匹配 URL 路径模式的正则表达式

Question

I have 2 URI Patterns with their regex:

1. "/api/orders/{id}/{version}"   ->      "^/api/orders/.+/.+$"
2. "/api/orders/{id}"             ->      "^/api/orders/.+$"

But with the above regex, String /api/orders/123-blah$-789/v1 matches both the regex patterns 1 and 2.

What regex can I use so that the above string matches only pattern 1's regex and not 2?

(The path may or may not be URL Encoded)

Answer 1

You could replace your second regex with this:

^\/api\/orders\/[[\p{Alnum}\p{Punct}]&&[^\/]]+$

Basically, it accepts every symbol or alphanum character following orders/ unless there is a slash in there. Like so, you're guaranteed that orders/ will be followed only by a single subpath.

Here is a link to test the regex:

https://regex101.com/r/KMEeXX/2

Similarly, the first regex could be rewritten on the same premises of the second one:

^\/api\/orders\/.+/[[\p{Alnum}\p{Punct}]&&[^\/]]+$

Here is a link to test it:

https://regex101.com/r/AW6cDh/1

Answer 2

You can restrict both pattern by using a negated character class [^/]+ matching 1+ characters excluding a forward slash.

For the first pattern you can use:

^/api/orders/[^/]+/[^/]+$

Regex demo

For the second pattern you can use

^/api/orders/[^/]+$

Regex demo

Answer 3

This might seem a little hardcoded but gets the job done.
And personally I prefer regex to be very specific.

    Pattern p = Pattern.compile("^/api/orders/.+/.+$");
    Pattern q = Pattern.compile("^/api/orders/[^/]+$");
    
    System.out.println(p.matcher("/api/orders/123-456-789/v1").matches());
    System.out.println(q.matcher("/api/orders/123-456-789/v1").matches());

    System.out.println(p.matcher("/api/orders/123-456-789").matches());
    System.out.println(q.matcher("/api/orders/123-456-789").matches());