Performance: Cache repeated computations in MappedData methods

[anthropic-code-agent]

Description

Cache repeated computations to avoid redundant calculations that occur multiple times within the same function or across related method calls.

Current Implementation Issues

R/MappedData.R - getAestheticsForGeom() (Lines 151-159)

• gsub() and setdiff() operations computed twice
• Called within loop at line 166 and again at line 171
• String manipulation repeated unnecessarily

R/MappedData.R - setLimits() (Lines 464-469)

• gsub() on "y" and "x" repeated multiple times
• Pattern replacement done in loop without caching

R/MappedDataTimeProfile.R - Multiple locations

• Lines 361-363: listOfAesthetics[which(...)]$aesthetic computed twice
• Lines 393-396: Same pattern repeated in setyLimits()
• Lines 318-321: Filtering operations in dataForPlot() repeated in loop

R/plotYVsX.R - ggplot_build() calls (Lines 405-433)

• ggplot_build() called and result analyzed multiple times
• Very expensive operation that should be cached
• Result can be reused for multiple analyses

R/plotYVsX.R - getCountsWithin() (Lines 692-707)

• Repeated rlang::eval_tidy() calls
• Evaluation results could be cached

R/utilities.R - tolower() calls (Lines 59, 177-180)

• tolower(dimension) computed multiple times
• Same transformation done repeatedly per function call

Suggested Implementation

1. Store intermediate results: Save computed values in local variables
2. Use R6 private fields: For class methods, cache expensive computations in private fields
3. Implement memoization: Use function memoization for pure functions with repeated calls
4. Hoist invariant computations: Move loop-invariant code outside loops

Example Refactoring

Before:

for (item in items) {
  result <- expensiveComputation(data)
  doSomething(result, item)
}

After:

cachedResult <- expensiveComputation(data)
for (item in items) {
  doSomething(cachedResult, item)
}

Expected Benefits

• Reduced redundant computations
• Faster method execution
• Lower CPU usage
• Better performance for complex plots

Implementation Notes

• Ensure cached values are invalidated when underlying data changes
• Consider using R6 active bindings for lazy evaluation
• Document which fields are cached and when they update
• Follow OSP R coding standards: https://dev.open-systems-pharmacology.org/r-development-resources/coding_standards_r

Files to Modify

• R/MappedData.R
• R/MappedDataTimeProfile.R
• R/plotYVsX.R
• R/utilities.R

Testing

• Run existing unit tests
• Verify cached values are correct
• Test that cache invalidation works properly
• Add tests for edge cases