Memoize instances typcasters in typecaster map #152
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # | ||
| # @since 0.6.0 | ||
| def typecaster_for(type) | ||
| typecaster = TYPECASTER_MAP[type] | ||
| typecaster.new if typecaster | ||
| typecaster if typecaster |
There was a problem hiding this comment.
This final line is no longer necessary since you're not calling anything. The method simply becomes:
def typecaster_for(type)
TYPECASTER_MAP[type]
end
All of the default typecasters are stateless by default, so allocating a new instance for each field becomes extremely costly, especially for larger models. To reduce this, we memoize instances by default, but still allow users to provide a custom stateful type.
film42
force-pushed
the
gt/optimize_typing
branch
from
f4ccaa1 to
bbf20e7
Compare
|
+1 I was just about to open a PR with the exact same change! |
|
🤖 Beep boop! Just sending a ping ❤️😄 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
All of the default typecasters are stateless, so allocating a new instance for each field becomes extremely costly, especially for larger models. To reduce this, we can memoize the
TYPECASTER_MAP.This should be a backwards compatible change with a speed boost. The only people who might have trouble are those who modified the typecaster map. This is a frozen constant, though, so nobody should be touching this.
This is currently about 10% quicker on a model with 93 fields. The benchmark creates a model and then calls
to_json. I also ran the benchmark on a smaller model with only 12 String fields and saw a 1-2% improvement.