This workflow stores Jira issue content in Pinecone:
The data is transformed to a nice JSON array with metadata:
Problem: the recursive Character Text Splitter produces lots of items with little content as all fields are splitted including metadata:
Useless output from metadata:
Is there a way to limit the text splitting to specific fields, e.g. only pageContent?
It looks like your topic is missing some important information. Could you provide the following if applicable.
Welcome to the community @Christian_Gick 
I’m not sure if I understand your question here but you should be able to load specific data and specify the data you want to split in the Default Data Loader node via Mode > Load Specific Data:
Could you maybe share your workflow or parts of it here by copy & pasting the workflow JSON content in between two block quotes (```)?
@aya thanks for pointing me to the solution! I load the specific data now from the variable pageContent and pass the metadata just like before. The metadata is not used in the splitter any more. Thank you!
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.
