// class template regex -*- C++ -*-
// Copyright (C) 2013-2018 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 3, 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.
// Under Section 7 of GPL version 3, you are granted additional
// permissions described in the GCC Runtime Library Exception, version
// 3.1, as published by the Free Software Foundation.
// You should have received a copy of the GNU General Public License and
// a copy of the GCC Runtime Library Exception along with this program;
// see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
// <http://www.gnu.org/licenses/>.
/**
* @file bits/regex_automaton.h
* This is an internal header file, included by other library headers.
* Do not attempt to use it directly. @headername{regex}
*/
// This macro defines the maximal state number a NFA can have.
#ifndef _GLIBCXX_REGEX_STATE_LIMIT
#define _GLIBCXX_REGEX_STATE_LIMIT 100000
#endif
namespace std _GLIBCXX_VISIBILITY(default)
{
_GLIBCXX_BEGIN_NAMESPACE_VERSION
namespace __detail
{
/**
* @defgroup regex-detail Base and Implementation Classes
* @ingroup regex
* @{
*/
typedef long _StateIdT;
static const _StateIdT _S_invalid_state_id = -1;
template<typename _CharT>
using _Matcher = std::function<bool (_CharT)>;
/// Operation codes that define the type of transitions within the base NFA
/// that represents the regular expression.
enum _Opcode : int
{
_S_opcode_unknown,
_S_opcode_alternative,
_S_opcode_repeat,
_S_opcode_backref,
_S_opcode_line_begin_assertion,
_S_opcode_line_end_assertion,
_S_opcode_word_boundary,
_S_opcode_subexpr_lookahead,
_S_opcode_subexpr_begin,
_S_opcode_subexpr_end,
_S_opcode_dummy,
_S_opcode_match,
_S_opcode_accept,
};
struct _State_base
{
protected:
_Opcode _M_opcode; // type of outgoing transition
public:
_StateIdT _M_next; // outgoing transition
union // Since they are mutually exclusive.
{
size_t _M_subexpr; // for _S_opcode_subexpr_*
size_t _M_backref_index; // for _S_opcode_backref
struct
{
// for _S_opcode_alternative, _S_opcode_repeat and
// _S_opcode_subexpr_lookahead
_StateIdT _M_alt;
// for _S_opcode_word_boundary or _S_opcode_subexpr_lookahead or
// quantifiers (ungreedy if set true)
bool _M_neg;
};
// For _S_opcode_match
__gnu_cxx::__aligned_membuf<_Matcher<char>> _M_matcher_storage;
};
protected:
explicit _State_base(_Opcode __opcode)
: _M_opcode(__opcode), _M_next(_S_invalid_state_id)
{ }
public:
bool
_M_has_alt()
{
return _M_opcode == _S_opcode_alternative
|| _M_opcode == _S_opcode_repeat
|| _M_opcode == _S_opcode_subexpr_lookahead;
}
#ifdef _GLIBCXX_DEBUG
std::ostream&
_M_print(std::ostream& ostr) const;
// Prints graphviz dot commands for state.
std::ostream&
_M_dot(std::ostream& __ostr, _StateIdT __id) const;
#endif
};
template<typename _Char_type>
struct _State : _State_base
{
typedef _Matcher<_Char_type> _MatcherT;
static_assert(sizeof(_MatcherT) == sizeof(_Matcher<char>),
"std::function<bool(T)> has the same size as "
"std::function<bool(char)>");
static_assert(alignof(_MatcherT) == alignof(_Matcher<char>),
"std::function<bool(T)> has the same alignment as "
"std::function<bool(char)>");
explicit
_State(_Opcode __opcode) : _State_base(__opcode)
{
if (_M_opcode() == _S_opcode_match)
new (this->_M_matcher_storage._M_addr()) _MatcherT();
}
_State(const _State& __rhs) : _State_base(__rhs)
{
if (__rhs._M_opcode() == _S_opcode_match)
new (this->_M_matcher_storage._M_addr())
_MatcherT(__rhs._M_get_matcher());
}
_State(_State&& __rhs) : _State_base(__rhs)
{
if (__rhs._M_opcode() == _S_opcode_match)
new (this->_M_matcher_storage._M_addr())
_MatcherT(std::move(__rhs._M_get_matcher()));
}
_State&
operator=(const _State&) = delete;
~_State()
{
if (_M_opcode() == _S_opcode_match)
_M_get_matcher().~_MatcherT();
}
// Since correct ctor and dtor rely on _M_opcode, it's better not to
// change it over time.
_Opcode
_M_opcode() const
{ return _State_base::_M_opcode; }
bool
_M_matches(_Char_type __char) const
{ return _M_get_matcher()(__char); }
_MatcherT&
_M_get_matcher()
{ return *static_cast<_MatcherT*>(this->_M_matcher_storage._M_addr()); }
const _MatcherT&
_M_get_matcher() const
{
return *static_cast<const _MatcherT*>(
this->_M_matcher_storage._M_addr());
}
};
struct _NFA_base
{
typedef size_t _SizeT;
typedef regex_constants::syntax_option_type _FlagT;
explicit
_NFA_base(_FlagT __f)
: _M_flags(__f), _M_start_state(0), _M_subexpr_count(0),
_M_has_backref(false)
{ }
_NFA_base(_NFA_base&&) = default;
protected:
~_NFA_base() = default;
public:
_FlagT
_M_options() const
{ return _M_flags; }
_StateIdT
_M_start() const
{ return _M_start_state; }
_SizeT
_M_sub_count() const
{ return _M_subexpr_count; }
std::vector<size_t> _M_paren_stack;
_FlagT _M_flags;
_StateIdT _M_start_state;
_SizeT _M_subexpr_count;
bool _M_has_backref;
};
template<typename _TraitsT>
struct _NFA
: _NFA_base, std::vector<_State<typename _TraitsT::char_type>>
{
typedef typename _TraitsT::char_type _Char_type;
typedef _State<_Char_type> _StateT;
typedef _Matcher<_Char_type> _MatcherT;
_NFA(const typename _TraitsT::locale_type& __loc, _FlagT __flags)
: _NFA_base(__flags)
{ _M_traits.imbue(__loc); }
// for performance reasons _NFA objects should only be moved not copied
_NFA(const _NFA&) = delete;
_NFA(_NFA&&) = default;
_StateIdT
_M_insert_accept()
{
auto __ret = _M_insert_state(_StateT(_S_opcode_accept));
return __ret;
}
_StateIdT
_M_insert_alt(_StateIdT __next, _StateIdT __alt,
bool __neg __attribute__((__unused__)))
{
_StateT __tmp(_S_opcode_alternative);
// It labels every quantifier to make greedy comparison easier in BFS
// approach.
__tmp._M_next = __next;
__tmp._M_alt = __alt;
return _M_insert_state(std::move(__tmp));
}
_StateIdT
_M_insert_repeat(_StateIdT __next, _StateIdT __alt, bool __neg)
{
_StateT __tmp(_S_opcode_repeat);
// It labels every quantifier to make greedy comparison easier in BFS
// approach.
__tmp._M_next = __next;
__tmp._M_alt = __alt;
__tmp._M_neg = __neg;
return _M_insert_state(std::move(__tmp));
}
_StateIdT
_M_insert_matcher(_MatcherT __m)
{
_StateT __tmp(_S_opcode_match);
__tmp._M_get_matcher() = std::move(__m);
return _M_insert_state(std::move(__tmp));
}
_StateIdT
_M_insert_subexpr_begin()
{
auto __id = this->_M_subexpr_count++;
this->_M_paren_stack.push_back(__id);
_StateT __tmp(_S_opcode_subexpr_begin);
__tmp._M_subexpr = __id;
return _M_insert_state(std::move(__tmp));
}
_StateIdT
_M_insert_subexpr_end()
{
_StateT __tmp(_S_opcode_subexpr_end);
__tmp._M_subexpr = this->_M_paren_stack.back();
this->_M_paren_stack.pop_back();
return _M_insert_state(std::move(__tmp));
}
_StateIdT
_M_insert_backref(size_t __index);
_StateIdT
_M_insert_line_begin()
{ return _M_insert_state(_StateT(_S_opcode_line_begin_assertion)); }
_StateIdT
_M_insert_line_end()
{ return _M_insert_state(_StateT(_S_opcode_line_end_assertion)); }
_StateIdT
_M_insert_word_bound(bool __neg)
{
_StateT __tmp(_S_opcode_word_boundary);
__tmp._M_neg = __neg;
return _M_insert_state(std::move(__tmp));
}
_StateIdT
_M_insert_lookahead(_StateIdT __alt, bool __neg)
{
_StateT __tmp(_S_opcode_subexpr_lookahead);
__tmp._M_alt = __alt;
__tmp._M_neg = __neg;
return _M_insert_state(std::move(__tmp));
}
_StateIdT
_M_insert_dummy()
{ return _M_insert_state(_StateT(_S_opcode_dummy)); }
_StateIdT
_M_insert_state(_StateT __s)
{
this->push_back(std::move(__s));
if (this->size() > _GLIBCXX_REGEX_STATE_LIMIT)
__throw_regex_error(
regex_constants::error_space,
"Number of NFA states exceeds limit. Please use shorter regex "
"string, or use smaller brace expression, or make "
"_GLIBCXX_REGEX_STATE_LIMIT larger.");
return this->size() - 1;
}
// Eliminate dummy node in this NFA to make it compact.
void
_M_eliminate_dummy();
#ifdef _GLIBCXX_DEBUG
std::ostream&
_M_dot(std::ostream& __ostr) const;
#endif
public:
_TraitsT _M_traits;
};
/// Describes a sequence of one or more %_State, its current start
/// and end(s). This structure contains fragments of an NFA during
/// construction.
template<typename _TraitsT>
class _StateSeq
{
public:
typedef _NFA<_TraitsT> _RegexT;
public:
_StateSeq(_RegexT& __nfa, _StateIdT __s)
: _M_nfa(__nfa), _M_start(__s), _M_end(__s)
{ }
_StateSeq(_RegexT& __nfa, _StateIdT __s, _StateIdT __end)
: _M_nfa(__nfa), _M_start(__s), _M_end(__end)
{ }
// Append a state on *this and change *this to the new sequence.
void
_M_append(_StateIdT __id)
{
_M_nfa[_M_end]._M_next = __id;
_M_end = __id;
}
// Append a sequence on *this and change *this to the new sequence.
void
_M_append(const _StateSeq& __s)
{
_M_nfa[_M_end]._M_next = __s._M_start;
_M_end = __s._M_end;
}
// Clones an entire sequence.
_StateSeq
_M_clone();
public:
_RegexT& _M_nfa;
_StateIdT _M_start;
_StateIdT _M_end;
};
//@} regex-detail
} // namespace __detail
_GLIBCXX_END_NAMESPACE_VERSION
} // namespace std
#include <bits/regex_automaton.tcc>
| Name | Type | Size | Permission | Actions |
|---|---|---|---|---|
| algorithmfwd.h | File | 21.23 KB | 0644 |
|
| alloc_traits.h | File | 19.6 KB | 0644 |
|
| allocated_ptr.h | File | 3.22 KB | 0644 |
|
| allocator.h | File | 7.39 KB | 0644 |
|
| atomic_base.h | File | 23.28 KB | 0644 |
|
| atomic_futex.h | File | 9.35 KB | 0644 |
|
| atomic_lockfree_defines.h | File | 2.2 KB | 0644 |
|
| basic_ios.h | File | 15.7 KB | 0644 |
|
| basic_ios.tcc | File | 5.94 KB | 0644 |
|
| basic_string.h | File | 237.01 KB | 0644 |
|
| basic_string.tcc | File | 52.5 KB | 0644 |
|
| boost_concept_check.h | File | 26.52 KB | 0644 |
|
| c++0x_warning.h | File | 1.44 KB | 0644 |
|
| char_traits.h | File | 20.42 KB | 0644 |
|
| codecvt.h | File | 20.79 KB | 0644 |
|
| concept_check.h | File | 3.34 KB | 0644 |
|
| cpp_type_traits.h | File | 9.56 KB | 0644 |
|
| cxxabi_forced.h | File | 1.77 KB | 0644 |
|
| cxxabi_init_exception.h | File | 2.17 KB | 0644 |
|
| deque.tcc | File | 33.32 KB | 0644 |
|
| enable_special_members.h | File | 12.1 KB | 0644 |
|
| exception.h | File | 2.23 KB | 0644 |
|
| exception_defines.h | File | 1.61 KB | 0644 |
|
| exception_ptr.h | File | 5.84 KB | 0644 |
|
| forward_list.h | File | 47.77 KB | 0644 |
|
| forward_list.tcc | File | 12.86 KB | 0644 |
|
| fs_dir.h | File | 14.38 KB | 0644 |
|
| fs_fwd.h | File | 10.04 KB | 0644 |
|
| fs_ops.h | File | 9.5 KB | 0644 |
|
| fs_path.h | File | 32.12 KB | 0644 |
|
| fstream.tcc | File | 32.03 KB | 0644 |
|
| functexcept.h | File | 3.18 KB | 0644 |
|
| functional_hash.h | File | 8.04 KB | 0644 |
|
| gslice.h | File | 5.39 KB | 0644 |
|
| gslice_array.h | File | 7.59 KB | 0644 |
|
| hash_bytes.h | File | 2.1 KB | 0644 |
|
| hashtable.h | File | 72.06 KB | 0644 |
|
| hashtable_policy.h | File | 66.38 KB | 0644 |
|
| indirect_array.h | File | 7.68 KB | 0644 |
|
| invoke.h | File | 3.57 KB | 0644 |
|
| ios_base.h | File | 30.3 KB | 0644 |
|
| istream.tcc | File | 30.36 KB | 0644 |
|
| list.tcc | File | 15.6 KB | 0644 |
|
| locale_classes.h | File | 24.31 KB | 0644 |
|
| locale_classes.tcc | File | 8.18 KB | 0644 |
|
| locale_conv.h | File | 15.72 KB | 0644 |
|
| locale_facets.h | File | 90.16 KB | 0644 |
|
| locale_facets.tcc | File | 38.62 KB | 0644 |
|
| locale_facets_nonio.h | File | 67.36 KB | 0644 |
|
| locale_facets_nonio.tcc | File | 44.22 KB | 0644 |
|
| localefwd.h | File | 5.51 KB | 0644 |
|
| mask_array.h | File | 7.42 KB | 0644 |
|
| memoryfwd.h | File | 2.4 KB | 0644 |
|
| move.h | File | 6.38 KB | 0644 |
|
| nested_exception.h | File | 4.69 KB | 0644 |
|
| node_handle.h | File | 8.02 KB | 0644 |
|
| ostream.tcc | File | 12.03 KB | 0644 |
|
| ostream_insert.h | File | 3.91 KB | 0644 |
|
| parse_numbers.h | File | 7.76 KB | 0644 |
|
| postypes.h | File | 8.02 KB | 0644 |
|
| predefined_ops.h | File | 8.87 KB | 0644 |
|
| ptr_traits.h | File | 6.47 KB | 0644 |
|
| quoted_string.h | File | 4.93 KB | 0644 |
|
| random.h | File | 171.14 KB | 0644 |
|
| random.tcc | File | 103.12 KB | 0644 |
|
| range_access.h | File | 9.79 KB | 0644 |
|
| refwrap.h | File | 11.61 KB | 0644 |
|
| regex.h | File | 95.12 KB | 0644 |
|
| regex.tcc | File | 16.18 KB | 0644 |
|
| regex_automaton.h | File | 10.47 KB | 0644 |
|
| regex_automaton.tcc | File | 7.65 KB | 0644 |
|
| regex_compiler.h | File | 17.63 KB | 0644 |
|
| regex_compiler.tcc | File | 18.84 KB | 0644 |
|
| regex_constants.h | File | 14.36 KB | 0644 |
|
| regex_error.h | File | 4.79 KB | 0644 |
|
| regex_executor.h | File | 7.31 KB | 0644 |
|
| regex_executor.tcc | File | 18.4 KB | 0644 |
|
| regex_scanner.h | File | 6.92 KB | 0644 |
|
| regex_scanner.tcc | File | 14.66 KB | 0644 |
|
| shared_ptr.h | File | 22.88 KB | 0644 |
|
| shared_ptr_atomic.h | File | 9.54 KB | 0644 |
|
| shared_ptr_base.h | File | 53.01 KB | 0644 |
|
| slice_array.h | File | 9.13 KB | 0644 |
|
| specfun.h | File | 45.95 KB | 0644 |
|
| sstream.tcc | File | 9.9 KB | 0644 |
|
| std_abs.h | File | 3.19 KB | 0644 |
|
| std_function.h | File | 22.71 KB | 0644 |
|
| std_mutex.h | File | 9.08 KB | 0644 |
|
| stl_algo.h | File | 209.27 KB | 0644 |
|
| stl_algobase.h | File | 49.31 KB | 0644 |
|
| stl_bvector.h | File | 32.94 KB | 0644 |
|
| stl_construct.h | File | 7.22 KB | 0644 |
|
| stl_deque.h | File | 76.73 KB | 0644 |
|
| stl_function.h | File | 40.77 KB | 0644 |
|
| stl_heap.h | File | 19.73 KB | 0644 |
|
| stl_iterator.h | File | 41.3 KB | 0644 |
|
| stl_iterator_base_funcs.h | File | 7.99 KB | 0644 |
|
| stl_iterator_base_types.h | File | 8.48 KB | 0644 |
|
| stl_list.h | File | 65.97 KB | 0644 |
|
| stl_map.h | File | 51.55 KB | 0644 |
|
| stl_multimap.h | File | 40.58 KB | 0644 |
|
| stl_multiset.h | File | 34.97 KB | 0644 |
|
| stl_numeric.h | File | 13.51 KB | 0644 |
|
| stl_pair.h | File | 18.21 KB | 0644 |
|
| stl_queue.h | File | 23.51 KB | 0644 |
|
| stl_raw_storage_iter.h | File | 3.74 KB | 0644 |
|
| stl_relops.h | File | 4.49 KB | 0644 |
|
| stl_set.h | File | 35.28 KB | 0644 |
|
| stl_stack.h | File | 11.66 KB | 0644 |
|
| stl_tempbuf.h | File | 8.15 KB | 0644 |
|
| stl_tree.h | File | 73.15 KB | 0644 |
|
| stl_uninitialized.h | File | 27.06 KB | 0644 |
|
| stl_vector.h | File | 59.12 KB | 0644 |
|
| stream_iterator.h | File | 6.5 KB | 0644 |
|
| streambuf.tcc | File | 4.81 KB | 0644 |
|
| streambuf_iterator.h | File | 13.44 KB | 0644 |
|
| string_view.tcc | File | 6.54 KB | 0644 |
|
| stringfwd.h | File | 2.55 KB | 0644 |
|
| uniform_int_dist.h | File | 9.84 KB | 0644 |
|
| unique_ptr.h | File | 25.38 KB | 0644 |
|
| unordered_map.h | File | 73.58 KB | 0644 |
|
| unordered_set.h | File | 57.76 KB | 0644 |
|
| uses_allocator.h | File | 6.37 KB | 0644 |
|
| valarray_after.h | File | 22.12 KB | 0644 |
|
| valarray_array.h | File | 21.3 KB | 0644 |
|
| valarray_array.tcc | File | 7.08 KB | 0644 |
|
| valarray_before.h | File | 18.08 KB | 0644 |
|
| vector.tcc | File | 28.95 KB | 0644 |
|