[GRASS5] Severe internal bugs in raster format

[Bill Brown]

>
>Dear developers,
>
>sorry for bad news: but it seems there is a lot
>of confusion in raster format read/write in several
>modules:
>
>Example:
> d.rast a_raster_map
> d.rast.zoom # we want to look at some pixels only
> d.rast.num  # show cell values
> d.what.rast # query with mouse: result is different
> r.out.ascii # result is different as well
>
>(from A. Lobo:)
>r.stats: some inconsistent results comparing to r.average
> *First, I compute the per-class averages:
>     r.average base=medv cover=test out=borra.me
> *Second, I check the result for, i.e., class 23:
>     d.what.rast map=medv,borra.me
>     2676500(E) 4409500(N)
>           medv in user1  (23)S_SUPRAMEDITERRANEAN_ZONE
>           borra.me in user1, quant   (111)
>           borra.me in user1, actual  (110.794075)
> *Third, I compute the statistics for each class, but the results are not
>  the same as given by d.what.rast. For example, for class 23:
>     r.stats -c in=medv,borra.me
>     r.stats:  100%
>       ...
>       23 110.299171-110.744906 6082
>       ...
> *While d.what.rast was giving 110.794075.

Was this the ONLY line containing 23 for medv - there
should have been another one or more following it probably with 
another quantized range. 

>
>We should check this quickly, otherwise we cannot trust any result.
>I guess the problem is due to ".5" (half raster cell) addition
>for query any some potential mixture of 
>  - save values at cell center
>  - save values at cell corner (which is wrong as far as I know but correct
>                                for ESRI).

Yes, the + 0.5 is correct in order to get the center of the cell.  
There could also be a problem in the quant routines, or just
some confusion about what r.stats reports.

>
>
>Another bug: r.in.arc reads in ARC/INFO grid files reversed.
>ESRI saves from lower left to upper right, GRASS from
>upper left to lower right. The module does not (yet) take
>care of this. I would be able to fix in GRASS 4.x but in GRASS 5
>the row number cannot be specified any more (G_put_raster_c_row()).
>So we have to flip the input file or something else... How to do this?
>

I've been using r.in.arc for weeks with no problems.

- Bill


---------------------------------------- 
If you want to unsubscribe from GRASS Development
Team internal mailing list write to:
minordomo at geog.uni-hannover.de with
subject 'unsubscribe grass5'
length: 3147
max: 0