I want to use Google Places Autocomplete API to provide search-while-typing for an Android app.
the Autocomplete API returns IDs and References, requiring an iteration through the results to get the details (such as name and geometry) for each place, this is slow and results in a bad user experience. also, many API calls are done for each key press.
is there a better way to provide such searching capability with Google Places ?
Is there a reason you need to have the geometry for each possible autocomplete place after each keypress of the API? Showing this seems like generally it would not be a great user experience, and is not what the Autocomplete API was designed for.
In theory the name (description) should be enough for the user to decide which item they’re interested in, and then after a user has actually selected something you can request additional details as needed.
On the other hand if you already have some information about the name of the place, you could use the Places API (not Autocomplete) to get additional details of matching nearby places:
https://developers.google.com/maps/documentation/places/#PlaceSearchResponses