Auto merge of #34485 - tbu-:pr_unicode_debug_str, r=alexcrichton

Escape fewer Unicode codepoints in `Debug` impl of `str`

Use the same procedure as Python to determine whether a character is
printable, described in [PEP 3138]. In particular, this means that the
following character classes are escaped:

- Cc (Other, Control)
- Cf (Other, Format)
- Cs (Other, Surrogate), even though they can't appear in Rust strings
- Co (Other, Private Use)
- Cn (Other, Not Assigned)
- Zl (Separator, Line)
- Zp (Separator, Paragraph)
- Zs (Separator, Space), except for the ASCII space `' '` `0x20`

This allows for user-friendly inspection of strings that are not
English (e.g. compare `"\u{e9}\u{e8}\u{ea}"` to `"éèê"`).

Fixes #34318.
CC #34422.

[PEP 3138]: https://www.python.org/dev/peps/pep-3138/

2 files changed, 55 insertions, 1 deletions

diff --git a/src/libcoretest/char.rs b/src/libcoretest/char.rs
index c8906fed3d2..4632419336d 100644
--- a/src/libcoretest/char.rs
+++ b/src/libcoretest/char.rs
@@ -124,6 +124,49 @@ fn test_is_digit() {
 }
 
 #[test]
+fn test_escape_debug() {
+    fn string(c: char) -> String {
+        c.escape_debug().collect()
+    }
+    let s = string('\n');
+    assert_eq!(s, "\\n");
+    let s = string('\r');
+    assert_eq!(s, "\\r");
+    let s = string('\'');
+    assert_eq!(s, "\\'");
+    let s = string('"');
+    assert_eq!(s, "\\\"");
+    let s = string(' ');
+    assert_eq!(s, " ");
+    let s = string('a');
+    assert_eq!(s, "a");
+    let s = string('~');
+    assert_eq!(s, "~");
+    let s = string('é');
+    assert_eq!(s, "é");
+    let s = string('\x00');
+    assert_eq!(s, "\\u{0}");
+    let s = string('\x1f');
+    assert_eq!(s, "\\u{1f}");
+    let s = string('\x7f');
+    assert_eq!(s, "\\u{7f}");
+    let s = string('\u{80}');
+    assert_eq!(s, "\\u{80}");
+    let s = string('\u{ff}');
+    assert_eq!(s, "\u{ff}");
+    let s = string('\u{11b}');
+    assert_eq!(s, "\u{11b}");
+    let s = string('\u{1d4b6}');
+    assert_eq!(s, "\u{1d4b6}");
+    let s = string('\u{200b}'); // zero width space
+    assert_eq!(s, "\\u{200b}");
+    let s = string('\u{e000}'); // private use 1
+    assert_eq!(s, "\\u{e000}");
+    let s = string('\u{100000}'); // private use 2
+    assert_eq!(s, "\\u{100000}");
+}
+
+#[test]
 fn test_escape_default() {
     fn string(c: char) -> String {
         c.escape_default().collect()
@@ -142,18 +185,28 @@ fn test_escape_default() {
     assert_eq!(s, "a");
     let s = string('~');
     assert_eq!(s, "~");
+    let s = string('é');
+    assert_eq!(s, "\\u{e9}");
     let s = string('\x00');
     assert_eq!(s, "\\u{0}");
     let s = string('\x1f');
     assert_eq!(s, "\\u{1f}");
     let s = string('\x7f');
     assert_eq!(s, "\\u{7f}");
+    let s = string('\u{80}');
+    assert_eq!(s, "\\u{80}");
     let s = string('\u{ff}');
     assert_eq!(s, "\\u{ff}");
     let s = string('\u{11b}');
     assert_eq!(s, "\\u{11b}");
     let s = string('\u{1d4b6}');
     assert_eq!(s, "\\u{1d4b6}");
+    let s = string('\u{200b}'); // zero width space
+    assert_eq!(s, "\\u{200b}");
+    let s = string('\u{e000}'); // private use 1
+    assert_eq!(s, "\\u{e000}");
+    let s = string('\u{100000}'); // private use 2
+    assert_eq!(s, "\\u{100000}");
 }
 
 #[test]
diff --git a/src/libcoretest/lib.rs b/src/libcoretest/lib.rs
index 1ef2b58351f..ef0042808f9 100644
--- a/src/libcoretest/lib.rs
+++ b/src/libcoretest/lib.rs
@@ -14,6 +14,7 @@
 #![feature(borrow_state)]
 #![feature(box_syntax)]
 #![feature(cell_extras)]
+#![feature(char_escape_debug)]
 #![feature(const_fn)]
 #![feature(core_private_bignum)]
 #![feature(core_private_diy_float)]
@@ -29,10 +30,10 @@
 #![feature(slice_patterns)]
 #![feature(step_by)]
 #![feature(test)]
+#![feature(try_from)]
 #![feature(unboxed_closures)]
 #![feature(unicode)]
 #![feature(unique)]
-#![feature(try_from)]
 
 extern crate core;
 extern crate test;