Auto merge of #6614 - servo:replace-surrogates, r=Ms2ger

Add a `-Z replace-surrogates` command-line option.

See #6564.

r? @Ms2ger

<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6614)
<!-- Reviewable:end -->

2 files changed, 28 insertions, 2 deletions

diff --git a/components/script/dom/bindings/conversions.rs b/components/script/dom/bindings/conversions.rs
index 9967b6cf9b7..9f6ed02a289 100644
--- a/components/script/dom/bindings/conversions.rs
+++ b/components/script/dom/bindings/conversions.rs
@@ -352,10 +352,34 @@ pub fn jsstring_to_str(cx: *mut JSContext, s: *mut JSString) -> DOMString {
             JS_GetTwoByteStringCharsAndLength(cx, ptr::null(), s, &mut length)
         };
         assert!(!chars.is_null());
-        let char_vec = unsafe {
+        let potentially_ill_formed_utf16 = unsafe {
             slice::from_raw_parts(chars as *const u16, length as usize)
         };
-        String::from_utf16(char_vec).unwrap()
+        let mut s = String::with_capacity(length as usize);
+        for item in ::rustc_unicode::str::utf16_items(potentially_ill_formed_utf16) {
+            use ::rustc_unicode::str::Utf16Item::*;
+            match item {
+                ScalarValue(c) => s.push(c),
+                LoneSurrogate(_) => {
+                    // FIXME: Add more info like document URL in the message?
+                    macro_rules! message {
+                        () => {
+                            "Found an unpaired surrogate in a DOM string. \
+                             If you see this in real web content, \
+                             please comment on https://github.com/servo/servo/issues/6564"
+                        }
+                    }
+                    if ::util::opts::get().replace_surrogates {
+                        error!(message!());
+                        s.push('\u{FFFD}');
+                    } else {
+                        panic!(concat!(message!(), " Use `-Z replace-surrogates` \
+                            on the command line to make this non-fatal."));
+                    }
+                }
+            }
+        }
+        s
     }
 }
 
diff --git a/components/script/lib.rs b/components/script/lib.rs
index 6b3f5ce8c8f..15a1c4c5db5 100644
--- a/components/script/lib.rs
+++ b/components/script/lib.rs
@@ -21,6 +21,7 @@
 #![feature(rc_unique)]
 #![feature(slice_chars)]
 #![feature(str_utf16)]
+#![feature(unicode)]
 #![feature(vec_push_all)]
 
 #![deny(unsafe_code)]
@@ -49,6 +50,7 @@ extern crate msg;
 extern crate net_traits;
 extern crate num;
 extern crate rustc_serialize;
+extern crate rustc_unicode;
 extern crate time;
 extern crate canvas;
 extern crate canvas_traits;