def configure_schema(schema, version):
# Generate relevant type mappings for entity properties so that
# we can do correct searches on each.
schema_mapping = {}
numeric_mapping = {registry.date.group: NUMERIC}
for prop in schema.properties.values():
config = dict(TYPE_MAPPINGS.get(prop.type, KEYWORD))
config["copy_to"] = ["text"]
schema_mapping[prop.name] = config
if prop.type in NUMERIC_TYPES:
numeric_mapping[prop.name] = NUMERIC
mapping = {
"date_detection": False,
"dynamic": False,
"_source": {"excludes": ["text", "fingerprints"]},
"properties": {
"caption": KEYWORD,
"schema": KEYWORD,
"schemata": KEYWORD,
registry.entity.group: KEYWORD,
registry.language.group: KEYWORD,
registry.country.group: KEYWORD,
registry.checksum.group: KEYWORD,
registry.ip.group: KEYWORD,
registry.url.group: KEYWORD,
registry.iban.group: KEYWORD,
registry.email.group: KEYWORD,
registry.phone.group: KEYWORD,
registry.mimetype.group: KEYWORD,
registry.identifier.group: KEYWORD,
registry.date.group: PARTIAL_DATE,
registry.address.group: KEYWORD,
registry.name.group: KEYWORD,
"fingerprints": {
"type": "keyword",
"normalizer": "latin_index",
"copy_to": "text",
"fields": {"text": LATIN_TEXT},
},
"text": {
"type": "text",
"analyzer": "latin_index",
"search_analyzer": "latin_query",
"search_quote_analyzer": "latin_index",
"term_vector": "with_positions_offsets",
},
"properties": {"type": "object", "properties": schema_mapping},
"numeric": {"type": "object", "properties": numeric_mapping},
"role_id": KEYWORD,
"collection_id": KEYWORD,
"origin": KEYWORD,
"created_at": {"type": "date"},
"updated_at": {"type": "date"},
},
}
index = schema_index(model.get(schema), version)
settings = index_settings(shards=get_shard_weight(schema))
return configure_index(index, mapping, settings)
def configure_schema(schema, version):
# Generate relevant type mappings for entity properties so that
# we can do correct searches on each.
schema_mapping = {}
for prop in schema.properties.values():
config = dict(TYPE_MAPPINGS.get(prop.type, KEYWORD))
config['copy_to'] = ['text']
schema_mapping[prop.name] = config
mapping = {
"date_detection": False,
"dynamic": False,
"_source": {
"excludes": ["text", "fingerprints"]
},
"properties": {
"name": {
"type": "text",
"analyzer": "icu_latin",
"fields": {"kw": KEYWORD},
"boost": 3.0,
"copy_to": "text"
},
"schema": KEYWORD,
"schemata": KEYWORD,
"foreign_id": KEYWORD,
"document_id": KEYWORD,
"collection_id": KEYWORD,
"uploader_id": KEYWORD,
"entities": KEYWORD,
"languages": KEYWORD,
"countries": KEYWORD,
"checksums": KEYWORD,
"keywords": KEYWORD,
"ips": KEYWORD,
"urls": KEYWORD,
"ibans": KEYWORD,
"emails": KEYWORD,
"phones": KEYWORD,
"mimetypes": KEYWORD,
"identifiers": KEYWORD,
"dates": PARTIAL_DATE,
"addresses": {
"type": "keyword",
"fields": {"text": LATIN_TEXT}
},
"names": {
"type": "keyword",
"fields": {"text": LATIN_TEXT},
"copy_to": "text"
},
"fingerprints": {
"type": "keyword",
"normalizer": "icu_latin",
"copy_to": "text",
"fields": {"text": LATIN_TEXT}
},
"text": {
"type": "text",
"analyzer": "icu_latin",
"term_vector": "with_positions_offsets",
"store": True
},
"properties": {
"type": "object",
"properties": schema_mapping
},
"updated_at": {"type": "date"},
}
}
index = schema_index(model.get(schema), version)
settings = index_settings(shards=get_shard_weight(schema))
return configure_index(index, mapping, settings)
def configure_schema(schema, version):
# Generate relevant type mappings for entity properties so that
# we can do correct searches on each.
schema_mapping = {}
for prop in schema.properties.values():
config = dict(TYPE_MAPPINGS.get(prop.type, KEYWORD))
config['copy_to'] = ['text']
schema_mapping[prop.name] = config
mapping = {
"date_detection": False,
"dynamic": False,
"_source": {
"excludes": ["text", "fingerprints"]
},
"properties": {
"name": {
"type": "text",
"analyzer": "icu_latin",
"fields": {"kw": KEYWORD},
"boost": 3.0,
"copy_to": "text"
},
"schema": KEYWORD,
"schemata": KEYWORD,
"bulk": {"type": "boolean"},
"status": KEYWORD,
"error_message": {
"type": "text",
"copy_to": "text",
"index": False
},
"foreign_id": KEYWORD,
"document_id": KEYWORD,
"collection_id": KEYWORD,
"uploader_id": KEYWORD,
"fingerprints": {
"type": "keyword",
"normalizer": "icu_latin",
"copy_to": "text",
"fields": {"text": LATIN_TEXT}
},
"entities": KEYWORD,
"languages": KEYWORD,
"countries": KEYWORD,
"checksums": KEYWORD,
"keywords": KEYWORD,
"ips": KEYWORD,
"urls": KEYWORD,
"ibans": KEYWORD,
"emails": KEYWORD,
"phones": KEYWORD,
"mimetypes": KEYWORD,
"identifiers": KEYWORD,
"addresses": {
"type": "keyword",
"fields": {"text": LATIN_TEXT}
},
"dates": PARTIAL_DATE,
"names": {
"type": "keyword",
"fields": {"text": LATIN_TEXT},
"copy_to": "text"
},
"created_at": {"type": "date"},
"updated_at": {"type": "date"},
"text": {
"type": "text",
"analyzer": "icu_latin",
"term_vector": "with_positions_offsets",
"store": True
},
"properties": {
"type": "object",
"properties": schema_mapping
}
}
}
index = schema_index(model.get(schema), version)
return configure_index(
index, mapping, index_settings(shards=get_shard_weight(schema))
)