I guess, for patterns, they meant that a number follows a specific pattern (that should be followed by your code too).
First of all, you should clarify if this is for integral (no fractions) numbers or floating point (fractions included) numbers.
Normally a pattern for a number is this [sign][integral part][decimal sign][fractional part]. (There is also the scientific notation, e.g. 1.234e3, but let's not focus on that right now.)
- A sign is a plus or minus sign.
- Both the integral and factional part are only series of numbers [0-9]
- Different regions may use different decimal signs. So this may be a variable (e.g. it can be a dot or a coma).
See the Scope Resolution Operator. That being said, I'm not sure why the use of the scope resolution operator is important here for this problem.
With this, you should be able to start your design and write your code. If you get stuck, please, post your existing solution so we can give you a hand.