deprecate Unicode functions that will be moved to crates.io

This deprecates the following functionality:

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():
   use unicode-normalization crate

Author: kwantam

src/libcollections/str.rs

@@ -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;
@@ -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;
@@ -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/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

@@ -531,6 +531,7 @@ Additional help:
               extra_help);
 }
 
+#[allow(deprecated)]
 fn describe_lints(lint_store: &lint::LintStore, loaded_plugins: bool) {
     println!("
 Available lint options:

src/libsyntax/diagnostic.rs

@@ -542,6 +542,7 @@ fn highlight_suggestion(err: &mut EmitterWriter,
     Ok(())
 }
 
+#[allow(deprecated)]
 fn highlight_lines(err: &mut EmitterWriter,
                    cm: &codemap::CodeMap,
                    sp: Span,
@@ -664,6 +665,7 @@ fn highlight_lines(err: &mut EmitterWriter,
 /// than 6 lines), `end_highlight_lines` will print the first line, then
 /// dot dot dot, then last line, whereas `highlight_lines` prints the first
 /// six lines.
+#[allow(deprecated)]
 fn end_highlight_lines(w: &mut EmitterWriter,
                           cm: &codemap::CodeMap,
                           sp: Span,

src/libunicode/char.rs

@@ -445,6 +445,8 @@ 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/normalize.rs

@@ -33,9 +33,17 @@ fn bsearch_table<T>(c: char, r: &'static [(char, &'static [T])]) -> Option<&'sta
 }
 
 /// Compute canonical Unicode decomposition for character
+#[deprecated(reason = "use the crates.io `unicode-normalization` library instead",
+             since = "1.0.0")]
+#[unstable(feature = "unicode",
+           reason = "this functionality will be moved to crates.io")]
 pub fn decompose_canonical<F>(c: char, mut i: F) where F: FnMut(char) { d(c, &mut i, false); }
 
 /// Compute canonical or compatible Unicode decomposition for character
+#[deprecated(reason = "use the crates.io `unicode-normalization` library instead",
+             since = "1.0.0")]
+#[unstable(feature = "unicode",
+           reason = "this functionality will be moved to crates.io")]
 pub fn decompose_compatible<F>(c: char, mut i: F) where F: FnMut(char) { d(c, &mut i, true); }
 
 // FIXME(#19596) This is a workaround, we should use `F` instead of `&mut F`
@@ -78,6 +86,10 @@ fn d<F>(c: char, i: &mut F, k: bool) where F: FnMut(char) {
     (*i)(c);
 }
 
+#[deprecated(reason = "use the crates.io `unicode-normalization` library instead",
+             since = "1.0.0")]
+#[unstable(feature = "unicode",
+           reason = "this functionality will be moved to crates.io")]
 pub fn compose(a: char, b: char) -> Option<char> {
     compose_hangul(a, b).or_else(|| {
         match bsearch_table(a, composition_table) {

src/libunicode/u_str.rs

@@ -75,6 +75,7 @@ impl UnicodeStr for str {
     #[inline]
     fn is_alphanumeric(&self) -> bool { self.chars().all(|c| c.is_alphanumeric()) }
 
+    #[allow(deprecated)]
     #[inline]
     fn width(&self, is_cjk: bool) -> usize {
         self.chars().map(|c| c.width(is_cjk).unwrap_or(0)).sum()