Fix anarchist URL where path starts with // (#817)

* Handle null host with leading empty path segment

* Disable failing test

* Fix invariants of anarchist URLs + empty segment

* Handle empty leading segment in relative paths

* Handle case where input starts with a slash

* Unify handling of leading empty path segment

* Avoid casting back and forth between u32 and usize

* Fix use of .get() that should never fail

Author: qsantos

url/src/lib.rs

@@ -683,7 +683,14 @@ impl Url {
             assert_eq!(self.host_end, self.scheme_end + 1);
             assert_eq!(self.host, HostInternal::None);
             assert_eq!(self.port, None);
-            assert_eq!(self.path_start, self.scheme_end + 1);
+            if self.path().starts_with("//") {
+                // special case when first path segment is empty
+                assert_eq!(self.byte_at(self.scheme_end + 1), b'/');
+                assert_eq!(self.byte_at(self.scheme_end + 2), b'.');
+                assert_eq!(self.path_start, self.scheme_end + 3);
+            } else {
+                assert_eq!(self.path_start, self.scheme_end + 1);
+            }
         }
         if let Some(start) = self.query_start {
             assert!(start >= self.path_start);

url/src/parser.rs

@@ -474,9 +474,8 @@ impl<'a> Parser<'a> {
         let host = HostInternal::None;
         let port = None;
         let remaining = if let Some(input) = input.split_prefix('/') {
-            let path_start = self.serialization.len();
             self.serialization.push('/');
-            self.parse_path(scheme_type, &mut false, path_start, input)
+            self.parse_path(scheme_type, &mut false, path_start as usize, input)
         } else {
             self.parse_cannot_be_a_base_path(input)
         };
@@ -1354,9 +1353,50 @@ impl<'a> Parser<'a> {
         host_end: u32,
         host: HostInternal,
         port: Option<u16>,
-        path_start: u32,
+        mut path_start: u32,
         remaining: Input<'_>,
     ) -> ParseResult<Url> {
+        // Special case for anarchist URL's with a leading empty path segment
+        // This prevents web+demo:/.//not-a-host/ or web+demo:/path/..//not-a-host/,
+        // when parsed and then serialized, from ending up as web+demo://not-a-host/
+        // (they end up as web+demo:/.//not-a-host/).
+        //
+        // If url’s host is null, url does not have an opaque path,
+        // url’s path’s size is greater than 1, and url’s path[0] is the empty string,
+        // then append U+002F (/) followed by U+002E (.) to output.
+        let scheme_end_as_usize = scheme_end as usize;
+        let path_start_as_usize = path_start as usize;
+        if path_start_as_usize == scheme_end_as_usize + 1 {
+            // Anarchist URL
+            if self.serialization[path_start_as_usize..].starts_with("//") {
+                // Case 1: The base URL did not have an empty path segment, but the resulting one does
+                // Insert the "/." prefix
+                self.serialization.insert_str(path_start_as_usize, "/.");
+                path_start += 2;
+            }
+            assert!(!self.serialization[scheme_end_as_usize..].starts_with("://"));
+        } else if path_start_as_usize == scheme_end_as_usize + 3
+            && &self.serialization[scheme_end_as_usize..path_start_as_usize] == ":/."
+        {
+            // Anarchist URL with leading empty path segment
+            // The base URL has a "/." between the host and the path
+            assert_eq!(self.serialization.as_bytes()[path_start_as_usize], b'/');
+            if self
+                .serialization
+                .as_bytes()
+                .get(path_start_as_usize + 1)
+                .copied()
+                != Some(b'/')
+            {
+                // Case 2: The base URL had an empty path segment, but the resulting one does not
+                // Remove the "/." prefix
+                self.serialization
+                    .replace_range(scheme_end_as_usize..path_start_as_usize, ":");
+                path_start -= 2;
+            }
+            assert!(!self.serialization[scheme_end_as_usize..].starts_with("://"));
+        }
+
         let (query_start, fragment_start) =
             self.parse_query_and_fragment(scheme_type, scheme_end, remaining)?;
         Ok(Url {

url/src/slicing.rs

@@ -149,7 +149,14 @@ impl Url {
                 }
             }
 
-            Position::AfterPort => self.path_start as usize,
+            Position::AfterPort => {
+                if let Some(port) = self.port {
+                    debug_assert!(self.byte_at(self.host_end) == b':');
+                    self.host_end as usize + ":".len() + port.to_string().len()
+                } else {
+                    self.host_end as usize
+                }
+            }
 
             Position::BeforePath => self.path_start as usize,
 

url/tests/unit.rs

@@ -954,6 +954,16 @@ fn no_panic() {
     url::quirks::set_hostname(&mut url, "//eom/datcom/\\\\t\\://eom/data.cs").unwrap();
 }
 
+#[test]
+fn test_null_host_with_leading_empty_path_segment() {
+    // since Note in item 3 of URL serializing in the URL Standard
+    // https://url.spec.whatwg.org/#url-serializing
+    let url = Url::parse("m:/.//\\").unwrap();
+    let encoded = url.as_str();
+    let reparsed = Url::parse(encoded).unwrap();
+    assert_eq!(reparsed, url);
+}
+
 #[test]
 fn pop_if_empty_in_bounds() {
     let mut url = Url::parse("m://").unwrap();

url/tests/urltestdata.json

@@ -7487,7 +7487,6 @@
     "hash": ""
   },
   "Serialize /. in path",
-  "skip next",
   {
     "input": "non-spec:/.//",
     "base": "about:blank",
@@ -7502,7 +7501,6 @@
     "search": "",
     "hash": ""
   },
-  "skip next",
   {
     "input": "non-spec:/..//",
     "base": "about:blank",
@@ -7517,7 +7515,6 @@
     "search": "",
     "hash": ""
   },
-  "skip next",
   {
     "input": "non-spec:/a/..//",
     "base": "about:blank",
@@ -7532,7 +7529,6 @@
     "search": "",
     "hash": ""
   },
-  "skip next",
   {
     "input": "non-spec:/.//path",
     "base": "about:blank",
@@ -7547,7 +7543,6 @@
     "search": "",
     "hash": ""
   },
-  "skip next",
   {
     "input": "non-spec:/..//path",
     "base": "about:blank",
@@ -7562,7 +7557,6 @@
     "search": "",
     "hash": ""
   },
-  "skip next",
   {
     "input": "non-spec:/a/..//path",
     "base": "about:blank",
@@ -7592,7 +7586,6 @@
     "search": "",
     "hash": ""
   },
-  "skip next",
   {
     "input": "/..//path",
     "base": "non-spec:/p",
@@ -7607,7 +7600,6 @@
     "search": "",
     "hash": ""
   },
-  "skip next",
   {
     "input": "..//path",
     "base": "non-spec:/p",
@@ -7622,7 +7614,6 @@
     "search": "",
     "hash": ""
   },
-  "skip next",
   {
     "input": "a/..//path",
     "base": "non-spec:/p",
@@ -7637,7 +7628,6 @@
     "search": "",
     "hash": ""
   },
-  "skip next",
   {
     "input": "",
     "base": "non-spec:/..//p",
@@ -7652,7 +7642,6 @@
     "search": "",
     "hash": ""
   },
-  "skip next",
   {
     "input": "path",
     "base": "non-spec:/..//p",