You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A user pointed out that our ES results serialization for rendering API requests might be slower than necessary because it involves converting dictionaries to Python objects in the DRF serializer and then back to dictionaries when sending data to the client.
This process could be slower than avoiding the conversion to Python objects by using custom code for our serializer.
Therefore, we should profile the performance of this related code to compare both approaches.
Additionally, we could further analyze the issue by considering that the ES response is already a Python object based on a custom AttrDict class. It might be possible to eliminate the Python objects earlier in the process and use dictionaries downstream if this improves performance and simplifies the code.
The text was updated successfully, but these errors were encountered:
A user pointed out that our ES results serialization for rendering API requests might be slower than necessary because it involves converting dictionaries to Python objects in the DRF serializer and then back to dictionaries when sending data to the client.
This process could be slower than avoiding the conversion to Python objects by using custom code for our serializer.
Therefore, we should profile the performance of this related code to compare both approaches.
Additionally, we could further analyze the issue by considering that the ES response is already a Python object based on a custom
AttrDict
class. It might be possible to eliminate the Python objects earlier in the process and use dictionaries downstream if this improves performance and simplifies the code.The text was updated successfully, but these errors were encountered: