Context - Why was this issue created?

Integration tests for WPSEO_Utils::sanitize_url method (tests/WP/Inc/Utils_Test.php) are failing on the following case:

'with_non_encoded_non_latin_url' => [
	'expected'        => 'https://example.com/%da%af%d8%b1%d9%88%d9%87-%d8%aa%d9%84%da%af%d8%b1%d8%a7%d9%85-%d8%b3%d8%a6%d9%88',
	'url_to_sanitize' => 'https://example.com/گروه-تلگرام-سئو',
]

The issue seems to be with wp_parse_url() call, which returns a corrupted string for the URL's path (گر��-ت�گرا�-سئ�_ instead of روه-تلگرام-سئو).
Considering this test case has been written in March 2020, it might be that something has changed in wp_parse_url() implementation.

What is the goal of this issue?

• Restore the original WPSEO_Utils::sanitize_url behaviour for URLs that have non-Unicode characters in their path.

What needs to be done to achieve the goal?

• Investigate wp_parse_url() current behaviour
• Change WPSEO_Utils::wp_parse_url() accordingly

Does the issue still need UX or research?

No

If available, what are the tips for fixing the problem or possible solutions?

• If wp_parse_url() needs its input to be encoded, change WPSEO_Utils::sanitize_url behaviour accordingly (I would say by limiting the change only in the specific case covered by the test (i.e., when non-encoded non-latin characters are present in the URL's path)

What is the expected result/behavior?

• WPSEO_Utils::sanitize_url() should return a correctly encoded URL, as expected in the integration test.

Should documentation be added or updated for this change? And if so, where?

No