How to add additional meta data to the index from json files

[svenfeld]

I have a data setup which contains the content insides pdfs and json files with addtional metadata (the jsons and pdfs share the same name).
I want to use the json files as addtional meta data for the pdfs. My approach was to parse the json as additional fields to the index but did not succeed.

I looked into this
https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/docs/customization.md#other-approaches-to-improve-search-results

and tried making the changes to the searchmanager.py but could not manage to receive the result I wanted. I added additional 'SimpleFields' but these were null after running'prepdocs.py'
https://github.com/Azure-Samples/azure-search-openai-demo/blob/main/app/backend/prepdocslib/searchmanager.py#L106

The json is structured like this.

{
    "title":"product",
    "isRelevant":true,
    "location":"en",
    "remarks":"comments",
    "thumbnailURL":"https://",
    "fileName":"file.pdf"
 }

Can someone help out on this to create the index correctly? Or suggest another approach that would add the json meta data to the pdf to improve the search?

[pamelafox]

If your search index already exists, then you'll need to add additional fields in this if block:

azure-search-openai-demo/app/backend/prepdocslib/searchmanager.py

Line 218 in 023dc1b

logger.info("Search index %s already exists", self.search_info.index_name)

Then, once those search fields exist, you need to change the update_content function to add the new fields to the uploaded document chunks:

azure-search-openai-demo/app/backend/prepdocslib/searchmanager.py

Line 264 in 023dc1b

{