Merge branch 'master' into frewsxcv-osstr

Author: frewsxcv

configure

@@ -445,6 +445,7 @@ opt dist-host-only 0 "only install bins for the host architecture"
 opt inject-std-version 1 "inject the current compiler version of libstd into programs"
 opt llvm-version-check 1 "check if the LLVM version is supported, build anyway"
 opt codegen-tests 1 "run the src/test/codegen tests"
+opt save-analysis 0 "save API analysis data"
 opt option-checking 1 "complain about unrecognized options in this configure script"
 opt ninja 0 "build LLVM using the Ninja generator (for MSVC, requires building in the correct environment)"
 opt locked-deps 0 "force Cargo.lock to be up to date"

src/bootstrap/config.rs

@@ -74,6 +74,7 @@ pub struct Config {
     pub rustc_default_ar: Option<String>,
     pub rust_optimize_tests: bool,
     pub rust_debuginfo_tests: bool,
+    pub rust_save_analysis: bool,
     pub rust_dist_src: bool,
 
     pub build: String,
@@ -225,6 +226,7 @@ struct Rust {
     optimize_tests: Option<bool>,
     debuginfo_tests: Option<bool>,
     codegen_tests: Option<bool>,
+    save_analysis: Option<bool>,
 }
 
 /// TOML representation of how each build target is configured.
@@ -350,6 +352,7 @@ impl Config {
             set(&mut config.rust_optimize_tests, rust.optimize_tests);
             set(&mut config.rust_debuginfo_tests, rust.debuginfo_tests);
             set(&mut config.codegen_tests, rust.codegen_tests);
+            set(&mut config.rust_save_analysis, rust.save_analysis);
             set(&mut config.rust_rpath, rust.rpath);
             set(&mut config.debug_jemalloc, rust.debug_jemalloc);
             set(&mut config.use_jemalloc, rust.use_jemalloc);
@@ -457,6 +460,7 @@ impl Config {
                 ("LOCAL_REBUILD", self.local_rebuild),
                 ("NINJA", self.ninja),
                 ("CODEGEN_TESTS", self.codegen_tests),
+                ("SAVE_ANALYSIS", self.rust_save_analysis),
                 ("LOCKED_DEPS", self.locked_deps),
                 ("VENDOR", self.vendor),
                 ("FULL_BOOTSTRAP", self.full_bootstrap),

src/bootstrap/config.toml.example

@@ -234,6 +234,9 @@
 # saying that the FileCheck executable is missing, you may want to disable this.
 #codegen-tests = true
 
+# Flag indicating whether the API analysis data should be saved.
+#save-analysis = false
+
 # =============================================================================
 # Options for specific targets
 #

src/bootstrap/dist.rs

@@ -311,18 +311,14 @@ pub fn rust_src_location(build: &Build) -> PathBuf {
 
 /// Creates a tarball of save-analysis metadata, if available.
 pub fn analysis(build: &Build, compiler: &Compiler, target: &str) {
+    if !build.config.rust_save_analysis {
+        return
+    }
+
     println!("Dist analysis");
 
-    if build.config.channel != "nightly" {
-        println!("\tskipping - not on nightly channel");
-        return;
-    }
     if compiler.host != build.config.build {
-        println!("\tskipping - not a build host");
-        return
-    }
-    if compiler.stage != 2 {
-        println!("\tskipping - not stage2");
+        println!("\tskipping, not a build host");
         return
     }
 

src/bootstrap/install.rs

@@ -49,6 +49,10 @@ pub fn install(build: &Build, stage: u32, host: &str) {
         install_sh(&build, "docs", "rust-docs", stage, host, &prefix,
                    &docdir, &libdir, &mandir, &empty_dir);
     }
+    if build.config.rust_save_analysis {
+        install_sh(&build, "analysis", "rust-analysis", stage, host, &prefix,
+                   &docdir, &libdir, &mandir, &empty_dir);
+    }
     install_sh(&build, "std", "rust-std", stage, host, &prefix,
                &docdir, &libdir, &mandir, &empty_dir);
     install_sh(&build, "rustc", "rustc", stage, host, &prefix,

src/bootstrap/lib.rs

@@ -524,7 +524,7 @@ impl Build {
                  .env(format!("CFLAGS_{}", target), self.cflags(target).join(" "));
         }
 
-        if self.config.channel == "nightly" && compiler.is_final_stage(self) {
+        if self.config.rust_save_analysis && compiler.is_final_stage(self) {
             cargo.env("RUSTC_SAVE_ANALYSIS", "api".to_string());
         }
 

src/ci/run.sh

@@ -43,6 +43,7 @@ fi
 if [ "$DEPLOY$DEPLOY_ALT" != "" ]; then
   RUST_CONFIGURE_ARGS="$RUST_CONFIGURE_ARGS --release-channel=nightly"
   RUST_CONFIGURE_ARGS="$RUST_CONFIGURE_ARGS --enable-llvm-static-stdcpp"
+  RUST_CONFIGURE_ARGS="$RUST_CONFIGURE_ARGS --enable-save-analysis"
 
   if [ "$NO_LLVM_ASSERTIONS" = "1" ]; then
     RUST_CONFIGURE_ARGS="$RUST_CONFIGURE_ARGS --disable-llvm-assertions"

src/libcollections/string.rs

@@ -1974,6 +1974,22 @@ impl<'a> From<&'a str> for String {
     }
 }
 
+// note: test pulls in libstd, which causes errors here
+#[cfg(not(test))]
+#[stable(feature = "string_from_box", since = "1.17.0")]
+impl From<Box<str>> for String {
+    fn from(s: Box<str>) -> String {
+        s.into_string()
+    }
+}
+
+#[stable(feature = "box_from_str", since = "1.17.0")]
+impl Into<Box<str>> for String {
+    fn into(self) -> Box<str> {
+        self.into_boxed_str()
+    }
+}
+
 #[stable(feature = "string_from_cow_str", since = "1.14.0")]
 impl<'a> From<Cow<'a, str>> for String {
     fn from(s: Cow<'a, str>) -> String {

src/libcollections/vec.rs

@@ -1897,6 +1897,22 @@ impl<'a, T> From<Cow<'a, [T]>> for Vec<T> where [T]: ToOwned<Owned=Vec<T>> {
     }
 }
 
+// note: test pulls in libstd, which causes errors here
+#[cfg(not(test))]
+#[stable(feature = "vec_from_box", since = "1.17.0")]
+impl<T> From<Box<[T]>> for Vec<T> {
+    fn from(s: Box<[T]>) -> Vec<T> {
+        s.into_vec()
+    }
+}
+
+#[stable(feature = "box_from_vec", since = "1.17.0")]
+impl<T> Into<Box<[T]>> for Vec<T> {
+    fn into(self) -> Box<[T]> {
+        self.into_boxed_slice()
+    }
+}
+
 #[stable(feature = "rust1", since = "1.0.0")]
 impl<'a> From<&'a str> for Vec<u8> {
     fn from(s: &'a str) -> Vec<u8> {

src/libcollectionstest/lib.rs

@@ -28,6 +28,7 @@
 #![feature(test)]
 #![feature(unboxed_closures)]
 #![feature(unicode)]
+#![feature(utf8_error_error_len)]
 
 extern crate collections;
 extern crate test;

src/libcollectionstest/str.rs

@@ -540,6 +540,36 @@ fn from_utf8_mostly_ascii() {
     }
 }
 
+#[test]
+fn from_utf8_error() {
+    macro_rules! test {
+        ($input: expr, $expected_valid_up_to: expr, $expected_error_len: expr) => {
+            let error = from_utf8($input).unwrap_err();
+            assert_eq!(error.valid_up_to(), $expected_valid_up_to);
+            assert_eq!(error.error_len(), $expected_error_len);
+        }
+    }
+    test!(b"A\xC3\xA9 \xFF ", 4, Some(1));
+    test!(b"A\xC3\xA9 \x80 ", 4, Some(1));
+    test!(b"A\xC3\xA9 \xC1 ", 4, Some(1));
+    test!(b"A\xC3\xA9 \xC1", 4, Some(1));
+    test!(b"A\xC3\xA9 \xC2", 4, None);
+    test!(b"A\xC3\xA9 \xC2 ", 4, Some(1));
+    test!(b"A\xC3\xA9 \xC2\xC0", 4, Some(1));
+    test!(b"A\xC3\xA9 \xE0", 4, None);
+    test!(b"A\xC3\xA9 \xE0\x9F", 4, Some(1));
+    test!(b"A\xC3\xA9 \xE0\xA0", 4, None);
+    test!(b"A\xC3\xA9 \xE0\xA0\xC0", 4, Some(2));
+    test!(b"A\xC3\xA9 \xE0\xA0 ", 4, Some(2));
+    test!(b"A\xC3\xA9 \xED\xA0\x80 ", 4, Some(1));
+    test!(b"A\xC3\xA9 \xF1", 4, None);
+    test!(b"A\xC3\xA9 \xF1\x80", 4, None);
+    test!(b"A\xC3\xA9 \xF1\x80\x80", 4, None);
+    test!(b"A\xC3\xA9 \xF1 ", 4, Some(1));
+    test!(b"A\xC3\xA9 \xF1\x80 ", 4, Some(2));
+    test!(b"A\xC3\xA9 \xF1\x80\x80 ", 4, Some(3));
+}
+
 #[test]
 fn test_as_bytes() {
     // no null

src/libcore/str/mod.rs

@@ -125,13 +125,14 @@ Section: Creating a string
 #[stable(feature = "rust1", since = "1.0.0")]
 pub struct Utf8Error {
     valid_up_to: usize,
+    error_len: Option<u8>,
 }
 
 impl Utf8Error {
     /// Returns the index in the given string up to which valid UTF-8 was
     /// verified.
     ///
-    /// It is the maximum index such that `from_utf8(input[..index])`
+    /// It is the maximum index such that `from_utf8(&input[..index])`
     /// would return `Ok(_)`.
     ///
     /// # Examples
@@ -152,6 +153,23 @@ impl Utf8Error {
     /// ```
     #[stable(feature = "utf8_error", since = "1.5.0")]
     pub fn valid_up_to(&self) -> usize { self.valid_up_to }
+
+    /// Provide more information about the failure:
+    ///
+    /// * `None`: the end of the input was reached unexpectedly.
+    ///   `self.valid_up_to()` is 1 to 3 bytes from the end of the input.
+    ///   If a byte stream (such as a file or a network socket) is being decoded incrementally,
+    ///   this could be a valid `char` whose UTF-8 byte sequence is spanning multiple chunks.
+    ///
+    /// * `Some(len)`: an unexpected byte was encountered.
+    ///   The length provided is that of the invalid byte sequence
+    ///   that starts at the index given by `valid_up_to()`.
+    ///   Decoding should resume after that sequence
+    ///   (after inserting a U+FFFD REPLACEMENT CHARACTER) in case of lossy decoding.
+    #[unstable(feature = "utf8_error_error_len", reason ="new", issue = "40494")]
+    pub fn error_len(&self) -> Option<usize> {
+        self.error_len.map(|len| len as usize)
+    }
 }
 
 /// Converts a slice of bytes to a string slice.
@@ -300,7 +318,12 @@ pub unsafe fn from_utf8_unchecked(v: &[u8]) -> &str {
 #[stable(feature = "rust1", since = "1.0.0")]
 impl fmt::Display for Utf8Error {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
-        write!(f, "invalid utf-8: invalid byte near index {}", self.valid_up_to)
+        if let Some(error_len) = self.error_len {
+            write!(f, "invalid utf-8 sequence of {} bytes from index {}",
+                   error_len, self.valid_up_to)
+        } else {
+            write!(f, "incomplete utf-8 byte sequence from index {}", self.valid_up_to)
+        }
     }
 }
 
@@ -1241,25 +1264,27 @@ fn run_utf8_validation(v: &[u8]) -> Result<(), Utf8Error> {
 
     while index < len {
         let old_offset = index;
-        macro_rules! err { () => {{
-            return Err(Utf8Error {
-                valid_up_to: old_offset
-            })
-        }}}
+        macro_rules! err {
+            ($error_len: expr) => {
+                return Err(Utf8Error {
+                    valid_up_to: old_offset,
+                    error_len: $error_len,
+                })
+            }
+        }
 
         macro_rules! next { () => {{
             index += 1;
             // we needed data, but there was none: error!
             if index >= len {
-                err!()
+                err!(None)
             }
             v[index]
         }}}
 
         let first = v[index];
         if first >= 128 {
             let w = UTF8_CHAR_WIDTH[first as usize];
-            let second = next!();
             // 2-byte encoding is for codepoints  \u{0080} to  \u{07ff}
             //        first  C2 80        last DF BF
             // 3-byte encoding is for codepoints  \u{0800} to  \u{ffff}
@@ -1279,25 +1304,36 @@ fn run_utf8_validation(v: &[u8]) -> Result<(), Utf8Error> {
             // UTF8-4      = %xF0 %x90-BF 2( UTF8-tail ) / %xF1-F3 3( UTF8-tail ) /
             //               %xF4 %x80-8F 2( UTF8-tail )
             match w {
-                2 => if second & !CONT_MASK != TAG_CONT_U8 {err!()},
+                2 => if next!() & !CONT_MASK != TAG_CONT_U8 {
+                    err!(Some(1))
+                },
                 3 => {
-                    match (first, second, next!() & !CONT_MASK) {
-                        (0xE0         , 0xA0 ... 0xBF, TAG_CONT_U8) |
-                        (0xE1 ... 0xEC, 0x80 ... 0xBF, TAG_CONT_U8) |
-                        (0xED         , 0x80 ... 0x9F, TAG_CONT_U8) |
-                        (0xEE ... 0xEF, 0x80 ... 0xBF, TAG_CONT_U8) => {}
-                        _ => err!()
+                    match (first, next!()) {
+                        (0xE0         , 0xA0 ... 0xBF) |
+                        (0xE1 ... 0xEC, 0x80 ... 0xBF) |
+                        (0xED         , 0x80 ... 0x9F) |
+                        (0xEE ... 0xEF, 0x80 ... 0xBF) => {}
+                        _ => err!(Some(1))
+                    }
+                    if next!() & !CONT_MASK != TAG_CONT_U8 {
+                        err!(Some(2))
                     }
                 }
                 4 => {
-                    match (first, second, next!() & !CONT_MASK, next!() & !CONT_MASK) {
-                        (0xF0         , 0x90 ... 0xBF, TAG_CONT_U8, TAG_CONT_U8) |
-                        (0xF1 ... 0xF3, 0x80 ... 0xBF, TAG_CONT_U8, TAG_CONT_U8) |
-                        (0xF4         , 0x80 ... 0x8F, TAG_CONT_U8, TAG_CONT_U8) => {}
-                        _ => err!()
+                    match (first, next!()) {
+                        (0xF0         , 0x90 ... 0xBF) |
+                        (0xF1 ... 0xF3, 0x80 ... 0xBF) |
+                        (0xF4         , 0x80 ... 0x8F) => {}
+                        _ => err!(Some(1))
+                    }
+                    if next!() & !CONT_MASK != TAG_CONT_U8 {
+                        err!(Some(2))
+                    }
+                    if next!() & !CONT_MASK != TAG_CONT_U8 {
+                        err!(Some(3))
                     }
                 }
-                _ => err!()
+                _ => err!(Some(1))
             }
             index += 1;
         } else {

src/libstd/ffi/c_str.rs

@@ -325,7 +325,7 @@ impl CString {
     }
 
     /// Converts this `CString` into a boxed `CStr`.
-    #[unstable(feature = "into_boxed_c_str", issue = "0")]
+    #[unstable(feature = "into_boxed_c_str", issue = "40380")]
     pub fn into_boxed_c_str(self) -> Box<CStr> {
         unsafe { mem::transmute(self.into_inner()) }
     }
@@ -415,6 +415,20 @@ impl<'a> From<&'a CStr> for Box<CStr> {
     }
 }
 
+#[stable(feature = "c_string_from_box", since = "1.17.0")]
+impl From<Box<CStr>> for CString {
+    fn from(s: Box<CStr>) -> CString {
+        s.into_c_string()
+    }
+}
+
+#[stable(feature = "box_from_c_string", since = "1.17.0")]
+impl Into<Box<CStr>> for CString {
+    fn into(self) -> Box<CStr> {
+        self.into_boxed_c_str()
+    }
+}
+
 #[stable(feature = "default_box_extra", since = "1.17.0")]
 impl Default for Box<CStr> {
     fn default() -> Box<CStr> {
@@ -728,6 +742,12 @@ impl CStr {
     pub fn to_string_lossy(&self) -> Cow<str> {
         String::from_utf8_lossy(self.to_bytes())
     }
+
+    /// Converts a `Box<CStr>` into a `CString` without copying or allocating.
+    #[unstable(feature = "into_boxed_c_str", issue = "40380")]
+    pub fn into_c_string(self: Box<CStr>) -> CString {
+        unsafe { mem::transmute(self) }
+    }
 }
 
 #[stable(feature = "rust1", since = "1.0.0")]
@@ -922,12 +942,11 @@ mod tests {
     fn into_boxed() {
         let orig: &[u8] = b"Hello, world!\0";
         let cstr = CStr::from_bytes_with_nul(orig).unwrap();
-        let cstring = cstr.to_owned();
-        let box1: Box<CStr> = Box::from(cstr);
-        let box2 = cstring.into_boxed_c_str();
-        assert_eq!(cstr, &*box1);
-        assert_eq!(box1, box2);
-        assert_eq!(&*box2, cstr);
+        let boxed: Box<CStr> = Box::from(cstr);
+        let cstring = cstr.to_owned().into_boxed_c_str().into_c_string();
+        assert_eq!(cstr, &*boxed);
+        assert_eq!(&*boxed, &*cstring);
+        assert_eq!(&*cstring, cstr);
     }
 
     #[test]