rollup of several PRs

CHANGELOG.md

@@ -1,3 +1,24 @@
+1.3.2 (2020-01-09)
+==================
+This is a small maintenance release with some house cleaning and bug fixes.
+
+New features:
+
+* [FEATURE #631](https://github.com/rust-lang/regex/issues/631):
+  Add a `Match::range` method an a `From<Match> for Range` impl.
+
+Bug fixes:
+
+* [BUG #521](https://github.com/rust-lang/regex/issues/521):
+  Corrects `/-/.splitn("a", 2)` to return `["a"]` instead of `["a", ""]`.
+* [BUG #594](https://github.com/rust-lang/regex/pull/594):
+  Improve error reporting when writing `\p\`.
+* [BUG #627](https://github.com/rust-lang/regex/issues/627):
+  Corrects `/-/.split("a-")` to return `["a", ""]` instead of `["a"]`.
+* [BUG #633](https://github.com/rust-lang/regex/pull/633):
+  Squash deprecation warnings for the `std::error::Error::description` method.
+
+
 1.3.1 (2019-09-04)
 ==================
 This is a maintenance release with no changes in order to try to work-around

bench/src/ffi/tcl.rs

@@ -2,12 +2,12 @@
 
 use std::mem;
 use std::ptr;
-use std::sync::{Once, ONCE_INIT};
+use std::sync::Once;
 
 use libc::{c_char, c_int, c_long, c_void};
 
 // Used to initialize the TCL interpreter exactly once.
-static ONCE: Once = ONCE_INIT;
+static ONCE: Once = Once::new();
 
 /// Text is a TCL string object backed by a Rust string.
 ///

regex-syntax/src/ast/mod.rs

@@ -156,6 +156,9 @@ pub enum ErrorKind {
     /// `(?i)*`. It is, however, possible to create a repetition operating on
     /// an empty sub-expression. For example, `()*` is still considered valid.
     RepetitionMissing,
+    /// The Unicode class is not valid. This typically occurs when a `\p` is
+    /// followed by something other than a `{`.
+    UnicodeClassInvalid,
     /// When octal support is disabled, this error is produced when an octal
     /// escape is used. The octal escape is assumed to be an invocation of
     /// a backreference, which is the common case.
@@ -176,6 +179,8 @@ pub enum ErrorKind {
 }
 
 impl error::Error for Error {
+    // TODO: Remove this method entirely on the next breaking semver release.
+    #[allow(deprecated)]
     fn description(&self) -> &str {
         use self::ErrorKind::*;
         match self.kind {
@@ -206,6 +211,7 @@ impl error::Error for Error {
             RepetitionCountInvalid => "invalid repetition count range",
             RepetitionCountUnclosed => "unclosed counted repetition",
             RepetitionMissing => "repetition operator missing expression",
+            UnicodeClassInvalid => "invalid Unicode character class",
             UnsupportedBackreference => "backreferences are not supported",
             UnsupportedLookAround => "look-around is not supported",
             _ => unreachable!(),
@@ -293,6 +299,9 @@ impl fmt::Display for ErrorKind {
             RepetitionMissing => {
                 write!(f, "repetition operator missing expression")
             }
+            UnicodeClassInvalid => {
+                write!(f, "invalid Unicode character class")
+            }
             UnsupportedBackreference => {
                 write!(f, "backreferences are not supported")
             }

regex-syntax/src/ast/parse.rs

@@ -2095,6 +2095,12 @@ impl<'s, P: Borrow<Parser>> ParserI<'s, P> {
         } else {
             let start = self.pos();
             let c = self.char();
+            if c == '\\' {
+                return Err(self.error(
+                    self.span_char(),
+                    ast::ErrorKind::UnicodeClassInvalid,
+                ));
+            }
             self.bump_and_bump_space();
             let kind = ast::ClassUnicodeKind::OneLetter(c);
             (start, kind)
@@ -5713,6 +5719,20 @@ bar
                 ],
             }))
         );
+        assert_eq!(
+            parser(r"\p\{").parse().unwrap_err(),
+            TestError {
+                span: span(2..3),
+                kind: ast::ErrorKind::UnicodeClassInvalid,
+            }
+        );
+        assert_eq!(
+            parser(r"\P\{").parse().unwrap_err(),
+            TestError {
+                span: span(2..3),
+                kind: ast::ErrorKind::UnicodeClassInvalid,
+            }
+        );
     }
 
     #[test]

regex-syntax/src/error.rs

@@ -40,6 +40,8 @@ impl From<hir::Error> for Error {
 }
 
 impl error::Error for Error {
+    // TODO: Remove this method entirely on the next breaking semver release.
+    #[allow(deprecated)]
     fn description(&self) -> &str {
         match *self {
             Error::Parse(ref x) => x.description(),

regex-syntax/src/hir/mod.rs

@@ -91,6 +91,8 @@ pub enum ErrorKind {
 }
 
 impl ErrorKind {
+    // TODO: Remove this method entirely on the next breaking semver release.
+    #[allow(deprecated)]
     fn description(&self) -> &str {
         use self::ErrorKind::*;
         match *self {
@@ -113,6 +115,8 @@ impl ErrorKind {
 }
 
 impl error::Error for Error {
+    // TODO: Remove this method entirely on the next breaking semver release.
+    #[allow(deprecated)]
     fn description(&self) -> &str {
         self.kind.description()
     }
@@ -126,6 +130,8 @@ impl fmt::Display for Error {
 
 impl fmt::Display for ErrorKind {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        // TODO: Remove this on the next breaking semver release.
+        #[allow(deprecated)]
         f.write_str(self.description())
     }
 }

regex-syntax/src/unicode.rs

@@ -277,7 +277,7 @@ enum CanonicalClassQuery {
 
 /// Looks up a Unicode class given a query. If one doesn't exist, then
 /// `None` is returned.
-pub fn class<'a>(query: ClassQuery<'a>) -> Result<hir::ClassUnicode> {
+pub fn class(query: ClassQuery) -> Result<hir::ClassUnicode> {
     use self::CanonicalClassQuery::*;
 
     match query.canonicalize()? {

regex-syntax/src/unicode_tables/property_bool.rs

@@ -8203,12 +8203,8 @@ pub const OTHER_GRAPHEME_EXTEND: &'static [(char, char)] = &[
 pub const OTHER_ID_CONTINUE: &'static [(char, char)] =
     &[('·', '·'), ('·', '·'), ('፩', '፱'), ('᧚', '᧚')];
 
-pub const OTHER_ID_START: &'static [(char, char)] = &[
-    ('\u{1885}', '\u{1886}'),
-    ('℘', '℘'),
-    ('℮', '℮'),
-    ('゛', '゜'),
-];
+pub const OTHER_ID_START: &'static [(char, char)] =
+    &[('\u{1885}', '\u{1886}'), ('℘', '℘'), ('℮', '℮'), ('゛', '゜')];
 
 pub const OTHER_LOWERCASE: &'static [(char, char)] = &[
     ('ª', 'ª'),
@@ -8370,13 +8366,8 @@ pub const OTHER_MATH: &'static [(char, char)] = &[
     ('𞺫', '𞺻'),
 ];
 
-pub const OTHER_UPPERCASE: &'static [(char, char)] = &[
-    ('Ⅰ', 'Ⅿ'),
-    ('Ⓐ', 'Ⓩ'),
-    ('🄰', '🅉'),
-    ('🅐', '🅩'),
-    ('🅰', '🆉'),
-];
+pub const OTHER_UPPERCASE: &'static [(char, char)] =
+    &[('Ⅰ', 'Ⅿ'), ('Ⓐ', 'Ⓩ'), ('🄰', '🅉'), ('🅐', '🅩'), ('🅰', '🆉')];
 
 pub const PATTERN_SYNTAX: &'static [(char, char)] = &[
     ('!', '/'),