pip install "unstructured[azure-cognitive-search]"
Run Locally
The upstream connector can be any of the ones supported, but for convenience here, showing a sample command using the upstream local connector.#!/usr/bin/env bash
EMBEDDING_PROVIDER=${EMBEDDING_PROVIDER:-"langchain-huggingface"}
unstructured-ingest \
local \
--input-path example-docs/book-war-and-peace-1225p.txt \
--output-dir local-output-to-azure-cog-search \
--strategy fast \
--chunk-elements \
--embedding-provider "$EMBEDDING_PROVIDER" \
--num-processes 2 \
--verbose \
azure-cognitive-search \
--key "$AZURE_SEARCH_API_KEY" \
--endpoint "$AZURE_SEARCH_ENDPOINT" \
--index utic-test-ingest-fixtures-output
import os
from unstructured.ingest.connector.azure_cognitive_search import (
AzureCognitiveSearchAccessConfig,
AzureCognitiveSearchWriteConfig,
SimpleAzureCognitiveSearchStorageConfig,
)
from unstructured.ingest.connector.local import SimpleLocalConfig
from unstructured.ingest.interfaces import (
ChunkingConfig,
EmbeddingConfig,
PartitionConfig,
ProcessorConfig,
ReadConfig,
)
from unstructured.ingest.runner import LocalRunner
from unstructured.ingest.runner.writers.azure_cognitive_search import (
AzureCognitiveSearchWriter,
)
from unstructured.ingest.runner.writers.base_writer import Writer
def get_writer() -> Writer:
return AzureCognitiveSearchWriter(
connector_config=SimpleAzureCognitiveSearchStorageConfig(
access_config=AzureCognitiveSearchAccessConfig(key=os.getenv("AZURE_SEARCH_API_KEY")),
endpoint=os.getenv("$AZURE_SEARCH_ENDPOINT"),
),
write_config=AzureCognitiveSearchWriteConfig(index="utic-test-ingest-fixtures-output"),
)
if __name__ == "__main__":
writer = get_writer()
runner = LocalRunner(
processor_config=ProcessorConfig(
verbose=True,
output_dir="local-output-to-azure-cog-search",
num_processes=2,
),
connector_config=SimpleLocalConfig(
input_path="example-docs/book-war-and-peace-1225p.txt",
),
read_config=ReadConfig(),
partition_config=PartitionConfig(),
chunking_config=ChunkingConfig(chunk_elements=True),
embedding_config=EmbeddingConfig(
provider="langchain-huggingface",
),
writer=writer,
writer_kwargs={},
)
runner.run()
unstructured-ingest <upstream connector> azure-cognitive-search --help.
NOTE: Keep in mind that you will need to have all the appropriate extras and dependencies for the file types of the documents contained in your data storage platform if you’re running this locally. You can find more information about this in the installation guide.
Sample Index Schema
To make sure the schema of the index matches the data being written to it, a sample schema json can be used:Object description
1{
2 "@odata.context": "https://utic-test-ingest-fixtures.search.windows.net/$metadata#indexes/$entity",
3 "@odata.etag": "\"0x8DBB93E09C8F4BD\"",
4 "name": "your-index-here",
5 "fields": [
6 {
7 "name": "id",
8 "type": "Edm.String",
9 "key": true
10 },
11 {
12 "name": "element_id",
13 "type": "Edm.String"
14 },
15 {
16 "name": "text",
17 "type": "Edm.String"
18 },
19 {
20 "name": "embeddings",
21 "type": "Collection(Edm.Single)",
22 "dimensions": 400,
23 "vectorSearchConfiguration": "embeddings-config"
24 },
25 {
26 "name": "type",
27 "type": "Edm.String"
28 },
29 {
30 "name": "metadata",
31 "type": "Edm.ComplexType",
32 "fields": [
33 {
34 "name": "category_depth",
35 "type": "Edm.Int32"
36 },
37 {
38 "name": "parent_id",
39 "type": "Edm.String"
40 },
41 {
42 "name": "attached_to_filename",
43 "type": "Edm.String"
44 },
45 {
46 "name": "filetype",
47 "type": "Edm.String"
48 },
49 {
50 "name": "last_modified",
51 "type": "Edm.DateTimeOffset"
52 },
53 {
54 "name": "file_directory",
55 "type": "Edm.String"
56 },
57 {
58 "name": "filename",
59 "type": "Edm.String"
60 },
61 {
62 "name": "data_source",
63 "type": "Edm.ComplexType",
64 "fields": [
65 {
66 "name": "url",
67 "type": "Edm.String"
68 },
69 {
70 "name": "version",
71 "type": "Edm.String"
72 },
73 {
74 "name": "date_created",
75 "type": "Edm.DateTimeOffset"
76 },
77 {
78 "name": "date_modified",
79 "type": "Edm.DateTimeOffset"
80 },
81 {
82 "name": "date_processed",
83 "type": "Edm.DateTimeOffset"
84 },
85 {
86 "name": "permissions_data",
87 "type": "Edm.String"
88 },
89 {
90 "name": "record_locator",
91 "type": "Edm.String"
92 }
93 ]
94 },
95 {
96 "name": "coordinates",
97 "type": "Edm.ComplexType",
98 "fields": [
99 {
100 "name": "system",
101 "type": "Edm.String"
102 },
103 {
104 "name": "layout_width",
105 "type": "Edm.Double"
106 },
107 {
108 "name": "layout_height",
109 "type": "Edm.Double"
110 },
111 {
112 "name": "points",
113 "type": "Edm.String"
114 }
115 ]
116 },
117 {
118 "name": "page_number",
119 "type": "Edm.String"
120 },
121 {
122 "name": "links",
123 "type": "Collection(Edm.String)"
124 },
125 {
126 "name": "url",
127 "type": "Edm.String"
128 },
129 {
130 "name": "link_urls",
131 "type": "Collection(Edm.String)"
132 },
133 {
134 "name": "link_texts",
135 "type": "Collection(Edm.String)"
136 },
137 {
138 "name": "sent_from",
139 "type": "Collection(Edm.String)"
140 },
141 {
142 "name": "sent_to",
143 "type": "Collection(Edm.String)"
144 },
145 {
146 "name": "subject",
147 "type": "Edm.String"
148 },
149 {
150 "name": "section",
151 "type": "Edm.String"
152 },
153 {
154 "name": "header_footer_type",
155 "type": "Edm.String"
156 },
157 {
158 "name": "emphasized_text_contents",
159 "type": "Collection(Edm.String)"
160 },
161 {
162 "name": "emphasized_text_tags",
163 "type": "Collection(Edm.String)"
164 },
165 {
166 "name": "text_as_html",
167 "type": "Edm.String"
168 },
169 {
170 "name": "regex_metadata",
171 "type": "Edm.String"
172 },
173 {
174 "name": "detection_class_prob",
175 "type": "Edm.Double"
176 }
177 ]
178 }
179 ],
180 "vectorSearch": {
181 "algorithmConfigurations": [
182 {
183 "name": "embeddings-config",
184 "kind": "hnsw",
185 "hnswParameters": {
186 "metric": "cosine",
187 "m": 4,
188 "efConstruction": 400,
189 "efSearch": 500
190 }
191 }
192 ]
193 }
194}