The path type has a feature called predefined regex that basically reads the re= field of the child zones and choses the record with most specific regex match. This feature enables you to have more advanced redirects and use different regexes in subzones to easily separate complex logic for path redirects.

When a path record contains a re= field and its value is record, TXTDirect will look for a list of numbered subzones that start from 1 and matches the request's path with each of these records’ re= field to find the most specific match. Take a look at the example below:

"_redirect.example.com.":     "v=txtv0;re=record;to=https://example.com;type=path",
"_redirect.1.example.com.":   "v=txtv0;re=\\/test1;to=https://example.com/first/predefined{1};type=host",
"_redirect.2.example.com.":   "v=txtv0;re=\\/test1\\/test2;to=https://example.com/second/predefined{1};type=host",

In the example above we have a path record which is _redirect.example.com and its re= field is set to record. When a request comes in for example.com, TXTDirect looks for the numbered subdomains which start from _redirect.1.example.com. TXTDirect will then match the request's path with the subzone's re= field to find a match. The order is defined by their numbered ordering. After finding a match, it will use that record's to= field to redirect the request.

If the incoming request's path is /test1/test2, TXTDirect will use the _redirect.2.example.com record and after parsing the placeholders, redirects the request to https://example.com/second/predefined/test1/test2.

For more information about the path records check out our specification and the full list of our placeholders.