ArcGIS Split Tool: How to change output name prefix based on polygon attribute

[LaraFaelivrin]

I want the output file names of the raster split tool (ArcGis;http://resources.arcgis.com/en/help/main/10.2/index.html#//00170000009v000000) to carry a specific attribute (not FID) of the polygon I used to split the raster file. How can I achieve this?

[NeilAyres]

The only way as far as I can see would be to run the Split Raster tool with your polygon grid, then rename each one afterwards with your attribute id. You would have to make sure that these are unique of course. Are these map sheet names / ids or something.

[LaraFaelivrin]

The attribute is an ID which is different from the FID. Can the rename be done automatically? Renaming all of them manually would be laborious as the Split Raster tool created more than 100 files.

[LaraFaelivrin]

The attribute is an ID which is different from the FID. Can the rename be done automatically? Renaming all of them manually would be laborious as the Split Raster tool created more than 100 files.

Maybe this can be done using an overlap query?

[LukeSturtevant]

You could use the iterate rasters with the rename tool in a model builder. I almost think it would be easier to make a tool that iterates features​ and then clips the rasters. You could use get field value​ tool to generate the naming output.

[LaraFaelivrin]

How can I rename the raster files based on a vector attribute with the iterate raster - rename approach?

The second approach worked (see script below) but using the Clip tool in a loop is substantially slower than the Split Raster tool:

cursor = arcpy.SearchCursor(Catchment)
for poly in cursor:
  print(poly.getValue(Outlet_ID))
  OutletID_poly = poly.getValue(Outlet_ID)
  Output_poly = Tile_chm_catchment2 + "/Tile_chm_catchment" + str(OutletID_poly) + ".tif "
  arcpy.Clip_management (CHM, "281898 706162 249670 761440", Output_poly, CHM, "", "NONE", "NO_MAINTAIN_EXTENT")
  print "Tile_chm_catchment_" + str(OutletID_poly) + " is done"

[LukeSturtevant]

The iterate raster is a model builder tool only. If you want to use python. You would need to use either List Rasters​ or walk​ to get a list of rasters you want to rename. You would need to run a Data Access Search Cursor​ on your polygon to match the FID row and get the field value that you want. Finally you would construct a string name using the field value and run rename​ for each raster.

Using the split tool your output rasters are named something like "Raster_0" correct? So you would need to split that name to get the FID number and then run a da search cursor on your polygon feature like this:

import arcpy
from arcpy import env

env.workspace = #workspace path for your raster folder
polygon = # path to your polygon
fieldName = # Name of the field you want the value from

for raster in arcpy.ListRasters():
     id = int(raster.split("_")[-1])
     oidField = arcpy.Describe(feature).oidFieldName
     fieldValue = arcpy.da.SearchCursor(feature,["OID@",fieldName],where_clause = '{0} =    {1}'.format(arcpy.AddFieldDelimiters(feature,oidField),id)).next()[1]
     newName = "Raster_"+fieldValue
     arcpy.Rename_management(raster,newName)

EDITED:I edited the code in hopes that it might clarify things a bit better. The raster.split("_") is assuming the raster names are something like 'Raster_0", but you may need to modify this to pull out the FID number.

[LaraFaelivrin]

I tried to do this but I am doing something wrong... I think i did not implement the arcpy.da.SearchCursor correctly:

# CHM = Raster Dataset
# Catchment = Vector Dataset
# Outlet_ID = Polygon_ID with I want to add to the Output of the Raster Split Tool

workspace = Tile_chm_catchment

walk = arcpy.da.Walk(workspace, datatype="RasterDataset", type="TIF")
feature_classes = []

for dirpath, dirnames, filenames in walk:
    for filename in filenames:
        feature_classes.append(os.path.join(dirpath, filename))
#print feature_classes # Output =  u'//.../Data_out/CHM/Tile_chm_catchment\\Tile_chm_catchment_99.TIF

# unicode to string
feature_class_string = [x.encode('UTF8') for x in feature_classes]
# print feature_class_string # Output = '/.../Data_out/CHM/Tile_chm_catchment\\Tile_chm_catchment_99.TIF'

for feature in feature_class_string:
    #print feature
    Raster_Number =  re.findall('\d+', feature) # Raster_99 would lead to Raster_Number = 99
    #print Raster_Number # Output = ['99']
    feature_class_string_NoNumber = ''.join([i for i in feature if not i.isdigit()])
    #print feature_class_string_NoNumber # Output = //.../Data_out/CHM/Tile_chm_catchment\Tile_chm_catchment_.TIF
    feature_class_string_NoNumber_NoTif = feature_class_string_NoNumber[:-4]
    #print feature_class_string_NoNumber_NoTif # Output = //.../Data_out/CHM/Tile_chm_catchment\Tile_chm_catchment_
    CHM_Input = feature
    oidField = arcpy.Describe(CHM_Input).oidFieldName
    #print oid_Field
    fieldValue = arcpy.da.SearchCursor(Catchment,["OID@","Outlet_ID"],where_clause = '{0} = {1}'.format(arcpy.AddFieldDelimiters(CHM_Input, oidField),id)) # ID@ —The value of the ObjectID field
    #printfieldValue
    Rename_Output = str(feature_class_string_NoNumber_NoTif) + str(fieldValue) + ".tif"
    arcpy.rename_management(fieldname_string, Rename_Output)

[LukeSturtevant]

It looks like you did not assign your 'id' for the search cursor. Try something like this:

Raster_Number =  int(re.findall('\d+', feature)[0])

then your search cursor would be

fieldValue = arcpy.da.SearchCursor(Catchment,["OID@","Outlet_ID"],where_clause = '{0} = {1}'.format(arcpy.AddFieldDelimiters(CHM_Input, oidField),Raster_Number)).next()[1]

I think you can skip a step or two in creating your output string like removing the '.tif'. You could just set

feature_class_string_NoNumber_NoTif = ''.join([i for i in feature if not i.isdigit()]).split(".")[0]

[LaraFaelivrin]

Thank you so much for your help!!

• When running the arcpy rename function I get the Error 9999: Error executing function

• When printing the Rename_Output I get the following string:
  
  //service.geo.uzh.ch/private/test/data/Desktop/Test/Data_out/CHM/Tile_chm_catchment\Tile_chm_catchment_<da.SearchCursor object at 0x00000000054E6870>.tif
  
  --> The slash before Tile_chm_catchment is in the wrong direction (this is the output of the walk function which I copied from the ArcGIS help). Why does this happen? Does this have to be changed?
  
  --> The da.SearchCursor is an object. Why is not a number?

• When applying "feature_class_string_NoNumber_NoTif = ''.join([i for i in feature if not i.isdigit()]).split(".")[0]" it somehow adds "//service" to the line. I have no clue why this happens.

See updated code:

# CHM = Raster Dataset

# Catchment = Vector Dataset

# Outlet_ID = Polygon_ID with I want to add to the Output of the Raster Split Tool

workspace = Tile_chm_catchment

walk = arcpy.da.Walk(workspace, datatype="RasterDataset", type="TIF")

feature_classes = []

for dirpath, dirnames, filenames in walk:

    for filename in filenames:

        feature_classes.append(os.path.join(dirpath, filename))

print feature_classes # Output =  u'///Data_out/CHM/Tile_chm_catchment\\Tile_chm_catchment_99.TIF

# unicode to string

feature_class_string = [x.encode('UTF8') for x in feature_classes]

# print feature_class_string # Output = '//.../Data_out/CHM/Tile_chm_catchment\\Tile_chm_catchment_99.TIF'

for feature in feature_class_string:

    #print feature # Output = //..../Data_out/CHM/Tile_chm_catchment\Tile_chm_catchment_99.TIF

    Raster_Number =  int(re.findall('\d+', feature)[0]) # Raster_99 would lead to Raster_Number = 99

    print Raster_Number # Output = ['99']

##    feature_class_string_NoNumber_NoTif = ''.join([i for i in feature if not i.isdigit()]).split(".")

##    print feature_class_string_NoNumber_NoTif[0]

    feature_class_string_NoNumber = ''.join([i for i in feature if not i.isdigit()])

    #print feature_class_string_NoNumber # Output = //.../Data_out/CHM/Tile_chm_catchment\Tile_chm_catchment_.TIF

    feature_class_string_NoNumber_NoTif = feature_class_string_NoNumber[:-4]

    #print feature_class_string_NoNumber_NoTif # Output = ///Data_out/CHM/Tile_chm_catchment\Tile_chm_catchment_

    CHM_Input = feature

    oidField = arcpy.Describe(Catchment).oidFieldName

    #print oidField

    fieldValue = arcpy.da.SearchCursor(Catchment,["OID@","Outlet_ID"],where_clause = '{0} = {1}'.format(arcpy.AddFieldDelimiters(CHM_Input, oidField),Raster_Number)) # ID@ —The value of the ObjectID field   fieldValue = arcpy.da.SearchCursor(Catchment,["OID@","Outlet_ID"],where_clause = '{0} = {1}'.format(arcpy.AddFieldDelimiters(CHM_Input, oidField),id)) # ID@ —The value of the ObjectID field

    #printfieldValue

    Rename_Output = str(feature_class_string_NoNumber_NoTif) + str(fieldValue) + ".tif"

    print Rename_Output

    arcpy.Rename_management(feature, Rename_Output)