From 387ed90b8136e71686d145a841273ab4d3b7b2f2 Mon Sep 17 00:00:00 2001
From: Simon Sapin <simon.sapin@exyr.org>
Date: Wed, 5 Nov 2014 16:34:10 -0800
Subject: [PATCH 1/2] RFC: ES6-style unicode string escaping.

---
 text/0000-es6-unicode-escapes.md | 62 ++++++++++++++++++++++++++++++++
 1 file changed, 62 insertions(+)
 create mode 100644 text/0000-es6-unicode-escapes.md

diff --git a/text/0000-es6-unicode-escapes.md b/text/0000-es6-unicode-escapes.md
new file mode 100644
index 00000000000..d429cd47803
--- /dev/null
+++ b/text/0000-es6-unicode-escapes.md
@@ -0,0 +1,62 @@
+- Start Date: 2014-11-05
+- RFC PR:
+- Rust Issue:
+
+# Summary
+
+Remove `\u203D` and `\U0001F4A9` unicode string escapes, and add
+[ECMAScript 6-style](https://mathiasbynens.be/notes/javascript-escapes#unicode-code-point)
+`\u{1F4A9}` escapes instead.
+
+# Motivation
+
+The syntax of `\u` followed by four hexadecimal digits dates from when Unicode
+was a 16-bit encoding, and only went up to U+FFFF.
+`\U` followed by eight hex digits was added as a band-aid
+when Unicode was extended to U+10FFFF,
+but neither four nor eight digits particularly make sense now.
+
+Having two different syntaxes with the same meaning but that apply
+to different ranges of values is inconsistent and arbitrary.
+This proposal unifies them into a single syntax that has a precedent
+in ECMAScript a.k.a. JavaScript.
+
+
+# Detailed design
+
+In terms of the grammar in [The Rust Reference](
+http://doc.rust-lang.org/reference.html#character-and-string-literals),
+replace:
+
+```
+unicode_escape : 'u' hex_digit 4
+               | 'U' hex_digit 8 ;
+```
+
+with
+
+```
+unicode_escape : 'u' '{' hex_digit+ 6 '}'
+```
+
+That is, `\u{` followed by one to six hexadecimal digits, followed by `}`.
+
+The behavior would otherwise be identical.
+
+
+# Drawbacks
+
+This is a breaking change and updating code for it manually is annoying.
+It is however very mechanical, and we could provide scripts to automate it.
+
+
+# Alternatives
+
+* Status quo: don’t change the escaping syntax.
+* Add the new `\u{…}` syntax, but also keep the existing `\u` and `\U` syntax.
+  This is what ES 6 does, but only to keep compatibility with ES 5.
+  We don’t have that constaint pre-1.0.
+
+# Unresolved questions
+
+None so far.

From b310c87c62347fdc2097fe4b867f771596e0c7ac Mon Sep 17 00:00:00 2001
From: Simon Sapin <simon.sapin@exyr.org>
Date: Fri, 7 Nov 2014 20:01:15 +0000
Subject: [PATCH 2/2] ES6-style escaping: add drawback: overriding {} with
 format!

---
 text/0000-es6-unicode-escapes.md | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/text/0000-es6-unicode-escapes.md b/text/0000-es6-unicode-escapes.md
index d429cd47803..d5a61221c5f 100644
--- a/text/0000-es6-unicode-escapes.md
+++ b/text/0000-es6-unicode-escapes.md
@@ -46,8 +46,15 @@ The behavior would otherwise be identical.
 
 # Drawbacks
 
-This is a breaking change and updating code for it manually is annoying.
-It is however very mechanical, and we could provide scripts to automate it.
+* This is a breaking change and updating code for it manually is annoying.
+  It is however very mechanical, and we could provide scripts to automate it.
+* Formatting templates already use curly braces.
+  Having multiple curly braces pairs in the same strings that have a very
+  different meaning can be surprising:
+  `format!("\u{e8}_{e8}", e8 = "é")` would be `"è_é"`.
+  However, there is a precedent of overriding characters:
+  `\` can start an escape sequence both in the Rust lexer for strings
+  and in regular expressions.
 
 
 # Alternatives