Skip to content

MODIS package's runGdal() returns error: dataFormat='GTiff', format not supported

6 messages · Hakim Abdi, Chris English

Hakim Abdi
# 
Same here. Loading rgdal didn't make a difference.

On Tue, May 17, 2016 at 9:05 AM, chris english <
englishchristophera at gmail.com> wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

  
  


  


    

    

      
      

        


  

        

      Chris English
    

    

      
      #
    

  


  

          

      
      
      
    

        
Checking availabillity of GDAL:
   OK, GDAL 2.1.0dev, released 2015/99/99 found!

            

      
      
      
    

        
[1] name        description extension
<0 rows> (or 0-length row.names)

            

      
      
      
    

        
Interestingly not finding even GTiff default. Well, more digging,  but
these outputs explain runGdal error output.

On Tue, May 17, 2016 at 10:07 AM, Hakim Abdi <hakim.abdi at nateko.lu.se>
wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      Chris English
    

    

      
      #
    

  


  

      
Hakim,
Your installation and mine share the same defect:

            

      
      
      
    

        
[1] TRUE

            

      
      
      
    

        
Error in correctPath(path) : object 'GTiff' not found

            

      
      
      
    

        
function (path = NULL)
{
    inW <- getOption("warn")
    on.exit(options(warn = inW))
    options(warn = -1)
    path <- correctPath(path)
    cmd <- paste0(path, "gdalinfo --formats")
    if (.Platform$OS == "windows") {
        driver <- try(shell(cmd, intern = TRUE), silent = TRUE)
    }
    else {
        driver <- try(system(cmd, intern = TRUE), silent = TRUE)
    }
    if (class(driver) == "try-error") {
        options(warn = inW)
        warning("No gdal installation found please install 'gdal' on your
system first!")
        return(FALSE)
    }
    if (length(grep(driver, pattern = "HDF4")) == 0) {
        return(FALSE)
    }
    else {
        return(TRUE)
    }
}
<bytecode: 0x7ddc970>
<environment: namespace:MODIS>

            

      
      
      
    

        
If we hard code:

            

      
      
      
    

        
[1] "Supported Formats:"

  [2] "  VRT -raster- (rw+v): Virtual Raster"

  [3] "  GTiff -raster- (rw+vs): GeoTIFF"

  [4] "  NITF -raster- (rw+vs): National Imagery Transmission Format"

  [5] "  RPFTOC -raster- (rovs): Raster Product Format TOC format"

  [6] "  ECRGTOC -raster- (rovs): ECRG TOC format"

  [7] "  HFA -raster- (rw+v): Erdas Imagine Images (.img)"

  [8] "  SAR_CEOS -raster- (rov): CEOS SAR Image"

  [9] "  CEOS -raster- (rov): CEOS Image"

 [10] "  JAXAPALSAR -raster- (rov): JAXA PALSAR Product Reader (Level
1.1/1.5)"

I am still looking for where correctPath() comes from as well as trying to
get my MRT recognized, but slowly slowly.
Seems we're looking at environment variables (MRT, and likely gdal_config)
and some other special sause that we've overlooked in installation of MODIS.
Chris

On Wed, May 18, 2016 at 8:53 AM, chris english <

        
englishchristophera at gmail.com> wrote:

        

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      Chris English
    

    

      
      #
    

  


  

      
Logging out (Log out?!?, do people still do that?), which is mentioned in
the manual, though actually means more than closing and opening a new
terminal, actually restarting the kernel, results in system update where
after MRT is recognized but still, in my case, drivers are not:

            

      
      
      
    

        
All suggested packages are installed
Detecting available write drivers!
Found: 64 candidate drivers, detecting file extensions...
ERROR 1: --format option given with format 'VRT-raster-', but that format
not
recognised.  Use the --formats option to get a list of available formats,
and use the short code (ie. GTiff or HFA) as the format identifier.

ERROR 1: --format option given with format 'GTiff-raster-', but that format
not
recognised.  Use the --formats option to get a list of available formats,
and use the short code (ie. GTiff or HFA) as the format identifier.

0 usable drivers detected!

STORAGE:
_______________
localArcPath : /home/chris/MODIS_ARC
outDirPath   : /home/chris/MODIS_ARC/PROCESSED


DOWNLOAD:
_______________
MODISserverOrder : LPDAAC, LAADS
dlmethod         : auto
stubbornness     : high


PROCESSING:
_______________
GDAL           : GDAL 2.1.0dev, released 2015/99/99
MRT            : Version 4.1 (March 2011)
pixelSize      : asIn
outProj        : asIn
resamplingType : NN
dataFormat     : GTiff


DEPENDENCIES:
_______________

            

      
      
      
    

        
So, a little more digging.
Chris


On Wed, May 18, 2016 at 2:13 PM, chris english <

        
englishchristophera at gmail.com> wrote:

        

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      Chris English
    

    

      
      #
    

  


  

      
Playing with MODISoptions.R lines 155-162:

            

      
      
      
    

        
defined at present ln 157

            

      
      
      
    

        
+ {
+ stop(paste0("The 'gdalPath' you have provided
'",normalizePath(gdalPath,"/",FALSE) ,
+ "' does not contain any gdal utilities, make sure to address the
+ folder with GDAL executables (ie: gdalinfo)!"))
+ }

            

      
      
      
    

        
The preceding executes without complaint or stop as gdalinfo was found in
/usr/local/bin to have length >0.

            

      
      
      
    

        
47 GTiff  GeoTIFF   TRUE  TRUE    FALSE

51  HDF4Image     HDF4 Dataset   TRUE FALSE    FALSE

We have the necessary drivers. Hard coding MODIS_Opts.R gdalPath doesn't
work. Ah well.


On Thu, May 19, 2016 at 7:10 AM, chris english <

        
englishchristophera at gmail.com> wrote:

        

            

      
      
      
    

        

      
      
    

        

  
  


  


      

    

    
5 days later

    

      
      

        


  

        

      Chris English
    

    

      
      #
    

  


  

      
Progress has certainly been made regards MODIS.

 > sessionInfo()
R version 3.2.2 (2015-08-14)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 14.04.4 LTS

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C
LC_TIME=en_US.UTF-8
 [4] LC_COLLATE=en_US.UTF-8     LC_MONETARY=en_US.UTF-8
LC_MESSAGES=en_US.UTF-8
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C
LC_ADDRESS=C
[10] LC_TELEPHONE=C             LC_MEASUREMENT=en_US.UTF-8
LC_IDENTIFICATION=C

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base

other attached packages:
[1] MODIS_0.10-34  raster_2.5-2   sp_1.2-3       devtools_1.9.1

loaded via a namespace (and not attached):
[1] rsconnect_0.4.1.11 tools_3.2.2        Rcpp_0.12.5
memoise_0.2.1      grid_3.2.2
[6] digest_0.6.8       lattice_0.20-33

Thomas Nauss, University of Marburg, examined the problem and proposed
a fix (https://github.com/MatMatt/MODIS/issues/5). It has not actually
been applied to the 'develop' branch currently the MODIS_0.10-34. But
cloning the develop branch and editing minorFuns.R with his suggested
fix and doing a local install from source gives me:

            

      
      
      
    

        
Loading required package: MODIS
Loading required package: raster
Loading required package: sp
MODIS_manual: https://ivfl-rio.boku.ac.at/owncloud/public.php?service=files&t=660dc830afb091237cc40b3dea2fdf6b


Attaching package: ?MODIS?

The following object is masked from ?package:base?:

    file.size

            

      
      
      
    

        
All suggested packages are installed
Detecting available write drivers!
Found: 64 candidate drivers, detecting file extensions...
44 usable drivers detected!

This is a darn sight better than zero drivers.

STORAGE:
_______________
localArcPath : /home/chris/MODIS_ARC
outDirPath   : /home/chris/MODIS_ARC/PROCESSED


DOWNLOAD:
_______________
MODISserverOrder : LPDAAC, LAADS
dlmethod         : auto
stubbornness     : high


PROCESSING:
_______________
GDAL           : GDAL 2.1.0dev, released 2015/99/99
MRT            : Version 4.1 (March 2011)
pixelSize      : asIn
outProj        : asIn
resamplingType : NN
dataFormat     : GTiff

Returning to Hakim's initial post:

            

      
      
      
    

        
+ h,tileV = v, SDSstring = bands, outProj="4326")
Loading required package: rgdal
rgdal: version: 1.1-3, (SVN revision 594)
 Geospatial Data Abstraction Library extensions to R successfully loaded
 Loaded GDAL runtime: GDAL 2.1.0dev, released 2015/99/99
 Path to GDAL shared files: /usr/local/share/gdal
 Loaded PROJ.4 runtime: Rel. 4.9.1, 04 March 2015, [PJ_VERSION: 491]
 Path to PROJ.4 shared files: (autodetected)
WARNING: no proj_defs.dat in PROJ.4 shared files
 Linking to sp version: 1.2-1
Loading required package: rgeos
rgeos version: 0.3-15, (SVN revision 515)
 GEOS runtime version: 3.4.2-CAPI-1.8.2 r3921
 Linking to sp version: 1.2-1
 Polygon checking: TRUE

########################
NOTE: rgdal::checkCRSArgs: no proj_defs.dat in PROJ.4 shared files
outProj          =  +init=epsg:4326 +proj=longlat +datum=WGS84
+no_defs +ellps=WGS84 +towgs84=0,0,0
pixelSize        =  asIn
resamplingType   =  near
Output directory =
/home/chris/MODIS_ARC/PROCESSED/MOD13Q1.005_20160525005632  (no 'job'
name specified, generated (date/time based))
########################
Loading required package: RCurl
Loading required package: bitops
Downloading structure on 'LPDAAC' for: MOD13Q1.005
Try: 1
Error in if (FtpDayDirs[1] == FALSE) { :
  missing value where TRUE/FALSE needed

Hummph.  Well, I'm very happy to have drivers. Uncertain if my 64
candidates vs 44 represents additional name changes as Thomas
suggested and the 44 apply to his fix and an additional 20 are
differences between Frank's and Evan's naming conventions as to those
other 20. And uncertain as to how to ask MODIS which 44 are viable vs
20 that were not viable, though this would probably be useful.

It appears, given Hakim's request code, that he won't be getting MODIS
data as Error after Try: 1 suggests a certain lack of stubbornness or
further problem. Nor will I of course.  I tried a debugonce(runGdal),
but the implications of the runGdal() are substantially above and
beyond my pay grade and current level in R.

So back to eyetracking data and Bagidis semi distances. I'm sure MODIS
will be sorted out by those more familiar with expected results.

Chris

On Thu, May 19, 2016 at 1:31 PM, chris english

        
<englishchristophera at gmail.com> wrote: