Is there a workaround to index SharePoint site pages in Azure AI Search?

Irene Gunning 0

We are currently developing a chatbot using Azure Open AI, and we plan to use SharePoint as a data source. We have pages containing the data that we want the chatbot to reference. However, according to the documentation SharePoint Online indexer (preview) - Azure AI Search | Microsoft Learn, it states that indexing .aspx files is not supported and it only index documents in document libraries. Is there a workaround for this?

VenkateshDodda-MSFT 21,566 Reputation points Microsoft Employee

2024-11-04T06:07:32.79+00:00
@Irene Gunning Thanks for posting your question in Microsoft Q&A, apologize for any inconvenience caused on this.

As called out in the indexer documentation limitations, yes you cannot load .aspx file in SharePoint to an index.

One workaround to load .aspx to an indexer in SharePoint is by following the below steps:

Use the SharePoint APIs/SDKs to get the ASPX data or the specific method available that you choose to crawl through the data.

You can export the crawled data to blob storage and then you just need to use blob indexer for the content to be indexed accordingly (https://learn.microsoft.com/en-us/azure/search/search-howto-indexing-azure-blob-storage).

Hope this helps, let me know if you have any further questions on this.
VenkateshDodda-MSFT 21,566 Reputation points Microsoft Employee

2024-11-05T06:59:43.64+00:00

@Irene Gunning Following up to see if you have a chance to check my previous response and helped, do let me know if you have any further questions on this.
Irene Gunning 0 Reputation points

2024-11-05T07:11:23.0366667+00:00

@VenkateshDodda-MSFT

Thank you for your response. So it means we need to develop a custom solution? Do you have any example projects or repo that shows how it is implemented through a custom solution?

Also the link you provided directs me to a "Bad Request" page:
Irene Gunning 0 Reputation points

2024-11-05T07:54:50.99+00:00

It's weird that my previous response is not posted. It looks like your site is bugged. I'll post it again:

Thank you for your response. So it means we need to develop a custom solution? Do you have any example projects or repo that shows how it is implemented through a custom solution? Also the link you provided directs me to a "Bad Request" page.

Edit: My previous comment showed up late, and I can't find a way to delete it. It looks like I need to post twice for my comments to appear. This also happened yesterday when my question didn’t show up right away so I thought it was not submitted but later appeared after I posted the same question thus the other question was a duplicate.
VenkateshDodda-MSFT 21,566 Reputation points Microsoft Employee

2024-11-05T13:03:27.5033333+00:00

@Irene Gunning Thanks for your follow questions on this.

Yes, you need to build a custom solution. Unfortunately, we don't have documentation or documentation steps for this.

Here is the link related to index blob storage which got broken in my previous message
https://learn.microsoft.com/en-us/azure/search/search-howto-indexing-azure-blob-storage

You can refer to this GitHub sample here which uses sharepointdataextractor class to extract the data from .docs and .pdf files and there is no specific callout that .aspx supports or not. If you face any issue, I would suggest you to report in that Github repo only.

Also note that this is beta stage not recommended for production.

Share via

Is there a workaround to index SharePoint site pages in Azure AI Search?

Your answer