Skipping index generation for Searchkick-enabled Model

[matteoredz]

Hi, we have an issue with Searchkick. We have a resource being indexed by another application, and the index name is fixed. We need Searchkick to access that index. We already managed to avoid the model being indexed (by setting should_index to false)
if we use the index_name option, Searchkick will append a timestamp to that
Will it still be able to access our existing index or is there anything else to do to achieve that?

[ankane]

Hey @matteoredz, can you provide more details about what you're seeing? Searchkick doesn't append a timestamp to the index name when searching.

[matteoredz]

Hi @ankane, the index structure is model_name_timestamp but I've an index injected from an external web app structured ad my_index_name without timestamp.

In my main application I've mapped self.index_name to match the index name injected by the other app.

My question is: Is there any - documented - way to skip indexing from Searchkick (which would generate my_index_name_timestamp) and use it as read-only using this gem?

[ankane]

You can turn callbacks off with callbacks: false, but someone could always reindex manually. The best way would be to give read-only access with X-Pack or your hosted provider.

[matteoredz]

Ok, at this stage I've overrided the rake task:

Searchkick.models.delete_if { |model| model.read_only_index? }.each do |model|
        puts "Reindexing #{model.name}..."
        model.reindex
end