Sorting a table based on if-then statement

Question

I have a large table with two attribute fields "layers" and "Meas" both contain numbers, I would like to sort "Meas" in the odd numbered "layers" in ascending order, and "Meas" in the even numbered "layers" in Descending order. This would be part of a larger model in modelbuilder. I am using Arc Desktop, so "If Value Is" tool is not available.

This feels like it should be easy to do, however I have tried unsuccessfully to code something in python and am at a loss about how to approach the issue, any suggestions?

What I have:

[ID,  Layer,  Meas],
[1,     1,     3],
[2,     1,     2],
[3,     1,     1],  
[4,     1,     5],
[5,     2,     3],
[6,     2,     2],
[7,     2,     5], 
[8,     3,     3],
[9,     3,     2],
[10,     3,     1],  
[11,     3,     5]

What I want:

[ID,  Layer,  Meas],
[3,     1,     1],
[2,     1,     2],
[1,     1,     3],
[4,     1,     5],
[7,     2,     5], 
[5,     2,     3],
[6,     2,     2],
[10,     3,     1], 
[9,     3,     2],
[8,     3,     3],
[11,     3,     5]

Answer 1

I'm giving you an idea, you can automate this in Python.

First, you can build a list for each value in the "layer" field, and you will get 3 lists:

list_layer_1 = [[1,     1,     3], [2,     1,     2], [3,     1,     1], [4,     1,     5]]
list_layer_2 = [[5,     2,     3], [6,     2,     2], [7,     2,     5]]
list_layer_3 = [[8,     3,     3], [9,     3,     2], [10,     3,     1], [11,     3,     5]]

In a second step, you will sort each list on your "Meas" field. To do this, you use the Python exited method, the "operator" module, you specify your list, the index of the sort field (Meas is the 3rd field so its index is equal to 2) and the hierarchical direction of the sort. You can execute the following commands with this syntax:

your_list = sorted(your_list, key = operator.itemgetter(index of your sort field), reverse = True or False)

# First list
list_layer_1 = sorted(list_layer_1, key = operator.itemgetter(2))
# Result 
[[3, 1, 1], [2, 1, 2], [1, 1, 3], [4, 1, 5]]

# Second list
list_layer_2 = sorted(list_layer_2, key = operator.itemgetter(2), reverse = True)
# Result
[[7, 2, 5], [5, 2, 3], [6, 2, 2]]

In a third step, you combine the 3 lists:

final_list = list_layer_1 + list_layer_2 + list_layer_3

You can automate this with a loop system with conditions to know the sorting direction.

Answer 2

Create two lists, one with even and one with odd numbers. Then sort each and combine. I copy-pasted your data into a csv file.

import arcpy, os

testfile = r"C:\GIS\data\testdata\sortme.csv"
outdb = r"C:\GIS\data\testdata\testdatabas.gdb"
outtable = 'sorted2'

arcpy.MakeTableView_management(in_table=testfile, out_view='tview')
arcpy.TableToTable_conversion(in_rows='tview', out_path=outdb, out_name=outtable) #Create table from view
arcpy.TruncateTable_management(os.path.join(outdb,outtable)) #Empty it

fields = ['Layer','Meas', 'ID']

rows = [i for i in arcpy.da.SearchCursor('tview', fields)] #List all rows
even = list(filter(lambda x: (x[0]%2==0), rows)) #filter out even Layer numbers (Layer is first element in fields list = index[0]
odd = list(filter(lambda x: (x[0]%2!=0), rows)) #odd

even_sorted  = sorted(even, key=lambda x: x[1], reverse=True)
odd_sorted  = sorted(odd, key=lambda x: x[1])
both = sorted(even_sorted+odd_sorted, key=lambda x: x[0])

icur = arcpy.da.InsertCursor(os.path.join(outdb,outtable), fields)
for row in both:
    icur.insertRow(row)

del icur