Rollup merge of rust-lang#105076 - mina86:a, r=scottmcm

Refactor core::char::EscapeDefault and co. structures

Change core::char::{EscapeUnicode, EscapeDefault and EscapeDebug}
structures from using a state machine to computing escaped sequence
upfront and during iteration just going through the characters.

This is arguably simpler since it’s easier to think about having
a buffer and start..end range to iterate over rather than thinking
about a state machine.

This also harmonises implementation of aforementioned iterators and
core::ascii::EscapeDefault struct.  This is done by introducing a new
helper EscapeIterInner struct which holds the buffer and offers simple
methods for iterating over range.

As a side effect, this probably optimises Display implementation for
those types since rather than calling write_char repeatedly, write_str
is invoked once.  On 64-bit platforms, it also reduces size of some of
the structs:

    | Struct                     | Before | After |
    |----------------------------+--------+-------+
    | core::char::EscapeUnicode  |     16 |    12 |
    | core::char::EscapeDefault  |     16 |    12 |
    | core::char::EscapeDebug    |     16 |    16 |

My ulterior motive and reason why I started looking into this is
addition of as_str method to the iterators.  With this change this
will became trivial.  It’s also going to be trivial to implement
DoubleEndedIterator if that’s ever desired.

Author: Dylan-DPC

library/core/src/ascii.rs

@@ -9,10 +9,10 @@
 
 #![stable(feature = "core_ascii", since = "1.26.0")]
 
+use crate::escape;
 use crate::fmt;
 use crate::iter::FusedIterator;
-use crate::ops::Range;
-use crate::str::from_utf8_unchecked;
+use crate::num::NonZeroUsize;
 
 /// An iterator over the escaped version of a byte.
 ///
@@ -21,10 +21,7 @@ use crate::str::from_utf8_unchecked;
 #[must_use = "iterators are lazy and do nothing unless consumed"]
 #[stable(feature = "rust1", since = "1.0.0")]
 #[derive(Clone)]
-pub struct EscapeDefault {
-    range: Range<u8>,
-    data: [u8; 4],
-}
+pub struct EscapeDefault(escape::EscapeIterInner<4>);
 
 /// Returns an iterator that produces an escaped version of a `u8`.
 ///
@@ -90,21 +87,9 @@ pub struct EscapeDefault {
 /// ```
 #[stable(feature = "rust1", since = "1.0.0")]
 pub fn escape_default(c: u8) -> EscapeDefault {
-    let (data, len) = match c {
-        b'\t' => ([b'\\', b't', 0, 0], 2),
-        b'\r' => ([b'\\', b'r', 0, 0], 2),
-        b'\n' => ([b'\\', b'n', 0, 0], 2),
-        b'\\' => ([b'\\', b'\\', 0, 0], 2),
-        b'\'' => ([b'\\', b'\'', 0, 0], 2),
-        b'"' => ([b'\\', b'"', 0, 0], 2),
-        b'\x20'..=b'\x7e' => ([c, 0, 0, 0], 1),
-        _ => {
-            let hex_digits: &[u8; 16] = b"0123456789abcdef";
-            ([b'\\', b'x', hex_digits[(c >> 4) as usize], hex_digits[(c & 0xf) as usize]], 4)
-        }
-    };
-
-    return EscapeDefault { range: 0..len, data };
+    let mut data = [0; 4];
+    let range = escape::escape_ascii_into(&mut data, c);
+    EscapeDefault(escape::EscapeIterInner::new(data, range))
 }
 
 #[stable(feature = "rust1", since = "1.0.0")]
@@ -113,33 +98,59 @@ impl Iterator for EscapeDefault {
 
     #[inline]
     fn next(&mut self) -> Option<u8> {
-        self.range.next().map(|i| self.data[i as usize])
+        self.0.next()
     }
+
+    #[inline]
     fn size_hint(&self) -> (usize, Option<usize>) {
-        self.range.size_hint()
+        let n = self.0.len();
+        (n, Some(n))
+    }
+
+    #[inline]
+    fn count(self) -> usize {
+        self.0.len()
     }
+
+    #[inline]
     fn last(mut self) -> Option<u8> {
-        self.next_back()
+        self.0.next_back()
+    }
+
+    #[inline]
+    fn advance_by(&mut self, n: usize) -> Result<(), NonZeroUsize> {
+        self.0.advance_by(n)
     }
 }
+
 #[stable(feature = "rust1", since = "1.0.0")]
 impl DoubleEndedIterator for EscapeDefault {
+    #[inline]
     fn next_back(&mut self) -> Option<u8> {
-        self.range.next_back().map(|i| self.data[i as usize])
+        self.0.next_back()
+    }
+
+    #[inline]
+    fn advance_back_by(&mut self, n: usize) -> Result<(), NonZeroUsize> {
+        self.0.advance_back_by(n)
     }
 }
+
 #[stable(feature = "rust1", since = "1.0.0")]
-impl ExactSizeIterator for EscapeDefault {}
+impl ExactSizeIterator for EscapeDefault {
+    #[inline]
+    fn len(&self) -> usize {
+        self.0.len()
+    }
+}
+
 #[stable(feature = "fused", since = "1.26.0")]
 impl FusedIterator for EscapeDefault {}
 
 #[stable(feature = "ascii_escape_display", since = "1.39.0")]
 impl fmt::Display for EscapeDefault {
     fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
-        // SAFETY: ok because `escape_default` created only valid utf-8 data
-        f.write_str(unsafe {
-            from_utf8_unchecked(&self.data[(self.range.start as usize)..(self.range.end as usize)])
-        })
+        f.write_str(self.0.as_str())
     }
 }
 

library/core/src/char/methods.rs

@@ -380,20 +380,7 @@ impl char {
     #[stable(feature = "rust1", since = "1.0.0")]
     #[inline]
     pub fn escape_unicode(self) -> EscapeUnicode {
-        let c = self as u32;
-
-        // or-ing 1 ensures that for c==0 the code computes that one
-        // digit should be printed and (which is the same) avoids the
-        // (31 - 32) underflow
-        let msb = 31 - (c | 1).leading_zeros();
-
-        // the index of the most significant hex digit
-        let ms_hex_digit = msb / 4;
-        EscapeUnicode {
-            c: self,
-            state: EscapeUnicodeState::Backslash,
-            hex_digit_idx: ms_hex_digit as usize,
-        }
+        EscapeUnicode::new(self)
     }
 
     /// An extended version of `escape_debug` that optionally permits escaping
@@ -403,21 +390,20 @@ impl char {
     /// characters, and double quotes in strings.
     #[inline]
     pub(crate) fn escape_debug_ext(self, args: EscapeDebugExtArgs) -> EscapeDebug {
-        let init_state = match self {
-            '\0' => EscapeDefaultState::Backslash('0'),
-            '\t' => EscapeDefaultState::Backslash('t'),
-            '\r' => EscapeDefaultState::Backslash('r'),
-            '\n' => EscapeDefaultState::Backslash('n'),
-            '\\' => EscapeDefaultState::Backslash(self),
-            '"' if args.escape_double_quote => EscapeDefaultState::Backslash(self),
-            '\'' if args.escape_single_quote => EscapeDefaultState::Backslash(self),
+        match self {
+            '\0' => EscapeDebug::backslash(b'0'),
+            '\t' => EscapeDebug::backslash(b't'),
+            '\r' => EscapeDebug::backslash(b'r'),
+            '\n' => EscapeDebug::backslash(b'n'),
+            '\\' => EscapeDebug::backslash(b'\\'),
+            '"' if args.escape_double_quote => EscapeDebug::backslash(b'"'),
+            '\'' if args.escape_single_quote => EscapeDebug::backslash(b'\''),
             _ if args.escape_grapheme_extended && self.is_grapheme_extended() => {
-                EscapeDefaultState::Unicode(self.escape_unicode())
+                EscapeDebug::from_unicode(self.escape_unicode())
             }
-            _ if is_printable(self) => EscapeDefaultState::Char(self),
-            _ => EscapeDefaultState::Unicode(self.escape_unicode()),
-        };
-        EscapeDebug(EscapeDefault { state: init_state })
+            _ if is_printable(self) => EscapeDebug::printable(self),
+            _ => EscapeDebug::from_unicode(self.escape_unicode()),
+        }
     }
 
     /// Returns an iterator that yields the literal escape code of a character
@@ -515,15 +501,14 @@ impl char {
     #[stable(feature = "rust1", since = "1.0.0")]
     #[inline]
     pub fn escape_default(self) -> EscapeDefault {
-        let init_state = match self {
-            '\t' => EscapeDefaultState::Backslash('t'),
-            '\r' => EscapeDefaultState::Backslash('r'),
-            '\n' => EscapeDefaultState::Backslash('n'),
-            '\\' | '\'' | '"' => EscapeDefaultState::Backslash(self),
-            '\x20'..='\x7e' => EscapeDefaultState::Char(self),
-            _ => EscapeDefaultState::Unicode(self.escape_unicode()),
-        };
-        EscapeDefault { state: init_state }
+        match self {
+            '\t' => EscapeDefault::backslash(b't'),
+            '\r' => EscapeDefault::backslash(b'r'),
+            '\n' => EscapeDefault::backslash(b'n'),
+            '\\' | '\'' | '"' => EscapeDefault::backslash(self as u8),
+            '\x20'..='\x7e' => EscapeDefault::printable(self as u8),
+            _ => EscapeDefault::from_unicode(self.escape_unicode()),
+        }
     }
 
     /// Returns the number of bytes this `char` would need if encoded in UTF-8.