How to use the API to query the knowledge base？

[H9990HH969]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit report (我已阅读并同意 Language Policy).
• Please do not modify this template :) and fill in all the required fields.

Provide a description of requested docs changes

I hope to operate the knowledge base through the API and use it within the workflow. However, I've noticed that only the API for managing the knowledge base has been released so far. Is there any way to query the content and related information of an article by the article name, database ID, or article ID?

[dosubot[bot]]

To query the content and related information of an article by the article name, database ID, or article ID using the API, you can use the Get Document List endpoint. Here is how you can do it:

curl --location --request GET 'https://api.dify.ai/v1/datasets/{dataset_id}/documents' \
--header 'Authorization: Bearer {api_key}'

Replace {dataset_id} with the ID of the dataset you are querying and {api_key} with your API key. This will return a list of documents within the specified dataset, from which you can find the article by name, database ID, or article ID ^[1].

Additionally, if you are working with PubMed articles, you can use the PubMedAPIWrapper class to conduct searches and fetch document summaries. Here is an example of how you can use it:

from pubmed_search import PubMedAPIWrapper

# Initialize the PubMed API Wrapper
pubmed_api = PubMedAPIWrapper(email="[email protected]")

# Run a search query
query = "your search query"
result = pubmed_api.run(query)

print(result)

This will return the document summaries of the top-k results for the given search query. The PubMedAPIWrapper class handles the search and retrieval of articles, including metadata such as title, abstract, and publication date ^[2].

Details

About Dosu
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

To continue the conversation, mention @dosu.

[H9990HH969]

This is about obtaining information from the knowledge base, but I need to access the full text of specific articles. How should I proceed?

[zfanswer]

+1, raised a similar issue, and has been put into discussion with no reply...

[crazywoola]

This is still under development.

[yong2khoo-lm]

Question: actually what is the 'dataset id' here? I only come across dataset name, are they the same?

[faizzed]

@yong2khoo-lm id and name are actually different.

You can find the document id in the url.

[zfanswer]

This is still under development.

any progress update？thx

[zhimoai]

这仍在开发中。

近期有进展吗？谢谢

[faizzed]

Would love to see the knowledge search endpoint implemented. This is going to be a huge help with integrating dify to other systems.

[harshtripathimaersk]

Is there an API endpoint that can be used to get the metadata of a document that we have uploaded onto dify