phone number parsing vs a-parser

Unlike e-mail's (for which there is even a built-in parser), phones are much more difficult to parse, because for them there is no universally accepted recording format and, accordingly, it will not be possible to write a regular expression that universally collects phones.
Usually, to collect phones, parsers are written for a specific resource (site). Or, if you need phones of only one country, you can bind to the phone code.
In general, the principle of operation is the same as for collecting any other data: you get the source code of the page (for example, via Net::HTTP) and collect numbers from it using a regular expression.