Elasticsearch
Connect Elasticsearch to your preprocessing pipeline, and batch process all your documents using unstructured-ingest
to store structured outputs locally on your filesystem.
First you’ll need to install the Elasticsearch dependencies as shown here.
pip install "unstructured[elasticsearch]"
Run Locally
#!/usr/bin/env bash
unstructured-ingest \
elasticsearch \
--metadata-exclude filename,file_directory,metadata.data_source.date_processed \
--url http://localhost:9200 \
--index-name movies \
--fields 'ethnicity, director, plot' \
--output-dir elasticsearch-ingest-output \
--num-processes 2
from unstructured.ingest.connector.elasticsearch import (
ElasticsearchAccessConfig,
SimpleElasticsearchConfig,
)
from unstructured.ingest.interfaces import PartitionConfig, ProcessorConfig, ReadConfig
from unstructured.ingest.runner import ElasticSearchRunner
if __name__ == "__main__":
runner = ElasticSearchRunner(
processor_config=ProcessorConfig(
verbose=True,
output_dir="elasticsearch-ingest-output",
num_processes=2,
),
read_config=ReadConfig(),
partition_config=PartitionConfig(
metadata_exclude=["filename", "file_directory", "metadata.data_source.date_processed"],
),
connector_config=SimpleElasticsearchConfig(
access_config=ElasticsearchAccessConfig(hosts=["http://localhost:9200"]),
index_name="movies",
fields=["ethnicity", "director", "plot"],
),
)
runner.run()
Run via the API
You can also use upstream connectors with the unstructured
API. For this you’ll need to use the --partition-by-api
flag and pass in your API key with --api-key
.
#!/usr/bin/env bash
unstructured-ingest \
elasticsearch \
--metadata-exclude filename,file_directory,metadata.data_source.date_processed \
--url http://localhost:9200 \
--index-name movies \
--fields 'ethnicity, director, plot' \
--output-dir elasticsearch-ingest-output \
--num-processes 2 \
--partition-by-api \
--api-key "<UNSTRUCTURED-API-KEY>"
import os
from unstructured.ingest.connector.elasticsearch import (
ElasticsearchAccessConfig,
SimpleElasticsearchConfig,
)
from unstructured.ingest.interfaces import PartitionConfig, ProcessorConfig, ReadConfig
from unstructured.ingest.runner import ElasticSearchRunner
if __name__ == "__main__":
runner = ElasticSearchRunner(
processor_config=ProcessorConfig(
verbose=True,
output_dir="elasticsearch-ingest-output",
num_processes=2,
),
read_config=ReadConfig(),
partition_config=PartitionConfig(
metadata_exclude=["filename", "file_directory", "metadata.data_source.date_processed"],
partition_by_api=True,
api_key=os.getenv("UNSTRUCTURED_API_KEY"),
),
connector_config=SimpleElasticsearchConfig(
access_config=ElasticsearchAccessConfig(hosts=["http://localhost:9200"]),
index_name="movies",
fields=["ethnicity", "director", "plot"],
),
)
runner.run()
Additionally, you will need to pass the --partition-endpoint
if you’re running the API locally. You can find more information about the unstructured
API here.
For a full list of the options the CLI accepts check unstructured-ingest elasticsearch --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.