[GRASS-user] r.mask: no MASK created when using many categories

Fri Apr 7 01:38:22 PDT 2017

On 07/04/17 09:56, Mira Kattwinkel wrote:
> Dear Anna,
>
> great! Thanks a lot. If I understand it correctly, the limit is now 1024.
>
> Btw, how would I update such a single function? Updating Grass does not
> work (linux' apt-get upgrade does not see a new version)?

You have to compile GRASS yourself [1] or wait for the next release 
which will then show up in your distribution.

But as Anna mentioned, in your case you might not need this as you could 
formulate your list of cats more efficiently using 'thru', i.e.:

r.mask raster=basins_new at PERMANENT maskcats="1 thru 107"

instead of listing them all. Or, if the series is not continuous:

r.mask raster=basins_new at PERMANENT maskcats="1 thru 25 53 thru 75"

Moritz

[1] https://grasswiki.osgeo.org/wiki/Compile_and_Install

>
>
> On 07/04/17 03:53, Anna Petrášová wrote:
>> On Thu, Apr 6, 2017 at 5:47 AM, Micha Silver <tsvibar at gmail.com> wrote:
>>> Interesting...
>>> I can duplicate the problem (but only up to 100 cats). Above that I get an
>>> error of "stack smashing":
>>>
>> I tried to fix it in r70847 (or at least allow longer sequences, the
>> arrays are still statically allocated). But in this case you can also
>> use "1 thru 106".
>>
>> Anna
>>
>>> GRASS 7.2.0 (ITM):~ > r.mask raster=farm_bas maskcats="`seq -s " " 1 100`"
>>> --o
>>> All subsequent raster operations will be limited to the MASK area. Removing
>>> or renaming raster map named 'MASK' will restore raster operations to
>>> normal.
>>> [Raster MASK present]
>>>
>>> GRASS 7.2.0 (ITM):~ > r.info -g MASK
>>> north=530000
>>> south=510000
>>> east=185000
>>> west=170000
>>> nsres=4
>>> ewres=4
>>> rows=5000
>>> cols=3750
>>> cells=18750000
>>> datatype=CELL
>>> ncats=0
>>> [Raster MASK present]
>>>
>>> But with 103 cats...
>>> GRASS 7.2.0 (ITM):~ > r.mask raster=farm_bas maskcats="`seq -s " " 1 103`"
>>> --o
>>> WARNING: MASK already exists and will be overwritten
>>> *** stack smashing detected ***: r.reclass terminated
>>> All subsequent raster operations will be limited to the MASK area. Removing
>>> or renaming raster map named 'MASK' will restore raster operations to
>>> normal.
>>> GRASS 7.2.0 (ITM):~ > r.info -g MASK
>>> ERROR: Raster map <MASK> not found
>>>
>>>
>>>
>>> On 04/06/2017 12:12 PM, Mira Kattwinkel wrote:
>>>
>>> Dear Anna, dear list
>>>
>>> I used the NC basic sample data set to replicate my case. Unfortunately, I
>>> get the same problem. Can anybody give me a hint? Thanks a lot!
>>>
>>> Here is what I did and the output:
>>>
>>> r.watershed elevation=elevation at PERMANENT threshold=500 basin=basins_new
>>> SECTION 1a (of 5): Initiating Memory.
>>> SECTION 1b (of 5): Determining Offmap Flow.
>>> SECTION 2: A* Search.
>>> SECTION 3a: Accumulating Surface Flow with MFD.
>>> SECTION 3b: Adjusting drainage directions.
>>> SECTION 4: Watershed determination.
>>> SECTION 5: Closing Maps.
>>>
>>> ## creating mask with 106 categories  works fine
>>> r.mask raster=basins_new at PERMANENT maskcats=1 2 3 4 5 6 7 8 9 10 11 12 13 14
>>> 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
>>> 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
>>> 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89
>>> 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106
>>> All subsequent raster operations will be limited to the MASK area. Removing
>>> or renaming raster map named 'MASK' will restore raster operations to
>>> normal.
>>>
>>> r.mapcalc expression=r106 = MASK
>>>
>>> r.mask -r
>>> Raster MASK removed
>>>
>>> ## creating mask with 107 categories seems to work without error but there
>>> is no MASK output
>>> r.mask raster=basins_new at PERMANENT maskcats=1 2 3 4 5 6 7 8 9 10 11 12 13 14
>>> 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
>>> 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
>>> 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89
>>> 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107
>>> All subsequent raster operations will be limited to the MASK area. Removing
>>> or renaming raster map named 'MASK' will restore raster operations to
>>> normal.
>>>
>>> r.mapcalc expression=m107 = MASK
>>> Invalid map <MASK>
>>> Parse error
>>> ERROR: parse error
>>>
>>> g.list type=raster
>>> basins
>>> basins_new
>>> elevation
>>> elevation_shade
>>> geology
>>> lakes
>>> landuse
>>> r106
>>> soils
>>>
>>>
>>>
>>> On 06/04/17 05:06, Anna Petrášová wrote:
>>>
>>> On Wed, Apr 5, 2017 at 8:22 AM, Mira Kattwinkel
>>> <kattwinkel-mira at uni-landau.de> wrote:
>>>
>>> Dear List,
>>>
>>> I am using r.mask to create a new raster map that only contains certain
>>> categories given in 'maskcats'. Then, I use r.mapcalc to save the map under
>>> a new name and (to be on the save side) delete the MASK with r.mask flag
>>> '-r'. I do this in a loop and it works fine until a case when the number of
>>> categories to combine is 213 (trail and error lead to 106 as the maximum
>>> number that works fine). Flag 'verbose' gives the message that a MASK was
>>> created, but none is there. The problem arises in both cases when I directly
>>> use Grass or through R using execGrass.
>>>
>>> I tried it on landsat raster from NC sample dataset and it worked.
>>> Perhaps you could try to replicate it with this sample dataset?
>>>
>>> Anna
>>>
>>> Is there a limit in the number of categories that can be passed to r.mask? I
>>> did not find any hint about that. Additionally, I wonder why there is no
>>> error message but in the contrary one that tells me that a MASK was created
>>> even when it failed.
>>>
>>> In case the details are important (or if anybody has a better idea how to
>>> achieve what I want): I have a raster map of subcatchemts belonging to
>>> stream segments created with r.stream.basins. For the endpoints of segments,
>>> I want to combine these subcatchments to a total catchment raster map
>>> containing all upstream catchments.
>>>
>>> Thanks a lot,
>>>
>>> Mira
>>>
>>>
>>> --
>>> Dr. Mira Kattwinkel
>>> Quantitative Landscape Ecology
>>> Institute for Environmental Sciences
>>> University of Koblenz-Landau
>>> Fortstraße 7
>>> 76829 Landau
>>> Germany
>>> Phone: + 49 6341 280-31553
>>> Office: Building I, Room 2.02
>>>
>>> _______________________________________________
>>> grass-user mailing list
>>> grass-user at lists.osgeo.org
>>> https://lists.osgeo.org/mailman/listinfo/grass-user
>>>
>>>
>>>
>>> --
>>> Micha Silver
>>> cell: +972-523-665918
>