Skip to main content

​
documents.upload

Uploads text documents to a text namespace. Moorcheh will process and embed these asynchronously.

​
Parameters

​
namespace_name
str
required
The name of the target text namespace.
​
documents
List[Dict]
required
A list of dictionaries. Each dict requires an id and text key.
Returns: Dict[str, Any] - A dictionary confirming the documents were queued. Raises: NamespaceNotFound, InvalidInputError.
Upload Documents Example
documents_to_upload = [
    {"id": "faq-1", "text": "To reset your password, go to the account settings page.", "category": "account"},
    {"id": "faq-2", "text": "Our return policy allows returns within 30 days of purchase.", "category": "shipping"}
]

status = client.documents.upload(
    namespace_name="my-faq-documents",
    documents=documents_to_upload
)
print(status)

​
vectors.upload

Uploads pre-computed vectors to a vector namespace. This is a synchronous operation.

​
Parameters

​
namespace_name
str
required
The name of the target vector namespace.
​
vectors
List[Dict]
required
A list of dictionaries. Each dict requires an id and a vector key.
Returns: Dict[str, Any] - A dictionary confirming the upload status. Raises: NamespaceNotFound, InvalidInputError.
Upload Vectors Example
vectors_to_upload = [
    {"id": "image_001.jpg", "vector": [0.12, -0.45, ...], "metadata": {"source": "product_database"}},
    {"id": "image_002.jpg", "vector": [-0.22, 0.81, ...], "metadata": {"source": "product_database"}}
]

status = client.vectors.upload(
    namespace_name="my-image-embeddings",
    vectors=vectors_to_upload
)
print(status)

​
documents.delete

Deletes specific documents from a text namespace by their IDs.

​
Parameters

​
namespace_name
str
required
The name of the target text namespace.
​
ids
List[Union[str, int]]
required
A list of document IDs to delete.
Returns: Dict[str, Any] - A dictionary confirming the deletion status.
Delete Documents Example
# Delete specific documents by ID
result = client.documents.delete(
    namespace_name="my-faq-documents",
    ids=["faq-1", "faq-3", "faq-5"]
)
print(f"Deletion result: {result}")

​
vectors.delete

Deletes specific vectors from a vector namespace by their IDs.

​
Parameters

​
namespace_name
str
required
The name of the target vector namespace.
​
ids
List[Union[str, int]]
required
A list of vector IDs to delete.
Returns: Dict[str, Any] - A dictionary confirming the deletion status.
Delete Vectors Example
# Delete specific vectors by ID
result = client.vectors.delete(
    namespace_name="my-image-embeddings",
    ids=["image_001.jpg", "image_002.jpg"]
)
print(f"Deletion result: {result}")

​
Complete Data Management Example

Complete Data Management Workflow
from moorcheh_sdk import MoorchehClient
import time

with MoorchehClient() as client:
    # 1. Create a namespace
    client.namespaces.create("my-data", type="text")

    # 2. Upload documents
    docs = [
        {
            "id": "doc-1",
            "text": "This is the first document",
            "category": "tutorial",
            "author": "John Doe"
        },
        {
            "id": "doc-2",
            "text": "This is the second document",
            "category": "guide",
            "author": "Jane Smith"
        }
    ]

    upload_result = client.documents.upload(namespace_name="my-data", documents=docs)
    print(f"Upload status: {upload_result}")

    # 3. Wait for processing (text documents need time to be embedded)
    print("Waiting for document processing...")
    time.sleep(5)

    # 4. Delete specific documents if needed
    delete_result = client.documents.delete(namespace_name="my-data", ids=["doc-1"])
    print(f"Delete result: {delete_result}")

​
Document Structure Best Practices

​
Required Fields

  • id: Unique identifier for the document (string or number)
  • text: The main content to be searched (string)

​
Optional Metadata

You can include any additional fields as metadata:
Well-Structured Documents
documents = [
    {
        "id": "article-123",
        "text": "Full article content here...",
        # Metadata fields : The other kwargs are considered as metadata
        "title": "Introduction to Machine Learning",
        "author": "Dr. Smith",
        "category": "education",
        "publish_date": "2024-01-15",
        "tags": ["ml", "ai", "tutorial"],
        "difficulty": "beginner"
    }
]

​
Vector Data Structure

For vector uploads, ensure your vectors match the namespace dimension:
Vector Structure
vectors = [
    {
        "id": "embedding-1",
        "vector": [0.1, 0.2, 0.3, ...],  # Must match namespace dimension
        # Optional metadata
        "source": "image_database",
        "category": "product",
        "confidence": 0.95
    }
]

​
Important Notes

Asynchronous Processing: Text documents are processed asynchronously. Allow a few seconds after upload before searching.
ID Uniqueness: Document and vector IDs must be unique within their namespace. Uploading with an existing ID will overwrite the previous entry.
Batch Processing: For large datasets, upload documents in batches of 100-1000 items for optimal performance.