feat: performance improvement deriveInitialInput #20
Conversation
By mutating the accumulator in the reduce function we no longer copy the entire object which is quite expensive. If you wish I can write a test displaying the difference in execution time.
Codecov Report
@@ Coverage Diff @@
## master #20 +/- ##
==========================================
+ Coverage 96.8% 96.83% +0.03%
==========================================
Files 17 17
Lines 250 253 +3
Branches 42 42
==========================================
+ Hits 242 245 +3
Misses 8 8
Continue to review full report at Codecov.
|
There was a problem hiding this comment.
Hey Pruxis,
Let me start off by thanking you for this PR, always nice to see things like these pop up.
The optimization is in line with what we are trying to achieve here, a performant form library with a low bundlesize footprint, this fix does not only decrease initial memory consumption but also does a slight decrease in bundlesize since it does not have to duplicate the _rest helper for the operations.
This makes for a mutation but in theory this is no big deal since it happens isolated on a fresh object, however it slightly decreases readability but I can live with that for the plusses I just summed up.
A test does not seem to be needed for me, I personally am looking to set up a benchmark in the near future though.
Thanks for the work on this and it will be released tonight in version 1.5.1, I'll include it in the CHANGELOG (which I planned to start on the first PR to give some kudo's to contributors).
PS: don't worry about the bundlesize check, I don't know why it is not completing.
By mutating the accumulator in the reduce function we no longer copy the entire object which is quite expensive.
If you wish I can write a test displaying the difference in execution time.