1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
|
// Copyright Yahoo. Licensed under the terms of the Apache 2.0 license. See LICENSE in the project root.
package com.yahoo.prelude.querytransform;
import java.util.*;
import com.yahoo.component.annotation.Inject;
import com.yahoo.component.chain.dependencies.After;
import com.yahoo.component.chain.dependencies.Provides;
import com.yahoo.prelude.Index;
import com.yahoo.prelude.IndexFacts;
import com.yahoo.prelude.IndexFacts.Session;
import com.yahoo.prelude.query.*;
import com.yahoo.prelude.query.WordAlternativesItem.Alternative;
import com.yahoo.search.Result;
import com.yahoo.search.Searcher;
import com.yahoo.language.Language;
import com.yahoo.language.Linguistics;
import com.yahoo.search.Query;
import com.yahoo.search.searchchain.Execution;
import com.yahoo.search.searchchain.PhaseNames;
import static com.yahoo.prelude.querytransform.StemmingSearcher.STEMMING;
/**
* Normalizes accents
*
* @author bratseth
*/
@After({ PhaseNames.UNBLENDED_RESULT, STEMMING })
@Provides(NormalizingSearcher.ACCENT_REMOVAL)
public class NormalizingSearcher extends Searcher {
public static final String ACCENT_REMOVAL = "AccentRemoval";
private final Linguistics linguistics;
@Inject
public NormalizingSearcher(Linguistics linguistics) {
this.linguistics = linguistics;
}
protected boolean handles(String command) {
return "normalize".equals(command);
}
public String getFunctionName() {
return "Normalizing accents";
}
@Override
public Result search(Query query, Execution execution) {
normalize(query, execution.context().getIndexFacts().newSession(query));
return execution.search(query);
}
protected void normalize(Query query, IndexFacts.Session indexFacts) {
String oldQuery = (query.getTrace().getLevel() >= 2) ? query.getModel().getQueryTree().getRoot().toString() : null;
normalizeBody(query, indexFacts);
if (query.getTrace().getLevel() >= 2 && ! query.getModel().getQueryTree().getRoot().toString().equals(oldQuery))
query.trace(getFunctionName(), true, 2);
}
private Query normalizeBody(Query query, IndexFacts.Session indexFacts) {
Item root = query.getModel().getQueryTree().getRoot();
Language language = query.getModel().getParsingLanguage();
if (root instanceof BlockItem) {
List<Item> rootItems = new ArrayList<>(1);
rootItems.add(root);
ListIterator<Item> i = rootItems.listIterator();
i.next();
normalizeBlocks(language, indexFacts, (BlockItem) root, i);
if ( ! rootItems.isEmpty()) // give up normalizing if the root was removed
query.getModel().getQueryTree().setRoot(rootItems.get(0));
} else if (root instanceof CompositeItem) {
query.getModel().getQueryTree().setRoot(normalizeComposite(language, indexFacts, (CompositeItem) root));
}
return query;
}
private Item normalizeComposite(Language language, IndexFacts.Session indexFacts, CompositeItem item) {
if (item instanceof PhraseItem) {
return normalizePhrase(language, indexFacts, (PhraseItem) item);
}
else {
for (ListIterator<Item> i = item.getItemIterator(); i.hasNext(); ) {
Item current = i.next();
if (current instanceof BlockItem) {
normalizeBlocks(language, indexFacts, (BlockItem) current, i);
} else if (current instanceof CompositeItem) {
Item currentProcessed = normalizeComposite(language, indexFacts, (CompositeItem) current);
i.set(currentProcessed);
}
}
return item;
}
}
private void normalizeBlocks(Language language, IndexFacts.Session indexFacts, BlockItem block, ListIterator<Item> i) {
if (block instanceof TermItem) {
if (block instanceof WordAlternativesItem) {
normalizeAlternatives(language, indexFacts, (WordAlternativesItem) block);
} else {
normalizeWord(language, indexFacts, (TermItem) block, i);
}
} else {
for (ListIterator<Item> j = ((SegmentItem) block).getItemIterator(); j.hasNext();)
normalizeWord(language, indexFacts, (TermItem) j.next(), j);
}
}
private void normalizeAlternatives(Language language, Session indexFacts, WordAlternativesItem block) {
if ( ! block.isNormalizable()) return;
Index index = indexFacts.getIndex(block.getIndexName());
if (index.isAttribute()) return;
if ( ! index.getNormalize()) return;
List<Alternative> terms = block.getAlternatives();
for (Alternative term : terms) {
String accentDropped = linguistics.getTransformer().accentDrop(term.word, language);
if ( ! term.word.equals(accentDropped) && accentDropped.length() > 0)
block.addTerm(accentDropped, term.exactness * .7d);
}
}
private Item normalizePhrase(Language language, IndexFacts.Session indexFacts, PhraseItem phrase) {
if ( ! indexFacts.getIndex(phrase.getIndexName()).getNormalize()) return phrase;
for (ListIterator<Item> i = phrase.getItemIterator(); i.hasNext();) {
IndexedItem content = (IndexedItem) i.next();
if (content instanceof TermItem) {
normalizeWord(language, indexFacts, (TermItem) content, i);
}
else {
PhraseSegmentItem segment = (PhraseSegmentItem) content;
for (ListIterator<Item> j = segment.getItemIterator(); j.hasNext();)
normalizeWord(language, indexFacts, (TermItem) j.next(), j);
}
}
return phrase;
}
private void normalizeWord(Language language, IndexFacts.Session indexFacts, TermItem term, ListIterator<Item> i) {
if ( ! (term instanceof WordItem)) return;
if ( ! term.isNormalizable()) return;
Index index = indexFacts.getIndex(term.getIndexName());
if (index.isAttribute()) return;
if ( ! index.getNormalize()) return;
WordItem word = (WordItem) term;
String accentDropped = linguistics.getTransformer().accentDrop(word.getWord(), language);
if (accentDropped.length() == 0)
i.remove();
else
word.setWord(accentDropped);
}
}
|
