@@ -0,0 +1,8 | |||
|
1 | using System; | |
|
2 | ||
|
3 | namespace Implab.Automaton.RegularExpressions { | |
|
4 | public interface ITaggedDFABuilder<TTag> : IDFATableBuilder { | |
|
5 | void SetStateTag(int s, TTag[] tags); | |
|
6 | } | |
|
7 | } | |
|
8 |
@@ -0,0 +1,89 | |||
|
1 | using System; | |
|
2 | using System.Collections.Generic; | |
|
3 | using System.Linq; | |
|
4 | ||
|
5 | namespace Implab.Automaton.RegularExpressions { | |
|
6 | public class RegularDFA<TInput, TTag> : DFATable, ITaggedDFABuilder<TTag> { | |
|
7 | ||
|
8 | readonly Dictionary<int,TTag[]> m_tags = new Dictionary<int, TTag[]>(); | |
|
9 | readonly IAlphabet<TInput> m_alphabet; | |
|
10 | ||
|
11 | public RegularDFA(IAlphabet<TInput> alphabet) { | |
|
12 | Safe.ArgumentNotNull(alphabet, "aplhabet"); | |
|
13 | ||
|
14 | m_alphabet = alphabet; | |
|
15 | } | |
|
16 | ||
|
17 | ||
|
18 | public IAlphabet<TInput> InputAlphabet { | |
|
19 | get { | |
|
20 | return m_alphabet; | |
|
21 | } | |
|
22 | } | |
|
23 | ||
|
24 | public void MarkFinalState(int s, TTag[] tags) { | |
|
25 | MarkFinalState(s); | |
|
26 | SetStateTag(s, tags); | |
|
27 | } | |
|
28 | ||
|
29 | public void SetStateTag(int s, TTag[] tags) { | |
|
30 | Safe.ArgumentNotNull(tags, "tags"); | |
|
31 | m_tags[s] = tags; | |
|
32 | } | |
|
33 | ||
|
34 | public TTag[] GetStateTag(int s) { | |
|
35 | TTag[] tags; | |
|
36 | return m_tags.TryGetValue(s, out tags) ? tags : new TTag[0]; | |
|
37 | } | |
|
38 | ||
|
39 | public new DFAStateDescriptor<TTag>[] CreateTransitionTable() { | |
|
40 | var table = new DFAStateDescriptor<TTag>[StateCount]; | |
|
41 | ||
|
42 | foreach (var t in this) { | |
|
43 | if (table[t.s1].transitions == null) | |
|
44 | table[t.s1] = new DFAStateDescriptor<TTag>(AlphabetSize, IsFinalState(t.s1), GetStateTag(t.s1)); | |
|
45 | if (table[t.s2].transitions == null) | |
|
46 | table[t.s2] = new DFAStateDescriptor<TTag>(AlphabetSize, IsFinalState(t.s2), GetStateTag(t.s2)); | |
|
47 | table[t.s1].transitions[t.edge] = t.s2; | |
|
48 | } | |
|
49 | ||
|
50 | return table; | |
|
51 | } | |
|
52 | ||
|
53 | /// <summary> | |
|
54 | /// Optimize the specified alphabet. | |
|
55 | /// </summary> | |
|
56 | /// <param name="alphabet">Пустой алфавит, который будет зполнен в процессе оптимизации.</param> | |
|
57 | public RegularDFA<TInput,TTag> Optimize(IAlphabetBuilder<TInput> alphabet) { | |
|
58 | Safe.ArgumentNotNull(alphabet, "alphabet"); | |
|
59 | ||
|
60 | var dfa = new RegularDFA<TInput, TTag>(alphabet); | |
|
61 | ||
|
62 | var states = new DummyAlphabet(StateCount); | |
|
63 | var alphaMap = new Dictionary<int,int>(); | |
|
64 | var stateMap = new Dictionary<int,int>(); | |
|
65 | ||
|
66 | Optimize(dfa, alphaMap, stateMap); | |
|
67 | ||
|
68 | // mark tags in the new DFA | |
|
69 | foreach (var g in m_tags.Where(x => x.Key < StateCount).GroupBy(x => stateMap[x.Key], x => x.Value )) | |
|
70 | dfa.SetStateTag(g.Key, g.SelectMany(x => x).ToArray()); | |
|
71 | ||
|
72 | // make the alphabet for the new DFA | |
|
73 | foreach (var pair in alphaMap) | |
|
74 | alphabet.DefineClass(m_alphabet.GetSymbols(pair.Key), pair.Value); | |
|
75 | ||
|
76 | return dfa; | |
|
77 | } | |
|
78 | ||
|
79 | protected override IEnumerable<HashSet<int>> GroupFinalStates() { | |
|
80 | var arrayComparer = new CustomEqualityComparer<TTag[]>( | |
|
81 | (x,y) => x.Length == y.Length && x.All(it => y.Contains(it)), | |
|
82 | x => x.Sum(it => x.GetHashCode()) | |
|
83 | ); | |
|
84 | return FinalStates.GroupBy(x => m_tags[x], arrayComparer).Select(g => new HashSet<int>(g)); | |
|
85 | } | |
|
86 | ||
|
87 | } | |
|
88 | } | |
|
89 |
@@ -0,0 +1,184 | |||
|
1 | using Implab; | |
|
2 | using System; | |
|
3 | using System.Collections.Generic; | |
|
4 | using System.Diagnostics; | |
|
5 | using System.Linq; | |
|
6 | ||
|
7 | namespace Implab.Automaton.RegularExpressions { | |
|
8 | /// <summary> | |
|
9 | /// Используется для построения ДКА по регулярному выражению, сначала обходит | |
|
10 | /// регулярное выражение и вычисляет followpos, затем используется метод | |
|
11 | /// <see cref="BuildDFA(IDFADefinition)"/> для построения автомата. | |
|
12 | /// </summary> | |
|
13 | public class RegularExpressionVisitor<TTag> : IVisitor<TTag> { | |
|
14 | int m_idx; | |
|
15 | Token<TTag> m_root; | |
|
16 | HashSet<int> m_firstpos; | |
|
17 | HashSet<int> m_lastpos; | |
|
18 | ||
|
19 | readonly Dictionary<int, HashSet<int>> m_followpos = new Dictionary<int, HashSet<int>>(); | |
|
20 | readonly Dictionary<int, int> m_indexes = new Dictionary<int, int>(); | |
|
21 | readonly Dictionary<int, TTag> m_ends = new Dictionary<int, TTag>(); | |
|
22 | ||
|
23 | public Dictionary<int, HashSet<int>> FollowposMap { | |
|
24 | get { return m_followpos; } | |
|
25 | } | |
|
26 | ||
|
27 | public HashSet<int> Followpos(int pos) { | |
|
28 | HashSet<int> set; | |
|
29 | return m_followpos.TryGetValue(pos, out set) ? set : m_followpos[pos] = new HashSet<int>(); | |
|
30 | } | |
|
31 | ||
|
32 | bool Nullable(object n) { | |
|
33 | if (n is EmptyToken<TTag> || n is StarToken<TTag>) | |
|
34 | return true; | |
|
35 | var altToken = n as AltToken<TTag>; | |
|
36 | if (altToken != null) | |
|
37 | return Nullable(altToken.Left) || Nullable(altToken.Right); | |
|
38 | var catToken = n as CatToken<TTag>; | |
|
39 | if (catToken != null) | |
|
40 | return Nullable(catToken.Left) && Nullable(catToken.Right); | |
|
41 | return false; | |
|
42 | } | |
|
43 | ||
|
44 | ||
|
45 | public void Visit(AltToken<TTag> token) { | |
|
46 | if (m_root == null) | |
|
47 | m_root = token; | |
|
48 | var firtspos = new HashSet<int>(); | |
|
49 | var lastpos = new HashSet<int>(); | |
|
50 | ||
|
51 | token.Left.Accept(this); | |
|
52 | firtspos.UnionWith(m_firstpos); | |
|
53 | lastpos.UnionWith(m_lastpos); | |
|
54 | ||
|
55 | token.Right.Accept(this); | |
|
56 | firtspos.UnionWith(m_firstpos); | |
|
57 | lastpos.UnionWith(m_lastpos); | |
|
58 | ||
|
59 | m_firstpos = firtspos; | |
|
60 | m_lastpos = lastpos; | |
|
61 | } | |
|
62 | ||
|
63 | public void Visit(StarToken<TTag> token) { | |
|
64 | if (m_root == null) | |
|
65 | m_root = token; | |
|
66 | token.Token.Accept(this); | |
|
67 | ||
|
68 | foreach (var i in m_lastpos) | |
|
69 | Followpos(i).UnionWith(m_firstpos); | |
|
70 | } | |
|
71 | ||
|
72 | public void Visit(CatToken<TTag> token) { | |
|
73 | if (m_root == null) | |
|
74 | m_root = token; | |
|
75 | ||
|
76 | var firtspos = new HashSet<int>(); | |
|
77 | var lastpos = new HashSet<int>(); | |
|
78 | token.Left.Accept(this); | |
|
79 | firtspos.UnionWith(m_firstpos); | |
|
80 | var leftLastpos = m_lastpos; | |
|
81 | ||
|
82 | token.Right.Accept(this); | |
|
83 | lastpos.UnionWith(m_lastpos); | |
|
84 | var rightFirstpos = m_firstpos; | |
|
85 | ||
|
86 | if (Nullable(token.Left)) | |
|
87 | firtspos.UnionWith(rightFirstpos); | |
|
88 | ||
|
89 | if (Nullable(token.Right)) | |
|
90 | lastpos.UnionWith(leftLastpos); | |
|
91 | ||
|
92 | m_firstpos = firtspos; | |
|
93 | m_lastpos = lastpos; | |
|
94 | ||
|
95 | foreach (var i in leftLastpos) | |
|
96 | Followpos(i).UnionWith(rightFirstpos); | |
|
97 | ||
|
98 | } | |
|
99 | ||
|
100 | public void Visit(EmptyToken<TTag> token) { | |
|
101 | if (m_root == null) | |
|
102 | m_root = token; | |
|
103 | } | |
|
104 | ||
|
105 | public void Visit(SymbolToken<TTag> token) { | |
|
106 | if (m_root == null) | |
|
107 | m_root = token; | |
|
108 | m_idx++; | |
|
109 | m_indexes[m_idx] = token.Value; | |
|
110 | m_firstpos = new HashSet<int>(new[] { m_idx }); | |
|
111 | m_lastpos = new HashSet<int>(new[] { m_idx }); | |
|
112 | } | |
|
113 | ||
|
114 | public void Visit(EndToken<TTag> token) { | |
|
115 | if (m_root == null) | |
|
116 | m_root = token; | |
|
117 | m_idx++; | |
|
118 | m_indexes[m_idx] = DFAConst.UNCLASSIFIED_INPUT; | |
|
119 | m_firstpos = new HashSet<int>(new[] { m_idx }); | |
|
120 | m_lastpos = new HashSet<int>(new[] { m_idx }); | |
|
121 | Followpos(m_idx); | |
|
122 | m_ends.Add(m_idx, token.Tag); | |
|
123 | } | |
|
124 | ||
|
125 | public void BuildDFA(ITaggedDFABuilder<TTag> dfa) { | |
|
126 | Safe.ArgumentNotNull(dfa,"dfa"); | |
|
127 | ||
|
128 | var states = new MapAlphabet<HashSet<int>>( | |
|
129 | false, | |
|
130 | new CustomEqualityComparer<HashSet<int>>( | |
|
131 | (x, y) => x.SetEquals(y), | |
|
132 | x => x.Sum(n => n.GetHashCode()) | |
|
133 | )); | |
|
134 | ||
|
135 | var initialState = states.DefineSymbol(m_firstpos); | |
|
136 | dfa.SetInitialState(initialState); | |
|
137 | ||
|
138 | var tags = GetStateTags(m_firstpos); | |
|
139 | if (tags != null && tags.Length > 0) | |
|
140 | dfa.MarkFinalState(initialState, tags); | |
|
141 | ||
|
142 | var inputMax = m_indexes.Values.Max(); | |
|
143 | var queue = new Queue<HashSet<int>>(); | |
|
144 | ||
|
145 | queue.Enqueue(m_firstpos); | |
|
146 | ||
|
147 | while (queue.Count > 0) { | |
|
148 | var state = queue.Dequeue(); | |
|
149 | var s1 = states.Translate(state); | |
|
150 | Debug.Assert(s1 != DFAConst.UNCLASSIFIED_INPUT); | |
|
151 | ||
|
152 | for (int a = 0; a <= inputMax; a++) { | |
|
153 | var next = new HashSet<int>(); | |
|
154 | foreach (var p in state) { | |
|
155 | if (m_indexes[p] == a) { | |
|
156 | next.UnionWith(Followpos(p)); | |
|
157 | } | |
|
158 | } | |
|
159 | if (next.Count > 0) { | |
|
160 | int s2 = states.Translate(next); | |
|
161 | if (s2 == DFAConst.UNCLASSIFIED_INPUT) { | |
|
162 | s2 = states.DefineSymbol(next); | |
|
163 | ||
|
164 | tags = GetStateTags(next); | |
|
165 | if (tags != null && tags.Length > 0) { | |
|
166 | dfa.MarkFinalState(s2); | |
|
167 | dfa.SetStateTag(s2, tags); | |
|
168 | } | |
|
169 | ||
|
170 | queue.Enqueue(next); | |
|
171 | } | |
|
172 | dfa.Add(new AutomatonTransition(s1, s2, a)); | |
|
173 | } | |
|
174 | } | |
|
175 | } | |
|
176 | } | |
|
177 | ||
|
178 | TTag[] GetStateTags(IEnumerable<int> state) { | |
|
179 | Debug.Assert(state != null); | |
|
180 | return state.Where(m_ends.ContainsKey).Select(pos => m_ends[pos]).ToArray(); | |
|
181 | } | |
|
182 | ||
|
183 | } | |
|
184 | } |
@@ -1,26 +1,26 | |||
|
1 | 1 | namespace Implab.Automaton { |
|
2 |
public struct DFAStateDescript |
|
|
2 | public struct DFAStateDescriptor { | |
|
3 | 3 | public readonly bool final; |
|
4 | 4 | public readonly int[] transitions; |
|
5 | 5 | |
|
6 | 6 | |
|
7 |
public DFAStateDescript |
|
|
7 | public DFAStateDescriptor(int[] transitions, bool final) { | |
|
8 | 8 | this.transitions = transitions; |
|
9 | 9 | this.final = final; |
|
10 | 10 | } |
|
11 | 11 | |
|
12 |
public DFAStateDescript |
|
|
12 | public DFAStateDescriptor(int[] transitions) : this(transitions, false) { | |
|
13 | 13 | } |
|
14 | 14 | |
|
15 |
public DFAStateDescript |
|
|
15 | public DFAStateDescriptor(int size, bool final) { | |
|
16 | 16 | Safe.ArgumentInRange(size, 0, int.MaxValue, "size"); |
|
17 | 17 | |
|
18 | 18 | this.final = final; |
|
19 | 19 | |
|
20 | 20 | transitions = new int[size]; |
|
21 | 21 | |
|
22 | 22 | for (int i = 0; i < size; i++) |
|
23 | 23 | transitions[i] = DFAConst.UNREACHABLE_STATE; |
|
24 | 24 | } |
|
25 | 25 | } |
|
26 | 26 | } |
@@ -1,291 +1,305 | |||
|
1 | 1 | using Implab; |
|
2 | 2 | using System; |
|
3 | 3 | using System.Collections.Generic; |
|
4 | 4 | using System.Linq; |
|
5 | 5 | |
|
6 | 6 | namespace Implab.Automaton { |
|
7 | 7 | public class DFATable : IDFATableBuilder { |
|
8 | 8 | int m_stateCount; |
|
9 | 9 | int m_symbolCount; |
|
10 | 10 | int m_initialState; |
|
11 | 11 | |
|
12 | 12 | readonly HashSet<int> m_finalStates = new HashSet<int>(); |
|
13 | 13 | readonly HashSet<AutomatonTransition> m_transitions = new HashSet<AutomatonTransition>(); |
|
14 | 14 | |
|
15 | 15 | |
|
16 | 16 | #region IDFADefinition implementation |
|
17 | 17 | |
|
18 | 18 | public bool IsFinalState(int s) { |
|
19 | 19 | Safe.ArgumentInRange(s, 0, m_stateCount, "s"); |
|
20 | 20 | |
|
21 | 21 | return m_finalStates.Contains(s); |
|
22 | 22 | } |
|
23 | 23 | |
|
24 | 24 | public IEnumerable<int> FinalStates { |
|
25 | 25 | get { |
|
26 | 26 | return m_finalStates; |
|
27 | 27 | } |
|
28 | 28 | } |
|
29 | 29 | |
|
30 | 30 | public int StateCount { |
|
31 | 31 | get { return m_stateCount; } |
|
32 | 32 | } |
|
33 | 33 | |
|
34 | 34 | public int AlphabetSize { |
|
35 | 35 | get { return m_symbolCount; } |
|
36 | 36 | } |
|
37 | 37 | |
|
38 | 38 | public int InitialState { |
|
39 | 39 | get { return m_initialState; } |
|
40 | 40 | } |
|
41 | 41 | |
|
42 | 42 | #endregion |
|
43 | 43 | |
|
44 | 44 | public void SetInitialState(int s) { |
|
45 | 45 | Safe.ArgumentAssert(s >= 0, "s"); |
|
46 | 46 | m_initialState = s; |
|
47 | 47 | } |
|
48 | 48 | |
|
49 | 49 | public void MarkFinalState(int state) { |
|
50 | 50 | m_finalStates.Add(state); |
|
51 | 51 | } |
|
52 | 52 | |
|
53 | 53 | public void Add(AutomatonTransition item) { |
|
54 | 54 | Safe.ArgumentAssert(item.s1 >= 0, "item"); |
|
55 | 55 | Safe.ArgumentAssert(item.s2 >= 0, "item"); |
|
56 | 56 | Safe.ArgumentAssert(item.edge >= 0, "item"); |
|
57 | 57 | |
|
58 | 58 | m_stateCount = Math.Max(m_stateCount, Math.Max(item.s1, item.s2) + 1); |
|
59 | 59 | m_symbolCount = Math.Max(m_symbolCount, item.edge); |
|
60 | 60 | |
|
61 | 61 | m_transitions.Add(item); |
|
62 | 62 | } |
|
63 | 63 | |
|
64 | 64 | public void Clear() { |
|
65 | 65 | m_stateCount = 0; |
|
66 | 66 | m_symbolCount = 0; |
|
67 | 67 | m_finalStates.Clear(); |
|
68 | 68 | m_transitions.Clear(); |
|
69 | 69 | } |
|
70 | 70 | |
|
71 | 71 | public bool Contains(AutomatonTransition item) { |
|
72 | 72 | return m_transitions.Contains(item); |
|
73 | 73 | } |
|
74 | 74 | |
|
75 | 75 | public void CopyTo(AutomatonTransition[] array, int arrayIndex) { |
|
76 | 76 | m_transitions.CopyTo(array, arrayIndex); |
|
77 | 77 | } |
|
78 | 78 | |
|
79 | 79 | public bool Remove(AutomatonTransition item) { |
|
80 | 80 | m_transitions.Remove(item); |
|
81 | 81 | } |
|
82 | 82 | |
|
83 | 83 | public int Count { |
|
84 | 84 | get { |
|
85 | 85 | return m_transitions.Count; |
|
86 | 86 | } |
|
87 | 87 | } |
|
88 | 88 | |
|
89 | 89 | public bool IsReadOnly { |
|
90 | 90 | get { |
|
91 | 91 | return false; |
|
92 | 92 | } |
|
93 | 93 | } |
|
94 | 94 | |
|
95 | 95 | public IEnumerator<AutomatonTransition> GetEnumerator() { |
|
96 | 96 | return m_transitions.GetEnumerator(); |
|
97 | 97 | } |
|
98 | 98 | |
|
99 | 99 | System.Collections.IEnumerator System.Collections.IEnumerable.GetEnumerator() { |
|
100 | 100 | return GetEnumerator(); |
|
101 | 101 | } |
|
102 | 102 | |
|
103 | public DFAStateDescriptor[] CreateTransitionTable() { | |
|
104 | var table = new DFAStateDescriptor[StateCount]; | |
|
105 | ||
|
106 | foreach (var t in this) { | |
|
107 | if (table[t.s1].transitions == null) | |
|
108 | table[t.s1] = new DFAStateDescriptor(AlphabetSize, IsFinalState(t.s1)); | |
|
109 | if (table[t.s2].transitions == null) | |
|
110 | table[t.s2] = new DFAStateDescriptor(AlphabetSize, IsFinalState(t.s2)); | |
|
111 | table[t.s1].transitions[t.edge] = t.s2; | |
|
112 | } | |
|
113 | ||
|
114 | return table; | |
|
115 | } | |
|
116 | ||
|
103 | 117 | /// <summary>Формирует множества конечных состояний перед началом работы алгоритма минимизации.</summary> |
|
104 | 118 | /// <remarks> |
|
105 | 119 | /// В процессе построения минимального автомата требуется разделить множество состояний, |
|
106 | 120 | /// на два подмножества - конечные состояния и все остальные, после чего эти подмножества |
|
107 | 121 | /// будут резделены на более мелкие. Иногда требуется гарантировать различия конечных сосотяний, |
|
108 | 122 | /// для этого необходимо переопределить даннцю фукнцию, для получения множеств конечных состояний. |
|
109 | 123 | /// </remarks> |
|
110 | 124 | /// <returns>The final states.</returns> |
|
111 | 125 | protected virtual IEnumerable<HashSet<int>> GroupFinalStates() { |
|
112 | 126 | return new HashSet<int>[] { m_finalStates }; |
|
113 | 127 | } |
|
114 | 128 | |
|
115 | 129 | protected void Optimize( |
|
116 | 130 | IDFATableBuilder optimalDFA, |
|
117 | 131 | IDictionary<int,int> alphabetMap, |
|
118 | 132 | IDictionary<int,int> stateMap |
|
119 | 133 | ) { |
|
120 | 134 | Safe.ArgumentNotNull(optimalDFA, "dfa"); |
|
121 | 135 | Safe.ArgumentNotNull(alphabetMap, "alphabetMap"); |
|
122 | 136 | Safe.ArgumentNotNull(stateMap, "stateMap"); |
|
123 | 137 | |
|
124 | 138 | |
|
125 | 139 | var setComparer = new CustomEqualityComparer<HashSet<int>>( |
|
126 | 140 | (x, y) => x.SetEquals(y), |
|
127 | 141 | s => s.Sum(x => x.GetHashCode()) |
|
128 | 142 | ); |
|
129 | 143 | |
|
130 | 144 | var optimalStates = new HashSet<HashSet<int>>(setComparer); |
|
131 | 145 | var queue = new HashSet<HashSet<int>>(setComparer); |
|
132 | 146 | |
|
133 | 147 | // получаем конечные состояния, сгруппированные по маркерам |
|
134 | 148 | optimalStates.UnionWith( |
|
135 | 149 | GroupFinalStates() |
|
136 | 150 | ); |
|
137 | 151 | |
|
138 | 152 | var state = new HashSet<int>( |
|
139 | 153 | Enumerable |
|
140 | 154 | .Range(0, m_stateCount - 1) |
|
141 | 155 | .Where(i => !m_finalStates.Contains(i)) |
|
142 | 156 | ); |
|
143 | 157 | |
|
144 | 158 | optimalStates.Add(state); |
|
145 | 159 | queue.Add(state); |
|
146 | 160 | |
|
147 | 161 | var rmap = m_transitions |
|
148 | 162 | .GroupBy(t => t.s2) |
|
149 | 163 | .ToLookup( |
|
150 | 164 | g => g.Key, // s2 |
|
151 | 165 | g => g.ToLookup(t => t.edge, t => t.s1) |
|
152 | 166 | ); |
|
153 | 167 | |
|
154 | 168 | while (queue.Count > 0) { |
|
155 | 169 | var stateA = queue.First(); |
|
156 | 170 | queue.Remove(stateA); |
|
157 | 171 | |
|
158 | 172 | for (int c = 0; c < m_symbolCount; c++) { |
|
159 | 173 | var stateX = new HashSet<int>(); |
|
160 | 174 | foreach(var a in stateA) |
|
161 | 175 | stateX.UnionWith(rmap[a][c]); // all states from wich 'c' leads to 'a' |
|
162 | 176 | |
|
163 | 177 | foreach (var stateY in optimalStates.ToArray()) { |
|
164 | 178 | if (stateX.Overlaps(stateY) && !stateY.IsSubsetOf(stateX)) { |
|
165 | 179 | var stateR1 = new HashSet<int>(stateY); |
|
166 | 180 | var stateR2 = new HashSet<int>(stateY); |
|
167 | 181 | |
|
168 | 182 | stateR1.IntersectWith(stateX); |
|
169 | 183 | stateR2.ExceptWith(stateX); |
|
170 | 184 | |
|
171 | 185 | optimalStates.Remove(stateY); |
|
172 | 186 | optimalStates.Add(stateR1); |
|
173 | 187 | optimalStates.Add(stateR2); |
|
174 | 188 | |
|
175 | 189 | if (queue.Contains(stateY)) { |
|
176 | 190 | queue.Remove(stateY); |
|
177 | 191 | queue.Add(stateR1); |
|
178 | 192 | queue.Add(stateR2); |
|
179 | 193 | } else { |
|
180 | 194 | queue.Add(stateR1.Count <= stateR2.Count ? stateR1 : stateR2); |
|
181 | 195 | } |
|
182 | 196 | } |
|
183 | 197 | } |
|
184 | 198 | } |
|
185 | 199 | } |
|
186 | 200 | |
|
187 | 201 | // карта получения оптимального состояния по соотвествующему ему простому состоянию |
|
188 | 202 | var nextState = 0; |
|
189 | 203 | foreach (var item in optimalStates) { |
|
190 | 204 | var id = nextState++; |
|
191 | 205 | foreach (var s in item) |
|
192 | 206 | stateMap[s] = id; |
|
193 | 207 | } |
|
194 | 208 | |
|
195 | 209 | // получаем минимальный алфавит |
|
196 | 210 | // входные символы не различимы, если Move(s,a1) == Move(s,a2), для любого s |
|
197 | 211 | // для этого используем алгоритм кластеризации, сначала |
|
198 | 212 | // считаем, что все символы не различимы |
|
199 | 213 | |
|
200 | 214 | var minClasses = new HashSet<HashSet<int>>(setComparer); |
|
201 | 215 | var alphaQueue = new Queue<HashSet<int>>(); |
|
202 | 216 | alphaQueue.Enqueue(new HashSet<int>(Enumerable.Range(0,AlphabetSize))); |
|
203 | 217 | |
|
204 | 218 | // для всех состояний, будем проверять каждый класс на различимость, |
|
205 | 219 | // т.е. символы различимы, если они приводят к разным состояниям |
|
206 | 220 | for (int s = 0 ; s < optimalStates.Count; s++) { |
|
207 | 221 | var newQueue = new Queue<HashSet<int>>(); |
|
208 | 222 | |
|
209 | 223 | foreach (var A in alphaQueue) { |
|
210 | 224 | // классы из одного символа делить бесполезно, переводим их сразу в |
|
211 | 225 | // результирующий алфавит |
|
212 | 226 | if (A.Count == 1) { |
|
213 | 227 | minClasses.Add(A); |
|
214 | 228 | continue; |
|
215 | 229 | } |
|
216 | 230 | |
|
217 | 231 | // различаем классы символов, которые переводят в различные оптимальные состояния |
|
218 | 232 | // optimalState -> alphaClass |
|
219 | 233 | var classes = new Dictionary<int, HashSet<int>>(); |
|
220 | 234 | |
|
221 | 235 | foreach (var term in A) { |
|
222 | 236 | // ищем все переходы класса по символу term |
|
223 | 237 | var res = m_transitions.Where(t => stateMap[t.s1] == s && t.edge == term).Select(t => stateMap[t.s2]).ToArray(); |
|
224 | 238 | |
|
225 | 239 | var s2 = res.Length > 0 ? res[0] : -1; |
|
226 | 240 | |
|
227 | 241 | HashSet<int> a2; |
|
228 | 242 | if (!classes.TryGetValue(s2, out a2)) { |
|
229 | 243 | a2 = new HashSet<int>(); |
|
230 | 244 | newQueue.Enqueue(a2); |
|
231 | 245 | classes[s2] = a2; |
|
232 | 246 | } |
|
233 | 247 | a2.Add(term); |
|
234 | 248 | } |
|
235 | 249 | } |
|
236 | 250 | |
|
237 | 251 | if (newQueue.Count == 0) |
|
238 | 252 | break; |
|
239 | 253 | alphaQueue = newQueue; |
|
240 | 254 | } |
|
241 | 255 | |
|
242 | 256 | // после окончания работы алгоритма в очереди останутся минимальные различимые классы |
|
243 | 257 | // входных символов |
|
244 | 258 | foreach (var A in alphaQueue) |
|
245 | 259 | minClasses.Add(A); |
|
246 | 260 | |
|
247 | 261 | // построение отображения алфавитов входных символов. |
|
248 | 262 | // поскольку символ DFAConst.UNCLASSIFIED_INPUT может иметь |
|
249 | 263 | // специальное значение, тогда сохраним минимальный класс, |
|
250 | 264 | // содержащий этот символ на томже месте. |
|
251 | 265 | |
|
252 | 266 | var nextCls = 0; |
|
253 | 267 | foreach (var item in minClasses) { |
|
254 | 268 | if (nextCls == DFAConst.UNCLASSIFIED_INPUT) |
|
255 | 269 | nextCls++; |
|
256 | 270 | |
|
257 | 271 | // сохраняем DFAConst.UNCLASSIFIED_INPUT |
|
258 | 272 | var cls = item.Contains(DFAConst.UNCLASSIFIED_INPUT) ? DFAConst.UNCLASSIFIED_INPUT : nextCls; |
|
259 | 273 | |
|
260 | 274 | foreach (var a in item) |
|
261 | 275 | alphabetMap[a] = cls; |
|
262 | 276 | |
|
263 | 277 | nextCls++; |
|
264 | 278 | } |
|
265 | 279 | |
|
266 | 280 | // построение автомата |
|
267 | 281 | optimalDFA.SetInitialState(stateMap[m_initialState]); |
|
268 | 282 | |
|
269 | 283 | foreach (var sf in m_finalStates.Select(s => stateMap[s]).Distinct()) |
|
270 | 284 | optimalDFA.MarkFinalState(sf); |
|
271 | 285 | |
|
272 | 286 | foreach (var t in m_transitions.Select(t => new AutomatonTransition(stateMap[t.s1],stateMap[t.s2],alphabetMap[t.edge])).Distinct()) |
|
273 | 287 | optimalDFA.Add(t); |
|
274 | 288 | } |
|
275 | 289 | |
|
276 | 290 | protected void PrintDFA<TInput, TState>(IAlphabet<TInput> inputAlphabet, IAlphabet<TState> stateAlphabet) { |
|
277 | 291 | Safe.ArgumentNotNull(inputAlphabet, "inputAlphabet"); |
|
278 | 292 | Safe.ArgumentNotNull(stateAlphabet, "stateAlphabet"); |
|
279 | 293 | |
|
280 | 294 | foreach(var t in m_transitions) |
|
281 | 295 | Console.WriteLine( |
|
282 | 296 | "[{0}] -{{{1}}}-> [{2}]{3}", |
|
283 | 297 | String.Join(",", stateAlphabet.GetSymbols(t.s1)), |
|
284 | 298 | String.Join("", inputAlphabet.GetSymbols(t.edge)), |
|
285 | 299 | String.Join(",", stateAlphabet.GetSymbols(t.s2)), |
|
286 | 300 | m_finalStates.Contains(t.s2) ? "$" : "" |
|
287 | 301 | ); |
|
288 | 302 | } |
|
289 | 303 | |
|
290 | 304 | } |
|
291 | 305 | } |
@@ -1,86 +1,94 | |||
|
1 | 1 | using Implab; |
|
2 | 2 | using System; |
|
3 | 3 | using System.Collections.Generic; |
|
4 | 4 | using System.Diagnostics; |
|
5 | 5 | using System.Linq; |
|
6 | 6 | |
|
7 | 7 | namespace Implab.Automaton { |
|
8 | 8 | /// <summary> |
|
9 | 9 | /// Indexed alphabet is the finite set of symbols where each symbol has a zero-based unique index. |
|
10 | 10 | /// </summary> |
|
11 | 11 | /// <remarks> |
|
12 | 12 | /// Indexed alphabets are usefull in bulting efficient translations from source alphabet |
|
13 | 13 | /// to the input alphabet of the automaton. It's assumed that the index to the symbol match |
|
14 | 14 | /// is well known and documented. |
|
15 | 15 | /// </remarks> |
|
16 | 16 | public abstract class IndexedAlphabetBase<T> : IAlphabetBuilder<T> { |
|
17 | 17 | int m_nextId = 1; |
|
18 | 18 | readonly int[] m_map; |
|
19 | 19 | |
|
20 | 20 | protected IndexedAlphabetBase(int mapSize) { |
|
21 | 21 | m_map = new int[mapSize]; |
|
22 | 22 | } |
|
23 | 23 | |
|
24 | 24 | protected IndexedAlphabetBase(int[] map) { |
|
25 | 25 | Debug.Assert(map != null && map.Length > 0); |
|
26 | 26 | Debug.Assert(map.All(x => x >= 0)); |
|
27 | 27 | |
|
28 | 28 | m_map = map; |
|
29 | 29 | m_nextId = map.Max() + 1; |
|
30 | 30 | } |
|
31 | 31 | |
|
32 | 32 | public int DefineSymbol(T symbol) { |
|
33 | 33 | var index = GetSymbolIndex(symbol); |
|
34 | 34 | if (m_map[index] == DFAConst.UNCLASSIFIED_INPUT) |
|
35 | 35 | m_map[index] = m_nextId++; |
|
36 | 36 | return m_map[index]; |
|
37 | 37 | } |
|
38 | 38 | |
|
39 | 39 | public int DefineSymbol(T symbol, int cls) { |
|
40 | 40 | var index = GetSymbolIndex(symbol); |
|
41 | 41 | m_map[index] = cls; |
|
42 | 42 | m_nextId = Math.Max(cls + 1, m_nextId); |
|
43 | 43 | return cls; |
|
44 | 44 | } |
|
45 | 45 | |
|
46 | 46 | public int DefineClass(IEnumerable<T> symbols) { |
|
47 | 47 | return DefineClass(symbols, m_nextId); |
|
48 | 48 | } |
|
49 | 49 | |
|
50 | 50 | public int DefineClass(IEnumerable<T> symbols, int cls) { |
|
51 | 51 | Safe.ArgumentNotNull(symbols, "symbols"); |
|
52 | 52 | symbols = symbols.Distinct(); |
|
53 | 53 | |
|
54 | 54 | foreach (var symbol in symbols) |
|
55 | 55 | m_map[GetSymbolIndex(symbol)] = cls; |
|
56 | 56 | |
|
57 | 57 | m_nextId = Math.Max(cls + 1, m_nextId); |
|
58 | 58 | |
|
59 | 59 | return cls; |
|
60 | 60 | } |
|
61 | 61 | |
|
62 | 62 | public virtual int Translate(T symbol) { |
|
63 | 63 | return m_map[GetSymbolIndex(symbol)]; |
|
64 | 64 | } |
|
65 | 65 | |
|
66 | 66 | public int Count { |
|
67 | 67 | get { return m_nextId; } |
|
68 | 68 | } |
|
69 | 69 | |
|
70 | 70 | public bool Contains(T symbol) { |
|
71 | 71 | return true; |
|
72 | 72 | } |
|
73 | 73 | |
|
74 | public IEnumerable<T> GetSymbols(int cls) { | |
|
75 | for (var i = 0; i < m_map.Length; i++) | |
|
76 | if (m_map[i] == cls) | |
|
77 | yield return GetSymbolByIndex(i); | |
|
78 | } | |
|
79 | ||
|
74 | 80 | public abstract int GetSymbolIndex(T symbol); |
|
75 | 81 | |
|
82 | public abstract T GetSymbolByIndex(int index); | |
|
83 | ||
|
76 | 84 | public abstract IEnumerable<T> InputSymbols { get; } |
|
77 | 85 | |
|
78 | 86 | /// <summary> |
|
79 | 87 | /// Gets the translation map from the index of the symbol to it's class this is usefull for the optimized input symbols transtaion. |
|
80 | 88 | /// </summary> |
|
81 | 89 | /// <returns>The translation map.</returns> |
|
82 | 90 | public int[] GetTranslationMap() { |
|
83 | 91 | return m_map; |
|
84 | 92 | } |
|
85 | 93 | } |
|
86 | 94 | } |
@@ -1,74 +1,77 | |||
|
1 | 1 | using System; |
|
2 | 2 | using System.Collections.Generic; |
|
3 | 3 | using System.Linq; |
|
4 | 4 | |
|
5 | 5 | namespace Implab.Automaton { |
|
6 | 6 | public class MapAlphabet<T> : IAlphabetBuilder<T> { |
|
7 | 7 | readonly Dictionary<T,int> m_map; |
|
8 | 8 | int m_nextCls; |
|
9 | 9 | readonly bool m_supportUnclassified; |
|
10 | 10 | |
|
11 | 11 | public MapAlphabet(bool supportUnclassified, IEqualityComparer<T> comparer) { |
|
12 | 12 | m_map = comparer != null ? new Dictionary<T, int>(comparer) : new Dictionary<T,int>(); |
|
13 | 13 | m_supportUnclassified = supportUnclassified; |
|
14 | 14 | m_nextCls = supportUnclassified ? 1 : 0; |
|
15 | 15 | } |
|
16 | 16 | |
|
17 | 17 | #region IAlphabetBuilder implementation |
|
18 | 18 | |
|
19 | 19 | public int DefineSymbol(T symbol) { |
|
20 | 20 | int cls; |
|
21 | 21 | return m_map.TryGetValue(symbol, out cls) ? cls : DefineSymbol(symbol, m_nextCls); |
|
22 | 22 | } |
|
23 | 23 | |
|
24 | 24 | public int DefineSymbol(T symbol, int cls) { |
|
25 | 25 | Safe.ArgumentAssert(cls >= 0, "cls"); |
|
26 | 26 | |
|
27 | 27 | m_nextCls = Math.Max(cls + 1, m_nextCls); |
|
28 | 28 | m_map.Add(symbol, cls); |
|
29 | 29 | return cls; |
|
30 | 30 | } |
|
31 | 31 | |
|
32 | 32 | public int DefineClass(IEnumerable<T> symbols) { |
|
33 | 33 | return DefineClass(symbols, m_nextCls); |
|
34 | 34 | } |
|
35 | 35 | |
|
36 | 36 | public int DefineClass(IEnumerable<T> symbols, int cls) { |
|
37 | 37 | Safe.ArgumentAssert(cls >= 0, "cls"); |
|
38 | 38 | Safe.ArgumentNotNull(symbols, "symbols"); |
|
39 | 39 | |
|
40 | 40 | m_nextCls = Math.Max(cls + 1, m_nextCls); |
|
41 | symbols = symbols.Distinct(); | |
|
42 | 41 | |
|
43 | 42 | foreach (var symbol in symbols) |
|
44 | 43 | m_map[symbol] = cls; |
|
45 | 44 | return cls; |
|
46 | 45 | } |
|
47 | 46 | |
|
48 | 47 | #endregion |
|
49 | 48 | |
|
50 | 49 | #region IAlphabet implementation |
|
51 | 50 | |
|
52 | 51 | public int Translate(T symbol) { |
|
53 | 52 | int cls; |
|
54 | 53 | if (m_map.TryGetValue(symbol, out cls)) |
|
55 | 54 | return cls; |
|
56 | 55 | if (!m_supportUnclassified) |
|
57 | 56 | throw new ArgumentOutOfRangeException("symbol", "The specified symbol isn't in the alphabet"); |
|
58 | 57 | return DFAConst.UNCLASSIFIED_INPUT; |
|
59 | 58 | } |
|
60 | 59 | |
|
61 | 60 | public int Count { |
|
62 | 61 | get { |
|
63 | 62 | return m_nextCls; |
|
64 | 63 | } |
|
65 | 64 | } |
|
66 | 65 | |
|
67 | 66 | public bool Contains(T symbol) { |
|
68 | 67 | return m_supportUnclassified || m_map.ContainsKey(symbol); |
|
69 | 68 | } |
|
70 | 69 | |
|
70 | ||
|
71 | public IEnumerable<T> GetSymbols(int cls) { | |
|
72 | return m_map.Where(p => p.Value == cls).Select(p => p.Key); | |
|
73 | } | |
|
71 | 74 | #endregion |
|
72 | 75 | } |
|
73 | 76 | } |
|
74 | 77 |
@@ -1,23 +1,23 | |||
|
1 | 1 | using System; |
|
2 | 2 | |
|
3 | 3 | namespace Implab.Automaton.RegularExpressions { |
|
4 |
public struct DFAStateDescriptor |
|
|
4 | public struct DFAStateDescriptor<T> { | |
|
5 | 5 | public readonly bool final; |
|
6 | 6 | |
|
7 | 7 | public readonly int[] transitions; |
|
8 | 8 | |
|
9 | 9 | public readonly T[] tags; |
|
10 | 10 | |
|
11 |
public DFAStateDescriptor |
|
|
11 | public DFAStateDescriptor(int size, bool final, T[] tags) { | |
|
12 | 12 | Safe.ArgumentAssert(size >= 0, "size"); |
|
13 | 13 | this.final = final; |
|
14 | 14 | this.tags = tags; |
|
15 | 15 | |
|
16 | 16 | transitions = new int[size]; |
|
17 | 17 | |
|
18 | 18 | for (int i = 0; i < size; i++) |
|
19 | 19 | transitions[i] = DFAConst.UNREACHABLE_STATE; |
|
20 | 20 | } |
|
21 | 21 | } |
|
22 | 22 | } |
|
23 | 23 |
@@ -1,91 +1,89 | |||
|
1 | 1 | using Implab; |
|
2 | 2 | using System; |
|
3 | 3 | using System.Collections.Generic; |
|
4 | 4 | using System.Linq; |
|
5 | 5 | |
|
6 | 6 | namespace Implab.Automaton.RegularExpressions { |
|
7 | 7 | /// <summary> |
|
8 | 8 | /// Базовый абстрактный класс. Грамматика, позволяет формулировать выражения над алфавитом типа <c>char</c>. |
|
9 | 9 | /// </summary> |
|
10 | 10 | public abstract class Grammar<TSymbol, TTag> { |
|
11 | 11 | |
|
12 | 12 | protected abstract IAlphabetBuilder<TSymbol> AlphabetBuilder { |
|
13 | 13 | get; |
|
14 | 14 | } |
|
15 | 15 | |
|
16 | 16 | protected SymbolToken<TTag> UnclassifiedToken() { |
|
17 | 17 | return new SymbolToken<TTag>(DFAConst.UNCLASSIFIED_INPUT); |
|
18 | 18 | } |
|
19 | 19 | |
|
20 | 20 | protected void DefineAlphabet(IEnumerable<TSymbol> alphabet) { |
|
21 | 21 | Safe.ArgumentNotNull(alphabet, "alphabet"); |
|
22 | 22 | |
|
23 | 23 | foreach (var ch in alphabet) |
|
24 | 24 | AlphabetBuilder.DefineSymbol(ch); |
|
25 | 25 | } |
|
26 | 26 | |
|
27 | 27 | protected Token<TTag> SymbolToken(TSymbol symbol) { |
|
28 | 28 | return Token<TTag>.New(TranslateOrAdd(symbol)); |
|
29 | 29 | } |
|
30 | 30 | |
|
31 | 31 | protected Token<TTag> SymbolToken(IEnumerable<TSymbol> symbols) { |
|
32 | 32 | Safe.ArgumentNotNull(symbols, "symbols"); |
|
33 | 33 | |
|
34 | 34 | return Token<TTag>.New(TranslateOrAdd(symbols).ToArray()); |
|
35 | 35 | } |
|
36 | 36 | |
|
37 | 37 | protected Token<TTag> SymbolSetToken(params TSymbol[] set) { |
|
38 | 38 | return SymbolToken(set); |
|
39 | 39 | } |
|
40 | 40 | |
|
41 | 41 | int TranslateOrAdd(TSymbol ch) { |
|
42 | 42 | var t = AlphabetBuilder.Translate(ch); |
|
43 | 43 | if (t == DFAConst.UNCLASSIFIED_INPUT) |
|
44 | 44 | t = AlphabetBuilder.DefineSymbol(ch); |
|
45 | 45 | return t; |
|
46 | 46 | } |
|
47 | 47 | |
|
48 | 48 | IEnumerable<int> TranslateOrAdd(IEnumerable<TSymbol> symbols) { |
|
49 | 49 | return symbols.Distinct().Select(TranslateOrAdd); |
|
50 | 50 | } |
|
51 | 51 | |
|
52 | 52 | int TranslateOrDie(TSymbol ch) { |
|
53 | 53 | var t = AlphabetBuilder.Translate(ch); |
|
54 | 54 | if (t == DFAConst.UNCLASSIFIED_INPUT) |
|
55 | 55 | throw new ApplicationException(String.Format("Symbol '{0}' is UNCLASSIFIED", ch)); |
|
56 | 56 | return t; |
|
57 | 57 | } |
|
58 | 58 | |
|
59 | 59 | IEnumerable<int> TranslateOrDie(IEnumerable<TSymbol> symbols) { |
|
60 | 60 | return symbols.Distinct().Select(TranslateOrDie); |
|
61 | 61 | } |
|
62 | 62 | |
|
63 | 63 | protected Token<TTag> SymbolTokenExcept(IEnumerable<TSymbol> symbols) { |
|
64 | 64 | Safe.ArgumentNotNull(symbols, "symbols"); |
|
65 | 65 | |
|
66 | 66 | return Token<TTag>.New( Enumerable.Range(0, AlphabetBuilder.Count).Except(TranslateOrDie(symbols)).ToArray() ); |
|
67 | 67 | } |
|
68 | 68 | |
|
69 |
protected |
|
|
70 | Safe.ArgumentNotNull(lang, "lang"); | |
|
71 | Safe.ArgumentNotNull(dfaAlphabet, "dfaAlphabet"); | |
|
69 | protected abstract IAlphabetBuilder<TSymbol> CreateAlphabet(); | |
|
72 | 70 | |
|
73 | var dfa = new RegularDFADefinition<TSymbol, TTag>(AlphabetBuilder); | |
|
71 | protected RegularDFA<TSymbol, TTag> BuildDFA(Token<TTag> regexp) { | |
|
74 | 72 | |
|
75 |
var |
|
|
73 | var dfa = new RegularDFA<TSymbol, TTag>(AlphabetBuilder); | |
|
76 | 74 | |
|
77 | lang.Accept( builder ); | |
|
75 | var visitor = new RegularExpressionVisitor<TTag>(); | |
|
76 | regexp.Accept( visitor ); | |
|
78 | 77 | |
|
79 |
|
|
|
78 | visitor.BuildDFA(dfa); | |
|
80 | 79 | |
|
81 | 80 | if (dfa.IsFinalState(dfa.InitialState)) |
|
82 | 81 | throw new ApplicationException("The specified language contains empty token"); |
|
83 | 82 | |
|
84 |
dfa.Optimize( |
|
|
85 | ||
|
83 | return dfa.Optimize(CreateAlphabet()); | |
|
86 | 84 | } |
|
87 | 85 | |
|
88 | 86 | } |
|
89 | 87 | |
|
90 | 88 | |
|
91 | 89 | } |
@@ -1,265 +1,255 | |||
|
1 | 1 | using Implab; |
|
2 | 2 | using System; |
|
3 | 3 | using System.Collections.Generic; |
|
4 | 4 | using System.IO; |
|
5 | 5 | using Implab.Components; |
|
6 | using Implab.Automaton.RegularExpressions; | |
|
6 | 7 | |
|
7 | 8 | namespace Implab.Automaton { |
|
8 | 9 | /// <summary> |
|
9 | 10 | /// Базовый класс для разбора потока входных символов на токены. |
|
10 | 11 | /// </summary> |
|
11 | 12 | /// <remarks> |
|
12 | 13 | /// Сканнер имеет внутри буффер с симолами входного текста, по которому перемещаются два |
|
13 | 14 | /// указателя, начала и конца токена, при перемещении искользуется ДКА для определения |
|
14 | 15 | /// конца токена и допустимости текущего символа. |
|
15 | 16 | /// </remarks> |
|
16 | 17 | public abstract class Scanner<TTag> : Disposable { |
|
17 | struct ScannerConfig { | |
|
18 |
public DFAStateDescript |
|
|
19 |
public int[] alphabet |
|
|
20 | public int initialState; | |
|
18 | protected struct ScannerConfig { | |
|
19 | public readonly DFAStateDescriptor<TTag>[] states; | |
|
20 | public readonly int[] alphabet; | |
|
21 | public readonly int initialState; | |
|
22 | ||
|
23 | public ScannerConfig(DFAStateDescriptor<TTag>[] states, int[] alphabet, int initialState) { | |
|
24 | this.initialState = initialState; | |
|
25 | this.alphabet = alphabet; | |
|
26 | this.states = states; | |
|
27 | } | |
|
21 | 28 | } |
|
22 | 29 | |
|
23 | 30 | Stack<ScannerConfig> m_defs = new Stack<ScannerConfig>(); |
|
24 | 31 | |
|
25 | DFAStateDescriptior<TTag>[] m_states; | |
|
26 | int[] m_alphabetMap; | |
|
27 | int m_initialState; | |
|
32 | ScannerConfig m_config; | |
|
28 | 33 | |
|
29 |
protected DFAStateDescript |
|
|
34 | protected DFAStateDescriptor<TTag> m_currentState; | |
|
30 | 35 | int m_previewCode; |
|
31 | 36 | |
|
32 | 37 | protected int m_tokenLen; |
|
33 | 38 | protected int m_tokenOffset; |
|
34 | 39 | |
|
35 | 40 | protected char[] m_buffer; |
|
36 | 41 | protected int m_bufferSize; |
|
37 | 42 | protected int m_pointer; |
|
38 | 43 | |
|
39 | 44 | TextReader m_reader; |
|
40 | 45 | bool m_disposeReader; |
|
41 | 46 | int m_chunkSize = 1024; // 1k |
|
42 | 47 | int m_limit = 10 * 1024 * 1024; // 10Mb |
|
43 | 48 | |
|
44 | protected Scanner(DFAStateDescriptior<TTag>[] states, int[] alphabet, int initialState) { | |
|
45 | Safe.ArgumentNotEmpty(states, "states"); | |
|
46 | Safe.ArgumentNotNull(alphabet, "alphabet"); | |
|
49 | protected Scanner(ScannerConfig config) { | |
|
50 | Safe.ArgumentNotEmpty(config.states, "config.states"); | |
|
51 | Safe.ArgumentNotNull(config.alphabet, "config.alphabet"); | |
|
47 | 52 | |
|
48 |
m_ |
|
|
49 | m_alphabetMap = alphabet; | |
|
50 | m_initialState = initialState; | |
|
51 | ||
|
52 | Feed(new char[0]); | |
|
53 | m_config = config; | |
|
53 | 54 | } |
|
54 | 55 | |
|
55 | 56 | /// <summary> |
|
56 | 57 | /// Заполняет входными данными буффер. |
|
57 | 58 | /// </summary> |
|
58 | 59 | /// <param name="data">Данные для обработки.</param> |
|
59 | 60 | /// <remarks>Копирование данных не происходит, переданный массив используется в |
|
60 | 61 | /// качестве входного буффера.</remarks> |
|
61 | 62 | public void Feed(char[] data) { |
|
62 | 63 | Safe.ArgumentNotNull(data, "data"); |
|
63 | 64 | |
|
64 | 65 | Feed(data, data.Length); |
|
65 | 66 | } |
|
66 | 67 | |
|
67 | 68 | /// <summary> |
|
68 | 69 | /// Заполняет буффур чтения входными данными. |
|
69 | 70 | /// </summary> |
|
70 | 71 | /// <param name="data">Данные для обработки.</param> |
|
71 | 72 | /// <param name="length">Длина данных для обработки.</param> |
|
72 | 73 | /// <remarks>Копирование данных не происходит, переданный массив используется в |
|
73 | 74 | /// качестве входного буффера.</remarks> |
|
74 | 75 | public void Feed(char[] data, int length) { |
|
75 | 76 | Safe.ArgumentNotNull(data, "data"); |
|
76 | 77 | Safe.ArgumentInRange(length, 0, data.Length, "length"); |
|
77 | 78 | AssertNotDisposed(); |
|
78 | 79 | |
|
79 | 80 | m_pointer = -1; |
|
80 | 81 | m_buffer = data; |
|
81 | 82 | m_bufferSize = length; |
|
82 | 83 | Shift(); |
|
83 | 84 | } |
|
84 | 85 | |
|
85 | 86 | public void Feed(TextReader reader, bool dispose) { |
|
86 | 87 | Safe.ArgumentNotNull(reader, "reader"); |
|
87 | 88 | AssertNotDisposed(); |
|
88 | 89 | |
|
89 | 90 | if (m_reader != null && m_disposeReader) |
|
90 | 91 | m_reader.Dispose(); |
|
91 | 92 | |
|
92 | 93 | m_reader = reader; |
|
93 | 94 | m_disposeReader = dispose; |
|
94 | 95 | m_pointer = -1; |
|
95 | 96 | m_buffer = new char[m_chunkSize]; |
|
96 | 97 | m_bufferSize = 0; |
|
97 | 98 | Shift(); |
|
98 | 99 | } |
|
99 | 100 | |
|
100 | 101 | /// <summary> |
|
101 | 102 | /// Получает текущий токен в виде строки. |
|
102 | 103 | /// </summary> |
|
103 | 104 | /// <returns></returns> |
|
104 | 105 | protected string GetTokenValue() { |
|
105 | 106 | return new String(m_buffer, m_tokenOffset, m_tokenLen); |
|
106 | 107 | } |
|
107 | 108 | |
|
108 | 109 | /// <summary> |
|
109 | 110 | /// Метки текущего токена, которые были назначены в регулярном выражении. |
|
110 | 111 | /// </summary> |
|
111 | 112 | protected TTag[] TokenTags { |
|
112 | 113 | get { |
|
113 | return m_currentState.tag; | |
|
114 | return m_currentState.tags; | |
|
114 | 115 | } |
|
115 | 116 | } |
|
116 | 117 | |
|
117 | 118 | /// <summary> |
|
118 | 119 | /// Признак конца данных |
|
119 | 120 | /// </summary> |
|
120 | 121 | public bool EOF { |
|
121 | 122 | get { |
|
122 | 123 | return m_pointer >= m_bufferSize; |
|
123 | 124 | } |
|
124 | 125 | } |
|
125 | 126 | |
|
126 | 127 | /// <summary> |
|
127 | 128 | /// Читает следующий токен, при этом <see cref="m_tokenOffset"/> указывает на начало токена, |
|
128 | 129 | /// <see cref="m_tokenLen"/> на длину токена, <see cref="m_buffer"/> - массив символов, в |
|
129 | 130 | /// котором находится токен. |
|
130 | 131 | /// </summary> |
|
131 | 132 | /// <returns><c>false</c> - достигнут конец данных, токен не прочитан.</returns> |
|
132 | 133 | protected bool ReadTokenInternal() { |
|
133 | 134 | if (m_pointer >= m_bufferSize) |
|
134 | 135 | return false; |
|
135 | 136 | |
|
136 | m_currentState = m_states[m_initialState]; | |
|
137 | m_currentState = m_config.states[m_config.initialState]; | |
|
137 | 138 | m_tokenLen = 0; |
|
138 | 139 | m_tokenOffset = m_pointer; |
|
139 | 140 | int nextState; |
|
140 | 141 | do { |
|
141 | 142 | nextState = m_currentState.transitions[m_previewCode]; |
|
142 | 143 | if (nextState == DFAConst.UNREACHABLE_STATE) { |
|
143 | 144 | if (m_currentState.final) |
|
144 | 145 | return true; |
|
145 | 146 | |
|
146 | 147 | throw new ParserException( |
|
147 | 148 | String.Format( |
|
148 | 149 | "Unexpected symbol '{0}', at pos {1}", |
|
149 | 150 | m_buffer[m_pointer], |
|
150 | 151 | Position |
|
151 | 152 | ) |
|
152 | 153 | ); |
|
153 | 154 | } |
|
154 | m_currentState = m_states[nextState]; | |
|
155 | m_currentState = m_config.states[nextState]; | |
|
155 | 156 | m_tokenLen++; |
|
156 | 157 | |
|
157 | 158 | } while (Shift()); |
|
158 | 159 | |
|
159 | 160 | // END OF DATA |
|
160 | 161 | if (!m_currentState.final) |
|
161 | 162 | throw new ParserException("Unexpected end of data"); |
|
162 | 163 | |
|
163 | 164 | return true; |
|
164 | 165 | } |
|
165 | 166 | |
|
166 | 167 | |
|
167 | 168 | bool Shift() { |
|
168 | 169 | m_pointer++; |
|
169 | 170 | |
|
170 | 171 | if (m_pointer >= m_bufferSize) { |
|
171 | 172 | if (!ReadNextChunk()) |
|
172 | 173 | return false; |
|
173 | 174 | } |
|
174 | 175 | |
|
175 |
m_previewCode = m_alphabet |
|
|
176 | m_previewCode = m_config.alphabet[m_buffer[m_pointer]]; | |
|
176 | 177 | |
|
177 | 178 | return true; |
|
178 | 179 | } |
|
179 | 180 | |
|
180 | 181 | bool ReadNextChunk() { |
|
181 | 182 | if (m_reader == null) |
|
182 | 183 | return false; |
|
183 | 184 | |
|
184 | 185 | // extend buffer if nesessary |
|
185 | 186 | if (m_pointer + m_chunkSize > m_buffer.Length) { |
|
186 | 187 | // trim unused buffer head |
|
187 | 188 | var size = m_tokenLen + m_chunkSize; |
|
188 | 189 | if (size >= m_limit) |
|
189 | 190 | throw new ParserException(String.Format("Input buffer {0} bytes limit exceeded", m_limit)); |
|
190 | 191 | var temp = new char[size]; |
|
191 | 192 | Array.Copy(m_buffer, m_tokenOffset, temp, 0, m_tokenLen); |
|
192 | 193 | m_pointer -= m_tokenOffset; |
|
193 | 194 | m_bufferSize -= m_tokenOffset; |
|
194 | 195 | m_tokenOffset = 0; |
|
195 | 196 | m_buffer = temp; |
|
196 | 197 | } |
|
197 | 198 | |
|
198 | 199 | var read = m_reader.Read(m_buffer, m_tokenLen, m_chunkSize); |
|
199 | 200 | if (read == 0) |
|
200 | 201 | return false; |
|
201 | 202 | |
|
202 | 203 | m_bufferSize += read; |
|
203 | 204 | |
|
204 | 205 | return true; |
|
205 | 206 | } |
|
206 | 207 | |
|
207 | 208 | /// <summary> |
|
208 | 209 | /// Позиция сканнера во входном буфере |
|
209 | 210 | /// </summary> |
|
210 | 211 | public int Position { |
|
211 | 212 | get { |
|
212 | 213 | return m_pointer + 1; |
|
213 | 214 | } |
|
214 | 215 | } |
|
215 | 216 | |
|
216 | 217 | /// <summary> |
|
217 | 218 | /// Преключает внутренний ДКА на указанный, позволяет реализовать подобие захватывающей |
|
218 | 219 | /// группировки. |
|
219 | 220 | /// </summary> |
|
220 |
/// <param name |
|
|
221 | /// <param name="alphabet">Таблица входных символов для нового ДКА</param> | |
|
222 | /// <param name = "initialState"></param> | |
|
223 | protected void Switch(DFAStateDescriptior<TTag>[] states, int[] alphabet, int initialState) { | |
|
224 | Safe.ArgumentNotNull(states, "dfa"); | |
|
221 | /// <param name = "config"></param> | |
|
222 | protected void Switch(ScannerConfig config) { | |
|
223 | Safe.ArgumentNotNull(config.states, "config.states"); | |
|
225 | 224 | |
|
226 |
m_defs.Push( |
|
|
227 | states = m_states, | |
|
228 | alphabetMap = m_alphabetMap, | |
|
229 | initialState = m_initialState | |
|
230 | }); | |
|
225 | m_defs.Push(m_config); | |
|
226 | m_config = config; | |
|
231 | 227 | |
|
232 | m_states = states; | |
|
233 | m_alphabetMap = alphabet; | |
|
234 | m_initialState = initialState; | |
|
235 | ||
|
236 | m_previewCode = m_alphabetMap[m_buffer[m_pointer]]; | |
|
228 | m_previewCode = m_config.alphabet[m_buffer[m_pointer]]; | |
|
237 | 229 | } |
|
238 | 230 | |
|
239 | 231 | /// <summary> |
|
240 | 232 | /// Восстанавливает предыдущей ДКА сканнера. |
|
241 | 233 | /// </summary> |
|
242 | 234 | protected void Restore() { |
|
243 | 235 | if (m_defs.Count == 0) |
|
244 | 236 | throw new InvalidOperationException(); |
|
245 |
|
|
|
246 | m_states = prev.states; | |
|
247 | m_alphabetMap = prev.alphabetMap; | |
|
248 | m_initialState = prev.initialState; | |
|
249 | m_previewCode = m_alphabetMap[m_buffer[m_pointer]]; | |
|
237 | m_config = m_defs.Pop(); | |
|
238 | ||
|
239 | m_previewCode = m_config.alphabet[m_buffer[m_pointer]]; | |
|
250 | 240 | } |
|
251 | 241 | |
|
252 | 242 | protected override void Dispose(bool disposing) { |
|
253 | 243 | if (disposing) { |
|
254 | 244 | if (m_reader != null && m_disposeReader) |
|
255 | 245 | m_reader.Dispose(); |
|
256 | 246 | m_buffer = null; |
|
257 | 247 | m_bufferSize = 0; |
|
258 | 248 | m_pointer = 0; |
|
259 | 249 | m_tokenLen = 0; |
|
260 | 250 | m_tokenOffset = 0; |
|
261 | 251 | } |
|
262 | 252 | base.Dispose(disposing); |
|
263 | 253 | } |
|
264 | 254 | } |
|
265 | 255 | } |
@@ -1,25 +1,29 | |||
|
1 | 1 | using System.Collections.Generic; |
|
2 | 2 | using System.Linq; |
|
3 | 3 | using Implab.Automaton; |
|
4 | 4 | |
|
5 | 5 | namespace Implab.Formats { |
|
6 | 6 | public class ByteAlphabet : IndexedAlphabetBase<byte> { |
|
7 | 7 | public ByteAlphabet() : base(byte.MaxValue + 1){ |
|
8 | 8 | } |
|
9 | 9 | |
|
10 | 10 | #region implemented abstract members of IndexedAlphabetBase |
|
11 | 11 | |
|
12 | 12 | public override int GetSymbolIndex(byte symbol) { |
|
13 | 13 | return (int)symbol; |
|
14 | 14 | } |
|
15 | 15 | |
|
16 | public override byte GetSymbolByIndex(int index) { | |
|
17 | return (byte)index; | |
|
18 | } | |
|
19 | ||
|
16 | 20 | public IEnumerable<byte> InputSymbols { |
|
17 | 21 | get { |
|
18 | 22 | return Enumerable.Range(byte.MinValue, byte.MaxValue).Cast<byte>(); |
|
19 | 23 | } |
|
20 | 24 | } |
|
21 | 25 | |
|
22 | 26 | #endregion |
|
23 | 27 | } |
|
24 | 28 | } |
|
25 | 29 |
@@ -1,20 +1,24 | |||
|
1 | 1 | using System.Collections.Generic; |
|
2 | 2 | using System.Linq; |
|
3 | 3 | using Implab.Automaton; |
|
4 | 4 | |
|
5 | 5 | namespace Implab.Formats { |
|
6 | 6 | public class CharAlphabet: IndexedAlphabetBase<char> { |
|
7 | 7 | |
|
8 | 8 | public CharAlphabet() |
|
9 | 9 | : base(char.MaxValue + 1) { |
|
10 | 10 | } |
|
11 | 11 | |
|
12 | 12 | public override int GetSymbolIndex(char symbol) { |
|
13 | 13 | return symbol; |
|
14 | 14 | } |
|
15 | 15 | |
|
16 | public override char GetSymbolByIndex(int index) { | |
|
17 | return (char)index; | |
|
18 | } | |
|
19 | ||
|
16 | 20 | public override IEnumerable<char> InputSymbols { |
|
17 | 21 | get { return Enumerable.Range(char.MinValue, char.MaxValue).Cast<char>(); } |
|
18 | 22 | } |
|
19 | 23 | } |
|
20 | 24 | } |
@@ -1,115 +1,116 | |||
|
1 | 1 | using System.Linq; |
|
2 | 2 | using Implab.Automaton.RegularExpressions; |
|
3 | 3 | using System; |
|
4 | using Implab.Automaton; | |
|
4 | 5 | |
|
5 | 6 | namespace Implab.Formats.JSON { |
|
6 | 7 | class JSONGrammar : Grammar<char,JSONGrammar.TokenType> { |
|
7 | 8 | public enum TokenType { |
|
8 | 9 | None, |
|
9 | 10 | BeginObject, |
|
10 | 11 | EndObject, |
|
11 | 12 | BeginArray, |
|
12 | 13 | EndArray, |
|
13 | 14 | String, |
|
14 | 15 | Number, |
|
15 | 16 | Literal, |
|
16 | 17 | NameSeparator, |
|
17 | 18 | ValueSeparator, |
|
18 | 19 | |
|
19 | 20 | StringBound, |
|
20 | 21 | EscapedChar, |
|
21 | 22 | UnescapedChar, |
|
22 | 23 | EscapedUnicode, |
|
23 | 24 | |
|
24 | 25 | Minus, |
|
25 | 26 | Plus, |
|
26 | 27 | Sign, |
|
27 | 28 | Integer, |
|
28 | 29 | Dot, |
|
29 | 30 | Exp |
|
30 | 31 | } |
|
31 | 32 | |
|
32 | 33 | static Lazy<JSONGrammar> _instance = new Lazy<JSONGrammar>(); |
|
33 | 34 | |
|
34 | 35 | public static JSONGrammar Instance { |
|
35 | 36 | get { return _instance.Value; } |
|
36 | 37 | } |
|
37 | 38 | |
|
38 |
readonly Regular |
|
|
39 |
readonly Regular |
|
|
39 | readonly RegularDFA<char, TokenType> m_jsonDFA; | |
|
40 | readonly RegularDFA<char, TokenType> m_stringDFA; | |
|
40 | 41 | |
|
41 | 42 | public JSONGrammar() { |
|
42 | 43 | DefineAlphabet(Enumerable.Range(0, 0x20).Select(x => (char)x)); |
|
43 | 44 | var hexDigit = SymbolRangeToken('a','f').Or(SymbolRangeToken('A','F')).Or(SymbolRangeToken('0','9')); |
|
44 | 45 | var digit9 = SymbolRangeToken('1', '9'); |
|
45 | 46 | var zero = SymbolToken('0'); |
|
46 | 47 | var digit = zero.Or(digit9); |
|
47 | 48 | var dot = SymbolToken('.'); |
|
48 | 49 | var minus = SymbolToken('-'); |
|
49 | 50 | var sign = SymbolSetToken('-', '+'); |
|
50 | 51 | var expSign = SymbolSetToken('e', 'E'); |
|
51 | 52 | var letters = SymbolRangeToken('a', 'z'); |
|
52 | 53 | var integer = zero.Or(digit9.Cat(digit.EClosure())); |
|
53 | 54 | var frac = dot.Cat(digit.Closure()); |
|
54 | 55 | var exp = expSign.Cat(sign.Optional()).Cat(digit.Closure()); |
|
55 | 56 | var quote = SymbolToken('"'); |
|
56 | 57 | var backSlash = SymbolToken('\\'); |
|
57 | 58 | var specialEscapeChars = SymbolSetToken('\\', '"', '/', 'b', 'f', 't', 'n', 'r'); |
|
58 | 59 | var unicodeEspace = SymbolToken('u').Cat(hexDigit.Repeat(4)); |
|
59 | 60 | var whitespace = SymbolSetToken('\n', '\r', '\t', ' ').EClosure(); |
|
60 | 61 | var beginObject = whitespace.Cat(SymbolToken('{')).Cat(whitespace); |
|
61 | 62 | var endObject = whitespace.Cat(SymbolToken('}')).Cat(whitespace); |
|
62 | 63 | var beginArray = whitespace.Cat(SymbolToken('[')).Cat(whitespace); |
|
63 | 64 | var endArray = whitespace.Cat(SymbolToken(']')).Cat(whitespace); |
|
64 | 65 | var nameSep = whitespace.Cat(SymbolToken(':')).Cat(whitespace); |
|
65 | 66 | var valueSep = whitespace.Cat(SymbolToken(',')).Cat(whitespace); |
|
66 | 67 | |
|
67 | 68 | var number = minus.Optional().Cat(integer).Cat(frac.Optional()).Cat(exp.Optional()); |
|
68 | 69 | var literal = letters.Closure(); |
|
69 | 70 | var unescaped = SymbolTokenExcept(Enumerable.Range(0, 0x20).Union(new int[] { '\\', '"' }).Select(x => (char)x)); |
|
70 | 71 | |
|
71 | 72 | var jsonExpression = |
|
72 | 73 | number.Tag(TokenType.Number) |
|
73 | 74 | .Or(literal.Tag(TokenType.Literal)) |
|
74 | 75 | .Or(quote.Tag(TokenType.StringBound)) |
|
75 | 76 | .Or(beginObject.Tag(TokenType.BeginObject)) |
|
76 | 77 | .Or(endObject.Tag(TokenType.EndObject)) |
|
77 | 78 | .Or(beginArray.Tag(TokenType.BeginArray)) |
|
78 | 79 | .Or(endArray.Tag(TokenType.EndArray)) |
|
79 | 80 | .Or(nameSep.Tag(TokenType.NameSeparator)) |
|
80 | 81 | .Or(valueSep.Tag(TokenType.ValueSeparator)); |
|
81 | 82 | |
|
82 | 83 | |
|
83 | 84 | var jsonStringExpression = |
|
84 | 85 | quote.Tag(TokenType.StringBound) |
|
85 | 86 | .Or(backSlash.Cat(specialEscapeChars).Tag(TokenType.EscapedChar)) |
|
86 | 87 | .Or(backSlash.Cat(unicodeEspace).Tag(TokenType.EscapedUnicode)) |
|
87 | 88 | .Or(unescaped.Closure().Tag(TokenType.UnescapedChar)); |
|
88 | 89 | |
|
89 | 90 | |
|
90 | m_jsonDFA = new RegularCharDFADefinition<TokenType>(new CharAlphabet()); | |
|
91 | BuildDFA(jsonExpression, m_jsonDFA, m_jsonDFA.InputAlphabet); | |
|
92 | ||
|
93 | ||
|
94 | m_stringDFA = new RegularCharDFADefinition<TokenType>(new CharAlphabet()); | |
|
95 | BuildDFA(jsonStringExpression, m_jsonDFA, m_jsonDFA.InputAlphabet); | |
|
91 | m_jsonDFA = BuildDFA(jsonExpression); | |
|
92 | m_stringDFA = BuildDFA(jsonStringExpression); | |
|
96 | 93 | } |
|
97 | 94 | |
|
98 |
public Regular |
|
|
95 | public RegularDFA<char, TokenType> JsonDFA { | |
|
99 | 96 | get { |
|
100 | 97 | return m_jsonDFA; |
|
101 | 98 | } |
|
102 | 99 | } |
|
103 | 100 | |
|
104 |
public RegularDFA |
|
|
101 | public RegularDFA<char,TokenType> JsonStringDFA { | |
|
105 | 102 | get { |
|
106 | 103 | return m_stringDFA; |
|
107 | 104 | } |
|
108 | 105 | } |
|
109 | 106 | |
|
110 | 107 | Token<TokenType> SymbolRangeToken(char start, char stop) { |
|
111 | 108 | return SymbolToken(Enumerable.Range(start,stop - start).Cast<char>()); |
|
112 | 109 | } |
|
113 | 110 | |
|
111 | protected override IAlphabetBuilder<char> CreateAlphabet() { | |
|
112 | return new CharAlphabet(); | |
|
113 | } | |
|
114 | ||
|
114 | 115 | } |
|
115 | 116 | } |
@@ -1,280 +1,280 | |||
|
1 | 1 | using System; |
|
2 | 2 | using System.Diagnostics; |
|
3 | 3 | using System.IO; |
|
4 | 4 | using Implab.Automaton; |
|
5 | 5 | using Implab.Automaton.RegularExpressions; |
|
6 | 6 | using System.Linq; |
|
7 | 7 | using Implab.Components; |
|
8 | 8 | |
|
9 | 9 | namespace Implab.Formats.JSON { |
|
10 | 10 | /// <summary> |
|
11 | 11 | /// internal |
|
12 | 12 | /// </summary> |
|
13 | 13 | public struct JSONParserContext { |
|
14 | 14 | public string memberName; |
|
15 | 15 | public JSONElementContext elementContext; |
|
16 | 16 | } |
|
17 | 17 | |
|
18 | 18 | /// <summary> |
|
19 | 19 | /// Pull парсер JSON данных. |
|
20 | 20 | /// </summary> |
|
21 | 21 | /// <remarks> |
|
22 | 22 | /// Следует отметить отдельную интерпретацию свойства <see cref="Level"/>, |
|
23 | 23 | /// оно означает текущий уровень вложенности объектов, однако закрывающий |
|
24 | 24 | /// элемент объекта и массива имеет уровень меньше, чем сам объект. |
|
25 | 25 | /// <code> |
|
26 | 26 | /// { // Level = 1 |
|
27 | 27 | /// "name" : "Peter", // Level = 1 |
|
28 | 28 | /// "address" : { // Level = 2 |
|
29 | 29 | /// city : "Stern" // Level = 2 |
|
30 | 30 | /// } // Level = 1 |
|
31 | 31 | /// } // Level = 0 |
|
32 | 32 | /// </code> |
|
33 | 33 | /// </remarks> |
|
34 | 34 | public class JSONParser : Disposable { |
|
35 | 35 | |
|
36 | 36 | enum MemberContext { |
|
37 | 37 | MemberName, |
|
38 | 38 | MemberValue |
|
39 | 39 | } |
|
40 | 40 | |
|
41 | 41 | struct ParserContext { |
|
42 | 42 | DFAStateDescriptior<object> |
|
43 | 43 | } |
|
44 | 44 | |
|
45 | 45 | static readonly EnumAlphabet<JsonTokenType> _alphabet = EnumAlphabet<JsonTokenType>.FullAlphabet; |
|
46 | 46 | static readonly DFAStateDescriptior<object>[] _jsonDFA; |
|
47 | 47 | static readonly int _jsonDFAInitialState; |
|
48 | 48 | static readonly DFAStateDescriptior<object>[] _objectDFA; |
|
49 | 49 | static readonly int _objectDFAInitialState; |
|
50 | 50 | static readonly DFAStateDescriptior<object>[] _arrayDFA; |
|
51 | 51 | static readonly int _arrayDFAInitialState; |
|
52 | 52 | |
|
53 | 53 | static JSONParser() { |
|
54 | 54 | |
|
55 | 55 | |
|
56 | 56 | var valueExpression = Token(JsonTokenType.BeginArray, JsonTokenType.BeginObject, JsonTokenType.Literal, JsonTokenType.Number, JsonTokenType.String); |
|
57 | 57 | var memberExpression = Token(JsonTokenType.String).Cat(Token(JsonTokenType.NameSeparator)).Cat(valueExpression); |
|
58 | 58 | |
|
59 | 59 | var objectExpression = memberExpression |
|
60 | 60 | .Cat( |
|
61 | 61 | Token(JsonTokenType.ValueSeparator) |
|
62 | 62 | .Cat(memberExpression) |
|
63 | 63 | .EClosure() |
|
64 | 64 | ) |
|
65 | 65 | .Optional() |
|
66 | 66 | .Cat(Token(JsonTokenType.EndObject)) |
|
67 | 67 | .Tag(null); |
|
68 | 68 | var arrayExpression = valueExpression |
|
69 | 69 | .Cat( |
|
70 | 70 | Token(JsonTokenType.ValueSeparator) |
|
71 | 71 | .Cat(valueExpression) |
|
72 | 72 | .EClosure() |
|
73 | 73 | ) |
|
74 | 74 | .Optional() |
|
75 | 75 | .Cat(Token(JsonTokenType.EndArray)) |
|
76 | 76 | .Tag(null); |
|
77 | 77 | |
|
78 | 78 | var jsonExpression = valueExpression.Tag(null); |
|
79 | 79 | |
|
80 | 80 | _jsonDFA = CreateDFA(jsonExpression).GetTransitionTable(); |
|
81 | 81 | _objectDFA = CreateDFA(objectExpression).GetTransitionTable(); |
|
82 | 82 | _arrayDFA = CreateDFA(arrayExpression).GetTransitionTable(); |
|
83 | 83 | } |
|
84 | 84 | |
|
85 | 85 | static Token<object> Token(params JsonTokenType[] input) { |
|
86 | 86 | return Token<object>.New(input.Select(t => _alphabet.Translate(t)).ToArray()); |
|
87 | 87 | } |
|
88 | 88 | |
|
89 |
static RegularDFA |
|
|
90 |
var builder = new Regular |
|
|
91 |
var dfa = new RegularDFA |
|
|
89 | static RegularDFA<JsonTokenType,object> CreateDFA(Token<object> expr) { | |
|
90 | var builder = new RegularExpressionVisitor<object>(); | |
|
91 | var dfa = new RegularDFA<JsonTokenType,object>(_alphabet); | |
|
92 | 92 | |
|
93 | 93 | expr.Accept(builder); |
|
94 | 94 | |
|
95 | 95 | builder.BuildDFA(dfa); |
|
96 | 96 | return dfa; |
|
97 | 97 | } |
|
98 | 98 | |
|
99 | 99 | JSONScanner m_scanner; |
|
100 | 100 | MemberContext m_memberContext; |
|
101 | 101 | |
|
102 | 102 | JSONElementType m_elementType; |
|
103 | 103 | object m_elementValue; |
|
104 | 104 | |
|
105 | 105 | /// <summary> |
|
106 | 106 | /// Создает новый парсер на основе строки, содержащей JSON |
|
107 | 107 | /// </summary> |
|
108 | 108 | /// <param name="text"></param> |
|
109 | 109 | public JSONParser(string text) |
|
110 | 110 | : base(_jsonDFA, INITIAL_STATE, new JSONParserContext { elementContext = JSONElementContext.None, memberName = String.Empty }) { |
|
111 | 111 | Safe.ArgumentNotEmpty(text, "text"); |
|
112 | 112 | m_scanner = new JSONScanner(); |
|
113 | 113 | m_scanner.Feed(text.ToCharArray()); |
|
114 | 114 | } |
|
115 | 115 | |
|
116 | 116 | /// <summary> |
|
117 | 117 | /// Создает новый экземпляр парсера, на основе текстового потока. |
|
118 | 118 | /// </summary> |
|
119 | 119 | /// <param name="reader">Текстовый поток.</param> |
|
120 | 120 | /// <param name="dispose">Признак того, что парсер должен конролировать время жизни входного потока.</param> |
|
121 | 121 | public JSONParser(TextReader reader, bool dispose) |
|
122 | 122 | : base(_jsonDFA, INITIAL_STATE, new JSONParserContext { elementContext = JSONElementContext.None, memberName = String.Empty }) { |
|
123 | 123 | Safe.ArgumentNotNull(reader, "reader"); |
|
124 | 124 | m_scanner = new JSONScanner(); |
|
125 | 125 | m_scanner.Feed(reader, dispose); |
|
126 | 126 | } |
|
127 | 127 | |
|
128 | 128 | /// <summary> |
|
129 | 129 | /// Тип текущего элемента на котором стоит парсер. |
|
130 | 130 | /// </summary> |
|
131 | 131 | public JSONElementType ElementType { |
|
132 | 132 | get { return m_elementType; } |
|
133 | 133 | } |
|
134 | 134 | |
|
135 | 135 | /// <summary> |
|
136 | 136 | /// Имя элемента - имя свойства родительского контейнера. Для элементов массивов и корневого всегда |
|
137 | 137 | /// пустая строка. |
|
138 | 138 | /// </summary> |
|
139 | 139 | public string ElementName { |
|
140 | 140 | get { return m_context.info.memberName; } |
|
141 | 141 | } |
|
142 | 142 | |
|
143 | 143 | /// <summary> |
|
144 | 144 | /// Значение элемента. Только для элементов типа <see cref="JSONElementType.Value"/>, для остальных <c>null</c> |
|
145 | 145 | /// </summary> |
|
146 | 146 | public object ElementValue { |
|
147 | 147 | get { return m_elementValue; } |
|
148 | 148 | } |
|
149 | 149 | |
|
150 | 150 | /// <summary> |
|
151 | 151 | /// Читает слеюудущий объект из потока |
|
152 | 152 | /// </summary> |
|
153 | 153 | /// <returns><c>true</c> - операция чтения прошла успешно, <c>false</c> - конец данных</returns> |
|
154 | 154 | public bool Read() { |
|
155 | 155 | if (m_context.current == UNREACHEBLE_STATE) |
|
156 | 156 | throw new InvalidOperationException("The parser is in invalid state"); |
|
157 | 157 | object tokenValue; |
|
158 | 158 | JsonTokenType tokenType; |
|
159 | 159 | m_context.info.memberName = String.Empty; |
|
160 | 160 | while (m_scanner.ReadToken(out tokenValue, out tokenType)) { |
|
161 | 161 | Move((int)tokenType); |
|
162 | 162 | if (m_context.current == UNREACHEBLE_STATE) |
|
163 | 163 | UnexpectedToken(tokenValue, tokenType); |
|
164 | 164 | switch (tokenType) { |
|
165 | 165 | case JsonTokenType.BeginObject: |
|
166 | 166 | Switch( |
|
167 | 167 | _objectDFA, |
|
168 | 168 | INITIAL_STATE, |
|
169 | 169 | new JSONParserContext { |
|
170 | 170 | memberName = m_context.info.memberName, |
|
171 | 171 | elementContext = JSONElementContext.Object |
|
172 | 172 | } |
|
173 | 173 | ); |
|
174 | 174 | m_elementValue = null; |
|
175 | 175 | m_memberContext = MemberContext.MemberName; |
|
176 | 176 | m_elementType = JSONElementType.BeginObject; |
|
177 | 177 | return true; |
|
178 | 178 | case JsonTokenType.EndObject: |
|
179 | 179 | Restore(); |
|
180 | 180 | m_elementValue = null; |
|
181 | 181 | m_elementType = JSONElementType.EndObject; |
|
182 | 182 | return true; |
|
183 | 183 | case JsonTokenType.BeginArray: |
|
184 | 184 | Switch( |
|
185 | 185 | _arrayDFA, |
|
186 | 186 | INITIAL_STATE, |
|
187 | 187 | new JSONParserContext { |
|
188 | 188 | memberName = m_context.info.memberName, |
|
189 | 189 | elementContext = JSONElementContext.Array |
|
190 | 190 | } |
|
191 | 191 | ); |
|
192 | 192 | m_elementValue = null; |
|
193 | 193 | m_memberContext = MemberContext.MemberValue; |
|
194 | 194 | m_elementType = JSONElementType.BeginArray; |
|
195 | 195 | return true; |
|
196 | 196 | case JsonTokenType.EndArray: |
|
197 | 197 | Restore(); |
|
198 | 198 | m_elementValue = null; |
|
199 | 199 | m_elementType = JSONElementType.EndArray; |
|
200 | 200 | return true; |
|
201 | 201 | case JsonTokenType.String: |
|
202 | 202 | if (m_memberContext == MemberContext.MemberName) { |
|
203 | 203 | m_context.info.memberName = (string)tokenValue; |
|
204 | 204 | break; |
|
205 | 205 | } |
|
206 | 206 | m_elementType = JSONElementType.Value; |
|
207 | 207 | m_elementValue = tokenValue; |
|
208 | 208 | return true; |
|
209 | 209 | case JsonTokenType.Number: |
|
210 | 210 | m_elementType = JSONElementType.Value; |
|
211 | 211 | m_elementValue = tokenValue; |
|
212 | 212 | return true; |
|
213 | 213 | case JsonTokenType.Literal: |
|
214 | 214 | m_elementType = JSONElementType.Value; |
|
215 | 215 | m_elementValue = ParseLiteral((string)tokenValue); |
|
216 | 216 | return true; |
|
217 | 217 | case JsonTokenType.NameSeparator: |
|
218 | 218 | m_memberContext = MemberContext.MemberValue; |
|
219 | 219 | break; |
|
220 | 220 | case JsonTokenType.ValueSeparator: |
|
221 | 221 | m_memberContext = m_context.info.elementContext == JSONElementContext.Object ? MemberContext.MemberName : MemberContext.MemberValue; |
|
222 | 222 | break; |
|
223 | 223 | default: |
|
224 | 224 | UnexpectedToken(tokenValue, tokenType); |
|
225 | 225 | break; |
|
226 | 226 | } |
|
227 | 227 | } |
|
228 | 228 | if (m_context.info.elementContext != JSONElementContext.None) |
|
229 | 229 | throw new ParserException("Unexpedted end of data"); |
|
230 | 230 | return false; |
|
231 | 231 | } |
|
232 | 232 | |
|
233 | 233 | object ParseLiteral(string literal) { |
|
234 | 234 | switch (literal) { |
|
235 | 235 | case "null": |
|
236 | 236 | return null; |
|
237 | 237 | case "false": |
|
238 | 238 | return false; |
|
239 | 239 | case "true": |
|
240 | 240 | return true; |
|
241 | 241 | default: |
|
242 | 242 | UnexpectedToken(literal, JsonTokenType.Literal); |
|
243 | 243 | return null; // avoid compliler error |
|
244 | 244 | } |
|
245 | 245 | } |
|
246 | 246 | |
|
247 | 247 | void UnexpectedToken(object value, JsonTokenType tokenType) { |
|
248 | 248 | throw new ParserException(String.Format("Unexpected token {0}: '{1}'", tokenType, value)); |
|
249 | 249 | } |
|
250 | 250 | |
|
251 | 251 | |
|
252 | 252 | /// <summary> |
|
253 | 253 | /// Признак конца потока |
|
254 | 254 | /// </summary> |
|
255 | 255 | public bool EOF { |
|
256 | 256 | get { |
|
257 | 257 | return m_scanner.EOF; |
|
258 | 258 | } |
|
259 | 259 | } |
|
260 | 260 | |
|
261 | 261 | protected override void Dispose(bool disposing) { |
|
262 | 262 | if (disposing) { |
|
263 | 263 | m_scanner.Dispose(); |
|
264 | 264 | } |
|
265 | 265 | } |
|
266 | 266 | |
|
267 | 267 | /// <summary> |
|
268 | 268 | /// Переходит в конец текущего объекта. |
|
269 | 269 | /// </summary> |
|
270 | 270 | public void SeekElementEnd() { |
|
271 | 271 | var level = Level - 1; |
|
272 | 272 | |
|
273 | 273 | Debug.Assert(level >= 0); |
|
274 | 274 | |
|
275 | 275 | while (Level != level) |
|
276 | 276 | Read(); |
|
277 | 277 | } |
|
278 | 278 | } |
|
279 | 279 | |
|
280 | 280 | } |
@@ -1,272 +1,272 | |||
|
1 | 1 | <?xml version="1.0" encoding="utf-8"?> |
|
2 | 2 | <Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003"> |
|
3 | 3 | <PropertyGroup> |
|
4 | 4 | <Configuration Condition=" '$(Configuration)' == '' ">Debug</Configuration> |
|
5 | 5 | <Platform Condition=" '$(Platform)' == '' ">AnyCPU</Platform> |
|
6 | 6 | <ProjectGuid>{F550F1F8-8746-4AD0-9614-855F4C4B7F05}</ProjectGuid> |
|
7 | 7 | <OutputType>Library</OutputType> |
|
8 | 8 | <RootNamespace>Implab</RootNamespace> |
|
9 | 9 | <AssemblyName>Implab</AssemblyName> |
|
10 | 10 | <TargetFrameworkVersion>v4.5</TargetFrameworkVersion> |
|
11 | 11 | <ReleaseVersion>0.2</ReleaseVersion> |
|
12 | 12 | <ProductVersion>8.0.30703</ProductVersion> |
|
13 | 13 | <SchemaVersion>2.0</SchemaVersion> |
|
14 | 14 | </PropertyGroup> |
|
15 | 15 | <PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Debug|AnyCPU' "> |
|
16 | 16 | <DebugSymbols>true</DebugSymbols> |
|
17 | 17 | <DebugType>full</DebugType> |
|
18 | 18 | <Optimize>false</Optimize> |
|
19 | 19 | <OutputPath>bin\Debug</OutputPath> |
|
20 | 20 | <DefineConstants>TRACE;DEBUG;</DefineConstants> |
|
21 | 21 | <ErrorReport>prompt</ErrorReport> |
|
22 | 22 | <WarningLevel>4</WarningLevel> |
|
23 | 23 | <ConsolePause>false</ConsolePause> |
|
24 | 24 | <RunCodeAnalysis>true</RunCodeAnalysis> |
|
25 | 25 | </PropertyGroup> |
|
26 | 26 | <PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Release|AnyCPU' "> |
|
27 | 27 | <DebugType>full</DebugType> |
|
28 | 28 | <Optimize>true</Optimize> |
|
29 | 29 | <OutputPath>bin\Release</OutputPath> |
|
30 | 30 | <ErrorReport>prompt</ErrorReport> |
|
31 | 31 | <WarningLevel>4</WarningLevel> |
|
32 | 32 | <ConsolePause>false</ConsolePause> |
|
33 | 33 | </PropertyGroup> |
|
34 | 34 | <PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Debug 4.5|AnyCPU' "> |
|
35 | 35 | <DebugSymbols>true</DebugSymbols> |
|
36 | 36 | <DebugType>full</DebugType> |
|
37 | 37 | <Optimize>false</Optimize> |
|
38 | 38 | <OutputPath>bin\Debug</OutputPath> |
|
39 | 39 | <DefineConstants>TRACE;DEBUG;NET_4_5</DefineConstants> |
|
40 | 40 | <ErrorReport>prompt</ErrorReport> |
|
41 | 41 | <WarningLevel>4</WarningLevel> |
|
42 | 42 | <RunCodeAnalysis>true</RunCodeAnalysis> |
|
43 | 43 | <ConsolePause>false</ConsolePause> |
|
44 | 44 | </PropertyGroup> |
|
45 | 45 | <PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Release 4.5|AnyCPU' "> |
|
46 | 46 | <Optimize>true</Optimize> |
|
47 | 47 | <OutputPath>bin\Release</OutputPath> |
|
48 | 48 | <ErrorReport>prompt</ErrorReport> |
|
49 | 49 | <WarningLevel>4</WarningLevel> |
|
50 | 50 | <ConsolePause>false</ConsolePause> |
|
51 | 51 | <DefineConstants>NET_4_5</DefineConstants> |
|
52 | 52 | </PropertyGroup> |
|
53 | 53 | <PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'DebugMono|AnyCPU' "> |
|
54 | 54 | <DebugSymbols>true</DebugSymbols> |
|
55 | 55 | <DebugType>full</DebugType> |
|
56 | 56 | <Optimize>false</Optimize> |
|
57 | 57 | <OutputPath>bin\Debug</OutputPath> |
|
58 | 58 | <DefineConstants>TRACE;DEBUG;NET_4_5;MONO</DefineConstants> |
|
59 | 59 | <ErrorReport>prompt</ErrorReport> |
|
60 | 60 | <WarningLevel>4</WarningLevel> |
|
61 | 61 | <RunCodeAnalysis>true</RunCodeAnalysis> |
|
62 | 62 | <ConsolePause>false</ConsolePause> |
|
63 | 63 | </PropertyGroup> |
|
64 | 64 | <PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'ReleaseMono|AnyCPU' "> |
|
65 | 65 | <Optimize>true</Optimize> |
|
66 | 66 | <OutputPath>bin\Release</OutputPath> |
|
67 | 67 | <DefineConstants>NET_4_5;MONO;</DefineConstants> |
|
68 | 68 | <ErrorReport>prompt</ErrorReport> |
|
69 | 69 | <WarningLevel>4</WarningLevel> |
|
70 | 70 | <ConsolePause>false</ConsolePause> |
|
71 | 71 | </PropertyGroup> |
|
72 | 72 | <ItemGroup> |
|
73 | 73 | <Reference Include="System" /> |
|
74 | 74 | <Reference Include="System.Xml" /> |
|
75 | 75 | <Reference Include="mscorlib" /> |
|
76 | 76 | </ItemGroup> |
|
77 | 77 | <ItemGroup> |
|
78 | 78 | <Compile Include="CustomEqualityComparer.cs" /> |
|
79 | 79 | <Compile Include="Diagnostics\ConsoleTraceListener.cs" /> |
|
80 | 80 | <Compile Include="Diagnostics\EventText.cs" /> |
|
81 | 81 | <Compile Include="Diagnostics\LogChannel.cs" /> |
|
82 | 82 | <Compile Include="Diagnostics\LogicalOperation.cs" /> |
|
83 | 83 | <Compile Include="Diagnostics\TextFileListener.cs" /> |
|
84 | 84 | <Compile Include="Diagnostics\TraceLog.cs" /> |
|
85 | 85 | <Compile Include="Diagnostics\TraceEvent.cs" /> |
|
86 | 86 | <Compile Include="Diagnostics\TraceEventType.cs" /> |
|
87 | 87 | <Compile Include="ICancellable.cs" /> |
|
88 | 88 | <Compile Include="IProgressHandler.cs" /> |
|
89 | 89 | <Compile Include="IProgressNotifier.cs" /> |
|
90 | 90 | <Compile Include="IPromiseT.cs" /> |
|
91 | 91 | <Compile Include="IPromise.cs" /> |
|
92 | 92 | <Compile Include="IServiceLocator.cs" /> |
|
93 | 93 | <Compile Include="ITaskController.cs" /> |
|
94 | 94 | <Compile Include="Parallels\DispatchPool.cs" /> |
|
95 | 95 | <Compile Include="Parallels\ArrayTraits.cs" /> |
|
96 | 96 | <Compile Include="Parallels\MTQueue.cs" /> |
|
97 | 97 | <Compile Include="Parallels\WorkerPool.cs" /> |
|
98 | 98 | <Compile Include="ProgressInitEventArgs.cs" /> |
|
99 | 99 | <Compile Include="Properties\AssemblyInfo.cs" /> |
|
100 | 100 | <Compile Include="Parallels\AsyncPool.cs" /> |
|
101 | 101 | <Compile Include="Safe.cs" /> |
|
102 | 102 | <Compile Include="ValueEventArgs.cs" /> |
|
103 | 103 | <Compile Include="PromiseExtensions.cs" /> |
|
104 | 104 | <Compile Include="SyncContextPromise.cs" /> |
|
105 | 105 | <Compile Include="Diagnostics\OperationContext.cs" /> |
|
106 | 106 | <Compile Include="Diagnostics\TraceContext.cs" /> |
|
107 | 107 | <Compile Include="Diagnostics\LogEventArgs.cs" /> |
|
108 | 108 | <Compile Include="Diagnostics\LogEventArgsT.cs" /> |
|
109 | 109 | <Compile Include="Diagnostics\Extensions.cs" /> |
|
110 | 110 | <Compile Include="PromiseEventType.cs" /> |
|
111 | 111 | <Compile Include="Parallels\AsyncQueue.cs" /> |
|
112 | 112 | <Compile Include="PromiseT.cs" /> |
|
113 | 113 | <Compile Include="IDeferred.cs" /> |
|
114 | 114 | <Compile Include="IDeferredT.cs" /> |
|
115 | 115 | <Compile Include="Promise.cs" /> |
|
116 | 116 | <Compile Include="PromiseTransientException.cs" /> |
|
117 | 117 | <Compile Include="Parallels\Signal.cs" /> |
|
118 | 118 | <Compile Include="Parallels\SharedLock.cs" /> |
|
119 | 119 | <Compile Include="Diagnostics\ILogWriter.cs" /> |
|
120 | 120 | <Compile Include="Diagnostics\ListenerBase.cs" /> |
|
121 | 121 | <Compile Include="Parallels\BlockingQueue.cs" /> |
|
122 | 122 | <Compile Include="AbstractEvent.cs" /> |
|
123 | 123 | <Compile Include="AbstractPromise.cs" /> |
|
124 | 124 | <Compile Include="AbstractPromiseT.cs" /> |
|
125 | 125 | <Compile Include="FuncTask.cs" /> |
|
126 | 126 | <Compile Include="FuncTaskBase.cs" /> |
|
127 | 127 | <Compile Include="FuncTaskT.cs" /> |
|
128 | 128 | <Compile Include="ActionChainTaskBase.cs" /> |
|
129 | 129 | <Compile Include="ActionChainTask.cs" /> |
|
130 | 130 | <Compile Include="ActionChainTaskT.cs" /> |
|
131 | 131 | <Compile Include="FuncChainTaskBase.cs" /> |
|
132 | 132 | <Compile Include="FuncChainTask.cs" /> |
|
133 | 133 | <Compile Include="FuncChainTaskT.cs" /> |
|
134 | 134 | <Compile Include="ActionTaskBase.cs" /> |
|
135 | 135 | <Compile Include="ActionTask.cs" /> |
|
136 | 136 | <Compile Include="ActionTaskT.cs" /> |
|
137 | 137 | <Compile Include="ICancellationToken.cs" /> |
|
138 | 138 | <Compile Include="SuccessPromise.cs" /> |
|
139 | 139 | <Compile Include="SuccessPromiseT.cs" /> |
|
140 | 140 | <Compile Include="PromiseAwaiterT.cs" /> |
|
141 | 141 | <Compile Include="PromiseAwaiter.cs" /> |
|
142 | 142 | <Compile Include="Components\ComponentContainer.cs" /> |
|
143 | 143 | <Compile Include="Components\Disposable.cs" /> |
|
144 | 144 | <Compile Include="Components\DisposablePool.cs" /> |
|
145 | 145 | <Compile Include="Components\ObjectPool.cs" /> |
|
146 | 146 | <Compile Include="Components\ServiceLocator.cs" /> |
|
147 | 147 | <Compile Include="Components\IInitializable.cs" /> |
|
148 | 148 | <Compile Include="TaskController.cs" /> |
|
149 | 149 | <Compile Include="Components\App.cs" /> |
|
150 | 150 | <Compile Include="Components\IRunnable.cs" /> |
|
151 | 151 | <Compile Include="Components\ExecutionState.cs" /> |
|
152 | 152 | <Compile Include="Components\RunnableComponent.cs" /> |
|
153 | 153 | <Compile Include="Components\IFactory.cs" /> |
|
154 | 154 | <Compile Include="Automaton\DFAStateDescriptor.cs" /> |
|
155 | 155 | <Compile Include="Automaton\EnumAlphabet.cs" /> |
|
156 | 156 | <Compile Include="Automaton\IAlphabet.cs" /> |
|
157 | 157 | <Compile Include="Automaton\ParserException.cs" /> |
|
158 | 158 | <Compile Include="Automaton\Scanner.cs" /> |
|
159 | 159 | <Compile Include="Automaton\IndexedAlphabetBase.cs" /> |
|
160 | 160 | <Compile Include="Automaton\IAlphabetBuilder.cs" /> |
|
161 | 161 | <Compile Include="Automaton\RegularExpressions\AltToken.cs" /> |
|
162 | 162 | <Compile Include="Automaton\RegularExpressions\BinaryToken.cs" /> |
|
163 | 163 | <Compile Include="Automaton\RegularExpressions\CatToken.cs" /> |
|
164 | 164 | <Compile Include="Automaton\DFAConst.cs" /> |
|
165 | 165 | <Compile Include="Automaton\RegularExpressions\Grammar.cs" /> |
|
166 | 166 | <Compile Include="Automaton\RegularExpressions\StarToken.cs" /> |
|
167 | 167 | <Compile Include="Automaton\RegularExpressions\SymbolToken.cs" /> |
|
168 | 168 | <Compile Include="Automaton\RegularExpressions\EmptyToken.cs" /> |
|
169 | 169 | <Compile Include="Automaton\RegularExpressions\EndToken.cs" /> |
|
170 | 170 | <Compile Include="Automaton\RegularExpressions\Token.cs" /> |
|
171 | 171 | <Compile Include="Automaton\RegularExpressions\IVisitor.cs" /> |
|
172 | 172 | <Compile Include="Automaton\AutomatonTransition.cs" /> |
|
173 | <Compile Include="Automaton\RegularExpressions\RegularDFABuilder.cs" /> | |
|
174 | 173 | <Compile Include="Formats\JSON\JSONElementContext.cs" /> |
|
175 | 174 | <Compile Include="Formats\JSON\JSONElementType.cs" /> |
|
176 | 175 | <Compile Include="Formats\JSON\JSONGrammar.cs" /> |
|
177 | 176 | <Compile Include="Formats\JSON\JSONParser.cs" /> |
|
178 | 177 | <Compile Include="Formats\JSON\JSONScanner.cs" /> |
|
179 | 178 | <Compile Include="Formats\JSON\JsonTokenType.cs" /> |
|
180 | 179 | <Compile Include="Formats\JSON\JSONWriter.cs" /> |
|
181 | 180 | <Compile Include="Formats\JSON\JSONXmlReader.cs" /> |
|
182 | 181 | <Compile Include="Formats\JSON\JSONXmlReaderOptions.cs" /> |
|
183 | 182 | <Compile Include="Formats\JSON\StringTranslator.cs" /> |
|
184 | 183 | <Compile Include="Automaton\MapAlphabet.cs" /> |
|
185 | 184 | <Compile Include="Automaton\DummyAlphabet.cs" /> |
|
186 | <Compile Include="Automaton\RegularExpressions\RegularDFADefinition.cs" /> | |
|
187 | 185 | <Compile Include="Formats\CharAlphabet.cs" /> |
|
188 | 186 | <Compile Include="Formats\ByteAlphabet.cs" /> |
|
189 | <Compile Include="Formats\RegularCharDFADefinition.cs" /> | |
|
190 | 187 | <Compile Include="Automaton\IDFATable.cs" /> |
|
191 | 188 | <Compile Include="Automaton\IDFATableBuilder.cs" /> |
|
192 | 189 | <Compile Include="Automaton\DFATable.cs" /> |
|
190 | <Compile Include="Automaton\RegularExpressions\RegularDFA.cs" /> | |
|
191 | <Compile Include="Automaton\RegularExpressions\RegularExpressionVisitor.cs" /> | |
|
192 | <Compile Include="Automaton\RegularExpressions\ITaggedDFABuilder.cs" /> | |
|
193 | 193 | <Compile Include="Automaton\RegularExpressions\DFAStateDescriptorT.cs" /> |
|
194 | 194 | </ItemGroup> |
|
195 | 195 | <Import Project="$(MSBuildBinPath)\Microsoft.CSharp.targets" /> |
|
196 | 196 | <ItemGroup /> |
|
197 | 197 | <ProjectExtensions> |
|
198 | 198 | <MonoDevelop> |
|
199 | 199 | <Properties> |
|
200 | 200 | <Policies> |
|
201 | 201 | <CSharpFormattingPolicy IndentSwitchBody="True" NamespaceBraceStyle="EndOfLine" ClassBraceStyle="EndOfLine" InterfaceBraceStyle="EndOfLine" StructBraceStyle="EndOfLine" EnumBraceStyle="EndOfLine" MethodBraceStyle="EndOfLine" ConstructorBraceStyle="EndOfLine" DestructorBraceStyle="EndOfLine" BeforeMethodDeclarationParentheses="False" BeforeMethodCallParentheses="False" BeforeConstructorDeclarationParentheses="False" NewLineBeforeConstructorInitializerColon="NewLine" NewLineAfterConstructorInitializerColon="SameLine" BeforeIndexerDeclarationBracket="False" BeforeDelegateDeclarationParentheses="False" NewParentheses="False" SpacesBeforeBrackets="False" inheritsSet="Mono" inheritsScope="text/x-csharp" scope="text/x-csharp" /> |
|
202 | 202 | <TextStylePolicy FileWidth="120" EolMarker="Unix" inheritsSet="VisualStudio" inheritsScope="text/plain" scope="text/x-csharp" /> |
|
203 | 203 | <DotNetNamingPolicy DirectoryNamespaceAssociation="PrefixedHierarchical" ResourceNamePolicy="MSBuild" /> |
|
204 | 204 | <TextStylePolicy FileWidth="120" TabsToSpaces="False" inheritsSet="VisualStudio" inheritsScope="text/plain" scope="application/xml" /> |
|
205 | 205 | <XmlFormattingPolicy inheritsSet="Mono" inheritsScope="application/xml" scope="application/xml" /> |
|
206 | 206 | <TextStylePolicy FileWidth="120" TabsToSpaces="False" inheritsSet="VisualStudio" inheritsScope="text/plain" scope="text/plain" /> |
|
207 | 207 | <NameConventionPolicy> |
|
208 | 208 | <Rules> |
|
209 | 209 | <NamingRule Name="Namespaces" AffectedEntity="Namespace" VisibilityMask="VisibilityMask" NamingStyle="PascalCase" IncludeInstanceMembers="True" IncludeStaticEntities="True" /> |
|
210 | 210 | <NamingRule Name="Types" AffectedEntity="Class, Struct, Enum, Delegate" VisibilityMask="VisibilityMask" NamingStyle="PascalCase" IncludeInstanceMembers="True" IncludeStaticEntities="True" /> |
|
211 | 211 | <NamingRule Name="Interfaces" AffectedEntity="Interface" VisibilityMask="VisibilityMask" NamingStyle="PascalCase" IncludeInstanceMembers="True" IncludeStaticEntities="True"> |
|
212 | 212 | <RequiredPrefixes> |
|
213 | 213 | <String>I</String> |
|
214 | 214 | </RequiredPrefixes> |
|
215 | 215 | </NamingRule> |
|
216 | 216 | <NamingRule Name="Attributes" AffectedEntity="CustomAttributes" VisibilityMask="VisibilityMask" NamingStyle="PascalCase" IncludeInstanceMembers="True" IncludeStaticEntities="True"> |
|
217 | 217 | <RequiredSuffixes> |
|
218 | 218 | <String>Attribute</String> |
|
219 | 219 | </RequiredSuffixes> |
|
220 | 220 | </NamingRule> |
|
221 | 221 | <NamingRule Name="Event Arguments" AffectedEntity="CustomEventArgs" VisibilityMask="VisibilityMask" NamingStyle="PascalCase" IncludeInstanceMembers="True" IncludeStaticEntities="True"> |
|
222 | 222 | <RequiredSuffixes> |
|
223 | 223 | <String>EventArgs</String> |
|
224 | 224 | </RequiredSuffixes> |
|
225 | 225 | </NamingRule> |
|
226 | 226 | <NamingRule Name="Exceptions" AffectedEntity="CustomExceptions" VisibilityMask="VisibilityMask" NamingStyle="PascalCase" IncludeInstanceMembers="True" IncludeStaticEntities="True"> |
|
227 | 227 | <RequiredSuffixes> |
|
228 | 228 | <String>Exception</String> |
|
229 | 229 | </RequiredSuffixes> |
|
230 | 230 | </NamingRule> |
|
231 | 231 | <NamingRule Name="Methods" AffectedEntity="Methods" VisibilityMask="VisibilityMask" NamingStyle="PascalCase" IncludeInstanceMembers="True" IncludeStaticEntities="True" /> |
|
232 | 232 | <NamingRule Name="Static Readonly Fields" AffectedEntity="ReadonlyField" VisibilityMask="Internal, Protected, Public" NamingStyle="CamelCase" IncludeInstanceMembers="False" IncludeStaticEntities="True" /> |
|
233 | 233 | <NamingRule Name="Fields (Non Private)" AffectedEntity="Field" VisibilityMask="Internal, Public" NamingStyle="CamelCase" IncludeInstanceMembers="True" IncludeStaticEntities="True" /> |
|
234 | 234 | <NamingRule Name="ReadOnly Fields (Non Private)" AffectedEntity="ReadonlyField" VisibilityMask="Internal, Public" NamingStyle="CamelCase" IncludeInstanceMembers="True" IncludeStaticEntities="False" /> |
|
235 | 235 | <NamingRule Name="Fields (Private)" AffectedEntity="Field, ReadonlyField" VisibilityMask="Private, Protected" NamingStyle="CamelCase" IncludeInstanceMembers="True" IncludeStaticEntities="False"> |
|
236 | 236 | <RequiredPrefixes> |
|
237 | 237 | <String>m_</String> |
|
238 | 238 | </RequiredPrefixes> |
|
239 | 239 | </NamingRule> |
|
240 | 240 | <NamingRule Name="Static Fields (Private)" AffectedEntity="Field" VisibilityMask="Private" NamingStyle="CamelCase" IncludeInstanceMembers="False" IncludeStaticEntities="True"> |
|
241 | 241 | <RequiredPrefixes> |
|
242 | 242 | <String>_</String> |
|
243 | 243 | </RequiredPrefixes> |
|
244 | 244 | </NamingRule> |
|
245 | 245 | <NamingRule Name="ReadOnly Fields (Private)" AffectedEntity="ReadonlyField" VisibilityMask="Private, Protected" NamingStyle="CamelCase" IncludeInstanceMembers="True" IncludeStaticEntities="False"> |
|
246 | 246 | <RequiredPrefixes> |
|
247 | 247 | <String>m_</String> |
|
248 | 248 | </RequiredPrefixes> |
|
249 | 249 | </NamingRule> |
|
250 | 250 | <NamingRule Name="Constant Fields" AffectedEntity="ConstantField" VisibilityMask="VisibilityMask" NamingStyle="AllUpper" IncludeInstanceMembers="True" IncludeStaticEntities="True" /> |
|
251 | 251 | <NamingRule Name="Properties" AffectedEntity="Property" VisibilityMask="VisibilityMask" NamingStyle="PascalCase" IncludeInstanceMembers="True" IncludeStaticEntities="True" /> |
|
252 | 252 | <NamingRule Name="Events" AffectedEntity="Event" VisibilityMask="VisibilityMask" NamingStyle="PascalCase" IncludeInstanceMembers="True" IncludeStaticEntities="True" /> |
|
253 | 253 | <NamingRule Name="Enum Members" AffectedEntity="EnumMember" VisibilityMask="VisibilityMask" NamingStyle="PascalCase" IncludeInstanceMembers="True" IncludeStaticEntities="True" /> |
|
254 | 254 | <NamingRule Name="Parameters" AffectedEntity="Parameter, LocalVariable" VisibilityMask="VisibilityMask" NamingStyle="CamelCase" IncludeInstanceMembers="True" IncludeStaticEntities="True" /> |
|
255 | 255 | <NamingRule Name="Type Parameters" AffectedEntity="TypeParameter" VisibilityMask="VisibilityMask" NamingStyle="PascalCase" IncludeInstanceMembers="True" IncludeStaticEntities="True"> |
|
256 | 256 | <RequiredPrefixes> |
|
257 | 257 | <String>T</String> |
|
258 | 258 | </RequiredPrefixes> |
|
259 | 259 | </NamingRule> |
|
260 | 260 | </Rules> |
|
261 | 261 | </NameConventionPolicy> |
|
262 | 262 | </Policies> |
|
263 | 263 | </Properties> |
|
264 | 264 | </MonoDevelop> |
|
265 | 265 | </ProjectExtensions> |
|
266 | 266 | <ItemGroup> |
|
267 | 267 | <Folder Include="Components\" /> |
|
268 | 268 | <Folder Include="Automaton\RegularExpressions\" /> |
|
269 | 269 | <Folder Include="Formats\" /> |
|
270 | 270 | <Folder Include="Formats\JSON\" /> |
|
271 | 271 | </ItemGroup> |
|
272 | 272 | </Project> No newline at end of file |
|
1 | NO CONTENT: file was removed |
|
1 | NO CONTENT: file was removed |
|
1 | NO CONTENT: file was removed |
General Comments 0
You need to be logged in to leave comments.
Login now