be52af70
 tangwang
first commit
|
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
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
|
self.config = config
def generate_mapping(self) -> Dict[str, Any]:
"""
Generate complete Elasticsearch index configuration including
settings and mappings.
Returns:
Dictionary containing index configuration
"""
return {
"settings": self._generate_settings(),
"mappings": self._generate_mappings()
}
def _generate_settings(self) -> Dict[str, Any]:
"""Generate index settings."""
settings = {
"number_of_shards": self.config.es_settings.get("number_of_shards", 1),
"number_of_replicas": self.config.es_settings.get("number_of_replicas", 0),
"refresh_interval": self.config.es_settings.get("refresh_interval", "30s"),
}
# Add similarity configuration (modified BM25)
similarity_config = get_default_similarity()
settings.update(similarity_config)
# Add analyzer configuration
analyzer_config = get_default_analyzers()
settings.update(analyzer_config)
# Merge any custom settings from config
for key, value in self.config.es_settings.items():
if key not in ["number_of_shards", "number_of_replicas", "refresh_interval"]:
settings[key] = value
return settings
def _generate_mappings(self) -> Dict[str, Any]:
"""Generate field mappings."""
properties = {}
for field in self.config.fields:
field_mapping = get_es_mapping_for_field(field)
properties[field.name] = field_mapping
return {
"properties": properties
}
def get_default_domain_fields(self) -> list:
"""
Get list of fields in the 'default' domain.
Returns:
List of field names
"""
for index in self.config.indexes:
if index.name == "default":
return index.fields
return []
def get_text_embedding_field(self) -> str:
"""
Get the primary text embedding field name.
Returns:
Field name or empty string if not found
"""
# Look for name_embedding or first text_embedding field
for field in self.config.fields:
if field.name == "name_embedding":
return field.name
# Otherwise return first text embedding field
for field in self.config.fields:
if "embedding" in field.name and "image" not in field.name:
return field.name
return ""
def get_image_embedding_field(self) -> str:
"""
Get the primary image embedding field name.
Returns:
Field name or empty string if not found
"""
for field in self.config.fields:
if "image" in field.name and "embedding" in field.name:
return field.name
return ""
def get_field_by_name(self, field_name: str) -> FieldConfig:
"""
Get field configuration by name.
Args:
field_name: Field name
Returns:
FieldConfig object or None if not found
"""
for field in self.config.fields:
if field.name == field_name:
return field
return None
def get_match_fields_for_domain(self, domain_name: str = "default") -> list:
"""
Get list of text fields for matching in a domain.
Args:
domain_name: Name of the query domain
Returns:
List of field names with optional boost (e.g., ["name^2.0", "category^1.5"])
"""
for index in self.config.indexes:
if index.name == domain_name:
result = []
for field_name in index.fields:
field = self.get_field_by_name(field_name)
if field and field.boost != 1.0:
result.append(f"{field_name}^{field.boost}")
else:
result.append(field_name)
return result
return []
def create_index_if_not_exists(es_client, index_name: str, mapping: Dict[str, Any]) -> bool:
"""
Create Elasticsearch index if it doesn't exist.
Args:
es_client: Elasticsearch client instance
index_name: Name of the index to create
mapping: Index mapping configuration
Returns:
True if index was created, False if it already exists
"""
if es_client.indices.exists(index=index_name):
print(f"Index '{index_name}' already exists")
return False
es_client.indices.create(index=index_name, body=mapping)
print(f"Index '{index_name}' created successfully")
return True
def delete_index_if_exists(es_client, index_name: str) -> bool:
"""
Delete Elasticsearch index if it exists.
Args:
es_client: Elasticsearch client instance
index_name: Name of the index to delete
Returns:
True if index was deleted, False if it didn't exist
"""
if not es_client.indices.exists(index=index_name):
print(f"Index '{index_name}' does not exist")
return False
es_client.indices.delete(index=index_name)
print(f"Index '{index_name}' deleted successfully")
return True
def update_mapping(es_client, index_name: str, new_fields: Dict[str, Any]) -> bool:
"""
Update mapping for existing index (only adding new fields).
Args:
es_client: Elasticsearch client instance
index_name: Name of the index
new_fields: New field mappings to add
Returns:
True if successful
"""
if not es_client.indices.exists(index=index_name):
print(f"Index '{index_name}' does not exist")
return False
mapping = {"properties": new_fields}
es_client.indices.put_mapping(index=index_name, body=mapping)
print(f"Mapping updated for index '{index_name}'")
return True
|