Assert::uuid(...) false positives on UUIDs with trailing newlines

[Malik12tree]

Assert::uuid(...) returns false positives on UUIDs with trailing newlines

Test Cases:

Assert::uuid("c9df9150-6536-42e0-802a-266f7ccc189c");     // Valid as expected
Assert::uuid("c9df9150-6536-42e0-802a-266f7ccc189c\r");   // Invalid as expected
Assert::uuid("c9df9150-6536-42e0-802a-266f7ccc189c\r\n"); // Invalid as expected
Assert::uuid("c9df9150-6536-42e0-802a-266f7ccc189c\n");   // Currently Valid (incorrect)

Proposed Cause:

assert/src/Assert.php

Line 1982 in f23349f

if (!\preg_match('/^[0-9A-Fa-f]{8}-[0-9A-Fa-f]{4}-[0-9A-Fa-f]{4}-[0-9A-Fa-f]{4}-[0-9A-Fa-f]{12}$/', $value)) {

The regular expression above checks if the test subject starts and ends with a UUID (by using ^ and $). However, the $ anchor not only anchors the end of the string, but also an immediate newline character. (as stated here)

Proposed Solution:

To prevent false positives, we can make use of the PCRE_DOLLAR_ENDONLY flag (D) that will constraint the $ anchor to match the end of the string only, not a new character.

- if (!\preg_match('/^[0-9A-Fa-f]{8}-[0-9A-Fa-f]{4}-[0-9A-Fa-f]{4}-[0-9A-Fa-f]{4}-[0-9A-Fa-f]{12}$/', $value)) { 
+ if (!\preg_match('/^[0-9A-Fa-f]{8}-[0-9A-Fa-f]{4}-[0-9A-Fa-f]{4}-[0-9A-Fa-f]{4}-[0-9A-Fa-f]{12}$/D', $value)) {