Chris@16: /*
Chris@16:  *
Chris@16:  * Copyright (c) 2003
Chris@16:  * John Maddock
Chris@16:  *
Chris@16:  * Use, modification and distribution are subject to the 
Chris@16:  * Boost Software License, Version 1.0. (See accompanying file 
Chris@16:  * LICENSE_1_0.txt or copy at http://www.boost.org/LICENSE_1_0.txt)
Chris@16:  *
Chris@16:  */
Chris@16: 
Chris@16:  /*
Chris@16:   *   LOCATION:    see http://www.boost.org for most recent version.
Chris@16:   *   FILE         u32regex_token_iterator.hpp
Chris@16:   *   VERSION      see <boost/version.hpp>
Chris@16:   *   DESCRIPTION: Provides u32regex_token_iterator implementation.
Chris@16:   */
Chris@16: 
Chris@16: #ifndef BOOST_REGEX_V4_U32REGEX_TOKEN_ITERATOR_HPP
Chris@16: #define BOOST_REGEX_V4_U32REGEX_TOKEN_ITERATOR_HPP
Chris@16: 
Chris@16: #if (BOOST_WORKAROUND(__BORLANDC__, >= 0x560) && BOOST_WORKAROUND(__BORLANDC__, BOOST_TESTED_AT(0x570)))\
Chris@16:       || BOOST_WORKAROUND(__MWERKS__, BOOST_TESTED_AT(0x3003))
Chris@16: //
Chris@16: // Borland C++ Builder 6, and Visual C++ 6,
Chris@16: // can't cope with the array template constructor
Chris@16: // so we have a template member that will accept any type as 
Chris@16: // argument, and then assert that is really is an array:
Chris@16: //
Chris@16: #include <boost/static_assert.hpp>
Chris@16: #include <boost/type_traits/is_array.hpp>
Chris@16: #endif
Chris@16: 
Chris@16: namespace boost{
Chris@16: 
Chris@16: #ifdef BOOST_HAS_ABI_HEADERS
Chris@16: #  include BOOST_ABI_PREFIX
Chris@16: #endif
Chris@101: #ifdef BOOST_MSVC
Chris@16: #  pragma warning(push)
Chris@16: #  pragma warning(disable:4700)
Chris@16: #endif
Chris@16: 
Chris@16: template <class BidirectionalIterator>
Chris@16: class u32regex_token_iterator_implementation 
Chris@16: {
Chris@16:    typedef u32regex                              regex_type;
Chris@16:    typedef sub_match<BidirectionalIterator>      value_type;
Chris@16: 
Chris@16:    match_results<BidirectionalIterator> what;   // current match
Chris@16:    BidirectionalIterator                end;    // end of search area
Chris@16:    BidirectionalIterator                base;   // start of search area
Chris@16:    const regex_type                     re;     // the expression
Chris@16:    match_flag_type                      flags;  // match flags
Chris@16:    value_type                           result; // the current string result
Chris@16:    int                                  N;      // the current sub-expression being enumerated
Chris@16:    std::vector<int>                     subs;   // the sub-expressions to enumerate
Chris@16: 
Chris@16: public:
Chris@16:    u32regex_token_iterator_implementation(const regex_type* p, BidirectionalIterator last, int sub, match_flag_type f)
Chris@16:       : end(last), re(*p), flags(f){ subs.push_back(sub); }
Chris@16:    u32regex_token_iterator_implementation(const regex_type* p, BidirectionalIterator last, const std::vector<int>& v, match_flag_type f)
Chris@16:       : end(last), re(*p), flags(f), subs(v){}
Chris@101: #if (BOOST_WORKAROUND(__BORLANDC__, >= 0x560) && BOOST_WORKAROUND(__BORLANDC__, BOOST_TESTED_AT(0x570)))\
Chris@16:       || BOOST_WORKAROUND(__MWERKS__, BOOST_TESTED_AT(0x3003)) \
Chris@16:       || BOOST_WORKAROUND(__HP_aCC, < 60700)
Chris@16:    template <class T>
Chris@16:    u32regex_token_iterator_implementation(const regex_type* p, BidirectionalIterator last, const T& submatches, match_flag_type f)
Chris@16:       : end(last), re(*p), flags(f)
Chris@16:    {
Chris@16:       // assert that T really is an array:
Chris@16:       BOOST_STATIC_ASSERT(::boost::is_array<T>::value);
Chris@16:       const std::size_t array_size = sizeof(T) / sizeof(submatches[0]);
Chris@16:       for(std::size_t i = 0; i < array_size; ++i)
Chris@16:       {
Chris@16:          subs.push_back(submatches[i]);
Chris@16:       }
Chris@16:    }
Chris@16: #else
Chris@16:    template <std::size_t CN>
Chris@16:    u32regex_token_iterator_implementation(const regex_type* p, BidirectionalIterator last, const int (&submatches)[CN], match_flag_type f)
Chris@16:       : end(last), re(*p), flags(f)
Chris@16:    {
Chris@16:       for(std::size_t i = 0; i < CN; ++i)
Chris@16:       {
Chris@16:          subs.push_back(submatches[i]);
Chris@16:       }
Chris@16:    }
Chris@16: #endif
Chris@16: 
Chris@16:    bool init(BidirectionalIterator first)
Chris@16:    {
Chris@16:       base = first;
Chris@16:       N = 0;
Chris@16:       if(u32regex_search(first, end, what, re, flags, base) == true)
Chris@16:       {
Chris@16:          N = 0;
Chris@16:          result = ((subs[N] == -1) ? what.prefix() : what[(int)subs[N]]);
Chris@16:          return true;
Chris@16:       }
Chris@16:       else if((subs[N] == -1) && (first != end))
Chris@16:       {
Chris@16:          result.first = first;
Chris@16:          result.second = end;
Chris@16:          result.matched = (first != end);
Chris@16:          N = -1;
Chris@16:          return true;
Chris@16:       }
Chris@16:       return false;
Chris@16:    }
Chris@16:    bool compare(const u32regex_token_iterator_implementation& that)
Chris@16:    {
Chris@16:       if(this == &that) return true;
Chris@16:       return (&re.get_data() == &that.re.get_data()) 
Chris@16:          && (end == that.end) 
Chris@16:          && (flags == that.flags) 
Chris@16:          && (N == that.N) 
Chris@16:          && (what[0].first == that.what[0].first) 
Chris@16:          && (what[0].second == that.what[0].second);
Chris@16:    }
Chris@16:    const value_type& get()
Chris@16:    { return result; }
Chris@16:    bool next()
Chris@16:    {
Chris@16:       if(N == -1)
Chris@16:          return false;
Chris@16:       if(N+1 < (int)subs.size())
Chris@16:       {
Chris@16:          ++N;
Chris@16:          result =((subs[N] == -1) ? what.prefix() : what[subs[N]]);
Chris@16:          return true;
Chris@16:       }
Chris@16:       //if(what.prefix().first != what[0].second)
Chris@16:       //   flags |= match_prev_avail | regex_constants::match_not_bob;
Chris@16:       BidirectionalIterator last_end(what[0].second);
Chris@16:       if(u32regex_search(last_end, end, what, re, ((what[0].first == what[0].second) ? flags | regex_constants::match_not_initial_null : flags), base))
Chris@16:       {
Chris@16:          N =0;
Chris@16:          result =((subs[N] == -1) ? what.prefix() : what[subs[N]]);
Chris@16:          return true;
Chris@16:       }
Chris@16:       else if((last_end != end) && (subs[0] == -1))
Chris@16:       {
Chris@16:          N =-1;
Chris@16:          result.first = last_end;
Chris@16:          result.second = end;
Chris@16:          result.matched = (last_end != end);
Chris@16:          return true;
Chris@16:       }
Chris@16:       return false;
Chris@16:    }
Chris@16: private:
Chris@16:    u32regex_token_iterator_implementation& operator=(const u32regex_token_iterator_implementation&);
Chris@16: };
Chris@16: 
Chris@16: template <class BidirectionalIterator>
Chris@16: class u32regex_token_iterator 
Chris@16: #ifndef BOOST_NO_STD_ITERATOR
Chris@16:    : public std::iterator<
Chris@16:          std::forward_iterator_tag, 
Chris@16:          sub_match<BidirectionalIterator>,
Chris@16:          typename re_detail::regex_iterator_traits<BidirectionalIterator>::difference_type,
Chris@16:          const sub_match<BidirectionalIterator>*,
Chris@16:          const sub_match<BidirectionalIterator>& >         
Chris@16: #endif
Chris@16: {
Chris@16: private:
Chris@16:    typedef u32regex_token_iterator_implementation<BidirectionalIterator> impl;
Chris@16:    typedef shared_ptr<impl> pimpl;
Chris@16: public:
Chris@16:    typedef          u32regex                                                regex_type;
Chris@16:    typedef          sub_match<BidirectionalIterator>                        value_type;
Chris@16:    typedef typename re_detail::regex_iterator_traits<BidirectionalIterator>::difference_type 
Chris@16:                                                                             difference_type;
Chris@16:    typedef          const value_type*                                       pointer;
Chris@16:    typedef          const value_type&                                       reference; 
Chris@16:    typedef          std::forward_iterator_tag                               iterator_category;
Chris@16:    
Chris@16:    u32regex_token_iterator(){}
Chris@16:    u32regex_token_iterator(BidirectionalIterator a, BidirectionalIterator b, const regex_type& re, 
Chris@16:                         int submatch = 0, match_flag_type m = match_default)
Chris@16:                         : pdata(new impl(&re, b, submatch, m))
Chris@16:    {
Chris@16:       if(!pdata->init(a))
Chris@16:          pdata.reset();
Chris@16:    }
Chris@16:    u32regex_token_iterator(BidirectionalIterator a, BidirectionalIterator b, const regex_type& re, 
Chris@16:                         const std::vector<int>& submatches, match_flag_type m = match_default)
Chris@16:                         : pdata(new impl(&re, b, submatches, m))
Chris@16:    {
Chris@16:       if(!pdata->init(a))
Chris@16:          pdata.reset();
Chris@16:    }
Chris@101: #if (BOOST_WORKAROUND(__BORLANDC__, >= 0x560) && BOOST_WORKAROUND(__BORLANDC__, BOOST_TESTED_AT(0x570)))\
Chris@16:       || BOOST_WORKAROUND(__MWERKS__, BOOST_TESTED_AT(0x3003)) \
Chris@16:       || BOOST_WORKAROUND(__HP_aCC, < 60700)
Chris@16:    template <class T>
Chris@16:    u32regex_token_iterator(BidirectionalIterator a, BidirectionalIterator b, const regex_type& re,
Chris@16:                         const T& submatches, match_flag_type m = match_default)
Chris@16:                         : pdata(new impl(&re, b, submatches, m))
Chris@16:    {
Chris@16:       if(!pdata->init(a))
Chris@16:          pdata.reset();
Chris@16:    }
Chris@16: #else
Chris@16:    template <std::size_t N>
Chris@16:    u32regex_token_iterator(BidirectionalIterator a, BidirectionalIterator b, const regex_type& re,
Chris@16:                         const int (&submatches)[N], match_flag_type m = match_default)
Chris@16:                         : pdata(new impl(&re, b, submatches, m))
Chris@16:    {
Chris@16:       if(!pdata->init(a))
Chris@16:          pdata.reset();
Chris@16:    }
Chris@16: #endif
Chris@16:    u32regex_token_iterator(const u32regex_token_iterator& that)
Chris@16:       : pdata(that.pdata) {}
Chris@16:    u32regex_token_iterator& operator=(const u32regex_token_iterator& that)
Chris@16:    {
Chris@16:       pdata = that.pdata;
Chris@16:       return *this;
Chris@16:    }
Chris@16:    bool operator==(const u32regex_token_iterator& that)const
Chris@16:    { 
Chris@16:       if((pdata.get() == 0) || (that.pdata.get() == 0))
Chris@16:          return pdata.get() == that.pdata.get();
Chris@16:       return pdata->compare(*(that.pdata.get())); 
Chris@16:    }
Chris@16:    bool operator!=(const u32regex_token_iterator& that)const
Chris@16:    { return !(*this == that); }
Chris@16:    const value_type& operator*()const
Chris@16:    { return pdata->get(); }
Chris@16:    const value_type* operator->()const
Chris@16:    { return &(pdata->get()); }
Chris@16:    u32regex_token_iterator& operator++()
Chris@16:    {
Chris@16:       cow();
Chris@16:       if(0 == pdata->next())
Chris@16:       {
Chris@16:          pdata.reset();
Chris@16:       }
Chris@16:       return *this;
Chris@16:    }
Chris@16:    u32regex_token_iterator operator++(int)
Chris@16:    {
Chris@16:       u32regex_token_iterator result(*this);
Chris@16:       ++(*this);
Chris@16:       return result;
Chris@16:    }
Chris@16: private:
Chris@16: 
Chris@16:    pimpl pdata;
Chris@16: 
Chris@16:    void cow()
Chris@16:    {
Chris@16:       // copy-on-write
Chris@16:       if(pdata.get() && !pdata.unique())
Chris@16:       {
Chris@16:          pdata.reset(new impl(*(pdata.get())));
Chris@16:       }
Chris@16:    }
Chris@16: };
Chris@16: 
Chris@16: typedef u32regex_token_iterator<const char*> utf8regex_token_iterator;
Chris@16: typedef u32regex_token_iterator<const UChar*> utf16regex_token_iterator;
Chris@16: typedef u32regex_token_iterator<const UChar32*> utf32regex_token_iterator;
Chris@16: 
Chris@16: // construction from an integral sub_match state_id:
Chris@16: inline u32regex_token_iterator<const char*> make_u32regex_token_iterator(const char* p, const u32regex& e, int submatch = 0, regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    return u32regex_token_iterator<const char*>(p, p+std::strlen(p), e, submatch, m);
Chris@16: }
Chris@16: #ifndef BOOST_NO_WREGEX
Chris@16: inline u32regex_token_iterator<const wchar_t*> make_u32regex_token_iterator(const wchar_t* p, const u32regex& e, int submatch = 0, regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    return u32regex_token_iterator<const wchar_t*>(p, p+std::wcslen(p), e, submatch, m);
Chris@16: }
Chris@16: #endif
Chris@16: #if !defined(U_WCHAR_IS_UTF16) && (U_SIZEOF_WCHAR_T != 2)
Chris@16: inline u32regex_token_iterator<const UChar*> make_u32regex_token_iterator(const UChar* p, const u32regex& e, int submatch = 0, regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    return u32regex_token_iterator<const UChar*>(p, p+u_strlen(p), e, submatch, m);
Chris@16: }
Chris@16: #endif
Chris@16: template <class charT, class Traits, class Alloc>
Chris@16: inline u32regex_token_iterator<typename std::basic_string<charT, Traits, Alloc>::const_iterator> make_u32regex_token_iterator(const std::basic_string<charT, Traits, Alloc>& p, const u32regex& e, int submatch = 0, regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    typedef typename std::basic_string<charT, Traits, Alloc>::const_iterator iter_type;
Chris@16:    return u32regex_token_iterator<iter_type>(p.begin(), p.end(), e, submatch, m);
Chris@16: }
Chris@16: inline u32regex_token_iterator<const UChar*> make_u32regex_token_iterator(const U_NAMESPACE_QUALIFIER UnicodeString& s, const u32regex& e, int submatch = 0, regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    return u32regex_token_iterator<const UChar*>(s.getBuffer(), s.getBuffer() + s.length(), e, submatch, m);
Chris@16: }
Chris@16: 
Chris@16: // construction from a reference to an array:
Chris@16: template <std::size_t N>
Chris@16: inline u32regex_token_iterator<const char*> make_u32regex_token_iterator(const char* p, const u32regex& e, const int (&submatch)[N], regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    return u32regex_token_iterator<const char*>(p, p+std::strlen(p), e, submatch, m);
Chris@16: }
Chris@16: #ifndef BOOST_NO_WREGEX
Chris@16: template <std::size_t N>
Chris@16: inline u32regex_token_iterator<const wchar_t*> make_u32regex_token_iterator(const wchar_t* p, const u32regex& e, const int (&submatch)[N], regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    return u32regex_token_iterator<const wchar_t*>(p, p+std::wcslen(p), e, submatch, m);
Chris@16: }
Chris@16: #endif
Chris@16: #if !defined(U_WCHAR_IS_UTF16) && (U_SIZEOF_WCHAR_T != 2)
Chris@16: template <std::size_t N>
Chris@16: inline u32regex_token_iterator<const UChar*> make_u32regex_token_iterator(const UChar* p, const u32regex& e, const int (&submatch)[N], regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    return u32regex_token_iterator<const UChar*>(p, p+u_strlen(p), e, submatch, m);
Chris@16: }
Chris@16: #endif
Chris@16: template <class charT, class Traits, class Alloc, std::size_t N>
Chris@16: inline u32regex_token_iterator<typename std::basic_string<charT, Traits, Alloc>::const_iterator> make_u32regex_token_iterator(const std::basic_string<charT, Traits, Alloc>& p, const u32regex& e, const int (&submatch)[N], regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    typedef typename std::basic_string<charT, Traits, Alloc>::const_iterator iter_type;
Chris@16:    return u32regex_token_iterator<iter_type>(p.begin(), p.end(), e, submatch, m);
Chris@16: }
Chris@16: template <std::size_t N>
Chris@16: inline u32regex_token_iterator<const UChar*> make_u32regex_token_iterator(const U_NAMESPACE_QUALIFIER UnicodeString& s, const u32regex& e, const int (&submatch)[N], regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    return u32regex_token_iterator<const UChar*>(s.getBuffer(), s.getBuffer() + s.length(), e, submatch, m);
Chris@16: }
Chris@16: 
Chris@16: // construction from a vector of sub_match state_id's:
Chris@16: inline u32regex_token_iterator<const char*> make_u32regex_token_iterator(const char* p, const u32regex& e, const std::vector<int>& submatch, regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    return u32regex_token_iterator<const char*>(p, p+std::strlen(p), e, submatch, m);
Chris@16: }
Chris@16: #ifndef BOOST_NO_WREGEX
Chris@16: inline u32regex_token_iterator<const wchar_t*> make_u32regex_token_iterator(const wchar_t* p, const u32regex& e, const std::vector<int>& submatch, regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    return u32regex_token_iterator<const wchar_t*>(p, p+std::wcslen(p), e, submatch, m);
Chris@16: }
Chris@16: #endif
Chris@16: #if !defined(U_WCHAR_IS_UTF16) && (U_SIZEOF_WCHAR_T != 2)
Chris@16: inline u32regex_token_iterator<const UChar*> make_u32regex_token_iterator(const UChar* p, const u32regex& e, const std::vector<int>& submatch, regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    return u32regex_token_iterator<const UChar*>(p, p+u_strlen(p), e, submatch, m);
Chris@16: }
Chris@16: #endif
Chris@16: template <class charT, class Traits, class Alloc>
Chris@16: inline u32regex_token_iterator<typename std::basic_string<charT, Traits, Alloc>::const_iterator> make_u32regex_token_iterator(const std::basic_string<charT, Traits, Alloc>& p, const u32regex& e, const std::vector<int>& submatch, regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    typedef typename std::basic_string<charT, Traits, Alloc>::const_iterator iter_type;
Chris@16:    return u32regex_token_iterator<iter_type>(p.begin(), p.end(), e, submatch, m);
Chris@16: }
Chris@16: inline u32regex_token_iterator<const UChar*> make_u32regex_token_iterator(const U_NAMESPACE_QUALIFIER UnicodeString& s, const u32regex& e, const std::vector<int>& submatch, regex_constants::match_flag_type m = regex_constants::match_default)
Chris@16: {
Chris@16:    return u32regex_token_iterator<const UChar*>(s.getBuffer(), s.getBuffer() + s.length(), e, submatch, m);
Chris@16: }
Chris@16: 
Chris@101: #ifdef BOOST_MSVC
Chris@16: #  pragma warning(pop)
Chris@16: #endif
Chris@16: #ifdef BOOST_HAS_ABI_HEADERS
Chris@16: #  include BOOST_ABI_SUFFIX
Chris@16: #endif
Chris@16: 
Chris@16: } // namespace boost
Chris@16: 
Chris@16: #endif // BOOST_REGEX_V4_REGEX_TOKEN_ITERATOR_HPP
Chris@16: 
Chris@16: 
Chris@16: 
Chris@16: