Replace ASCII control chars with Unicode Control Pictures

```
error: bare CR not allowed in doc-comment
  --> $DIR/lex-bare-cr-string-literal-doc-comment.rs:3:32
   |
LL | /// doc comment with bare CR: '␍'
   |                                ^
```

Author: estebank

compiler/rustc_errors/src/emitter.rs

@@ -677,10 +677,7 @@ impl HumanEmitter {
             .skip(left)
             .take_while(|ch| {
                 // Make sure that the trimming on the right will fall within the terminal width.
-                // FIXME: `unicode_width` sometimes disagrees with terminals on how wide a `char`
-                // is. For now, just accept that sometimes the code line will be longer than
-                // desired.
-                let next = unicode_width::UnicodeWidthChar::width(*ch).unwrap_or(1);
+                let next = char_width(*ch);
                 if taken + next > right - left {
                     return false;
                 }
@@ -742,11 +739,7 @@ impl HumanEmitter {
         let left = margin.left(source_string.len());
 
         // Account for unicode characters of width !=0 that were removed.
-        let left = source_string
-            .chars()
-            .take(left)
-            .map(|ch| unicode_width::UnicodeWidthChar::width(ch).unwrap_or(1))
-            .sum();
+        let left = source_string.chars().take(left).map(|ch| char_width(ch)).sum();
 
         self.draw_line(
             buffer,
@@ -2039,7 +2032,7 @@ impl HumanEmitter {
                     let sub_len: usize =
                         if is_whitespace_addition { &part.snippet } else { part.snippet.trim() }
                             .chars()
-                            .map(|ch| unicode_width::UnicodeWidthChar::width(ch).unwrap_or(1))
+                            .map(|ch| char_width(ch))
                             .sum();
 
                     let offset: isize = offsets
@@ -2076,11 +2069,8 @@ impl HumanEmitter {
                     }
 
                     // length of the code after substitution
-                    let full_sub_len = part
-                        .snippet
-                        .chars()
-                        .map(|ch| unicode_width::UnicodeWidthChar::width(ch).unwrap_or(1))
-                        .sum::<usize>() as isize;
+                    let full_sub_len =
+                        part.snippet.chars().map(|ch| char_width(ch)).sum::<usize>() as isize;
 
                     // length of the code to be substituted
                     let snippet_len = span_end_pos as isize - span_start_pos as isize;
@@ -2580,6 +2570,40 @@ const OUTPUT_REPLACEMENTS: &[(char, &str)] = &[
     ('\u{2068}', ""),
     ('\u{202C}', ""),
     ('\u{2069}', ""),
+    // In terminals without Unicode support the following will be garbled, but in *all* terminals
+    // the underlying codepoint will be as well. We could gate this replacement behind a "unicode
+    // support" gate.
+    ('\u{0000}', "␀"),
+    ('\u{0001}', "␁"),
+    ('\u{0002}', "␂"),
+    ('\u{0003}', "␃"),
+    ('\u{0004}', "␄"),
+    ('\u{0005}', "␅"),
+    ('\u{0006}', "␆"),
+    ('\u{0007}', "␇"),
+    ('\u{0008}', "␈"),
+    ('\u{000B}', "␋"),
+    ('\u{000C}', "␌"),
+    ('\u{000D}', "␍"),
+    ('\u{000E}', "␎"),
+    ('\u{000F}', "␏"),
+    ('\u{0010}', "␐"),
+    ('\u{0011}', "␑"),
+    ('\u{0012}', "␒"),
+    ('\u{0013}', "␓"),
+    ('\u{0014}', "␔"),
+    ('\u{0015}', "␕"),
+    ('\u{0016}', "␖"),
+    ('\u{0017}', "␗"),
+    ('\u{0018}', "␘"),
+    ('\u{0019}', "␙"),
+    ('\u{001A}', "␚"),
+    ('\u{001B}', "␛"),
+    ('\u{001C}', "␜"),
+    ('\u{001D}', "␝"),
+    ('\u{001E}', "␞"),
+    ('\u{001F}', "␟"),
+    ('\u{007F}', "␡"),
 ];
 
 fn normalize_whitespace(str: &str) -> String {
@@ -2590,6 +2614,21 @@ fn normalize_whitespace(str: &str) -> String {
     s
 }
 
+fn char_width(ch: char) -> usize {
+    // FIXME: `unicode_width` sometimes disagrees with terminals on how wide a `char` is. For now,
+    // just accept that sometimes the code line will be longer than desired.
+    match ch {
+        '\t' => 4,
+        '\u{0000}' | '\u{0001}' | '\u{0002}' | '\u{0003}' | '\u{0004}' | '\u{0005}'
+        | '\u{0006}' | '\u{0007}' | '\u{0008}' | '\u{000B}' | '\u{000C}' | '\u{000D}'
+        | '\u{000E}' | '\u{000F}' | '\u{0010}' | '\u{0011}' | '\u{0012}' | '\u{0013}'
+        | '\u{0014}' | '\u{0015}' | '\u{0016}' | '\u{0017}' | '\u{0018}' | '\u{0019}'
+        | '\u{001A}' | '\u{001B}' | '\u{001C}' | '\u{001D}' | '\u{001E}' | '\u{001F}'
+        | '\u{007F}' => 1,
+        _ => unicode_width::UnicodeWidthChar::width(ch).unwrap_or(1),
+    }
+}
+
 fn draw_col_separator(buffer: &mut StyledBuffer, line: usize, col: usize) {
     buffer.puts(line, col, "| ", Style::LineNumber);
 }

tests/ui/lexer/lex-bare-cr-string-literal-doc-comment.stderr

@@ -1,31 +1,31 @@
 error: bare CR not allowed in doc-comment
   --> $DIR/lex-bare-cr-string-literal-doc-comment.rs:3:32
    |
-LL | /// doc comment with bare CR: ''
+LL | /// doc comment with bare CR: '␍'
    |                                ^
 
 error: bare CR not allowed in block doc-comment
   --> $DIR/lex-bare-cr-string-literal-doc-comment.rs:7:38
    |
-LL | /** block doc comment with bare CR: '' */
+LL | /** block doc comment with bare CR: '␍' */
    |                                      ^
 
 error: bare CR not allowed in doc-comment
   --> $DIR/lex-bare-cr-string-literal-doc-comment.rs:12:36
    |
-LL |     //! doc comment with bare CR: ''
+LL |     //! doc comment with bare CR: '␍'
    |                                    ^
 
 error: bare CR not allowed in block doc-comment
   --> $DIR/lex-bare-cr-string-literal-doc-comment.rs:15:42
    |
-LL |     /*! block doc comment with bare CR: '' */
+LL |     /*! block doc comment with bare CR: '␍' */
    |                                          ^
 
 error: bare CR not allowed in string, use `\r` instead
   --> $DIR/lex-bare-cr-string-literal-doc-comment.rs:19:18
    |
-LL |     let _s = "foobar";
+LL |     let _s = "foo␍bar";
    |                  ^
    |
 help: escape the character
@@ -36,13 +36,13 @@ LL |     let _s = "foo\rbar";
 error: bare CR not allowed in raw string
   --> $DIR/lex-bare-cr-string-literal-doc-comment.rs:22:19
    |
-LL |     let _s = r"barfoo";
+LL |     let _s = r"bar␍foo";
    |                   ^
 
 error: unknown character escape: `\r`
   --> $DIR/lex-bare-cr-string-literal-doc-comment.rs:25:19
    |
-LL |     let _s = "foo\bar";
+LL |     let _s = "foo\␍bar";
    |                   ^ unknown character escape
    |
    = help: this is an isolated carriage return; consider checking your editor and version control settings

tests/ui/parser/bad-char-literals.rs

@@ -25,16 +25,27 @@ LL |     '\n';
 error: character constant must be escaped: `\r`
   --> $DIR/bad-char-literals.rs:15:6
    |
-LL |     '';
+LL |     '␍';
    |      ^
    |
 help: escape the character
    |
 LL |     '\r';
    |      ++
 
+error: character literal may only contain one codepoint
+  --> $DIR/bad-char-literals.rs:18:5
+   |
+LL |     '-␀-';
+   |     ^^^^
+   |
+help: if you meant to write a string literal, use double quotes
+   |
+LL |     "-␀-";
+   |     ~  ~
+
 error: character constant must be escaped: `\t`
-  --> $DIR/bad-char-literals.rs:18:6
+  --> $DIR/bad-char-literals.rs:21:6
    |
 LL |     '    ';
    |      ^^^^
@@ -44,5 +55,5 @@ help: escape the character
 LL |     '\t';
    |      ++
 
-error: aborting due to 4 previous errors
+error: aborting due to 5 previous errors
 

tests/ui/parser/bad-char-literals.stderr

@@ -1,7 +1,7 @@
 error: bare CR not allowed in raw string
   --> $DIR/raw-byte-string-literals.rs:4:9
    |
-LL |     br"a";
+LL |     br"a␍";
    |         ^
 
 error: non-ASCII character in raw byte string literal

tests/ui/parser/issues/issue-66473.stderr

@@ -1,19 +1,19 @@
 error: bare CR not allowed in doc-comment
   --> $DIR/several-carriage-returns-in-doc-comment.rs:6:12
    |
-LL | /// This doc comment contains three isolated `\r` symbols
+LL | /// This do␍c comment contains ␍three isolated `\r`␍ symbols
    |            ^
 
 error: bare CR not allowed in doc-comment
   --> $DIR/several-carriage-returns-in-doc-comment.rs:6:32
    |
-LL | /// This doc comment contains three isolated `\r` symbols
+LL | /// This do␍c comment contains ␍three isolated `\r`␍ symbols
    |                               ^
 
 error: bare CR not allowed in doc-comment
   --> $DIR/several-carriage-returns-in-doc-comment.rs:6:52
    |
-LL | /// This doc comment contains three isolated `\r` symbols
+LL | /// This do␍c comment contains ␍three isolated `\r`␍ symbols
    |                                                  ^
 
 error: aborting due to 3 previous errors

tests/ui/parser/issues/issue-68629.stderr

@@ -1,7 +1,7 @@
 error: unknown character escape: `\r`
   --> $DIR/trailing-carriage-return-in-string.rs:10:25
    |
-LL |     let bad = "This is \ a test";
+LL |     let bad = "This is \␍ a test";
    |                         ^ unknown character escape
    |
    = help: this is an isolated carriage return; consider checking your editor and version control settings

tests/ui/parser/issues/issue-68730.stderr

@@ -22,7 +22,7 @@ warning: whitespace symbol '\u{c}' is not skipped
    |
 LL |       let s = b"a\
    |  ________________^
-LL | |     b";
+LL | |     ␌b";
    | |    ^- whitespace symbol '\u{c}' is not skipped
    | |____|
    |