deprecate Unicode functions that will be moved to crates.io

This patch
1. renames libunicode to librustc_unicode,
2. deprecates several pieces of libunicode (see below), and
3. removes references to deprecated functions from
   librustc_driver and libsyntax. This may change pretty-printed
   output from these modules in cases involving wide or combining
   characters used in filenames, identifiers, etc.

The following functions are marked deprecated:

1. char.width() and str.width():
   --> use unicode-width crate

2. str.graphemes() and str.grapheme_indices():
   --> use unicode-segmentation crate

3. str.nfd_chars(), str.nfkd_chars(), str.nfc_chars(), str.nfkc_chars(),
   char.compose(), char.decompose_canonical(), char.decompose_compatible(),
   char.canonical_combining_class():
   --> use unicode-normalization crate

mk/crates.mk

@@ -52,7 +52,7 @@
 TARGET_CRATES := libc std flate arena term \
                  serialize getopts collections test rand \
                  log graphviz core rbml alloc \
-                 unicode rustc_bitflags
+                 rustc_unicode rustc_bitflags
 RUSTC_CRATES := rustc rustc_typeck rustc_borrowck rustc_resolve rustc_driver \
                 rustc_trans rustc_back rustc_llvm rustc_privacy rustc_lint
 HOST_CRATES := syntax $(RUSTC_CRATES) rustdoc fmt_macros
@@ -61,9 +61,9 @@ TOOLS := compiletest rustdoc rustc rustbook
 
 DEPS_core :=
 DEPS_libc := core
-DEPS_unicode := core
+DEPS_rustc_unicode := core
 DEPS_alloc := core libc native:jemalloc
-DEPS_std := core libc rand alloc collections unicode \
+DEPS_std := core libc rand alloc collections rustc_unicode \
 	native:rust_builtin native:backtrace native:rustrt_native \
 	rustc_bitflags
 DEPS_graphviz := std
@@ -94,7 +94,7 @@ DEPS_serialize := std log
 DEPS_rbml := std log serialize
 DEPS_term := std log
 DEPS_getopts := std
-DEPS_collections := core alloc unicode
+DEPS_collections := core alloc rustc_unicode
 DEPS_num := std
 DEPS_test := std getopts serialize rbml term native:rust_test_helpers
 DEPS_rand := core
@@ -115,11 +115,11 @@ ONLY_RLIB_libc := 1
 ONLY_RLIB_alloc := 1
 ONLY_RLIB_rand := 1
 ONLY_RLIB_collections := 1
-ONLY_RLIB_unicode := 1
+ONLY_RLIB_rustc_unicode := 1
 ONLY_RLIB_rustc_bitflags := 1
 
 # Documented-by-default crates
-DOC_CRATES := std alloc collections core libc unicode
+DOC_CRATES := std alloc collections core libc rustc_unicode
 
 ################################################################################
 # You should not need to edit below this line

src/etc/unicode.py

@@ -518,11 +518,14 @@ def comp_pfun(char):
     emit_table(f, "combining_class_table", combine, "&'static [(char, char, u8)]", is_pub=False,
             pfun=lambda x: "(%s,%s,%s)" % (escape_char(x[0]), escape_char(x[1]), x[2]))
 
-    f.write("    pub fn canonical_combining_class(c: char) -> u8 {\n"
-        + "        bsearch_range_value_table(c, combining_class_table)\n"
-        + "    }\n")
+    f.write("""    #[deprecated(reason = "use the crates.io `unicode-normalization` lib instead",
+                 since = "1.0.0")]
+    #[unstable(feature = "unicode",
+               reason = "this functionality will be moved to crates.io")]
+    pub fn canonical_combining_class(c: char) -> u8 {
+        bsearch_range_value_table(c, combining_class_table)
+    }
 
-    f.write("""
 }
 
 """)

src/libcollections/lib.rs

@@ -50,7 +50,7 @@
 #[macro_use]
 extern crate core;
 
-extern crate unicode;
+extern crate rustc_unicode;
 extern crate alloc;
 
 #[cfg(test)] #[macro_use] extern crate std;

src/libcollections/str.rs

@@ -59,13 +59,13 @@ use core::result::Result;
 use core::str as core_str;
 use core::str::pattern::Pattern;
 use core::str::pattern::{Searcher, ReverseSearcher, DoubleEndedSearcher};
-use unicode::str::{UnicodeStr, Utf16Encoder};
+use rustc_unicode::str::{UnicodeStr, Utf16Encoder};
 
 use core::convert::AsRef;
 use vec_deque::VecDeque;
 use borrow::{Borrow, ToOwned};
 use string::String;
-use unicode;
+use rustc_unicode;
 use vec::Vec;
 use slice::SliceConcatExt;
 
@@ -78,7 +78,7 @@ pub use core::str::{Matches, RMatches};
 pub use core::str::{MatchIndices, RMatchIndices};
 pub use core::str::{from_utf8, Chars, CharIndices, Bytes};
 pub use core::str::{from_utf8_unchecked, ParseBoolError};
-pub use unicode::str::{Words, Graphemes, GraphemeIndices};
+pub use rustc_unicode::str::{Words, Graphemes, GraphemeIndices};
 pub use core::str::pattern;
 
 /*
@@ -161,6 +161,9 @@ enum DecompositionType {
 /// External iterator for a string decomposition's characters.
 ///
 /// For use with the `std::iter` module.
+#[allow(deprecated)]
+#[deprecated(reason = "use the crates.io `unicode-normalization` library instead",
+             since = "1.0.0")]
 #[derive(Clone)]
 #[unstable(feature = "unicode",
            reason = "this functionality may be replaced with a more generic \
@@ -172,6 +175,7 @@ pub struct Decompositions<'a> {
     sorted: bool
 }
 
+#[allow(deprecated)]
 #[stable(feature = "rust1", since = "1.0.0")]
 impl<'a> Iterator for Decompositions<'a> {
     type Item = char;
@@ -198,7 +202,7 @@ impl<'a> Iterator for Decompositions<'a> {
                 {
                     let callback = |d| {
                         let class =
-                            unicode::char::canonical_combining_class(d);
+                            rustc_unicode::char::canonical_combining_class(d);
                         if class == 0 && !*sorted {
                             canonical_sort(buffer);
                             *sorted = true;
@@ -207,10 +211,10 @@ impl<'a> Iterator for Decompositions<'a> {
                     };
                     match self.kind {
                         Canonical => {
-                            unicode::char::decompose_canonical(ch, callback)
+                            rustc_unicode::char::decompose_canonical(ch, callback)
                         }
                         Compatible => {
-                            unicode::char::decompose_compatible(ch, callback)
+                            rustc_unicode::char::decompose_compatible(ch, callback)
                         }
                     }
                 }
@@ -254,6 +258,9 @@ enum RecompositionState {
 /// External iterator for a string recomposition's characters.
 ///
 /// For use with the `std::iter` module.
+#[allow(deprecated)]
+#[deprecated(reason = "use the crates.io `unicode-normalization` library instead",
+             since = "1.0.0")]
 #[derive(Clone)]
 #[unstable(feature = "unicode",
            reason = "this functionality may be replaced with a more generic \
@@ -266,6 +273,7 @@ pub struct Recompositions<'a> {
     last_ccc: Option<u8>
 }
 
+#[allow(deprecated)]
 #[stable(feature = "rust1", since = "1.0.0")]
 impl<'a> Iterator for Recompositions<'a> {
     type Item = char;
@@ -276,7 +284,7 @@ impl<'a> Iterator for Recompositions<'a> {
             match self.state {
                 Composing => {
                     for ch in self.iter.by_ref() {
-                        let ch_class = unicode::char::canonical_combining_class(ch);
+                        let ch_class = rustc_unicode::char::canonical_combining_class(ch);
                         if self.composee.is_none() {
                             if ch_class != 0 {
                                 return Some(ch);
@@ -288,7 +296,7 @@ impl<'a> Iterator for Recompositions<'a> {
 
                         match self.last_ccc {
                             None => {
-                                match unicode::char::compose(k, ch) {
+                                match rustc_unicode::char::compose(k, ch) {
                                     Some(r) => {
                                         self.composee = Some(r);
                                         continue;
@@ -316,7 +324,7 @@ impl<'a> Iterator for Recompositions<'a> {
                                     self.last_ccc = Some(ch_class);
                                     continue;
                                 }
-                                match unicode::char::compose(k, ch) {
+                                match rustc_unicode::char::compose(k, ch) {
                                     Some(r) => {
                                         self.composee = Some(r);
                                         continue;
@@ -465,6 +473,9 @@ impl str {
 
     /// Returns an iterator over the string in Unicode Normalization Form D
     /// (canonical decomposition).
+    #[allow(deprecated)]
+    #[deprecated(reason = "use the crates.io `unicode-normalization` library instead",
+             since = "1.0.0")]
     #[inline]
     #[unstable(feature = "unicode",
                reason = "this functionality may be replaced with a more generic \
@@ -480,6 +491,9 @@ impl str {
 
     /// Returns an iterator over the string in Unicode Normalization Form KD
     /// (compatibility decomposition).
+    #[allow(deprecated)]
+    #[deprecated(reason = "use the crates.io `unicode-normalization` library instead",
+             since = "1.0.0")]
     #[inline]
     #[unstable(feature = "unicode",
                reason = "this functionality may be replaced with a more generic \
@@ -495,6 +509,9 @@ impl str {
 
     /// An Iterator over the string in Unicode Normalization Form C
     /// (canonical decomposition followed by canonical composition).
+    #[allow(deprecated)]
+    #[deprecated(reason = "use the crates.io `unicode-normalization` library instead",
+             since = "1.0.0")]
     #[inline]
     #[unstable(feature = "unicode",
                reason = "this functionality may be replaced with a more generic \
@@ -511,6 +528,9 @@ impl str {
 
     /// An Iterator over the string in Unicode Normalization Form KC
     /// (compatibility decomposition followed by canonical composition).
+    #[allow(deprecated)]
+    #[deprecated(reason = "use the crates.io `unicode-normalization` library instead",
+             since = "1.0.0")]
     #[inline]
     #[unstable(feature = "unicode",
                reason = "this functionality may be replaced with a more generic \
@@ -1690,6 +1710,8 @@ impl str {
     ///
     /// assert_eq!(&gr2[..], b);
     /// ```
+    #[deprecated(reason = "use the crates.io `unicode-segmentation` library instead",
+             since = "1.0.0")]
     #[unstable(feature = "unicode",
                reason = "this functionality may only be provided by libunicode")]
     pub fn graphemes(&self, is_extended: bool) -> Graphemes {
@@ -1709,6 +1731,8 @@ impl str {
     ///
     /// assert_eq!(&gr_inds[..], b);
     /// ```
+    #[deprecated(reason = "use the crates.io `unicode-segmentation` library instead",
+             since = "1.0.0")]
     #[unstable(feature = "unicode",
                reason = "this functionality may only be provided by libunicode")]
     pub fn grapheme_indices(&self, is_extended: bool) -> GraphemeIndices {
@@ -1749,6 +1773,8 @@ impl str {
     /// recommends that these
     /// characters be treated as 1 column (i.e., `is_cjk = false`) if the
     /// locale is unknown.
+    #[deprecated(reason = "use the crates.io `unicode-width` library instead",
+                 since = "1.0.0")]
     #[unstable(feature = "unicode",
                reason = "this functionality may only be provided by libunicode")]
     pub fn width(&self, is_cjk: bool) -> usize {

src/libcollections/string.rs

@@ -25,8 +25,8 @@ use core::ops::{self, Deref, Add, Index};
 use core::ptr;
 use core::slice;
 use core::str::pattern::Pattern;
-use unicode::str as unicode_str;
-use unicode::str::Utf16Item;
+use rustc_unicode::str as unicode_str;
+use rustc_unicode::str::Utf16Item;
 
 use borrow::{Cow, IntoCow};
 use str::{self, FromStr, Utf8Error};

src/libcollectionstest/str.rs

@@ -19,6 +19,7 @@ fn test_le() {
     assert!("foo" != "bar");
 }
 
+#[allow(deprecated)]
 #[test]
 fn test_len() {
     assert_eq!("".len(), 0);
@@ -944,6 +945,7 @@ fn test_words() {
     assert_eq!(words, ["Märy", "häd", "ä", "little", "lämb", "Little", "lämb"])
 }
 
+#[allow(deprecated)]
 #[test]
 fn test_nfd_chars() {
     macro_rules! t {
@@ -963,6 +965,7 @@ fn test_nfd_chars() {
     t!("\u{ac1c}", "\u{1100}\u{1162}");
 }
 
+#[allow(deprecated)]
 #[test]
 fn test_nfkd_chars() {
     macro_rules! t {
@@ -982,6 +985,7 @@ fn test_nfkd_chars() {
     t!("\u{ac1c}", "\u{1100}\u{1162}");
 }
 
+#[allow(deprecated)]
 #[test]
 fn test_nfc_chars() {
     macro_rules! t {
@@ -1002,6 +1006,7 @@ fn test_nfc_chars() {
     t!("a\u{300}\u{305}\u{315}\u{5ae}b", "\u{e0}\u{5ae}\u{305}\u{315}b");
 }
 
+#[allow(deprecated)]
 #[test]
 fn test_nfkc_chars() {
     macro_rules! t {
@@ -1033,6 +1038,7 @@ fn test_lines() {
     assert_eq!(lines, ["", "Märy häd ä little lämb", "", "Little lämb"]);
 }
 
+#[allow(deprecated)]
 #[test]
 fn test_graphemes() {
     use std::iter::order;

src/libcoretest/char.rs

@@ -210,6 +210,7 @@ fn test_len_utf16() {
     assert!('\u{1f4a9}'.len_utf16() == 2);
 }
 
+#[allow(deprecated)]
 #[test]
 fn test_width() {
     assert_eq!('\x00'.width(false),Some(0));

src/librustc_driver/lib.rs

@@ -35,7 +35,6 @@
 #![feature(staged_api)]
 #![feature(exit_status)]
 #![feature(set_stdio)]
-#![feature(unicode)]
 
 extern crate arena;
 extern crate flate;
@@ -574,7 +573,7 @@ Available lint options:
     let builtin_groups = sort_lint_groups(builtin_groups);
 
     let max_name_len = plugin.iter().chain(builtin.iter())
-        .map(|&s| s.name.width(true))
+        .map(|&s| s.name.chars().count())
         .max().unwrap_or(0);
     let padded = |x: &str| {
         let mut s = repeat(" ").take(max_name_len - x.chars().count())
@@ -601,7 +600,7 @@ Available lint options:
 
 
     let max_name_len = plugin_groups.iter().chain(builtin_groups.iter())
-        .map(|&(s, _)| s.width(true))
+        .map(|&(s, _)| s.chars().count())
         .max().unwrap_or(0);
     let padded = |x: &str| {
         let mut s = repeat(" ").take(max_name_len - x.chars().count())
@@ -790,7 +789,6 @@ fn parse_crate_attrs(sess: &Session, input: &Input) ->
 ///
 /// The diagnostic emitter yielded to the procedure should be used for reporting
 /// errors of the compiler.
-#[allow(deprecated)]
 pub fn monitor<F:FnOnce()+Send+'static>(f: F) {
     const STACK_SIZE: usize = 8 * 1024 * 1024; // 8MB
 

src/libunicode/char.rs → src/librustc_unicode/char.rs

@@ -37,7 +37,9 @@ use tables::{derived_property, property, general_category, conversions, charwidt
 pub use core::char::{MAX, from_u32, from_digit, EscapeUnicode, EscapeDefault};
 
 // unstable reexports
+#[allow(deprecated)]
 pub use normalize::{decompose_canonical, decompose_compatible, compose};
+#[allow(deprecated)]
 pub use tables::normalization::canonical_combining_class;
 pub use tables::UNICODE_VERSION;
 
@@ -445,6 +447,8 @@ impl char {
     /// [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
     /// recommends that these characters be treated as 1 column (i.e.,
     /// `is_cjk` = `false`) if the context cannot be reliably determined.
+    #[deprecated(reason = "use the crates.io `unicode-width` library instead",
+                 since = "1.0.0")]
     #[unstable(feature = "unicode",
                reason = "needs expert opinion. is_cjk flag stands out as ugly")]
     pub fn width(self, is_cjk: bool) -> Option<usize> { charwidth::width(self, is_cjk) }

src/libunicode/lib.rs → src/librustc_unicode/lib.rs

@@ -22,7 +22,7 @@
 
 // Do not remove on snapshot creation. Needed for bootstrap. (Issue #22364)
 #![cfg_attr(stage0, feature(custom_attribute))]
-#![crate_name = "unicode"]
+#![crate_name = "rustc_unicode"]
 #![unstable(feature = "unicode")]
 #![feature(lang_items)]
 #![feature(staged_api)]