* Refactor and structuring optimizations to be less computationally heavy
* Forgot to commit the new file containing the optimization do over
* Point to correct functions in shrinker2
* Split out inline_constr_ops since it adds in builtins that can then be swept up by the builtin force reduction
* Fix: issue where identity reducer was always returning true
* Forward inlining on lambdas produces better results. This is due to a forward pass being able to apply an argument that may have no_inline at the top where as vice-versa would reduce the arg first.
* Clippy and test fixes
* Clear no_inlines when inlining a function
* Convert shrinker2 to replace shrinker and update tests
Before this commit, we would require those target a specific handler,
whereas they are in fact global to the entire validator.
So now, we recover the behaviour from before where we default to the
only available validator when there's no ambiguity.
Note that this also solves the need for repeatedly applying parameters
to each handler of a parameterized validator. The command now
rightfully apply parameters to each corresponding handler.
- Ensure consistency between overall summary and individual summaries.
- Remove 'max_*' properties, since they are only padding numbers used
for formatting the terminal output.
- Rename a few fields to be closer to the existing naming (name ->
title, memory -> mem, etc..)
- Remove duplicated outputs
- Re-order fields such that summaries come first
For the program to be consistent, the 'EventListener' target that we
pass to a Project should be responsible for the output format.
Otherwise, we are contingent on developers to remember passing the
option at call-site. Plus, it overloads the project code with an extra
boolean option.
Instead, since the behaviour is solely driven by the execution
context, we can instantiate a different event target upfront, and
simply hold on to it throughout the program.
As a nice side-effect, we can gently re-organize the code to keep the
terminal printing logic and the json printing logic separate.
It has somehow always annoyed me that we have --trace_level (with 'trace' first) and --filter_traces (with traces, plural and last). Given that these two options are almost always used together, the inconsistency is frustrating.
Kasey