tr1/regex

Go to the documentation of this file.

// class template regex -*- C++ -*-

// Copyright (C) 2007 Free Software Foundation, Inc.
//
// This file is part of the GNU ISO C++ Library.  This library is free
// software; you can redistribute it and/or modify it under the
// terms of the GNU General Public License as published by the
// Free Software Foundation; either version 2, or (at your option)
// any later version.

// This library is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
// GNU General Public License for more details.

// You should have received a copy of the GNU General Public License along
// with this library; see the file COPYING.  If not, write to the Free
// Software Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301,
// USA.

// As a special exception, you may use this file as part of a free software
// library without restriction.  Specifically, if other files instantiate
// templates or use macros or inline functions from this file, or you compile
// this file and link it with other files to produce an executable, this
// file does not by itself cause the resulting executable to be covered by
// the GNU General Public License.  This exception does not however
// invalidate any other reasons why the executable file might be covered by
// the GNU General Public License.

/**
 * @file tr1/regex
 * @author Stephen M. Webb  <stephen.webb@bregmasoft.ca>
 * This is a TR1 C++ Library header. 
 */

#ifndef _TR1_REGEX
#define _TR1_REGEX 1

#include <bits/c++config.h>
#include <algorithm>
#include <bitset>
#include <iterator>
#include <locale>
#include <stdexcept>
#include <string>
#include <utility>
#include <vector>

namespace std
{
_GLIBCXX_BEGIN_NAMESPACE(_GLIBCXX_TR1)

/**
 * @addtogroup tr1_regex Regular Expressions
 * A facility for performing regular expression pattern matching.
 * @{
 */

namespace regex_constants
{
  // [7.5.1] Bitmask Type syntax_option_type
  enum __syntax_option
    {
      _S_icase,
      _S_nosubs,
      _S_optimize,
      _S_collate,
      _S_ECMAScript,
      _S_basic,
      _S_extended,
      _S_awk,
      _S_grep,
      _S_egrep,
      _S_syntax_last
    };

  /**
   * @brief This is a bitmask type indicating how to interpret the regex.
   *
   * The @c syntax_option_type is implementation defined but it is valid to
   * perform bitwise operations on these values and expect the right thing to
   * happen.
   *
   * A valid value of type syntax_option_type shall have exactly one of the
   * elements @c ECMAScript, @c basic, @c extended, @c awk, @c grep, @c egrep
   * set.
   */
  typedef unsigned int syntax_option_type;

  /// Specifies that the matching of regular expressions against a character
  /// sequence shall be performed without regard to case.
  static const syntax_option_type icase      = 1 << _S_icase;

  /// Specifies that when a regular expression is matched against a character
  /// container sequence, no sub-expression matches are to be stored in the
  /// supplied match_results structure.
  static const syntax_option_type nosubs     = 1 << _S_nosubs;

  /// Specifies that the regular expression engine should pay more attention to
  /// the speed with which regular expressions are matched, and less to the
  /// speed with which regular expression objects are constructed. Otherwise
  /// it has no detectable effect on the program output.
  static const syntax_option_type optimize   = 1 << _S_optimize;

  /// Specifies that character ranges of the form [a-b] should be locale
  /// sensitive.
  static const syntax_option_type collate    = 1 << _S_collate;

  /// Specifies that the grammar recognized by the regular expression engine is
  /// that used by ECMAScript in ECMA-262 [Ecma International, ECMAScript
  /// Language Specification, Standard Ecma-262, third edition, 1999], as
  /// modified in tr1 section [7.13].  This grammar is similar to that defined
  /// in the PERL scripting language but extended with elements found in the
  /// POSIX regular expression grammar.
  static const syntax_option_type ECMAScript = 1 << _S_ECMAScript;

  /// Specifies that the grammar recognized by the regular expression engine is
  /// that used by POSIX basic regular expressions in IEEE Std 1003.1-2001,
  /// Portable Operating System Interface (POSIX), Base Definitions and
  /// Headers, Section 9, Regular Expressions [IEEE, Information Technology --
  /// Portable Operating System Interface (POSIX), IEEE Standard 1003.1-2001].
  static const syntax_option_type basic      = 1 << _S_basic;

  /// Specifies that the grammar recognized by the regular expression engine is
  /// that used by POSIX extended regular expressions in IEEE Std 1003.1-2001,
  /// Portable Operating System Interface (POSIX), Base Definitions and Headers,
  /// Section 9, Regular Expressions.
  static const syntax_option_type extended   = 1 << _S_extended;

  /// Specifies that the grammar recognized by the regular expression engine is
  /// that used by POSIX utility awk in IEEE Std 1003.1-2001.  This option is
  /// identical to syntax_option_type extended, except that C-style escape
  /// sequences are supported.  These sequences are, explicitly, '\\', '\a',
  /// '\b', '\f', '\n', '\r', '\t' , '\v', '\"', '\\', and '\ddd' (where ddd is
  /// one, two, or three octal digits).  
  static const syntax_option_type awk        = 1 << _S_awk;

  /// Specifies that the grammar recognized by the regular expression engine is
  /// that used by POSIX utility grep in IEEE Std 1003.1-2001.  This option is
  /// identical to syntax_option_type basic, except that newlines are treated
  /// as whitespace.
  static const syntax_option_type grep       = 1 << _S_grep;

  /// Specifies that the grammar recognized by the regular expression engine is
  /// that used by POSIX utility grep when given the -E option in
  /// IEEE Std 1003.1-2001.  This option is identical to syntax_option_type 
  /// extended, except that newlines are treated as whitespace.
  static const syntax_option_type egrep      = 1 << _S_egrep;


  // [7.5.2] Bitmask Type match_flag_type
  enum __match_flag
    {
      _S_not_bol,
      _S_not_eol,
      _S_not_bow,
      _S_not_eow,
      _S_any,
      _S_not_null,
      _S_continuous,
      _S_prev_avail,
      _S_sed,
      _S_no_copy,
      _S_first_only,
      _S_match_flag_last
    };

  /**
   * @brief This is a bitmask type indicating regex matching rules.
   *
   * Matching a regular expression against a sequence of characters [first,
   * last) proceeds according to the rules of the grammar specified for the
   * regular expression object, modified according to the effects listed
   * below for any bitmask elements set.
   *
   * The @c match_flag_type is implementation defined but it is valid to
   * perform bitwise operations on these values and expect the right thing to
   * happen.
   */
  typedef std::bitset<_S_match_flag_last> match_flag_type;

  static const match_flag_type match_default     = 0;

  /// The first character in the sequence [first, last) is treated as though it
  /// is not at the beginning of a line, so the character "^" in the regular
  /// expression shall not match [first, first).
  static const match_flag_type match_not_bol     = 1 << _S_not_bol;

  /// The last character in the sequence [first, last) is treated as though it
  /// is not at the end of a line, so the character "$" in the regular
  /// expression shall not match [last, last).
  static const match_flag_type match_not_eol     = 1 << _S_not_eol;
   
  /// The expression "\b" is not matched against the sub-sequence
  /// [first,first).
  static const match_flag_type match_not_bow     = 1 << _S_not_bow;
   
  /// The expression "\b" should not be matched against the sub-sequence
  /// [last,last).
  static const match_flag_type match_not_eow     = 1 << _S_not_eow;
   
  /// If more than one match is possible then any match is an acceptable
  /// result.
  static const match_flag_type match_any         = 1 << _S_any;
   
  /// The expression does not match an empty sequence.
  static const match_flag_type match_not_null    = 1 << _S_not_null;
   
  /// The expression only matchs a sub-sequence that begins at first .
  static const match_flag_type match_continuous  = 1 << _S_continuous;
   
  /// --first is a valid iterator position.  When this flag is set then the
  /// flags match_not_bol and match_not_bow are ignored by the regular
  /// expression algorithms 7.11 and iterators 7.12.
  static const match_flag_type match_prev_avail  = 1 << _S_prev_avail;

  /// When a regular expression match is to be replaced by a new string, the
  /// new string is constructed using the rules used by the ECMAScript replace
  /// function in ECMA- 262 [Ecma International, ECMAScript Language
  /// Specification, Standard Ecma-262, third edition, 1999], part 15.5.4.11
  /// String.prototype.replace. In addition, during search and replace
  /// operations all non-overlapping occurrences of the regular expression
  /// are located and replaced, and sections of the input that did not match
  /// the expression are copied unchanged to the output string.
  ///
  /// Format strings (from ECMA-262 [15.5.4.11]):
  /// $$  $
  /// $&  The matched substring.
  /// $`  The portion of <em>string</em> that preceeds the matched substring.
  /// $'  The portion of <em>string</em> that follows the matched substring.
  /// $n  The nth capture, where n is in [1,9] and $n is not followed by a
  ///     decimal digit.  If n <= m and the nth capture is undefined, use the
  ///     empty string
  ///     instead. If n > m, the result is implementation-defined.
  /// $nn The nnth capture, where nn is a two-digit decimal number on [01, 99].
  ///     If nn <= m and the nth capture is undefined, use the empty string
  ///     instead. If nn > m, the result is implementation-defined.
  ///
  static const match_flag_type format_default    = 0;

  /// When a regular expression match is to be replaced by a new string, the
  /// new string is constructed using the rules used by the POSIX sed utility
  /// in IEEE Std 1003.1- 2001 [IEEE, Information Technology -- Portable
  /// Operating System Interface (POSIX), IEEE Standard 1003.1-2001].
  static const match_flag_type format_sed        = 1 << _S_sed;

  /// During a search and replace operation, sections of the character
  /// container sequence being searched that do not match the regular
  /// expression shall not be copied to the output string.
  static const match_flag_type format_no_copy    = 1 << _S_no_copy;

  /// When specified during a search and replace operation, only the first
  /// occurrence of the regular expression shall be replaced.
  static const match_flag_type format_first_only = 1 << _S_first_only;


  // [7.5.3] implementation-defined error type
  enum error_type
    {
      _S_error_collate,
      _S_error_ctype,
      _S_error_escape,
      _S_error_backref,
      _S_error_brack,
      _S_error_paren,
      _S_error_brace,
      _S_error_badbrace,
      _S_error_range,
      _S_error_space,
      _S_error_badrepeat,
      _S_error_complexity,
      _S_error_stack,
      _S_error_last
    };

  /// The expression contained an invalid collating element name.
  static const error_type error_collate(_S_error_collate);

  /// The expression contained an invalid character class name.
  static const error_type error_ctype(_S_error_ctype);

  /// The expression contained an invalid escaped character, or a trailing
  /// escape.
  static const error_type error_escape(_S_error_escape);

  /// The expression contained an invalid back reference.
  static const error_type error_backref(_S_error_backref);

  /// The expression contained mismatched [ and ].
  static const error_type error_brack(_S_error_brack);

  /// The expression contained mismatched ( and ).
  static const error_type error_paren(_S_error_paren);

  /// The expression contained mismatched { and }
  static const error_type error_brace(_S_error_brace);

  /// The expression contained an invalid range in a {} expression.
  static const error_type error_badbrace(_S_error_badbrace);

  /// The expression contained an invalid character range,
  /// such as [b-a] in most encodings.
  static const error_type error_range(_S_error_range);

  /// There was insufficient memory to convert the expression into a
  /// finite state machine.
  static const error_type error_space(_S_error_space);

  /// One of *?+{ was not preceded by a valid regular expression.
  static const error_type error_badrepeat(_S_error_badrepeat);

  /// The complexity of an attempted match against a regular expression
  /// exceeded a pre-set level.
  static const error_type error_complexity(_S_error_complexity);

  /// There was insufficient memory to determine whether the
  /// regular expression could match the specified character sequence.
  static const error_type error_stack(_S_error_stack);
}


  // [7.8] Class regex_error
  /**
   * Defines the type of objects thrown as exceptions to report errors from the
   * regular expression library.
   */
  class regex_error
  : public std::runtime_error
  {
  public:
    /**
     * @brief constructs a regex_error object.
     *
     * @param ecode the regex error code.
     */
    explicit
    regex_error(regex_constants::error_type __ecode)
    : std::runtime_error("regex_error"), _M_code(__ecode)
    { }

    /**
     * @brief gets the regex error code.
     *
     * @returns the regex error code.
     */
    regex_constants::error_type
    code() const
    { return _M_code; }

  protected:
    regex_constants::error_type _M_code;
  };


  // [7.7] Class regex_traits
  /**
   * A regular expression traits class that satisfies the requirements of tr1
   * section [7.2].
   *
   * The class %regex is parameterized around a set of related types and
   * functions used to complete the definition of its semantics.  This class
   * satisfies the requirements of such a traits class.
   */
  template<typename _Ch_type>
    struct regex_traits
    {
    public:
      typedef _Ch_type                     char_type;
      typedef std::basic_string<char_type> string_type;
      typedef std::locale                  locale_type;
      typedef std::ctype_base::mask        char_class_type;

    public:
      /**
       * @brief Constructs a default traits object.
       */
      regex_traits()
      { }
      
      /**
       * @brief Gives the length of a C-style string starting at @p __p.
       *
       * @param __p a pointer to the start of a character sequence.
       *
       * @returns the number of characters between @p *__p and the first
       * default-initialized value of type @p char_type.  In other words, uses
       * the C-string algorithm for determiining the length of a sequence of
       * characters.
       */
      static std::size_t
      length(const char_type* __p)
      { return string_type::traits_type::length(__p); }

      /**
       * @brief Performs the identity translation.
       *
       * @param c A character to the locale-specific character set.
       *
       * @returns c.
       */
      char_type
      translate(char_type __c) const
      { return __c; }
      
      /**
       * @brief Translates a character into a case-insensitive equivalent.
       *
       * @param c A character to the locale-specific character set.
       *
       * @returns the locale-specific lower-case equivalent of c.
       * @throws std::bad_cast if the imbued locale does not support the ctype
       *         facet.
       */
      char_type
      translate_nocase(char_type __c) const
      {
    using std::ctype;
    using std::use_facet;
    return use_facet<ctype<char_type> >(_M_locale).tolower(__c);
      }
      
      /**
       * @brief Gets a sort key for a character sequence.
       *
       * @param first beginning of the character sequence.
       * @param last  one-past-the-end of the character sequence.
       *
       * Returns a sort key for the character sequence designated by the
       * iterator range [F1, F2) such that if the character sequence [G1, G2)
       * sorts before the character sequence [H1, H2) then
       * v.transform(G1, G2) < v.transform(H1, H2).
       *
       * What this really does is provide a more efficient way to compare a
       * string to multiple other strings in locales with fancy collation
       * rules and equivalence classes.
       *
       * @returns a locale-specific sort key equivalent to the input range.
       *
       * @throws std::bad_cast if the current locale does not have a collate
       *         facet.
       */
      template<typename _Fwd_iter>
        string_type
        transform(_Fwd_iter __first, _Fwd_iter __last) const
        {
      using std::collate;
      using std::use_facet;
      const collate<_Ch_type>& __c(use_facet<
                       collate<_Ch_type> >(_M_locale));
      string_type __s(__first, __last);
      return __c.transform(__s.data(), __s.data() + __s.size());
    }

      /**
       * @brief Dunno.
       *
       * @param first beginning of the character sequence.
       * @param last  one-past-the-end of the character sequence.
       *
       * Effects: if typeid(use_facet<collate<_Ch_type> >) ==
       * typeid(collate_byname<_Ch_type>) and the form of the sort key
       * returned by collate_byname<_Ch_type>::transform(first, last) is known
       * and can be converted into a primary sort key then returns that key,
       * otherwise returns an empty string. WTF??
       *
       * @todo Implement this function.
       */
      template<typename _Fwd_iter>
        string_type
        transform_primary(_Fwd_iter __first, _Fwd_iter __last) const
        { return string_type(); }

      /**
       * @breief Gets a collation element by name.
       *
       * @param first beginning of the collation element name.
       * @param last  one-past-the-end of the collation element name.
       * 
       * @returns a sequence of one or more characters that represents the
       * collating element consisting of the character sequence designated by
       * the iterator range [first, last). Returns an empty string if the
       * character sequence is not a valid collating element.
       *
       * @todo Implement this function.
       */
      template<typename _Fwd_iter>
        string_type
        lookup_collatename(_Fwd_iter __first, _Fwd_iter __last) const
        { return string_type(); }

      /**
       * @brief Maps one or mire characters to a named character
       *        classification.
       *
       * @param first beginning of the character sequence.
       * @param last  one-past-the-end of the character sequence.
       *
       * @returns an unspecified value that represents the character
       * classification named by the character sequence designated by the
       * iterator range [first, last). The value returned shall be independent
       * of the case of the characters in the character sequence. If the name
       * is not recognized then returns a value that compares equal to 0.
       *
       * At least the following names (or their wide-character equivalent) are
       * supported.
       * - d
       * - w
       * - s
       * - alnum
       * - alpha
       * - blank
       * - cntrl
       * - digit
       * - graph
       * - lower
       * - print
       * - punct
       * - space
       * - upper
       * - xdigit
       *
       * @todo Implement this function.
       */
      template<typename _Fwd_iter>
        char_class_type
        lookup_classname(_Fwd_iter __first, _Fwd_iter __last) const
        { return 0; }

      /**
       * @brief Determines if @p c is a member of an identified class.
       *
       * @param c a character.
       * @param f a class type (as returned from lookup_classname).
       *
       * @returns true if the character @p c is a member of the classification
       * represented by @p f, false otherwise.
       *
       * @throws std::bad_cast if the current locale does not have a ctype
       *         facet.
       */
      bool
      isctype(_Ch_type __c, char_class_type __f) const
      {
    using std::ctype;
    using std::use_facet;
    const ctype<_Ch_type>& __ctype(use_facet<
                       ctype<_Ch_type> >(_M_locale));
    
    if (__ctype.is(__c, __f))
      return true;
    
    // special case of underscore in [[:w:]]
    if (__c == __ctype.widen('_'))
      {
        const char* const __wb[] = "w";
        char_class_type __wt = this->lookup_classname(__wb,
                              __wb + sizeof(__wb));
        if (__f | __wt)
          return true;
      }
      
    // special case of [[:space:]] in [[:blank:]]
    if (__c == __ctype.isspace(__c))
      {
        const char* const __bb[] = "blank";
        char_class_type __bt = this->lookup_classname(__bb,
                              __bb + sizeof(__bb));
        if (__f | __bt)
          return true;
      }
    
    return false;
      }

      /**
       * @brief Converts a digit to an int.
       *
       * @param ch    a character representing a digit.
       * @param radix the radix if the numeric conversion (limited to 8, 10,
       *              or 16).
       * 
       * @returns the value represented by the digit ch in base radix if the
       * character ch is a valid digit in base radix; otherwise returns -1.
       *
       * @todo Implement this function.
       */
      int
      value(_Ch_type __ch, int __radix) const;
      
      /**
       * @brief Imbues the regex_traits object with a copy of a new locale.
       *
       * @param loc A locale.
       *
       * @returns a copy of the previous locale in use by the regex_traits
       *          object.
       *
       * @note Calling imbue with a different locale than the one currently in
       *       use invalidates all cached data held by *this.
       */
      locale_type
      imbue(locale_type __loc)
      {
    std::swap(_M_locale, __loc);
    return __loc;
      }
      
      /**
       * @brief Gets a copy of the current locale in use by the regex_traits
       * object.
       */
      locale_type
      getloc() const
      { return _M_locale; }
      
    protected:
      locale_type _M_locale;
    };


  // [7.8] Class basic_regex
  /**
   * Objects of specializations of this class represent regular expressions
   * constructed from sequences of character type @p _Ch_type.
   *
   * Storage for the regular expression is allocated and deallocated as
   * necessary by the member functions of this class.
   */
  template<typename _Ch_type, typename _Rx_traits = regex_traits<_Ch_type> >
    class basic_regex
    {
    public:
      // types:
      typedef _Ch_type                              value_type;
      typedef regex_constants::syntax_option_type flag_type;
      typedef typename _Rx_traits::locale_type  locale_type;
      typedef typename _Rx_traits::string_type  string_type;

      // [7.8.1] constants
      static const regex_constants::syntax_option_type icase
        = regex_constants::icase;
      static const regex_constants::syntax_option_type nosubs
        = regex_constants::nosubs;
      static const regex_constants::syntax_option_type optimize
        = regex_constants::optimize;
      static const regex_constants::syntax_option_type collate
        = regex_constants::collate;
      static const regex_constants::syntax_option_type ECMAScript
        = regex_constants::ECMAScript;
      static const regex_constants::syntax_option_type basic
        = regex_constants::basic;
      static const regex_constants::syntax_option_type extended
        = regex_constants::extended;
      static const regex_constants::syntax_option_type awk
        = regex_constants::awk;
      static const regex_constants::syntax_option_type grep
        = regex_constants::grep;
      static const regex_constants::syntax_option_type egrep
        = regex_constants::egrep;

      // [7.8.2] construct/copy/destroy
      /**
       * Constructs a basic regular expression that does not match any
       * character sequence.
       */
      basic_regex()
      : _M_flags(regex_constants::ECMAScript), _M_pattern(), _M_mark_count(0)
      { _M_compile(); }

      /**
       * @brief Constructs a basic regular expression from the sequence
       * [p, p + char_traits<_Ch_type>::length(p)) interpreted according to the
       * flags in @p f.
       *
       * @param p A pointer to the start of a C-style null-terminated string
       *          containing a regular expression.
       * @param f Flags indicating the syntax rules and options.
       *
       * @throws regex_error if @p p is not a valid regular expression.
       */
      explicit
      basic_regex(const _Ch_type* __p,
          flag_type __f = regex_constants::ECMAScript)
      : _M_flags(__f), _M_pattern(__p), _M_mark_count(0)
      { _M_compile(); }

      /**
       * @brief Constructs a basic regular expression from the sequence
       * [p, p + len) interpreted according to the flags in @p f.
       *
       * @param p   A pointer to the start of a string containing a regular
       *            expression.
       * @param len The length of the string containing the regular expression.
       * @param f   Flags indicating the syntax rules and options.
       *
       * @throws regex_error if @p p is not a valid regular expression.
       */
      basic_regex(const _Ch_type* __p, std::size_t __len, flag_type __f)
      : _M_flags(__f) , _M_pattern(__p, __len), _M_mark_count(0)
      { _M_compile(); }

      /**
       * @brief Copy-contructs a basic regular expression.
       *
       * @param rhs A @p regex object.
     */
      basic_regex(const basic_regex& __rhs)
      : _M_flags(__rhs._M_flags), _M_pattern(__rhs._M_pattern),
    _M_mark_count(__rhs._M_mark_count)
      { _M_compile(); }

      /**
       * @brief Constructs a basic regular expression from the string
       * @p interpreted according to the flags in @p f.
       *
       * @param p A string containing a regular expression.
       * @param f Flags indicating the syntax rules and options.
       *
       * @throws regex_error if @p p is not a valid regular expression.
       */
      template<typename _Ch_traits, typename _Ch_alloc>
        explicit
        basic_regex(const basic_string<_Ch_type, _Ch_traits, _Ch_alloc>& __s,
            flag_type __f = regex_constants::ECMAScript)
    : _M_flags(__f), _M_pattern(__s), _M_mark_count(0)
        { _M_compile(); }

      /**
       * @brief Constructs a basic regular expression from the range
       * [first, last) interpreted according to the flags in @p f.
       *
       * @param first The start of arange containing a valid regular
       *              expression.
       * @param last  The end of a range containing a valid regular
       *              expression.
       * @param f     The format flags of the regular expression.
       *
       * @throws regex_error if @p p is not a valid regular expression.
       */
      template<typename _InputIterator>
        basic_regex(_InputIterator __first, _InputIterator __last, 
            flag_type __f = regex_constants::ECMAScript)
    : _M_flags(__f), _M_pattern(__first, __last), _M_mark_count(0)
        { _M_compile(); }

      /**
       * @brief Destroys a basic regular expression.
       */
      ~basic_regex()
      { }
      
      /**
       * @brief Assigns one regular expression to another.
       */
      basic_regex&
      operator=(const basic_regex& __rhs)
      { return this->assign(__rhs); }

      /**
       * @brief Replaces a regular expression with a new one constructed from
       * a C-style null-terminated string.
       *
       * @param A pointer to the start of a null-terminated C-style string
       *        containing a regular expression.
       */
      basic_regex&
      operator=(const _Ch_type* __p)
      { return this->assign(__p, flags()); }
      
      /**
       * @brief Replaces a regular expression with a new one constructed from
       * a string.
       *
       * @param A pointer to a string containing a regular expression.
       */
      template<typename _Ch_typeraits, typename _Allocator>
        basic_regex&
        operator=(const basic_string<_Ch_type, _Ch_typeraits, _Allocator>& __s)
        { return this->assign(__s, flags()); }

      // [7.8.3] assign
      /**
       * @brief the real assignment operator.
       *
       * @param that Another regular expression object.
       */
      basic_regex&
      assign(const basic_regex& __that)
      {
    basic_regex __tmp(__that);
    this->swap(__tmp);
    return *this;
      }
      
      /**
       * @brief Assigns a new regular expression to a regex object from a
       * C-style null-terminated string containing a regular expression
       * pattern.
       *
       * @param p     A pointer to a C-style null-terminated string containing
       *              a regular expression pattern.
       * @param flags Syntax option flags.
       *
       * @throws regex_error if p does not contain a valid regular expression
       * pattern interpreted according to @p flags.  If regex_error is thrown,
       * *this remains unchanged.
       */
      basic_regex&
      assign(const _Ch_type* __p,
         flag_type __flags = regex_constants::ECMAScript)
      { return this->assign(string_type(__p), __flags); }

      /**
       * @brief Assigns a new regular expression to a regex object from a
       * C-style string containing a regular expression pattern.
       *
       * @param p     A pointer to a C-style string containing a
       *              regular expression pattern.
       * @param len   The length of the regular expression pattern string.
       * @param flags Syntax option flags.
       *
       * @throws regex_error if p does not contain a valid regular expression
       * pattern interpreted according to @p flags.  If regex_error is thrown,
       * *this remains unchanged.
       */
      basic_regex&
      assign(const _Ch_type* __p, std::size_t __len, flag_type __flags)
      { return this->assign(string_type(__p, __len), __flags); }

      /**
       * @brief Assigns a new regular expression to a regex object from a 
       * string containing a regular expression pattern.
       *
       * @param s     A string containing a regular expression pattern.
       * @param flags Syntax option flags.
       *
       * @throws regex_error if p does not contain a valid regular expression
       * pattern interpreted according to @p flags.  If regex_error is thrown,
       * *this remains unchanged.
       */
      template<typename _Ch_typeraits, typename _Allocator>
        basic_regex&
        assign(const basic_string<_Ch_type, _Ch_typeraits, _Allocator>& __s,
           flag_type __f = regex_constants::ECMAScript)
        { 
      basic_regex __tmp(__s, __f);
      this->swap(__tmp);
      return *this;
    }

      /**
       * @brief Assigns a new regular expression to a regex object.
       *
       * @param first The start of a range containing a valid regular
       *              expression.
       * @param last  The end of a range containing a valid regular
       *              expression.
       * @param flags Syntax option flags.
       *
       * @throws regex_error if p does not contain a valid regular expression
       * pattern interpreted according to @p flags.  If regex_error is thrown,
       * *this remains unchanged.
       */
      template<typename _InputIterator>
        basic_regex&
        assign(_InputIterator __first, _InputIterator __last,
           flag_type __flags = regex_constants::ECMAScript)
        { return this->assign(string_type(__first, __last), __flags); }

      // [7.8.4] const operations
      /**
       * @brief Gets the number of marked subexpressions within the regular
       * expresison.
       */
      unsigned int
      mark_count() const
      { return _M_mark_count; }
      
      /**
       * @brief Gets the flags used to construct the regular expression
       * or in the last call to assign().
       */
      flag_type
      flags() const
      { return _M_flags; }
      
      // [7.8.5] locale
      /**
       * @brief Imbues the regular expression object with the given locale.
       *
       * @param loc A locale.
       */
      locale_type
      imbue(locale_type __loc)
      { return _M_traits.imbue(__loc); }
      
      /**
       * @brief Gets the locale currently imbued in the regular expression
       *        object.
       */
      locale_type
      getloc() const
      { return _M_traits.getloc(); }
      
      // [7.8.6] swap
      /**
       * @brief Swaps the contents of two regular expression obects.
       *
       * @param rhs Another regular expression object.
       */
      void
      swap(basic_regex& __rhs)
      {
    std::swap(_M_flags,      __rhs._M_flags);
    std::swap(_M_pattern,    __rhs._M_pattern);
    std::swap(_M_mark_count, __rhs._M_mark_count);
    std::swap(_M_traits,     __rhs._M_traits);
      }
      
    private:
      /**
       * @brief Compiles a regular expression pattern into a NFA.
       * @todo Implement this function.
       */
      void _M_compile()
      { }

    protected:
      flag_type    _M_flags;
      string_type  _M_pattern;
      unsigned int _M_mark_count;
      _Rx_traits   _M_traits;
    };
  
  typedef basic_regex<char>    regex;
#ifdef _GLIBCXX_USE_WCHAR_T
  typedef basic_regex<wchar_t> wregex;
#endif


  // [7.8.6] basic_regex swap
  /**
   * @brief Swaps the contents of two regular expression objects.
   * @param lhs First regular expression.
   * @param rhs Second regular expression.
   */
  template<typename _Ch_type, typename _Rx_traits>
    inline void
    swap(basic_regex<_Ch_type, _Rx_traits>& __lhs,
     basic_regex<_Ch_type, _Rx_traits>& __rhs)
    { return __lhs.swap(__rhs); }


  // [7.9] Class template sub_match
  /**
   * A sequence of characters matched by a particular marked sub-expression.
   *
   * An object of this class is essentially a pair of iterators marking a
   * matched subexpression within a regular expression pattern match. Such
   * objects can be converted to and compared with std::basic_string objects
   * of a similar base character type as the pattern matched by the regular
   * expression.
   *
   * The iterators that make up the pair are the usual half-open interval
   * referencing the actual original pattern matched.
   */
  template<typename _BiIter>
    class sub_match : public std::pair<_BiIter, _BiIter>
    {
    public:
      typedef typename iterator_traits<_BiIter>::value_type      value_type;
      typedef typename iterator_traits<_BiIter>::difference_type
                                                            difference_type;
      typedef _BiIter                                              iterator;

    public:
      bool matched;
      
      /**
       * Gets the length of the matching sequence.
       */
      difference_type
      length() const
      { return this->matched ? std::distance(this->first, this->second) : 0; }

      /**
       * @brief Gets the matching sequence as a string.
       *
       * @returns the matching sequence as a string.
       *
       * This is the implicit conversion operator.  It is identical to the
       * str() member function except that it will want to pop up in
       * unexpected places and cause a great deal of confusion and cursing
       * from the unwary.
       */
      operator basic_string<value_type>() const
      {
    return this->matched
      ? std::basic_string<value_type>(this->first, this->second)
      : std::basic_string<value_type>();
      }
      
      /**
       * @brief Gets the matching sequence as a string.
       *
       * @returns the matching sequence as a string.
       */
      basic_string<value_type>
      str() const
      {
    return this->matched
      ? std::basic_string<value_type>(this->first, this->second)
      : std::basic_string<value_type>();
      }
      
      /**
       * @brief Compares this and another matched sequence.
       *
       * @param s Another matched sequence to compare to this one.
       *
       * @retval <0 this matched sequence will collate before @p s.
       * @retval =0 this matched sequence is equivalent to @p s.
       * @retval <0 this matched sequence will collate after @p s.
       */
      int
      compare(const sub_match& __s) const
      { return this->str().compare(__s.str()); }

      /**
       * @brief Compares this sub_match to a string.
       *
       * @param s A string to compare to this sub_match.
       *
       * @retval <0 this matched sequence will collate before @p s.
       * @retval =0 this matched sequence is equivalent to @p s.
       * @retval <0 this matched sequence will collate after @p s.
       */
      int
      compare(const basic_string<value_type>& __s) const
      { return this->str().compare(__s); }
      
      /**
       * @brief Compares this sub_match to a C-style string.
       *
       * @param s A C-style string to compare to this sub_match.
       *
       * @retval <0 this matched sequence will collate before @p s.
       * @retval =0 this matched sequence is equivalent to @p s.
       * @retval <0 this matched sequence will collate after @p s.
       */
      int
      compare(const value_type* __s) const
      { return this->str().compare(__s); }
    };
  
  
  typedef sub_match<const char*>             csub_match;
  typedef sub_match<string::const_iterator>  ssub_match;
#ifdef _GLIBCXX_USE_WCHAR_T
  typedef sub_match<const wchar_t*>          wcsub_match;
  typedef sub_match<wstring::const_iterator> wssub_match;
#endif

  // [7.9.2] sub_match non-member operators
  
  /**
   * @brief Tests the equivalence of two regular expression submatches.
   * @param lhs First regular expression submatch.
   * @param rhs Second regular expression submatch.
   * @returns true if @a lhs  is equivalent to @a rhs, false otherwise.
   */
  template<typename _BiIter>
    inline bool
    operator==(const sub_match<_BiIter>& __lhs,
           const sub_match<_BiIter>& __rhs)
    { return __lhs.compare(__rhs) == 0; }

  /**
   * @brief Tests the inequivalence of two regular expression submatches.
   * @param lhs First regular expression submatch.
   * @param rhs Second regular expression submatch.
   * @returns true if @a lhs  is not equivalent to @a rhs, false otherwise.
   */
  template<typename _BiIter>
    inline bool
    operator!=(const sub_match<_BiIter>& __lhs,
           const sub_match<_BiIter>& __rhs)
    { return __lhs.compare(__rhs) != 0; }

  /**
   * @brief Tests the ordering of two regular expression submatches.
   * @param lhs First regular expression submatch.
   * @param rhs Second regular expression submatch.
   * @returns true if @a lhs precedes @a rhs, false otherwise.
   */
  template<typename _BiIter>
    inline bool