fix: proper hostname canonicalization of badhostnames (#65)

Author: edgul

src/canonicalize_and_process.rs

@@ -46,7 +46,8 @@ pub fn canonicalize_hostname(value: &str) -> Result<String, Error> {
     return Ok(String::new());
   }
   let mut url = url::Url::parse("http://dummy.test").unwrap();
-  url.set_host(Some(value)).map_err(Error::Url)?;
+  url::quirks::set_hostname(&mut url, value)
+    .map_err(|_| Error::Url(url::ParseError::InvalidDomainCharacter))?;
   Ok(url::quirks::hostname(&url).to_string())
 }
 

src/testdata/urlpatterntestdata.json

@@ -1883,7 +1883,17 @@
   {
     "pattern": [ "https://{sub.}?example{.com/}foo" ],
     "inputs": [ "https://example.com/foo" ],
-    "expected_obj": "error"
+    "exactly_empty_components": [ "port" ],
+    "expected_obj": {
+      "protocol": "https",
+      "hostname": "{sub.}?example.com",
+      "pathname": "*"
+    },
+    "expected_match": {
+      "protocol": { "input": "https", "groups": {} },
+      "hostname": { "input": "example.com", "groups": {} },
+      "pathname": { "input": "/foo", "groups": { "0": "/foo" } }
+    }
   },
   {
     "pattern": [ "{https://}example.com/foo" ],
@@ -2441,15 +2451,27 @@
   },
   {
     "pattern": [{ "hostname": "bad#hostname" }],
-    "expected_obj": "error"
+    "inputs": [{ "hostname":  "bad" }],
+    "expected_obj": {
+      "hostname": "bad"
+    },
+    "expected_match": {
+      "hostname": { "input": "bad", "groups": {} }
+    }
   },
   {
     "pattern": [{ "hostname": "bad%hostname" }],
     "expected_obj": "error"
   },
   {
     "pattern": [{ "hostname": "bad/hostname" }],
-    "expected_obj": "error"
+    "inputs": [{ "hostname": "bad" }],
+    "expected_obj": {
+      "hostname": "bad"
+    },
+    "expected_match": {
+      "hostname": { "input": "bad", "groups": {} }
+    }
   },
   {
     "skip": "likely a bug in rust-url",
@@ -2482,7 +2504,11 @@
   },
   {
     "pattern": [{ "hostname": "bad\\\\hostname" }],
-    "expected_obj": "error"
+    "inputs": [{ "hostname": "badhostname" }],
+    "expected_obj": {
+      "hostname": "bad"
+    },
+    "expected_match": null
   },
   {
     "pattern": [{ "hostname": "bad^hostname" }],
@@ -2494,15 +2520,33 @@
   },
   {
     "pattern": [{ "hostname": "bad\nhostname" }],
-    "expected_obj": "error"
+    "inputs": [{ "hostname": "badhostname" }],
+    "expected_obj": {
+      "hostname": "badhostname"
+    },
+    "expected_match": {
+      "hostname": { "input": "badhostname", "groups": {} }
+    }
   },
   {
     "pattern": [{ "hostname": "bad\rhostname" }],
-    "expected_obj": "error"
+    "inputs": [{ "hostname": "badhostname" }],
+    "expected_obj": {
+      "hostname": "badhostname"
+    },
+    "expected_match": {
+      "hostname": { "input": "badhostname", "groups": {} }
+    }
   },
   {
     "pattern": [{ "hostname": "bad\thostname" }],
-    "expected_obj": "error"
+    "inputs": [{ "hostname": "badhostname" }],
+    "expected_obj": {
+      "hostname": "badhostname"
+    },
+    "expected_match": {
+      "hostname": { "input": "badhostname", "groups": {} }
+    }
   },
   {
     "pattern": [{}],