[BUG]: OperationFailure: $listSearchIndexes is not allowed or the syntax is incorrect

[makkarss929]

Contact Details [Optional]

[email protected]

System Information

OperationFailure: $listSearchIndexes is not allowed or the syntax is incorrect, see the Atlas documentation for more information, full error: {'ok': 0, 'errmsg': '$listSearchIndexes is not allowed or the syntax is incorrect, see the Atlas documentation for more information', 'code': 8000, 'codeName': 'AtlasError'}

What happened?

Running below code database - mongodb + atlas

from superduperdb.backends.mongodb import Collection
from superduperdb import Document as D
from IPython.display import *

# Define the query for the search
# query = 'Code snippet how to create a `VectorIndex` with a torchvision model'
query = 'can you explain vector-indexes with `superduperdb`?'

# Execute a search using SuperDuperDB to find documents containing the specified query
result = db.execute(
    collection
        .like(D({'txt': query}), vector_index='my-index', n=5)
        .find()
)

# Display a horizontal rule to separate results
display(Markdown('---'))

# Display each document's 'txt' field and separate them with a horizontal rule
for r in result:
    display(Markdown(r['txt']))
    display(r['link'])
    display(Markdown('---'))

It's not creating atlas search index

Steps to reproduce

...

Relevant log output

No response

[blythed]

@makkarss929 did you set up the index with db.add(VectorIndex(...))? Please provide information about your system.

[makkarss929]

system : Mac m1 pro 14 inch, 16 GB RAM python : 3.11

# Import the Listener class from the superduperdb module
from superduperdb import Listener


# Create a Listener instance with the specified model, key, and selection criteria
listener = Listener(
    model=model,          # The model to be used for listening
    key='txt',            # The key field in the documents to be processed by the model
    select=collection.find()  # The selection criteria for the documents
)

# Import the VectorIndex class from the superduperdb module
from superduperdb import VectorIndex

# Add a VectorIndex to the SuperDuperDB database with the specified identifier and indexing listener
_ = db.add(
    VectorIndex(
        identifier='my-index',        # Unique identifier for the VectorIndex
        indexing_listener=listener    # Listener to be used for indexing documents
    )
)