere else on their system. It
also
creates a lot of extra work for the package maintainer.
This is probably done for Windows because a run-of-the-mill Windows
system
is utterly incapable of building anything from source. Linux and OS X
systems are perfectly capable, so it is expected that you can build
the
package from source and use your installed version of GDAL.
Pablo Alvarez-2 wrote:
I have also tried to solve this problem by looking on the net for
an old
question, and though I have found it, the answers do not help very
much
because our porgraming skills are embryonic.
Our GIS professor has suggested we ask you guys how to proceed with
the
CRAN (source). Any suggestions or direction to an old document
dealing
with this?
Thank you very much for your help,
Pablo and Margarita
PS. We have already installed: GDAL (for QGIS) and "sp" (for R)
I've built the GDAL library and installed rgdal a few times-- it's
typically
one of the first things I do after wiping or reinstalling my OS X
system or
a Linux system. Building software from source can be a tricky
business with
a steep learning curve-- but once you figure out some of the
patterns it
becomes quite doable (as long as those darn platform-specific issues
don't
cut you off at the kneecaps).
The rgdal package contains a Configure script that is run during
installation and attempts to locate the GDAL library. When building
software
on Unix/Linux, there are three kinds of folders containing various
items of
interest that compilers and linkers will frequently pester you
about. These
are:
* "lib" folders-- these contain libraries that hold code you are
trying to
use.
* "include" folders-- these contain header files that tell the
compiler
how to use
the code in the libraries.
* "bin" folders-- these contain scripts and programs that may help
the
Configure script,
or the program you are building do the things they do.
I suspect the Configure script for the rgdal package only checks
standard
Unix/Linux folders such as /usr/bin or /usr/local/lib. The Framework
package
provided by William Kyngesburye installs GDAL Mac-Style, which means
these
folders are located in:
/Library/Frameworks/GDAL.framework/unix/
And
/Library/Frameworks/PROJ.framework/unix/
Fortunately, R allows us to pass hints to configure during package
installation in order to help it find what it is looking for.
So, grab the source tarball of rgdal from:
http://cran.r-project.org/web/packages/rgdal/index.html
I'm also using the PROJ4 and GDAL frameworks installed by the GDAL
Complete
framework, available from:
http://www.kyngchaos.com/software:frameworks/
Also, make sure you have installed the "Developer Tools" contained
on your
OS X System Software CD as these provide C compilers and other tools
needed
to build the code in the rdal package.
Now pop open a Terminal. Assuming you saved the package in your
Downloads
folder, execute the following.
cd ~/Downloads
R CMD INSTALL rgdal_0.6-20.tar.gz
R will start to install the package, and then wipe out with a
message that
includes:
Error: gdal-config not found
The gdal-config script distributed with GDAL could not be found.
If you have not installed the GDAL libraries, you can
download the source from http://www.gdal.org/
If you have installed the GDAL libraries, then make sure that
gdal-config is in your path. Try typing gdal-config at a
shell prompt and see if it runs. If not, use:
--configure-args='--with-gdal-config=/usr/local/bin/gdal-config'
echo with
appropriate values for your installation.
gdal-config is a little script who's sole purpose is to tell Configure
scripts whatever they want to know about a GDAL installation.
Scripts and
programs that accompany a library are commonly installed in a "bin"
folder.
So given what I've noted above about the install location used by
the GDAL
framework, we can give rgdal's Configure script a hint:
R CMD INSTALL --configure-args=\
"--with-gdal-config=/Library/Frameworks/GDAL.framework/unix/bin/gdal-
config"
\
rgdal_0.6-20.tar.gz
The R installer now spits out another error:
Error: proj_api.h not found.
If the PROJ.4 library is installed in a non-standard location,
use --configure-args='--with-proj-include=/opt/local/include'
for example, replacing /opt/local/* with appropriate values
for your installation. If PROJ.4 is not installed, install it.
PROJ4 has no "proj4-config" script, so we will probably need to tell
Configure where both the lib and include folders are. The following
command:
R CMD INSTALL --configure-args=\
"--with-gdal-config=/Library/Frameworks/GDAL.framework/unix/bin/gdal-
config
\
--with-proj-lib=/Library/Frameworks/PROJ.framework/unix/lib \
--with-proj-include=/Library/Frameworks/PROJ.framework/unix/
include" \
rgdal_0.6-20.tar.gz
Should provide Configure with enough information to build and
install the
package.
When smashed onto one line (for easier copying and pasting) the above
command is:
R CMD INSTALL
--configure-args="--with-gdal-config=/Library/Frameworks/
GDAL.framework/unix/bin/gdal-config
--with-proj-lib=/Library/Frameworks/PROJ.framework/unix/lib
--with-proj-include=/Library/Frameworks/PROJ.framework/unix/include"
rgdal_0.6-20.tar.gz
Hope this helps!
-Charlie
-----
Charlie Sharpsteen
Undergraduate
Environmental Resources Engineering
Humboldt State University
--
View this message in context:
http://old.nabble.com/Help-with-RGDAL-tp26135289p26137651.html
Sent from the R help mailing list archive at Nabble.com.
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
