When building this site, I encountered a requirement: [Need to get the IDs of all articles]

At first, I thought it was easy to do, just do a loop traversal, and then skip the duplicates, and end the search when the total number is the same as the total number in the database. However, when the number of articles increased, I found that such a loop traversal was a waste of performance, and each retrieval would take a lot of time.

So I studied the python library documentation of ElasticSearch and found that the filter_path parameter can be used to control the field name obtained. This took me a lot of time to find the documentation. I think someone may be able to use it, so record it here.

# use filter_path

The following code searches for all documents and returns the IDs of all documents:

res = es.search(index="db_name", body={"query":{"match_all":{}}}, filter_path="hits.hits._id")