Separate the unescape functions for string, byte string and C string, but avoid duplicating code via macro_rules.

Also plays with NonZero, since C strings cannot contain null bytes, which can be captured in the type system.

Author: hkBst

compiler/rustc_ast/src/util/literal.rs

@@ -3,7 +3,7 @@
 use std::{ascii, fmt, str};
 
 use literal_escaper::{
-    MixedUnit, Mode, byte_from_char, unescape_byte, unescape_char, unescape_mixed, unescape_unicode,
+    MixedUnit, unescape_byte, unescape_byte_str, unescape_char, unescape_cstr, unescape_str,
 };
 use rustc_span::{Span, Symbol, kw, sym};
 use tracing::debug;
@@ -87,11 +87,10 @@ impl LitKind {
                     // Force-inlining here is aggressive but the closure is
                     // called on every char in the string, so it can be hot in
                     // programs with many long strings containing escapes.
-                    unescape_unicode(
+                    unescape_str(
                         s,
-                        Mode::Str,
                         &mut #[inline(always)]
-                        |_, c| match c {
+                        |_, res| match res {
                             Ok(c) => buf.push(c),
                             Err(err) => {
                                 assert!(!err.is_fatal(), "failed to unescape string literal")
@@ -111,8 +110,8 @@ impl LitKind {
             token::ByteStr => {
                 let s = symbol.as_str();
                 let mut buf = Vec::with_capacity(s.len());
-                unescape_unicode(s, Mode::ByteStr, &mut |_, c| match c {
-                    Ok(c) => buf.push(byte_from_char(c)),
+                unescape_byte_str(s, &mut |_, res| match res {
+                    Ok(b) => buf.push(b),
                     Err(err) => {
                         assert!(!err.is_fatal(), "failed to unescape string literal")
                     }
@@ -128,11 +127,11 @@ impl LitKind {
             token::CStr => {
                 let s = symbol.as_str();
                 let mut buf = Vec::with_capacity(s.len());
-                unescape_mixed(s, Mode::CStr, &mut |_span, c| match c {
+                unescape_cstr(s, &mut |_span, c| match c {
                     Ok(MixedUnit::Char(c)) => {
-                        buf.extend_from_slice(c.encode_utf8(&mut [0; 4]).as_bytes())
+                        buf.extend_from_slice(c.get().encode_utf8(&mut [0; 4]).as_bytes())
                     }
-                    Ok(MixedUnit::HighByte(b)) => buf.push(b),
+                    Ok(MixedUnit::HighByte(b)) => buf.push(b.get()),
                     Err(err) => {
                         assert!(!err.is_fatal(), "failed to unescape C string literal")
                     }

compiler/rustc_parse/src/lexer/mod.rs

@@ -985,8 +985,8 @@ impl<'psess, 'src> Lexer<'psess, 'src> {
         prefix_len: u32,
         postfix_len: u32,
     ) -> (token::LitKind, Symbol) {
-        self.cook_common(kind, mode, start, end, prefix_len, postfix_len, |src, mode, callback| {
-            literal_escaper::unescape_mixed(src, mode, &mut |span, result| {
+        self.cook_common(kind, mode, start, end, prefix_len, postfix_len, |src, _mode, callback| {
+            literal_escaper::unescape_cstr(src, &mut |span, result| {
                 callback(span, result.map(drop))
             })
         })

library/core/src/num/niche_types.rs

@@ -131,6 +131,8 @@ define_valid_range_type! {
     pub struct NonZeroI32Inner(i32 as u32 in 1..=0xffff_ffff);
     pub struct NonZeroI64Inner(i64 as u64 in 1..=0xffffffff_ffffffff);
     pub struct NonZeroI128Inner(i128 as u128 in 1..=0xffffffffffffffff_ffffffffffffffff);
+
+    pub struct NonZeroCharInner(char as u32 in 1..=0x10ffff);
 }
 
 #[cfg(target_pointer_width = "16")]

library/core/src/num/nonzero.rs

@@ -79,6 +79,7 @@ impl_zeroable_primitive!(
     NonZeroI64Inner(i64),
     NonZeroI128Inner(i128),
     NonZeroIsizeInner(isize),
+    NonZeroCharInner(char),
 );
 
 /// A value that is known not to equal zero.