Skip to content

fix: only percent-encode characters in the userinfo encode set #1852

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Oct 29, 2025
Merged

fix: only percent-encode characters in the userinfo encode set #1852

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
fa2ae72
fix: only percent-encode characters in the userinfo encode set
davidhewitt Oct 23, 2025
26aa072
fix type hint for 3.9
davidhewitt Oct 23, 2025
ad320a7
also percent-encode `%`
davidhewitt Oct 29, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
32 changes: 30 additions & 2 deletions src/url.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@ use std::sync::OnceLock;

use idna::punycode::decode_to_string;
use jiter::{PartialMode, StringCacheMode};
use percent_encoding::{percent_encode, NON_ALPHANUMERIC};
use percent_encoding::{percent_encode, AsciiSet, CONTROLS};
use pyo3::exceptions::PyValueError;
use pyo3::pyclass::CompareOp;
use pyo3::sync::OnceLockExt;
Expand Down Expand Up @@ -602,8 +602,36 @@ fn is_punnycode_domain(lib_url: &Url, domain: &str) -> bool {
scheme_is_special(lib_url.scheme()) && domain.split('.').any(|part| part.starts_with(PUNYCODE_PREFIX))
}

/// See <https://url.spec.whatwg.org/#userinfo-percent-encode-set>
///
/// Note that this doesn't actually include % itself - see the note in
/// https://url.spec.whatwg.org/#string-percent-encode-after-encoding
Copy link
Member

@Viicos Viicos Oct 29, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

RFC 3986 says:

Because the percent ("%") character serves as the indicator for
percent-encoded octets, it must be percent-encoded as "%25" for that
octet to be used as data within a URI.

And the note in the WHATWG spec isn't clear to me:

The other values for the percentEncodeSet argument — which happen to be used by the URL parser — leave U+0025 (%) untouched and as such it needs to be percent-encoded first in order to be properly represented.

As I understand it, it says that some encoding sets don't encode %, and as such % need special-casing and must be percent-encoded?

Before #1829, it seems that % wasn't encoded:

import pydantic as py

scheme = "mysql+pymysql"
host = "my_host"
port = 3306
path = "my_db"
username = "my_user"
password = "my_passwor%d"

url_1 = py.AnyUrl(f"{scheme}://{username}:{password}@{host}:{port}/{path}")
print(url_1)
#> mysql+pymysql://my_user:my_passwor%d@my_host:3306/my_db

But following what I said previously this should be a bug.

Copy link
Contributor Author

@davidhewitt davidhewitt Oct 29, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ad320a7

const USERINFO_ENCODE_SET: &AsciiSet = &CONTROLS
Copy link
Member

@Viicos Viicos Oct 29, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Note to self: ~ is no included in the set. RFC 1738 used to consider it an unsafe character, but RFC 3986 doesn't.

// query percent-encodes is controls plus the below
.add(b' ')
.add(b'"')
.add(b'#')
.add(b'<')
.add(b'>')
// path percent-encodes is query percent-encodes plus the below
.add(b'?')
.add(b'^')
.add(b'`')
.add(b'{')
.add(b'}')
// userinfo percent-encodes is path percent-encodes plus the below
.add(b'/')
.add(b':')
.add(b';')
.add(b'=')
.add(b'@')
.add(b'[')
.add(b'\\')
.add(b']')
.add(b'|');

fn encode_userinfo_component(value: &str) -> Cow<'_, str> {
let encoded = percent_encode(value.as_bytes(), NON_ALPHANUMERIC).to_string();
let encoded = percent_encode(value.as_bytes(), USERINFO_ENCODE_SET).to_string();
if encoded == value {
Cow::Borrowed(value)
} else {
Expand Down
33 changes: 20 additions & 13 deletions tests/validators/test_url.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1318,31 +1318,38 @@ def test_multi_url_build() -> None:
assert str(url) == 'postgresql://testuser:[email protected]:5432/database?sslmode=require#test'


def test_multi_url_build_encodes_credentials() -> None:
url = MultiHostUrl.build(
@pytest.mark.parametrize('url_type', [Url, MultiHostUrl])
def test_url_build_encodes_credentials(url_type: type[Union[Url, MultiHostUrl]]) -> None:
url = url_type.build(
scheme='postgresql',
username='user name',
password='p@ss/word?#',
password='p@ss/word?#__',
host='example.com',
port=5432,
)
assert url == MultiHostUrl('postgresql://user%20name:p%40ss%2Fword%3F%[email protected]:5432')
assert str(url) == 'postgresql://user%20name:p%40ss%2Fword%3F%[email protected]:5432'
assert url.hosts() == [
{'username': 'user%20name', 'password': 'p%40ss%2Fword%3F%23', 'host': 'example.com', 'port': 5432}
]
assert url == url_type('postgresql://user%20name:p%40ss%2Fword%3F%[email protected]:5432')
assert str(url) == 'postgresql://user%20name:p%40ss%2Fword%3F%[email protected]:5432'
if url_type is Url:
assert url.username == 'user%20name'
assert url.password == 'p%40ss%2Fword%3F%23__'
else:
assert url.hosts() == [
{'username': 'user%20name', 'password': 'p%40ss%2Fword%3F%23__', 'host': 'example.com', 'port': 5432}
]


def test_multi_url_build_hosts_encodes_credentials() -> None:
hosts = [
{'host': 'example.com', 'password': 'p@ss/word?#', 'username': 'user name', 'port': 5431},
{'host': 'example.org', 'password': 'pa%ss', 'username': 'other', 'port': 5432},
{'host': 'example.com', 'password': 'p@ss/word?#__', 'username': 'user name', 'port': 5431},
{'host': 'example.org', 'password': 'p@%ss__', 'username': 'other', 'port': 5432},
]
url = MultiHostUrl.build(scheme='postgresql', hosts=hosts)
assert str(url) == 'postgresql://user%20name:p%40ss%2Fword%3F%[email protected]:5431,other:pa%[email protected]:5432'
assert (
str(url) == 'postgresql://user%20name:p%40ss%2Fword%3F%[email protected]:5431,other:p%40%[email protected]:5432'
)
assert url.hosts() == [
{'username': 'user%20name', 'password': 'p%40ss%2Fword%3F%23', 'host': 'example.com', 'port': 5431},
{'username': 'other', 'password': 'pa%25ss', 'host': 'example.org', 'port': 5432},
{'username': 'user%20name', 'password': 'p%40ss%2Fword%3F%23__', 'host': 'example.com', 'port': 5431},
{'username': 'other', 'password': 'p%40%ss__', 'host': 'example.org', 'port': 5432},
]


Expand Down
Loading