One has to go the numpy route?
import numpy as np
#with some fiddling
# Setup some rasters of rows and columns
# (like old $$ROWMAP and $$COLMAP)
arcpy.env.workspace = "c:/temp"
arcpy.env.overwriteOutput = 1
nprows = np.indices((10,10))
npcols = np.indices((10,10))
# Convert the numpy arrays to ESRI rasters (ie Raster objects)
# called 'rows' and 'cols'
row_ras = arcpy.NumPyArrayToRaster(nprows)
col_ras = arcpy.NumPyArrayToRaster(npcols)
gp = arcgisscripting.create(9.3) #This works in ArcGIS 10!!!
Luke -- arcgisscripting is still there under arcpy. (You probably know this by know as this post was 2011!)
So this works (for now anyway).
arcpy.env.scratchWorkspace = r"D:\Users\cprice\work"
arcpy.env.extent = arcpy.Extent(0, 0, 10, 10)
arcpy.env.cellSize = 1
The best workaround I found in that environment was to compute flow accumulations for constant direction grids (and unit weights): $$ColMap is the flow accumulation of a grid filled with 1's and $$RowMap is the flow accumulation of a grid filled with 4's.
I did an implementation of this idea in arcpy map algebra here:arcpy map algebra $$ROWMAP, $$COLMAP
Retrieving data ...