[GH-ISSUE #1769] [FEAT]: Should support the ingestion of CSV, JSON and XML. #1154

Closed
opened 2026-02-22 18:23:23 -05:00 by yindo · 3 comments
Owner

Originally created by @avinashkurup on GitHub (Jun 27, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/1769

What would you like to see?

Would like to see the ingestion of multiple related CSV, JSON and XML files.

Thank you.

Originally created by @avinashkurup on GitHub (Jun 27, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/1769 ### What would you like to see? Would like to see the ingestion of multiple related CSV, JSON and XML files. Thank you.
yindo added the enhancementfeature request labels 2026-02-22 18:23:23 -05:00
yindo closed this issue 2026-02-22 18:23:23 -05:00
Author
Owner

@avinashkurup commented on GitHub (Jun 27, 2024):

What would you like to see?

Would like to see the ingestion of multiple related CSV, JSON and XML files.

Thank you.

I came across an excellent feature in the pandas-ai project that allows the use of multiple DataFrames for data analysis. This feature is particularly useful for complex data operations that require the integration and manipulation of multiple datasets.

@avinashkurup commented on GitHub (Jun 27, 2024): > ### What would you like to see? > Would like to see the ingestion of multiple related CSV, JSON and XML files. > > Thank you. I came across an excellent feature in the [pandas-ai](https://github.com/sinaptik-ai/pandas-ai) project that allows the use of multiple DataFrames for data analysis. This feature is particularly useful for complex data operations that require the integration and manipulation of multiple datasets.
Author
Owner

@avinashkurup commented on GitHub (Oct 4, 2024):

I noticed that CSV and JSON files can be uploaded and very simple queries work on them (XML does not work at all), I however wish to be able to upload multiple CSV, JSON, XML files from the Synthea dataset and be able to do RAG as we do it with unstructured textual documents (PDF, DOCX and TXT). What would be the best approach to achieve this, or can a generic implementation be included in the tool? Help with this @timothycarambat would be greatly appreciated !

@avinashkurup commented on GitHub (Oct 4, 2024): I noticed that CSV and JSON files can be uploaded and very simple queries work on them (XML does not work at all), I however wish to be able to upload multiple CSV, JSON, XML files from the [Synthea](https://synthea.mitre.org/downloads) dataset and be able to do RAG as we do it with unstructured textual documents (PDF, DOCX and TXT). What would be the best approach to achieve this, or can a generic implementation be included in the tool? Help with this @timothycarambat would be greatly appreciated !
Author
Owner

@timothycarambat commented on GitHub (Jan 3, 2025):

RAG on datasets is always a complex thing since typically, an LLM requires "whole context" to be able to make a 100% accurate inference since data may be in various chunks across the vector db.

In this case, uploading and document pinning is the best general solution

@timothycarambat commented on GitHub (Jan 3, 2025): RAG on datasets is always a complex thing since typically, an LLM requires "whole context" to be able to make a 100% accurate inference since data may be in various chunks across the vector db. In this case, uploading and [document pinning](https://docs.anythingllm.com/llm-not-using-my-docs#document-pinning) is the best general solution
yindo changed title from [FEAT]: Should support the ingestion of CSV, JSON and XML. to [GH-ISSUE #1769] [FEAT]: Should support the ingestion of CSV, JSON and XML. 2026-06-05 14:39:15 -04:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#1154