Magic numbers
Magic numbers are my newest favorite thing in software. I mean, they aren’t new and I was vaguely aware of them prior but I’ve had recent opportunities to implement them and use them.
Basically certain binary file formats always start with a sequence of bytes that specify what kind of file it is.
This allows reading the first 2-6 bytes or so and determining - with reasonable certainty - the kind of file.
For example, gzip archives start with 1F, and 8B; PNGs start with 89, 50, 40, 47, 0D, 0A, 1A, and 0A - bytes 1-3 are “PNG”; and PDFs start with 25, 50, 44, 46, and 2D - or “%PDF-“.
Knowing some of the common ones in your field can be a simple and useful way to figure out if some file or stream of data or web response is in the format you expect.