fix(adapters): post-set_path normalization invariant check (FWD-13)

- assemble_request_url now verifies after Url::set_path that the decoded
  URL path segments are byte-identical to base_dir + decoded rendered
  segments; a mismatch (any future normalizer rewrite in the url crate)
  fails loudly with INTERNAL instead of silently re-routing an
  authenticated request
- property-style corpus test over dot/percent/binary values: accepted
  values must survive byte-identical as one literal segment with no lone
  dot segments; failures may only be INVALID_INPUT

Verification: scripts/verify.sh (383 passed) and --all-features (499
passed), clippy -D warnings, fmt --check all pass.
This commit is contained in:
2026-08-31 00:08:12 +00:00
parent d81e7ef319
commit 58bdba35cf
+116 -1
View File
@@ -118,7 +118,7 @@ use alkcall::registry::context::OperationContext;
use alkcall::registry::registration::ResponseStream; use alkcall::registry::registration::ResponseStream;
use futures::stream; use futures::stream;
use futures::StreamExt; use futures::StreamExt;
use percent_encoding::{utf8_percent_encode, AsciiSet, CONTROLS}; use percent_encoding::{percent_decode_str, utf8_percent_encode, AsciiSet, CONTROLS};
use reqwest::header::{HeaderMap, HeaderName, HeaderValue, ACCEPT, AUTHORIZATION, CONTENT_TYPE}; use reqwest::header::{HeaderMap, HeaderName, HeaderValue, ACCEPT, AUTHORIZATION, CONTENT_TYPE};
use reqwest::Method; use reqwest::Method;
use serde_json::Value; use serde_json::Value;
@@ -655,6 +655,7 @@ fn assemble_request_url(base_url: &str, rendered_path: &str) -> Result<Url, Call
full_path.push_str(&request_path); full_path.push_str(&request_path);
let mut url = base.clone(); let mut url = base.clone();
url.set_path(&full_path); url.set_path(&full_path);
assert_untouched_by_normalization(&url, base_dir, rendered_path, base_url)?;
let same_origin = url.scheme() == base.scheme() let same_origin = url.scheme() == base.scheme()
&& url.host() == base.host() && url.host() == base.host()
&& url.port_or_known_default() == base.port_or_known_default(); && url.port_or_known_default() == base.port_or_known_default();
@@ -668,6 +669,48 @@ fn assemble_request_url(base_url: &str, rendered_path: &str) -> Result<Url, Call
Ok(url) Ok(url)
} }
/// Post-`set_path` invariant (FWD-13): the decoded segments of the
/// final URL path must equal `base_dir`'s segments plus the decoded
/// rendered segments, byte-identical. `set_path` is the one step this
/// crate does not control, and its parser normalizes lone dot segments
/// away (`/../x` → `/x`); this check turns any future normalizer
/// surprise into a loud `INTERNAL` error instead of a silently
/// re-routed authenticated request.
fn assert_untouched_by_normalization(
url: &Url,
base_dir: &str,
rendered_path: &str,
base_url: &str,
) -> Result<(), CallError> {
let rendered_segments = rendered_path
.trim_matches('/')
.split('/')
.filter(|s| !s.is_empty())
.map(|s| percent_decode_str(s).decode_utf8_lossy().into_owned());
let base_segments: Vec<&str> = base_dir.split('/').filter(|s| !s.is_empty()).collect();
let expected: Vec<String> = base_segments
.iter()
.copied()
.map(str::to_string)
.chain(rendered_segments)
.collect();
let actual: Vec<String> = url
.path()
.split('/')
.filter(|s| !s.is_empty())
.map(|s| percent_decode_str(s).decode_utf8_lossy().into_owned())
.collect();
let matches = actual.len() == expected.len()
&& std::iter::Iterator::zip(actual.iter(), expected.iter()).all(|(a, e)| a == e);
if matches {
Ok(())
} else {
Err(CallError::internal(format!(
"request path `{rendered_path}` resolved against `{base_url}` was rewritten by URL normalization: expected segments {expected:?}, got {actual:?}"
)))
}
}
#[derive(Debug, thiserror::Error)] #[derive(Debug, thiserror::Error)]
pub(crate) enum BodyReadError { pub(crate) enum BodyReadError {
#[error("upstream response body exceeds the {RESPONSE_BODY_CAP}-byte response cap")] #[error("upstream response body exceeds the {RESPONSE_BODY_CAP}-byte response cap")]
@@ -1823,6 +1866,78 @@ mod tests {
} }
} }
/// FWD-13 belt-and-braces: across a dot/percent/binary corpus the
/// post-`set_path` invariant holds — the decoded URL path segments
/// equal the base dir plus the decoded rendered segments,
/// byte-identical — so a normalizer rewrite (today's lone-dot
/// removal, tomorrow's regression in the `url` crate) can never
/// silently re-route an authenticated request. Lone-dot values are
/// rejected outright; values containing `/` or `\` are rejected as
/// smuggled separators only if the encode-set layer ever regressed.
#[test]
fn post_set_path_invariant_holds_across_dot_percent_binary_corpus() {
let corpus = [
"v1.2.3",
".hidden-file",
"..",
".",
"%2e",
"%2E",
"%2e%2e",
"%2E%2E",
"%252e",
"hidden..",
"a..b",
"...",
"a%2Fb",
"a b",
"h\\éllo→世界",
"line\nbreak",
"tab\tchar",
"\u{7f}\u{1f600}",
];
for value in corpus {
let outcome = request_url(
"https://api.example.com/v1",
"/files/{name}",
json!({ "name": value }),
);
match outcome {
Ok(url) => {
assert_eq!(url.host_str(), Some("api.example.com"));
let segments: Vec<String> = url
.path_segments()
.map(|s| {
s.map(|seg| percent_decode_str(seg).decode_utf8_lossy().into_owned())
.collect()
})
.unwrap_or_default();
assert_eq!(
segments.len(),
3,
"value `{value:?}` must render as one literal segment under /v1/files/"
);
assert_eq!(
segments.get(2).map(String::as_str),
Some(value),
"value `{value:?}` must survive byte-identical as the final segment"
);
assert!(
!segments.iter().any(|s| s == "." || s == ".."),
"value `{value:?}` must not leave lone dot segments: {:?}",
segments
);
}
Err(err) => {
assert_eq!(
err.code, "INVALID_INPUT",
"value `{value:?}` may only fail as INVALID_INPUT"
);
}
}
}
}
fn ctx_with_capability(namespace: &str, value: String) -> OperationContext { fn ctx_with_capability(namespace: &str, value: String) -> OperationContext {
let mut ctx = noop_context(); let mut ctx = noop_context();
ctx.capabilities = Capabilities::new().with_http_token(namespace, value); ctx.capabilities = Capabilities::new().with_http_token(namespace, value);