Replace unchanging args with one config arg within pretty-format

[pedrottimark]

Summary

Benefits of distinguishing config properties which don’t change (depend only on options and defaults) from arguments which do change during recursive descent through data structures:

• concise clear signature for core functions
• correct complete consistent API for plugins (but one step at a time :)

Converting from options to config in an even more functional lazy evaluation pattern solves a recent accidental performance regression (always iterating theme colors, even for top-level basic values).

Removed a readbump in printList, printMap, printObject, printSet: logic for final comma had been combined with edge spacing and indent instead of in parallel with logic for non-final comma.

Make yourself a cup of coffee or tea before you review this diff.

Test plan

Jest, if the Continuous Irritation system allows

[cpojer]

Woah, that is significantly simpler! Nice work. Do you have any perf numbers?

[codecov-io]

Codecov Report

Merging #4076 into master will increase coverage by 0.02%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master    #4076      +/-   ##
==========================================
+ Coverage   60.42%   60.45%   +0.02%     
==========================================
  Files         196      196              
  Lines        6769     6769              
  Branches        6        6              
==========================================
+ Hits         4090     4092       +2     
+ Misses       2676     2674       -2     
  Partials        3        3

Impacted Files	Coverage Δ
packages/pretty-format/src/index.js	98.03% <100%> (+0.04%)	⬆️
packages/jest-config/src/normalize.js	80.97% <0%> (-0.11%)	⬇️
packages/jest-config/src/index.js	0% <0%> (ø)	⬆️
packages/jest-cli/src/get_changed_files_promise.js	100% <0%> (ø)	⬆️
packages/jest-haste-map/src/index.js	92.96% <0%> (+0.33%)	⬆️
packages/jest-changed-files/src/git.js	92.59% <0%> (+2.93%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 45ceff8...632afb7. Read the comment docs.

[pedrottimark]

Thank you for reminding. Measurements of an experimental version in which proposed with this change is compared to baseline with all other upcoming proposed changes except this change.

• The first row was accidental discovery that there had been a recent performance regression.
• Other than that, I breathed a sigh of relief that accessing object properties doesn’t hurt.

scenario	plugins	proposed	baseline	ratio
basic	0	5863	22992	0.255
basic	9	23869	65253	0.366
complex	0	192674	237179	0.812
complex	9	584595	624248	0.936
geo	0	27962412	27013068	1.035
geo	9	60198020	60391597	0.997
react-element	9	772529	822932	0.939
react-object	9	658507	724195	0.909

Measuring performance is still spooky stuff to me. Even taking geometric mean of 3 runs from restart without wifi occasionally has surprising variation.

Because pretty-format has always seemed very careful not to allocate unnecessary memory, I looked at heapUsed for this batch but sometimes it varied wildly when the times didn’t.

[cpojer]

Yap, it's close enough.

[github-actions]

This pull request has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.
Please note this issue tracker is not a help forum. We recommend using StackOverflow or our discord channel for questions.