From 91be07b52446d8e468671d3dbd749b939acac00f Mon Sep 17 00:00:00 2001
From: lenamonj <lenamonj@yahoo.com>
Date: Sun, 9 Aug 2026 09:04:07 -0400
Subject: [PATCH] Fix Windows drive letter normalization in hostless file: URLs

parse_path implemented the spec's "url's path is empty" test as
segment_start == path_start + 1, which assumes path[0] begins one byte
after path_start. A hostless file: URL carries a run of slashes there,
because parse_path_start finds "file://" already ending in a slash and
returns without adding one, so the drive letter in file:///C|/a was
never recognised and the | was never rewritten to :.

Derive where path[0] starts instead of assuming it.

Removes <file:///w|/m> from url/tests/expected_failures.txt, which the
WPT harness now passes.

Fixes #889.
---
 url/src/parser.rs               | 13 ++++++++++++-
 url/tests/expected_failures.txt |  1 -
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/url/src/parser.rs b/url/src/parser.rs
index 6ccfa9b46..dbfb92c63 100644
--- a/url/src/parser.rs
+++ b/url/src/parser.rs
@@ -1323,7 +1323,8 @@ impl Parser<'_> {
                 _ => {
                     // If url’s scheme is "file", url’s path is empty, and buffer is a Windows drive letter, then
                     if scheme_type.is_file()
-                        && segment_start == path_start + 1
+                        && segment_start
+                            == Parser::file_first_segment_start(&self.serialization, path_start)
                         && is_windows_drive_letter(segment_before_slash)
                     {
                         // Replace the second code point in buffer with U+003A (:).
@@ -1361,6 +1362,16 @@ impl Parser<'_> {
         input
     }
 
+    /// The offset at which `path[0]` of a `file:` URL begins.
+    ///
+    /// While a hostless `file:` URL is being parsed the serialization carries a
+    /// run of slashes at `path_start` rather than a single one, so the first
+    /// segment does not always begin at `path_start + 1`.
+    fn file_first_segment_start(serialization: &str, path_start: usize) -> usize {
+        let path = &serialization[path_start..];
+        path_start + (path.len() - path.trim_start_matches('/').len())
+    }
+
     fn last_slash_can_be_removed(serialization: &str, path_start: usize) -> bool {
         let url_before_segment = &serialization[..serialization.len() - 1];
         if let Some(segment_before_start) = url_before_segment.rfind('/') {
diff --git a/url/tests/expected_failures.txt b/url/tests/expected_failures.txt
index 314adad59..b5d818ee5 100644
--- a/url/tests/expected_failures.txt
+++ b/url/tests/expected_failures.txt
@@ -42,7 +42,6 @@
 <non-spec:/> set pathname to </..//p>
 <non-spec:/> set pathname to <//p>
 <non-spec:/.//> set pathname to <p>
-<file:///w|/m>
 <non-special:opaque  ?hi>
 <non-special:opaque  #hi>
 <non-special:opaque \t\t  \t#hi>