---

yaml
---
r: 77655
b: refs/heads/master
c: 6f9c68a
h: refs/heads/master
i:
  77653: 288e9fc
  77651: f2f34c5
  77647: b8bb182
v: v3

Author: lilyball

[refs]

@@ -1,5 +1,5 @@
 ---
-refs/heads/master: 9d8897ba603b044a82a4cbcf97e5aadcb77ef661
+refs/heads/master: 6f9c68af2e85f1b0460f0f5f885aa3d9322a1584
 refs/heads/snap-stage1: e33de59e47c5076a89eadeb38f4934f58a3618a6
 refs/heads/snap-stage3: 60fba4d7d677ec098e6a43014132fe99f7547363
 refs/heads/try: ebfe63cd1c0b5d23f7ea60c69b4fde2e30cfd42a

trunk/configure

@@ -381,7 +381,8 @@ opt mingw-cross 0 "cross-compile for win32 using mingw"
 opt clang 0 "prefer clang to gcc for building the runtime"
 opt ccache 0 "invoke gcc/clang via ccache to reuse object files between builds"
 opt local-rust 0 "use an installed rustc rather than downloading a snapshot"
-opt pax-flags 0 "apply PaX flags to rustc binaries (required for GRSecurity/PaX-patched kernels)"
+opt pax-flags 0 "apply PaX flags to rustc binaries (required for GRSecurity/PaX-patched
+ kernels)"
 valopt prefix "/usr/local" "set installation prefix"
 valopt local-rust-root "/usr/local" "set prefix for local rust binary"
 valopt llvm-root "" "set LLVM root"

trunk/src/etc/emacs/rust-mode.el

@@ -225,27 +225,4 @@ The initializer is `DEFAULT-TAB-WIDTH'.")
 
 (provide 'rust-mode)
 
-;; Issue #6887: Rather than inheriting the 'gnu compilation error
-;; regexp (which is broken on a few edge cases), add our own 'rust
-;; compilation error regexp and use it instead.
-(defvar rustc-compilation-regexps
-  (let ((file "^\\([^\n]+\\)")
-        (start-line "\\([0-9]+\\)")
-        (start-col  "\\([0-9]+\\)")
-        (end-line   "\\([0-9]+\\)")
-        (end-col    "\\([0-9]+\\)")
-        (error-or-warning "\\(?:[Ee]rror\\|\\([Ww]arning\\)\\)"))
-    (let ((re (concat "^" file ":" start-line ":" start-col
-                      ": " end-line ":" end-col
-                      " \\(?:[Ee]rror\\|\\([Ww]arning\\)\\):")))
-      (cons re '(1 (2 . 4) (3 . 5) (6)))))
-  "Specifications for matching errors in rustc invocations.
-See `compilation-error-regexp-alist for help on their format.")
-
-(eval-after-load 'compile
-  '(progn
-     (add-to-list 'compilation-error-regexp-alist-alist
-                  (cons 'rustc rustc-compilation-regexps))
-     (add-to-list 'compilation-error-regexp-alist 'rustc)))
-
 ;;; rust-mode.el ends here

trunk/src/libextra/ebml.rs

@@ -12,6 +12,8 @@
 
 
 use std::str;
+use std::cast;
+use std::vec;
 
 // Simple Extensible Binary Markup Language (ebml) reader and writer on a
 // cursor model. See the specification here:
@@ -29,9 +31,42 @@ struct EbmlState {
     data_pos: uint,
 }
 
+#[deriving(Clone)]
+pub enum EbmlData {
+    SafeData(@~[u8]),
+    UnsafeData(*u8, uint)
+}
+
+impl EbmlData {
+    #[inline]
+    pub fn slice<'a>(&'a self, start: uint, end: uint) -> &'a [u8] {
+        match *self {
+            SafeData(@ref v) => v.slice(start, end),
+            UnsafeData(buf, len) => unsafe {
+                do vec::raw::buf_as_slice(buf, len) |s| {
+                    cast::transmute(s.slice(start, end))
+                }
+            }
+        }
+    }
+
+    #[inline]
+    pub fn as_slice<'a>(&'a self) -> &'a [u8] {
+        self.slice(0, self.len())
+    }
+
+    #[inline]
+    pub fn len(&self) -> uint {
+        match *self {
+            SafeData(@ref v) => v.len(),
+            UnsafeData(_, len) => len
+        }
+    }
+}
+
 #[deriving(Clone)]
 pub struct Doc {
-    data: @~[u8],
+    data: EbmlData,
     start: uint,
     end: uint,
 }
@@ -185,24 +220,28 @@ pub mod reader {
     }
 
     pub fn Doc(data: @~[u8]) -> Doc {
-        Doc { data: data, start: 0u, end: data.len() }
+        Doc { data: SafeData(data), start: 0u, end: data.len() }
+    }
+
+    pub fn unsafe_Doc(buf: *u8, len: uint) -> Doc {
+        Doc { data: UnsafeData(buf, len), start: 0u, end: len }
     }
 
-    pub fn doc_at(data: @~[u8], start: uint) -> TaggedDoc {
-        let elt_tag = vuint_at(*data, start);
-        let elt_size = vuint_at(*data, elt_tag.next);
+    pub fn doc_at(data: &EbmlData, start: uint) -> TaggedDoc {
+        let elt_tag = vuint_at(data.as_slice(), start);
+        let elt_size = vuint_at(data.as_slice(), elt_tag.next);
         let end = elt_size.next + elt_size.val;
         TaggedDoc {
             tag: elt_tag.val,
-            doc: Doc { data: data, start: elt_size.next, end: end }
+            doc: Doc { data: data.clone(), start: elt_size.next, end: end }
         }
     }
 
     pub fn maybe_get_doc(d: Doc, tg: uint) -> Option<Doc> {
         let mut pos = d.start;
         while pos < d.end {
-            let elt_tag = vuint_at(*d.data, pos);
-            let elt_size = vuint_at(*d.data, elt_tag.next);
+            let elt_tag = vuint_at(d.data.as_slice(), pos);
+            let elt_size = vuint_at(d.data.as_slice(), elt_tag.next);
             pos = elt_size.next + elt_size.val;
             if elt_tag.val == tg {
                 return Some(Doc { data: d.data, start: elt_size.next,
@@ -225,8 +264,8 @@ pub mod reader {
     pub fn docs(d: Doc, it: &fn(uint, Doc) -> bool) -> bool {
         let mut pos = d.start;
         while pos < d.end {
-            let elt_tag = vuint_at(*d.data, pos);
-            let elt_size = vuint_at(*d.data, elt_tag.next);
+            let elt_tag = vuint_at(d.data.as_slice(), pos);
+            let elt_size = vuint_at(d.data.as_slice(), elt_tag.next);
             pos = elt_size.next + elt_size.val;
             let doc = Doc { data: d.data, start: elt_size.next, end: pos };
             if !it(elt_tag.val, doc) {
@@ -239,8 +278,8 @@ pub mod reader {
     pub fn tagged_docs(d: Doc, tg: uint, it: &fn(Doc) -> bool) -> bool {
         let mut pos = d.start;
         while pos < d.end {
-            let elt_tag = vuint_at(*d.data, pos);
-            let elt_size = vuint_at(*d.data, elt_tag.next);
+            let elt_tag = vuint_at(d.data.as_slice(), pos);
+            let elt_size = vuint_at(d.data.as_slice(), elt_tag.next);
             pos = elt_size.next + elt_size.val;
             if elt_tag.val == tg {
                 let doc = Doc { data: d.data, start: elt_size.next,
@@ -260,22 +299,22 @@ pub mod reader {
 
     pub fn doc_as_u8(d: Doc) -> u8 {
         assert_eq!(d.end, d.start + 1u);
-        (*d.data)[d.start]
+        d.data.as_slice()[d.start]
     }
 
     pub fn doc_as_u16(d: Doc) -> u16 {
         assert_eq!(d.end, d.start + 2u);
-        io::u64_from_be_bytes(*d.data, d.start, 2u) as u16
+        io::u64_from_be_bytes(d.data.as_slice(), d.start, 2u) as u16
     }
 
     pub fn doc_as_u32(d: Doc) -> u32 {
         assert_eq!(d.end, d.start + 4u);
-        io::u64_from_be_bytes(*d.data, d.start, 4u) as u32
+        io::u64_from_be_bytes(d.data.as_slice(), d.start, 4u) as u32
     }
 
     pub fn doc_as_u64(d: Doc) -> u64 {
         assert_eq!(d.end, d.start + 8u);
-        io::u64_from_be_bytes(*d.data, d.start, 8u)
+        io::u64_from_be_bytes(d.data.as_slice(), d.start, 8u)
     }
 
     pub fn doc_as_i8(d: Doc) -> i8 { doc_as_u8(d) as i8 }
@@ -298,8 +337,7 @@ pub mod reader {
     impl Decoder {
         fn _check_label(&mut self, lbl: &str) {
             if self.pos < self.parent.end {
-                let TaggedDoc { tag: r_tag, doc: r_doc } =
-                    doc_at(self.parent.data, self.pos);
+                let TaggedDoc { tag: r_tag, doc: r_doc } = doc_at(&self.parent.data, self.pos);
 
                 if r_tag == (EsLabel as uint) {
                     self.pos = r_doc.end;
@@ -316,8 +354,7 @@ pub mod reader {
             if self.pos >= self.parent.end {
                 fail!("no more documents in current node!");
             }
-            let TaggedDoc { tag: r_tag, doc: r_doc } =
-                doc_at(self.parent.data, self.pos);
+            let TaggedDoc { tag: r_tag, doc: r_doc } = doc_at(&self.parent.data, self.pos);
             debug!("self.parent=%?-%? self.pos=%? r_tag=%? r_doc=%?-%?",
                    self.parent.start,
                    self.parent.end,

trunk/src/libextra/getopts.rs

@@ -689,9 +689,9 @@ pub mod groups {
                 }
             }
 
-            // FIXME: #5516
+            // FIXME: #5516 should be graphemes not codepoints
             // here we just need to indent the start of the description
-            let rowlen = row.len();
+            let rowlen = row.char_len();
             if rowlen < 24 {
                 do (24 - rowlen).times {
                     row.push_char(' ')
@@ -707,14 +707,14 @@ pub mod groups {
                 desc_normalized_whitespace.push_char(' ');
             }
 
-            // FIXME: #5516
+            // FIXME: #5516 should be graphemes not codepoints
             let mut desc_rows = ~[];
             do each_split_within(desc_normalized_whitespace, 54) |substr| {
                 desc_rows.push(substr.to_owned());
                 true
             };
 
-            // FIXME: #5516
+            // FIXME: #5516 should be graphemes not codepoints
             // wrapped description
             row.push_str(desc_rows.connect(desc_sep));
 
@@ -798,7 +798,7 @@ pub mod groups {
             cont
         };
 
-        ss.iter().enumerate().advance(|x| machine(x));
+        ss.char_offset_iter().advance(|x| machine(x));
 
         // Let the automaton 'run out' by supplying trailing whitespace
         while cont && match state { B | C => true, A => false } {
@@ -1580,4 +1580,31 @@ Options:
         debug!("generated: <<%s>>", usage);
         assert!(usage == expected)
     }
+
+    #[test]
+    fn test_groups_usage_description_multibyte_handling() {
+        let optgroups = ~[
+            groups::optflag("k", "k\u2013w\u2013",
+                "The word kiwi is normally spelled with two i's"),
+            groups::optflag("a", "apple",
+                "This \u201Cdescription\u201D has some characters that could \
+confuse the line wrapping; an apple costs 0.51€ in some parts of Europe."),
+        ];
+
+        let expected =
+~"Usage: fruits
+
+Options:
+    -k --k–w–           The word kiwi is normally spelled with two i's
+    -a --apple          This “description” has some characters that could
+                        confuse the line wrapping; an apple costs 0.51€ in
+                        some parts of Europe.
+";
+
+        let usage = groups::usage("Usage: fruits", optgroups);
+
+        debug!("expected: <<%s>>", expected);
+        debug!("generated: <<%s>>", usage);
+        assert!(usage == expected)
+    }
 }

trunk/src/librustc/driver/driver.rs

@@ -763,7 +763,7 @@ pub fn build_session_options(binary: @str,
         parse_only: parse_only,
         no_trans: no_trans,
         debugging_opts: debugging_opts,
-        android_cross_path: android_cross_path
+        android_cross_path: android_cross_path,
     };
     return sopts;
 }

trunk/src/librustc/metadata/creader.rs

@@ -15,6 +15,7 @@ use metadata::cstore;
 use metadata::decoder;
 use metadata::filesearch::FileSearch;
 use metadata::loader;
+use metadata::loader::MetadataSection;
 
 use std::hashmap::HashMap;
 use syntax::ast;
@@ -308,7 +309,7 @@ fn resolve_crate(e: @mut Env,
 }
 
 // Go through the crate metadata and load any crates that it references
-fn resolve_crate_deps(e: @mut Env, cdata: @~[u8]) -> cstore::cnum_map {
+fn resolve_crate_deps(e: @mut Env, cdata: MetadataSection) -> cstore::cnum_map {
     debug!("resolving deps of external crate");
     // The map from crate numbers in the crate we're resolving to local crate
     // numbers

trunk/src/librustc/metadata/csearch.rs

@@ -188,7 +188,7 @@ pub fn get_field_type(tcx: ty::ctxt, class_id: ast::def_id,
                       def: ast::def_id) -> ty::ty_param_bounds_and_ty {
     let cstore = tcx.cstore;
     let cdata = cstore::get_crate_data(cstore, class_id.crate);
-    let all_items = reader::get_doc(reader::Doc(cdata.data), tag_items);
+    let all_items = reader::get_doc(decoder::section_to_ebml_doc(cdata.data), tag_items);
     debug!("Looking up %?", class_id);
     let class_doc = expect(tcx.diag,
                            decoder::maybe_find_item(class_id.node, all_items),

trunk/src/librustc/metadata/cstore.rs

@@ -15,6 +15,7 @@
 
 use metadata::cstore;
 use metadata::decoder;
+use metadata::loader::MetadataSection;
 
 use std::hashmap::HashMap;
 use extra;
@@ -29,7 +30,7 @@ pub type cnum_map = @mut HashMap<ast::CrateNum, ast::CrateNum>;
 
 pub struct crate_metadata {
     name: @str,
-    data: @~[u8],
+    data: MetadataSection,
     cnum_map: cnum_map,
     cnum: ast::CrateNum
 }