From 73c88ced926bbdbdab4c123c62e79dfed54fdc2f Mon Sep 17 00:00:00 2001
From: binarycat <binarycat@envs.net>
Date: Fri, 9 Aug 2024 17:38:10 -0400
Subject: [PATCH 1/2] Hint on unknown escape of Unicode quotation marks in
 string literal

Fixes #128858

I opted not to produce a suggestion, since it's not obvious what the user meant to do.
---
 .../src/lexer/unescape_error_reporting.rs     | 26 +++++++++++++++++++
 tests/ui/unicode-quote.rs                     |  3 +++
 tests/ui/unicode-quote.stderr                 | 20 ++++++++++++++
 3 files changed, 49 insertions(+)
 create mode 100644 tests/ui/unicode-quote.rs
 create mode 100644 tests/ui/unicode-quote.stderr

diff --git a/compiler/rustc_parse/src/lexer/unescape_error_reporting.rs b/compiler/rustc_parse/src/lexer/unescape_error_reporting.rs
index efa53f0962b78..15b57e6c76926 100644
--- a/compiler/rustc_parse/src/lexer/unescape_error_reporting.rs
+++ b/compiler/rustc_parse/src/lexer/unescape_error_reporting.rs
@@ -158,6 +158,12 @@ pub(crate) fn emit_unescape_error(
                     "this is an isolated carriage return; consider checking your editor and \
                      version control settings",
                 );
+            } else if looks_like_quote(c) {
+                diag.help(
+                    format!("{ec} is not an ascii quote, \
+                             but may look like one in some fonts.\n\
+                             consider writing it in its \
+                             escaped form for clarity."));
             } else {
                 if mode == Mode::Str || mode == Mode::Char {
                     diag.span_suggestion(
@@ -295,3 +301,23 @@ pub(crate) fn escaped_char(c: char) -> String {
         _ => c.escape_default().to_string(),
     }
 }
+
+/// Returns true if `c` may look identical to `"` in some fonts.
+fn looks_like_quote(c: char) -> bool {
+    // list of homoglyphs generated using the following wikidata query:
+    // SELECT ?u WHERE {
+    //   wd:Q87495536 wdt:P2444+ ?c.
+    //   ?c wdt:P4213 ?u.
+    // }
+    match c {
+        '\u{2033}' | 
+        '\u{02BA}' |
+        '\u{02DD}' |
+        '\u{030B}' |
+        '\u{030E}' |
+        '\u{05F4}' |
+        '\u{201C}' |
+        '\u{201D}' => true,
+        _ => false,
+    }
+}
diff --git a/tests/ui/unicode-quote.rs b/tests/ui/unicode-quote.rs
new file mode 100644
index 0000000000000..0421fdb85041d
--- /dev/null
+++ b/tests/ui/unicode-quote.rs
@@ -0,0 +1,3 @@
+fn main() {
+    dbg!("since when is \“THIS\” not allowed in a string literal");
+}
diff --git a/tests/ui/unicode-quote.stderr b/tests/ui/unicode-quote.stderr
new file mode 100644
index 0000000000000..f418bbb65fa2e
--- /dev/null
+++ b/tests/ui/unicode-quote.stderr
@@ -0,0 +1,20 @@
+error: unknown character escape: `\u{201c}`
+  --> $DIR/unicode-quote.rs:2:26
+   |
+LL |     dbg!("since when is \“THIS\” not allowed in a string literal");
+   |                          ^ unknown character escape
+   |
+   = help: \u{201c} is not an ascii quote, but may look like one in some fonts.
+           consider writing it in its escaped form for clarity.
+
+error: unknown character escape: `\u{201d}`
+  --> $DIR/unicode-quote.rs:2:32
+   |
+LL |     dbg!("since when is \“THIS\” not allowed in a string literal");
+   |                                ^ unknown character escape
+   |
+   = help: \u{201d} is not an ascii quote, but may look like one in some fonts.
+           consider writing it in its escaped form for clarity.
+
+error: aborting due to 2 previous errors
+

From 775cb88ad6f7bcd9d80a40140324c2cf44a08015 Mon Sep 17 00:00:00 2001
From: binarycat <binarycat@envs.net>
Date: Fri, 9 Aug 2024 18:01:57 -0400
Subject: [PATCH 2/2] fmt

---
 .../src/lexer/unescape_error_reporting.rs       | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/compiler/rustc_parse/src/lexer/unescape_error_reporting.rs b/compiler/rustc_parse/src/lexer/unescape_error_reporting.rs
index 15b57e6c76926..266cb01d5b534 100644
--- a/compiler/rustc_parse/src/lexer/unescape_error_reporting.rs
+++ b/compiler/rustc_parse/src/lexer/unescape_error_reporting.rs
@@ -159,11 +159,12 @@ pub(crate) fn emit_unescape_error(
                      version control settings",
                 );
             } else if looks_like_quote(c) {
-                diag.help(
-                    format!("{ec} is not an ascii quote, \
+                diag.help(format!(
+                    "{ec} is not an ascii quote, \
                              but may look like one in some fonts.\n\
                              consider writing it in its \
-                             escaped form for clarity."));
+                             escaped form for clarity."
+                ));
             } else {
                 if mode == Mode::Str || mode == Mode::Char {
                     diag.span_suggestion(
@@ -310,14 +311,8 @@ fn looks_like_quote(c: char) -> bool {
     //   ?c wdt:P4213 ?u.
     // }
     match c {
-        '\u{2033}' | 
-        '\u{02BA}' |
-        '\u{02DD}' |
-        '\u{030B}' |
-        '\u{030E}' |
-        '\u{05F4}' |
-        '\u{201C}' |
-        '\u{201D}' => true,
+        '\u{2033}' | '\u{02BA}' | '\u{02DD}' | '\u{030B}' | '\u{030E}' | '\u{05F4}'
+        | '\u{201C}' | '\u{201D}' => true,
         _ => false,
     }
 }