Implement int_format_into feature

Author: GuillaumeGomez

library/alloc/src/string.rs

@@ -2875,7 +2875,8 @@ macro_rules! impl_to_string {
                     out = String::with_capacity(SIZE);
                 }
 
-                out.push_str(self.unsigned_abs()._fmt(&mut buf));
+                // SAFETY: `buf` is always big enough to contain all the digits.
+                unsafe { out.push_str(self.unsigned_abs()._fmt(&mut buf)); }
                 out
             }
         }
@@ -2887,7 +2888,8 @@ macro_rules! impl_to_string {
                 const SIZE: usize = $unsigned::MAX.ilog10() as usize + 1;
                 let mut buf = [core::mem::MaybeUninit::<u8>::uninit(); SIZE];
 
-                self._fmt(&mut buf).to_string()
+                // SAFETY: `buf` is always big enough to contain all the digits.
+                unsafe { self._fmt(&mut buf).to_string() }
             }
         }
         )*

library/core/src/fmt/mod.rs

@@ -15,6 +15,7 @@ mod float;
 #[cfg(no_fp_fmt_parse)]
 mod nofloat;
 mod num;
+mod num_buffer;
 mod rt;
 
 #[stable(feature = "fmt_flags_align", since = "1.28.0")]
@@ -33,6 +34,9 @@ pub enum Alignment {
     Center,
 }
 
+#[unstable(feature = "int_format_into", issue = "138215")]
+pub use num_buffer::{NumBuffer, NumBufferTrait};
+
 #[stable(feature = "debug_builders", since = "1.2.0")]
 pub use self::builders::{DebugList, DebugMap, DebugSet, DebugStruct, DebugTuple};
 #[unstable(feature = "debug_closure_helpers", issue = "117729")]

library/core/src/fmt/num.rs

@@ -1,5 +1,6 @@
 //! Integer and floating-point number formatting
 
+use crate::fmt::NumBuffer;
 use crate::mem::MaybeUninit;
 use crate::num::fmt as numfmt;
 use crate::ops::{Div, Rem, Sub};
@@ -199,6 +200,20 @@ static DEC_DIGITS_LUT: &[u8; 200] = b"\
       6061626364656667686970717273747576777879\
       8081828384858687888990919293949596979899";
 
+/// This function converts a slice of ascii characters into a `&str` starting from `offset`.
+///
+/// # Safety
+///
+/// `buf` content starting from `offset` index MUST BE initialized and MUST BE ascii
+/// characters.
+unsafe fn slice_buffer_to_str(buf: &[MaybeUninit<u8>], offset: usize) -> &str {
+    // SAFETY: `offset` is always included between 0 and `buf`'s length.
+    let written = unsafe { buf.get_unchecked(offset..) };
+    // SAFETY: (`assume_init_ref`) All buf content since offset is set.
+    // SAFETY: (`from_utf8_unchecked`) Writes use ASCII from the lookup table exclusively.
+    unsafe { str::from_utf8_unchecked(written.assume_init_ref()) }
+}
+
 macro_rules! impl_Display {
     ($($signed:ident, $unsigned:ident,)* ; as $u:ident via $conv_fn:ident named $gen_name:ident) => {
 
@@ -212,7 +227,8 @@ macro_rules! impl_Display {
                     // Buffer decimals for $unsigned with right alignment.
                     let mut buf = [MaybeUninit::<u8>::uninit(); MAX_DEC_N];
 
-                    f.pad_integral(true, "", self._fmt(&mut buf))
+                    // SAFETY: `buf` is always big enough to contain all the digits.
+                    unsafe { f.pad_integral(true, "", self._fmt(&mut buf)) }
                 }
                 #[cfg(feature = "optimize_for_size")]
                 {
@@ -230,7 +246,8 @@ macro_rules! impl_Display {
                     // Buffer decimals for $unsigned with right alignment.
                     let mut buf = [MaybeUninit::<u8>::uninit(); MAX_DEC_N];
 
-                    f.pad_integral(*self >= 0, "", self.unsigned_abs()._fmt(&mut buf))
+                    // SAFETY: `buf` is always big enough to contain all the digits.
+                    unsafe { f.pad_integral(*self >= 0, "", self.unsigned_abs()._fmt(&mut buf)) }
                 }
                 #[cfg(feature = "optimize_for_size")]
                 {
@@ -247,7 +264,14 @@ macro_rules! impl_Display {
                 reason = "specialized method meant to only be used by `SpecToString` implementation",
                 issue = "none"
             )]
-            pub fn _fmt<'a>(self, buf: &'a mut [MaybeUninit::<u8>]) -> &'a str {
+            pub unsafe fn _fmt<'a>(self, buf: &'a mut [MaybeUninit::<u8>]) -> &'a str {
+                // SAFETY: `buf` will always be big enough to contain all digits.
+                let offset = unsafe { self._fmt_inner(buf) };
+                // SAFETY: Starting from `offset`, all elements of the slice have been set.
+                unsafe { slice_buffer_to_str(buf, offset) }
+            }
+
+            unsafe fn _fmt_inner(self, buf: &mut [MaybeUninit::<u8>]) -> usize {
                 // Count the number of bytes in buf that are not initialized.
                 let mut offset = buf.len();
                 // Consume the least-significant decimals from a working copy.
@@ -309,47 +333,123 @@ macro_rules! impl_Display {
                     // not used: remain = 0;
                 }
 
-                // SAFETY: All buf content since offset is set.
-                let written = unsafe { buf.get_unchecked(offset..) };
-                // SAFETY: Writes use ASCII from the lookup table exclusively.
+                offset
+            }
+        }
+
+        impl $signed {
+            /// Allows users to write an integer (in signed decimal format) into a variable `buf` of
+            /// type [`NumBuffer`] that is passed by the caller by mutable reference.
+            ///
+            /// # Examples
+            ///
+            /// ```
+            /// #![feature(int_format_into)]
+            /// use core::fmt::NumBuffer;
+            ///
+            #[doc = concat!("let n = 0", stringify!($signed), ";")]
+            /// let mut buf = NumBuffer::new();
+            /// assert_eq!(n.format_into(&mut buf), "0");
+            ///
+            #[doc = concat!("let n1 = 32", stringify!($signed), ";")]
+            /// assert_eq!(n1.format_into(&mut buf), "32");
+            ///
+            #[doc = concat!("let n2 = ", stringify!($signed::MAX), ";")]
+            #[doc = concat!("assert_eq!(n2.format_into(&mut buf), ", stringify!($signed::MAX), ".to_string());")]
+            /// ```
+            #[unstable(feature = "int_format_into", issue = "138215")]
+            pub fn format_into(self, buf: &mut NumBuffer<Self>) -> &str {
+                let mut offset;
+
+                #[cfg(not(feature = "optimize_for_size"))]
+                // SAFETY: `buf` will always be big enough to contain all digits.
+                unsafe {
+                    offset = self.unsigned_abs()._fmt_inner(&mut buf.buf);
+                }
+                #[cfg(feature = "optimize_for_size")]
+                {
+                    offset = _inner_slow_integer_to_str(self.unsigned_abs().$conv_fn(), &mut buf.buf);
+                }
+                // Only difference between signed and unsigned are these 4 lines.
+                if self < 0 {
+                    offset -= 1;
+                    buf.buf[offset].write(b'-');
+                }
+                // SAFETY: Starting from `offset`, all elements of the slice have been set.
+                unsafe { slice_buffer_to_str(&buf.buf, offset) }
+            }
+        }
+
+        impl $unsigned {
+            /// Allows users to write an integer (in signed decimal format) into a variable `buf` of
+            /// type [`NumBuffer`] that is passed by the caller by mutable reference.
+            ///
+            /// # Examples
+            ///
+            /// ```
+            /// #![feature(int_format_into)]
+            /// use core::fmt::NumBuffer;
+            ///
+            #[doc = concat!("let n = 0", stringify!($unsigned), ";")]
+            /// let mut buf = NumBuffer::new();
+            /// assert_eq!(n.format_into(&mut buf), "0");
+            ///
+            #[doc = concat!("let n1 = 32", stringify!($unsigned), ";")]
+            /// assert_eq!(n1.format_into(&mut buf), "32");
+            ///
+            #[doc = concat!("let n2 = ", stringify!($unsigned::MAX), ";")]
+            #[doc = concat!("assert_eq!(n2.format_into(&mut buf), ", stringify!($unsigned::MAX), ".to_string());")]
+            /// ```
+            #[unstable(feature = "int_format_into", issue = "138215")]
+            pub fn format_into(self, buf: &mut NumBuffer<Self>) -> &str {
+                let offset;
+
+                #[cfg(not(feature = "optimize_for_size"))]
+                // SAFETY: `buf` will always be big enough to contain all digits.
                 unsafe {
-                    str::from_utf8_unchecked(slice::from_raw_parts(
-                          MaybeUninit::slice_as_ptr(written),
-                          written.len(),
-                    ))
+                    offset = self._fmt_inner(&mut buf.buf);
                 }
+                #[cfg(feature = "optimize_for_size")]
+                {
+                    offset = _inner_slow_integer_to_str(self.$conv_fn(), &mut buf.buf);
+                }
+                // SAFETY: Starting from `offset`, all elements of the slice have been set.
+                unsafe { slice_buffer_to_str(&buf.buf, offset) }
             }
-        })*
+        }
+
+
+        )*
 
         #[cfg(feature = "optimize_for_size")]
-        fn $gen_name(mut n: $u, is_nonnegative: bool, f: &mut fmt::Formatter<'_>) -> fmt::Result {
-            const MAX_DEC_N: usize = $u::MAX.ilog10() as usize + 1;
-            let mut buf = [MaybeUninit::<u8>::uninit(); MAX_DEC_N];
-            let mut curr = MAX_DEC_N;
-            let buf_ptr = MaybeUninit::slice_as_mut_ptr(&mut buf);
+        fn _inner_slow_integer_to_str(mut n: $u, buf: &mut [MaybeUninit::<u8>]) -> usize {
+            let mut curr = buf.len();
 
             // SAFETY: To show that it's OK to copy into `buf_ptr`, notice that at the beginning
             // `curr == buf.len() == 39 > log(n)` since `n < 2^128 < 10^39`, and at
             // each step this is kept the same as `n` is divided. Since `n` is always
             // non-negative, this means that `curr > 0` so `buf_ptr[curr..curr + 1]`
             // is safe to access.
-            unsafe {
-                loop {
-                    curr -= 1;
-                    buf_ptr.add(curr).write((n % 10) as u8 + b'0');
-                    n /= 10;
+            loop {
+                curr -= 1;
+                buf[curr].write((n % 10) as u8 + b'0');
+                n /= 10;
 
-                    if n == 0 {
-                        break;
-                    }
+                if n == 0 {
+                    break;
                 }
             }
+            curr
+        }
 
-            // SAFETY: `curr` > 0 (since we made `buf` large enough), and all the chars are valid UTF-8
-            let buf_slice = unsafe {
-                str::from_utf8_unchecked(
-                    slice::from_raw_parts(buf_ptr.add(curr), buf.len() - curr))
-            };
+        #[cfg(feature = "optimize_for_size")]
+        fn $gen_name(n: $u, is_nonnegative: bool, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+            const MAX_DEC_N: usize = $u::MAX.ilog(10) as usize + 1;
+            let mut buf = [MaybeUninit::<u8>::uninit(); MAX_DEC_N];
+
+            let offset = _inner_slow_integer_to_str(n, &mut buf);
+            // SAFETY: Starting from `offset`, all elements of the slice have been set.
+            let buf_slice = unsafe { slice_buffer_to_str(&buf, offset) };
             f.pad_integral(is_nonnegative, "", buf_slice)
         }
     };
@@ -572,7 +672,8 @@ impl fmt::Display for u128 {
     fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
         let mut buf = [MaybeUninit::<u8>::uninit(); U128_MAX_DEC_N];
 
-        f.pad_integral(true, "", self._fmt(&mut buf))
+        // SAFETY: `buf` is always big enough to contain all the digits.
+        unsafe { f.pad_integral(true, "", self._fmt(&mut buf)) }
     }
 }
 
@@ -584,7 +685,8 @@ impl fmt::Display for i128 {
         let mut buf = [MaybeUninit::<u8>::uninit(); U128_MAX_DEC_N];
 
         let is_nonnegative = *self >= 0;
-        f.pad_integral(is_nonnegative, "", self.unsigned_abs()._fmt(&mut buf))
+        // SAFETY: `buf` is always big enough to contain all the digits.
+        unsafe { f.pad_integral(is_nonnegative, "", self.unsigned_abs()._fmt(&mut buf)) }
     }
 }
 
@@ -597,13 +699,21 @@ impl u128 {
         reason = "specialized method meant to only be used by `SpecToString` implementation",
         issue = "none"
     )]
-    pub fn _fmt<'a>(self, buf: &'a mut [MaybeUninit<u8>]) -> &'a str {
+    pub unsafe fn _fmt<'a>(self, buf: &'a mut [MaybeUninit<u8>]) -> &'a str {
+        // SAFETY: `buf` will always be big enough to contain all digits.
+        let offset = unsafe { self._fmt_inner(buf) };
+        // SAFETY: Starting from `offset`, all elements of the slice have been set.
+        unsafe { slice_buffer_to_str(buf, offset) }
+    }
+
+    unsafe fn _fmt_inner(self, buf: &mut [MaybeUninit<u8>]) -> usize {
         // Optimize common-case zero, which would also need special treatment due to
         // its "leading" zero.
         if self == 0 {
-            return "0";
+            let offset = buf.len() - 1;
+            buf[offset].write(b'0');
+            return offset;
         }
-
         // Take the 16 least-significant decimals.
         let (quot_1e16, mod_1e16) = div_rem_1e16(self);
         let (mut remain, mut offset) = if quot_1e16 == 0 {
@@ -677,16 +787,86 @@ impl u128 {
             buf[offset].write(DEC_DIGITS_LUT[last * 2 + 1]);
             // not used: remain = 0;
         }
+        offset
+    }
 
-        // SAFETY: All buf content since offset is set.
-        let written = unsafe { buf.get_unchecked(offset..) };
-        // SAFETY: Writes use ASCII from the lookup table exclusively.
-        unsafe {
-            str::from_utf8_unchecked(slice::from_raw_parts(
-                MaybeUninit::slice_as_ptr(written),
-                written.len(),
-            ))
+    /// Allows users to write an integer (in signed decimal format) into a variable `buf` of
+    /// type [`NumBuffer`] that is passed by the caller by mutable reference.
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// #![feature(int_format_into)]
+    /// use core::fmt::NumBuffer;
+    ///
+    /// let n = 0u128;
+    /// let mut buf = NumBuffer::new();
+    /// assert_eq!(n.format_into(&mut buf), "0");
+    ///
+    /// let n1 = 32u128;
+    /// let mut buf1 = NumBuffer::new();
+    /// assert_eq!(n1.format_into(&mut buf1), "32");
+    ///
+    /// let n2 = u128::MAX;
+    /// let mut buf2 = NumBuffer::new();
+    /// assert_eq!(n2.format_into(&mut buf2), u128::MAX.to_string());
+    /// ```
+    #[unstable(feature = "int_format_into", issue = "138215")]
+    pub fn format_into(self, buf: &mut NumBuffer<Self>) -> &str {
+        let diff = buf.capacity() - U128_MAX_DEC_N;
+        // FIXME: Once const generics are better, use `NumberBufferTrait::BUF_SIZE` as generic const
+        // for `fmt_u128_inner`.
+        //
+        // In the meantime, we have to use a slice starting at index 1 and add 1 to the returned
+        // offset to ensure the number is correctly generated at the end of the buffer.
+        // SAFETY: `diff` will always be between 0 and its initial value.
+        unsafe { self._fmt(buf.buf.get_unchecked_mut(diff..)) }
+    }
+}
+
+impl i128 {
+    /// Allows users to write an integer (in signed decimal format) into a variable `buf` of
+    /// type [`NumBuffer`] that is passed by the caller by mutable reference.
+    ///
+    /// # Examples
+    ///
+    /// ```
+    /// #![feature(int_format_into)]
+    /// use core::fmt::NumBuffer;
+    ///
+    /// let n = 0i128;
+    /// let mut buf = NumBuffer::new();
+    /// assert_eq!(n.format_into(&mut buf), "0");
+    ///
+    /// let n1 = i128::MIN;
+    /// assert_eq!(n1.format_into(&mut buf), i128::MIN.to_string());
+    ///
+    /// let n2 = i128::MAX;
+    /// assert_eq!(n2.format_into(&mut buf), i128::MAX.to_string());
+    /// ```
+    #[unstable(feature = "int_format_into", issue = "138215")]
+    pub fn format_into(self, buf: &mut NumBuffer<Self>) -> &str {
+        let diff = buf.capacity() - U128_MAX_DEC_N;
+        // FIXME: Once const generics are better, use `NumberBufferTrait::BUF_SIZE` as generic const
+        // for `fmt_u128_inner`.
+        //
+        // In the meantime, we have to use a slice starting at index 1 and add 1 to the returned
+        // offset to ensure the number is correctly generated at the end of the buffer.
+        let mut offset =
+            // SAFETY: `buf` will always be big enough to contain all digits.
+            unsafe { self.unsigned_abs()._fmt_inner(buf.buf.get_unchecked_mut(diff..)) };
+        // We put back the offset at the right position.
+        offset += diff;
+        // Only difference between signed and unsigned are these 4 lines.
+        if self < 0 {
+            offset -= 1;
+            // SAFETY: `buf` will always be big enough to contain all digits plus the minus sign.
+            unsafe {
+                buf.buf.get_unchecked_mut(offset).write(b'-');
+            }
         }
+        // SAFETY: Starting from `offset`, all elements of the slice have been set.
+        unsafe { slice_buffer_to_str(&buf.buf, offset) }
     }
 }