ArcGIS 2.9.2. - How to calculate a CRC checksum in geodatabase?

Question

Hi,

I wonder if there is a geoprocessing tool to calculate a CRC checksum for a given set of attributes in a file and/or enterprise GDB?

Ideally, the CRC checksum should be generated with alphanumeric attributes and spatial attributes to output a robust checksum to detect any changes on feature after editing... If running same CRC checksum later on, it will be obvious to detect updated feature...

Any help appreciated

Zartico-GIS · Answer

Hey thanks @VincentLaunstorfer  for the example. It was useful to me.

You may want to update your code block to say "return hash", not attribute. 🙂 Otherwise it just returns the parameter given.

VincentLaunstorfer · Answer

Hi,

I managed to compute a checksum on attribute with Calculate Field code-block, using the hashlib module

import hashlib
def hash(attribute):
  hash = hashlib.md5(attribute.encode()).hexdigest()
  return attribute

Here I used MD5 algorithm. But SHA256 is just as good (or better).

Just as simple as that!

Ideally, I would like to compute this MD5 checksum in Arcade in attribute rule...

VincentLaunstorfer · Answer

Correct, I have seen the GenerateCheckSum from the Aviation Charting extension... However, this needs extra licensing and I am unsure if it exists in ArcGIS Pro. Also, attribute rule cannot use Python script.

However, if I were going with Python, I found:

# Python program to find SHA256 hash string of a file
import hashlib

filename = input("Enter the input file name: ")
sha256_hash = hashlib.sha256()
with open(filename,"rb") as f:
# Read and update hash string value in blocks of 4K
for byte_block in iter(lambda: f.read(4096),b""):
sha256_hash.update(byte_block)
print(sha256_hash.hexdigest())

in https://www.codegrepper.com/code-examples/python/frameworks/file-path-in-python/python+sha256+checksum+file

I wonder if it is doable to loop through selected attributes (including geometry), instead of byte_block, to calculate a SHA256 checksum...although computing on alphanumeric character instead of bytes would make the algorithm not as robust!