There is a draft version of the standard document here. Note that this is a download link for a Microsoft Word document.

• 02/xx xx xx xx (inside Czech Republic)
• +420 (0)2/xx xx xx xx (outside Czech Republic)

Actually not. The 0 has been dropped from the phone numbers some years ago. The original scheme was 4 to 8 numbers not starting with 0 for a local number (different districts had different number of phones and thus different number length) or 0 area_code local_number. 02 used to mean Prague. They increased the number of digits as necessary and made the area code required and thus the 0 unnecessary. So 02 12 34 56 78 became 212 345 678.

Regarding +420 (0)2 ... the original country code was 42 so the numbers were +42 0 area_code local_number. After we split with the slovaks our code was "changed" to 420, they were assigned 421. So the international numbers became +420 area_code local_number. In either case +420 02/xx xx xx xx would have one too many zeroes I believe. (Sorry for this lengthy and largely pointless post.)

In either case I think this is a bit too tricky task. Especially if you wanted them to accept the special numbers. 911 in US, 112,150,155,158 and probably a few others in Czech Rep and quite a few more in other countries. I think it's best not to be too strict while validating phone numbers. At least unless you can restrict them to a single country.

But it does make a good teaching example, by starting with the most basic rules (is it the right length?) then add more and more complex ones - start with why the A and D digits can't be 0 or 1 (getting the students to figure out why this is the case is a useful exercise in itself, even without then encoding the rule in a program); then add the other special cases (eg, the BC digits can't be 11 and again get the students to figure out why) and finally wrap up with "and you don't have to do any of that work if you use perl because there's a CPAN module for it". That module is Number::Phone::NANP.