Milvus vector store, when create collection, can i set the type of langchain_text

Questions
1

Describe the problem/error/question

when i have a lots of items to insert into the store, the first item’s content length will be set to the langchain_text’s varchar length, if the next item’s content length longer then the first, it will not be able to insert.

image
image1332×600 62.2 KB

can i change this type to text,

What is the error message (if any)?

Error inserting data: {“succ_index”:,“err_index”:[0,1],“status”:{“extra_info”:{},“error_code”:“IllegalArgument”,“reason”:“length of varchar field langchain_text exceeds max length, row number: 1, length: 718, max length: 631: invalid parameter”,“code”:1100,“retriable”:false,“detail”:“length of varchar field langchain_text exceeds max length, row number: 1, length: 718, max length: 631: invalid parameter”},“IDs”:null,“acknowledged”:false,“insert_cnt”:“0”,“delete_cnt”:“0”,“upsert_cnt”:“0”,“timestamp”:“0”}

Please share your workflow

image
image1782×1206 198 KB

Share the output returned by the last node

Information on your n8n setup

  • n8n version:
  • Database (default: SQLite):
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app):
  • Operating system:
1 Like
2

I think you would need to create the schema before ingesting data. If none was provided upfront, it will create it, based on the first data ingestion.
But i would also be interested in a workflow that generates this schema properly for langchain.