Dear all,
Now, I am very happy with a programme running (this is my first real programme in GRASS with a GUI, etc) that can sweep multiple radius buffers over points selected one by one from a point vector map and save the pixel count of each category of the overlapping raster in a text file. However, the point selection process is based on the category values of the vector file. Example:
v.extract -t in=$vector out=essaisV list=$count type=point --o --q 2>/dev/null # point extraction
where $count is incremented from 1 to the number of points.
This could be messy if two different points have the same category value (I currently manage to get category value unique and ranging from 1 to the number of points, so that each point is extracted alone).
I wonder if it would be possible to make it in a safer way, e.g. to extract vectors as does v.extract but with a selection grounded on the unique FID number of each point (and not on its category value as e.g. with 'list=1' in v.extract). This means to read the FID numbers of a point vector map and then to extract stepwise each vector point to a new vector map calling it via its FID.
Any hint appreciated,
Cheers,
Patrick