Basic optimizations #2
Description
TODO:
- During response decode we store iterators pointing to the start of data and to the end of data (tuples). Firstly, we need only one iterator pointing to the start of data (assuming that the response stored continuously in the input buffer it's enough to decode response). This allows us to get rid of skipping tuple in decoder (which turns out to be another one traverse over the tuple) since we know size of data section in response (
package size - header size).