Think of yourself in the shoes of a sales agent. ☎️
You arrive at the office, get ready to make calls, but your call list only has about half of the phone numbers available. Some of them are blank, some others are formatted differently, while the index jumps from 3 to 6 randomly. You are not working on the data here. Instead, you are struggling against it.
This problem was what I solved in my current project. Raw data with customers' call lists became structured information that could actually be used. Open the document -> dial.
But there were things that I didn't expect when working on it. 🧐
Turns out, fixing a 'simple' NaN issue isn't always simple. Between a 'NaN' that looks like text and one that actually is text is the difference between code that works and code that smiles at you while doing absolutely nothing.