17 ms·
Very interesting reading, as almost always from CJ. I was impressed by the precision in the data, i.e. being able to tell exactly how many lines of code deal w
by unwind 10y ago
Very interesting reading, as almost always from CJ.
I was impressed by the precision in the data, i.e. being able to tell exactly how many lines of code deal with "fraud detection". Perhaps there's a "frauddetection.py" file, but then there has to be someone importing it and using it, and those lines are harder to count, I'd expect. But perhaps the integrations are small enough to deal with manually.
- bemmu 10y agoThe code is short enough that what I did was actually just go through every line of code and try to assign it to a category. It did have some which were hard, such as the import statements you mention that multiple parts might use. But I figured being off by a few dozen lines wouldn't be that important for seeing the bigger picture.