fix(javascript): keep directory-form HTML URL references distinct from files (#4646)

* fix(javascript): keep directory-form HTML URL references distinct from files

Signed-off-by: Vinícius DS Andrade <95101505+viniciusdsandrade@users.noreply.github.com>

* test(javascript): cover directory-form HTML URL references

Signed-off-by: Vinícius DS Andrade <95101505+viniciusdsandrade@users.noreply.github.com>

---------

Signed-off-by: Vinícius DS Andrade <95101505+viniciusdsandrade@users.noreply.github.com>

src/application/javascript/JavaScriptArtifactPathResolution.test.ts

@@ -3184,18 +3184,20 @@ describe("url-reference directory and empty-segment semantics", () => {
     "/dist/worker.js/.",
     "./worker.js/.?cache=1#entry",
   ])("keeps directory-form URLs distinct from files: %s", (declaredPath) => {
-    expect(
-      resolve({
-        declaredPath,
-        sourcePath: "dist/app.mjs",
-        context: "url-reference",
-        files: fileMap([file("dist/worker.js", "root")]),
-      }),
-    ).toMatchObject({
-      declared_path: declaredPath,
-      resolution_status: "not-found",
-      resolved_path: null,
-    });
+    for (const context of ["url-reference", "html-reference"] as const)
+      expect(
+        resolve({
+          declaredPath,
+          sourcePath: "dist/app.mjs",
+          context,
+          files: fileMap([file("dist/worker.js", "root")]),
+        }),
+        context,
+      ).toMatchObject({
+        declared_path: declaredPath,
+        resolution_status: "not-found",
+        resolved_path: null,
+      });
   });
 
   it.each([

src/application/javascript/JavaScriptArtifactPathResolution.ts

@@ -1315,11 +1315,19 @@ const htmlCandidate = (
       "The HTML reference path percent-decodes to NUL or to bytes that are not UTF-8.",
     ]);
   if (declaredPath.startsWith("/"))
-    return urlRelativeCandidate("", declaredPath);
+    return exactUrlPathCandidate(
+      declaredPath,
+      urlRelativeCandidate("", declaredPath),
+      "HTML",
+    );
   if (base === undefined || base === null || base === "")
     return declaredPath === ""
       ? input.sourcePath
-      : urlRelativeCandidate(posix.dirname(input.sourcePath), declaredPath);
+      : exactUrlPathCandidate(
+          declaredPath,
+          urlRelativeCandidate(posix.dirname(input.sourcePath), declaredPath),
+          "HTML",
+        );
   // A local base href is a second untrusted path input; apply the same
   // admission rules a declared path gets so it cannot smuggle traversal or
   // separator syntax past canonicalization.
@@ -1338,23 +1346,28 @@ const htmlCandidate = (
     urlDirectoryState(posix.dirname(input.sourcePath)),
     decodedBase,
   );
-  if (declaredPath === "") {
-    const basePath = urlPathCandidate(baseState);
-    return typeof basePath === "string"
-      ? exactUrlPathCandidate(decodedBase, basePath, "HTML")
-      : basePath;
-  }
-  return urlPathCandidate(
-    applyUrlReference(urlBaseDirectory(baseState), declaredPath),
+  if (declaredPath === "")
+    return exactUrlPathCandidate(
+      decodedBase,
+      urlPathCandidate(baseState),
+      "HTML",
+    );
+  return exactUrlPathCandidate(
+    declaredPath,
+    urlPathCandidate(
+      applyUrlReference(urlBaseDirectory(baseState), declaredPath),
+    ),
+    "HTML",
   );
 };
 
 const exactUrlPathCandidate = (
   declared: string,
-  candidate: string,
+  candidate: string | CandidateResolution,
   kind: "HTML" | "URL" | "module URL",
 ): string | CandidateResolution => {
-  if (!urlPathIsDirectory(declared)) return candidate;
+  if (typeof candidate !== "string" || !urlPathIsDirectory(declared))
+    return candidate;
   const confined = confineCandidate(candidate);
   if (typeof confined !== "string") return confined;
   return unresolvedCandidate("not-found", [

src/application/javascript/JavaScriptHtmlBaseResolution.test.ts

@@ -105,6 +105,23 @@ describe("HTML base href dot segments", () => {
       });
     },
   );
+  it.each([
+    ["app.js/", "/assets/", "assets/app.js"],
+    ["app.js/.", "/assets/", "assets/app.js"],
+    ["?cache=1", "/assets/app.js/", "assets/app.js"],
+  ])(
+    "keeps directory-form HTML reference %s with base %s distinct from %s",
+    (declared, base, directory) => {
+      expect(resolve(declared, base)).toMatchObject({
+        declared_path: declared,
+        resolution_status: "not-found",
+        resolved_path: null,
+        limitations: [
+          `The HTML reference resolves to directory ${directory}, not an exact selected application file.`,
+        ],
+      });
+    },
+  );
 });
 describe("HTML base href empty segments", () => {
   it("lets the script reference remove an empty segment from the base", () => {