David Potter 76e0d10e61
feat: add MongoDB source connector (#2393)
Adds MongoDB as a source (we already had it as a destination connector)

---------

Co-authored-by: potter-potter <david.potter@gmail.com>
2024-01-16 20:56:29 +00:00

27 lines
875 B
Python

import os
from unstructured.ingest.connector.mongodb import (
SimpleMongoDBConfig,
)
from unstructured.ingest.interfaces import PartitionConfig, ProcessorConfig, ReadConfig
from unstructured.ingest.runner import MongoDBRunner
if __name__ == "__main__":
runner = MongoDBRunner(
processor_config=ProcessorConfig(
verbose=True,
output_dir="mongodb-ingest-output",
num_processes=2,
),
read_config=ReadConfig(),
partition_config=PartitionConfig(
metadata_exclude=["filename", "file_directory", "metadata.data_source.date_processed"],
),
connector_config=SimpleMongoDBConfig(
uri=os.getenv("MONGODB_URI"),
database=os.getenv("MONGODB_DATABASE_NAME"),
collection=os.getenv("DESTINATION_MONGO_COLLECTION"),
),
)
runner.run()