Code Review

[samreid]

To familiarize myself better with the codebase, and better prepare for code review, I'll take a look around.

UPDATE:

To begin a code review, the responsible developer should:

• Copy this checklist to a new GitHub issue titled "Code Review" and labeled dev:code-review.
• Delete the Table of Contents section, since the links will be incorrect.
• Complete (or delete) the Specific Instructions section.
• Delete (or strikethrough) checklist items and sections that are not relevant.
• Replace {{GITHUB_ISSUE_LINK}} with links to GitHub issues.
• Assign the GitHub issue to the developer who is doing the code review.
  @samreid adds: I'll check off an item if the review of that item is complete, even if it has been moved to a side issue. The code review will be complete when all 84 tasks are checked off.

---

PhET Code-Review Checklist (a.k.a "CRC")

• The responsible dev is responsible for removing the irrelevant parts
• A checked-off item doesn't mean "no problem here", it means "it was reviewed"
• Problems can be noted in side issues that reference this issue, or through // REVIEW comments in the code

Specific Instructions

There are 10 screens across 3 sims. We are publishing with keyboard support but without sonification or interactive description. This simulation uses 2 third party engines. Three.js for 3d rendering and p2 for physics. Review mobius usages as necessary.

GitHub Issues

The following standard GitHub issues should exist. If these issues are missing, or have not been completed, pause code
review until the issues have been created and addressed by the responsible dev.

• Write model.md #166
  describe the model, in terms appropriate for teachers? Has it been reviewed by the sim designer?
• Write implementation-notes.md #167
  implementation-notes.md. Does it provide an overview that will be useful to future maintainers?
• Memory testing issue #168
• results of memory testing for brands=phet-io (if the sim is instrumented for PhET-iO), see {{GITHUB_ISSUE_LINK}}
• performance testing and sign-off #169
• review of pointer areas #170
• credits (add right before RC.1) #171

Build and Run Checks

If any of these items fail, pause code review.

• Does the sim build without warnings or errors?
• Does the html file size seem reasonable, compared to other similar sims? #172
• Does the sim start up? (unbuilt and built versions)
• Does the sim experience any assertion failures? (run with query parameter ea)
• Does the sim pass a scenery fuzz test? (run with query parameters fuzz&ea)
• Does the sim behave correctly when listener order is shuffled? (run with query #173
  parameters ea&listenerOrder=random and ea&listenerOrder=random&fuzz)
• Does the sim output any deprecation warnings? Run with ?deprecationWarnings. Do not use deprecated methods in
  new code.

Memory Leaks

See also #168

• Does a heap comparison using Chrome Developer Tools indicate a memory leak? (This process is
  described here.) Test on a
  version built using grunt --minify.mangle=false. Compare to testing results done by the responsible developer.
  Results can be found in Memory testing issue #168.
• For each common-code component (sun, scenery-phet, vegas, …) that opaquely registers observers or listeners, is #283
  there a call to that component’s dispose function, or is it obvious why it isn't necessary, or is there
  documentation about why dispose isn't called? An example of why no call to dispose is needed is if the component
  is used in a ScreenView that would never be removed from the scene graph. Note that it's also acceptable (and
  encouraged!) to describe what needs to be disposed in implementation-notes.md.
• Are there leaks due to registering observers or listeners? The following guidelines should be followed unless
  documentation (in-line or in implementation-notes.md) describes why following them is not necessary.
  • AXON: Property.link or lazyLink is accompanied by unlink.
  • AXON: Multilink.multilink is accompanied by unmultilink.
  • AXON: Creation of Multilink is accompanied by dispose.
  • AXON: Creation of DerivedProperty is accompanied by dispose.
  • AXON: Emitter.addListener is accompanied by removeListener.
  • AXON: ObservableArrayDef.element*Emitter.addListener is accompanied
    by ObservableArrayDef.element*Emitter.removeListener
  • SCENERY: Node.addInputListener is accompanied by removeInputListener
  • TANDEM: Creation of an instrumented PhetioObject is accompanied by dispose.
  • UPDATE: Will be covered in Memory testing issue #168 and For each common-code component (sun, scenery-phet, vegas, …) that opaquely registers observers or listeners, is #283
• All classes that require a dispose function should have one. This should expose a public dispose function that
  calls this.dispose{{CLASS_NAME}}(), where dispose{{CLASS_NAME}} is a private function declared in the
  constructor. {{CLASS_NAME}} should exactly match the class name.. Will be covered in Change the code review checklist details regarding dispose{{CLASS_NAME}} phet-info#224 and Memory testing issue #168
• All classes that do not properly override dispose should either (a) use isDisposable: false, or (b) implement a dispose method that calls Disposable.assertNotDisposable. #284 Use (a) for classes that inherit a dispose method
  from Disposable. Use (b) for classes that inherit a dispose method from something other than Disposable. The goal
  here is to prevent clients from calling dispose for something that does not properly clean itself up, creating a
  memory leak. This is especially important for common code, but is good defensive programming for sim-specific code.

Performance

• Play with sim, identify any obvious performance issues. Examples: animation that slows down with large numbers of
  objects; animation that pauses or "hitches" during garbage collection.
• If the sim uses WebGL, does it have a fallback? Does the fallback perform reasonably well? (run with query #174
  parameter webgl=false)

Usability

• Are UI components sufficiently responsive? (especially continuous UI components, such as sliders)
• Are pointer areas optimized, especially for touch? (run with query parameter showPointerAreas). UPDATE: Will be addressed in review of pointer areas #170
• Do pointer areas overlap? (run with query parameter showPointerAreas) Overlap may be OK in some cases, depending
  on the z-ordering (if the front-most object is supposed to occlude pointer areas) and whether objects can be moved. UPDATE: Will be addressed in review of pointer areas #170

Internationalization

• Does the sim behave correctly with dynamic layout, to support dynamic locale? Run with stringTest=dynamic and
  use the left/right arrow keys.
• Are there any strings that are not internationalized, and does the sim layout gracefully handle internationalized
  strings that are shorter than the English strings? (run with query parameter stringTest=X. You should see nothing
  but 'X' strings.)
• Does the sim layout gracefully handle internationalized strings that are longer than the English strings? (run
  with query parameters stringTest=double and stringTest=long)
• Does the sim stay on the sim page (doesn't redirect to an external page) when running with the query parameter
  stringTest=xss? This test passes if sim does not redirect, OK if sim crashes or fails to fully start. Only test on
  one desktop platform. For PhET-iO sims, additionally test ?stringTest=xss in Studio to make sure i18n strings didn't
  leak to phetioDocumentation, see https://github.com/phetsims/phet-io/issues/1377
• Avoid using concatenation to create strings that will be visible in the user interface. Use StringUtils.fillIn
  and a string pattern to ensure that strings are properly localized. This is relevant in cases where order should be
  translatable.
• Use named placeholders (e.g. "{{value}} {{units}}") instead of numbered placeholders (e.g. "{0} {1}").
• Inspect ${REPO}-strings_en.json and verify that all string keys conform #183
  to string key conventions. String
  keys are difficult to change after a sim has been published, and they appear in the PhET-iO API (and Studio) as the
  phetioIDs for StringProperties.
• If the sim was already released, make sure none of the original string keys have changed. If they have changed,
  make sure any changes have a good reason and have been discussed with @jbphet (it is likely that an issue
  like Translations will lose some strings the next time this is published off of master gravity-force-lab#166 should be created). UPDATE: @zepumph investigated this in missingStringPhetioElements "density.general.model.strings.densityBuoyancyCommon.averageStringProperty" #175 (comment)

Repository Structure

• The repository name should correspond to the sim title. For example, if the sim title is "Wave Interference", then
  the repository name should be "wave-interference".

• Are all required files and directories present? For a sim repository named “my-repo”, the general structure should
  look like this (where assets/, images/, mipmaps/ or sounds/ may be omitted if the sim doesn’t have those types of
  resource files).

  my-repo/
    assets/
    doc/
      images/
        *see annotation
      model.md
      implementation-notes.md
    images/
      license.json
    js/
      (see section below)
    mipmaps/
      license.json
    sound/
      license.json
    dependencies.json
    .gitignore
    my-repo_en.html
    my-repo-strings_en.json
    Gruntfile.js
    LICENSE
    package.json
    README.md
  

  *Any images used in model.md or implementation-notes.md should be added here. Images specific to aiding with
  documentation do not need their own license.

• Verify that the same image file is not present in both images/ and mipmaps/. If you need a mipmap, use it for all
  occurrences of the image.

• Is the js/ directory properly structured? All JavaScript source should be in the js/ directory. There should be a
  subdirectory for each screen (this also applies for single-screen sims, where the subdirectory matches the repo name).
  For a multi-screen sim, code shared by 2 or more screens should be in a js/common/ subdirectory. Model and view code
  should be in model/ and view/ subdirectories for each screen and common/. For example, for a sim with screens
  “Introduction” and “Lab”, the general directory structure should look like this:

  my-repo/
    js/
    common/
      model/
      view/
    introduction/
      model/
      view/
    lab/
      model/
      view/
    my-repo-main.js
    myRepo.js
    myRepoStrings.js
  

UPDATE: We discussed this in #125 and approved a variation

• Do filenames use an appropriate prefix? Some filenames may be prefixed with the repository name,
  e.g. MolarityConstants.js in molarity. If the repository name is long, the developer may choose to abbreviate the
  repository name, e.g. EEConstants.js in expression-exchange. If the abbreviation is already used by another
  repository, then the full name must be used. For example, if the "EE" abbreviation is already used by
  expression-exchange, then it should not be used in equality-explorer. Whichever convention is used, it should be used
  consistently within a repository - don't mix abbreviations and full names. The abbreviation should be all uppercase
  letters; e.g. MOTHAConstants, not MotHAConstants for "Model of the Hydrogen Atom".
• Is there a file in assets/ for every resource file in sound/ and images/? #285 Note that there is not necessarily a Is there a file in assets/ for every resource file in sound/ and images/? #285
  1:1 correspondence between asset and resource files; for example, several related images may be in the same .ai file.
  Check license.json for possible documentation of why some resources might not have a corresponding asset file.
• For simulations, was the README.md generated by grunt published-README or grunt unpublished-README? Common
  code repos can have custom README files.
• Is the LICENSE file correct? (Generally GPL v3 for sims and MIT for common code,
  see this thread for additional information).
  Note that "common" repos that are mostly sim-specific code (like circuit-construction-kit-common and
  density-buoyancy-common) should be GPL v3,
  see other thread.
• Does .gitignore match the one in simula-rasa?
• In GitHub, verify that all non-release branches have an associated issue that describes their purpose.
• Are there any GitHub branches that are no longer needed and should be deleted?
• Sim-specific query parameters (if any) should be identified and documented in one .js file in js/common/ or js/ (
  if there is no common/). The .js file should be named {{PREFIX}}QueryParameters.js, for example
  ArithmeticQueryParameters.js for the arithmetic repository, or FBQueryParameters.js for Function Builder (where
  the FB prefix is used).
• Query parameters that are public-facing should be identified using public: true in the schema.
• All sims should use a color file named MyRepoColors.ts or, if using abbreviations, MRColors.ts, and
  use ProfileColorProperty where appropriate, even if they have a single (default) profile (to support color editing
  and PhET-iO Studio). The ColorProfile pattern was converted to *Colors.ts files in
  PhET-iO instrumentation for ColorProfile scenery-phet#515. Please see
  GasPropertiesColors.ts
  for a good example.
• Sim-specific preferences (which appear in the Simulation tab of the Preferences dialog) should appear as
  Properties in
  {{PREFIX}}Preferences.ts, for example KeplersLawsPreferences.ts. The initial value of each such preferences Property
  should be set via a query parameter from {{PREFIX}}QueryParameters.js.
• Does package.json refer to any dependencies that are not used by the sim?
• Does include any config that was only needed for development? For example #286
  • Lint rules turned off or made more graceful
  • Certain testing or simFeatures turned off with flags

Coding Conventions

• Are coding conventions outlined in
  PhET's Coding Conventions document
  followed and adhered to? This document deals with PhET coding conventions. You do not need to exhaustively check every
  item in this section, nor do you necessarily need to check these items one at a time. The goal is to determine whether
  the code generally meets PhET standards.

TypeScript Conventions

• Are TypeScript conventions outlined in
  PhET's TypeScript Conventions
  document followed and adhered to?

Math Libraries

• DOT/Utils.toFixed or DOT/Utils.toFixedNumber should be used instead of toFixed. JavaScript's toFixed is
  notoriously buggy. Behavior differs depending on browser, because the spec doesn't specify whether to round or floor.

IE11

• IE is no longer supported. With that in mind remove IE-specific workarounds
• Use string.includes and string.startsWith where possible.

Organization, Readability, and Maintainability

• Does the organization and structure of the code make sense? Do the model and view contain types that you would
  expect (or guess!) by looking at the sim? Do the names of things correspond to the names that you see in the user
  interface?
• Are appropriate design patterns used?
  See phet-software-design-patterns.md.
  If new or inappropriate patterns are identified, create an issue.
• Is inheritance used where appropriate? Does the type hierarchy make sense?
• Is composition favored over inheritance where appropriate?
  See https://en.wikipedia.org/wiki/Composition_over_inheritance.
• Is there any unnecessary coupling? (e.g., by passing large objects to constructors, or exposing unnecessary
  properties/functions). If you only need a few fields from a large object, pass them in as separate parameters. The
  threshold for the number of parameters is up to you - use your judgement. Alternatively in TypeScript, you can
  decouple by narrowing the API using Pick, but this is a bit of a hack. Here's an example:

class MyClass {
  public constructor( tickMarksVisibleProperty: Property<boolean>,
                      model: Pick<IntroModel, 'changeWaterLevel'>, // <-- Note the call site can pass the whole model, but we declare we will only use this part of it
                      waterCup: WaterCup, modelViewTransform: ModelViewTransform2,
                      providedOptions?: WaterCup3DNodeOptions ) {}
}

• Is there too much unnecessary decoupling? (e.g. by passing all of the properties of an object independently
  instead of passing the object itself)?
• Are the source files reasonable in size? Scrutinize large files with too many responsibilities - can
  responsibilities be broken into smaller delegates? To see file sizes for TypeScript sims, run this shell command:

cd {{repo}}/js ; wc -l `find . -name "*.ts" -print` | sort

• Are any significant chunks of code duplicated? In addition to manual identification, tools include: WebStorm #291
  Code > Analyze Code > Locate Duplicates and https://github.com/danielstjules/jsinspect.
• Is there anything that should be generalized and migrated to common code?
• Are there any TODO or FIXME or REVIEW comments in the code? They should be addressed or promoted to GitHub
  issues.
  • This means that there should be no TODOs pointing to One issue to hold every TODO in the project tasks#1129
  • This means that your repo should have the "todo-should-have-repo" lint rule enabled.
• Are there any magic numbers that should be factored
  out as constants and documented?
• Are there any constants that are duplicated in multiple files that should be factored out into
  a {{REPO}}Constants.js file?
• Does the implementation rely on any specific constant values that are likely to change in the future? Identify
  constants that might be changed in the future. (Use your judgement about which constants are likely candidates.) Does
  changing the values of these constants break the sim? For example,
  see allow minimum rows to go to "1" and address dependency on current minimum of "5" plinko-probability#84.
• Is PhetColorScheme used where
  appropriate? Verify that the sim is not inventing/creating its own colors for things that have been standardized
  in PhetColorScheme. Identify any colors that might be worth adding to PhetColorScheme.
• Are all dependent Properties modeled as DerivedProperty instead of Property? #288
• All dynamics should be called from Sim.step(dt), do not use window.setTimeout or window.setInterval. This will
  help support Legends of Learning and PhET-iO.

Accessibility

This section may be omitted if the sim has not been instrumented with accessibility features. Accessibility includes
various features, not all are always include. Ignore sections that do not apply.

General

• Are accessibility features integrated well into the code? They should be added in a maintainable way, even if that
  requires upfront refactoring. UPDATE: Existing features are in good shape. More will be done in Blocks should use GrabDragInteraction for keyboard input #209

Alternative Input

• Does the sim pass an accessibility fuzz test? (run with query parameters fuzzBoard&ea)
• Does the sim use ScreenView.pdomPlayAreaNode.pdomOrder and Screenview.pdomControlAreaNode.pdomOrder to control
  the traversal order and categorize Nodes in the PDOM? See alternative-input-quickstart-guide.md for more information.
• Does this sim use specific keyboard shortcuts that overlap with global shortcuts? This includes the use of
  modifier keys like in Unintended zoom/keyboard behavior ratio-and-proportion#287. NOTE: There is currently no list
  of global shortcuts, and therefore no way to complete this checklist item.
  See Create a list of global keyboard shortcuts. phet-info#188.. UPDATE: The design did not call for any global keystrokes that I am aware of.

Interactive Description

Section deleted because Interactive Description is not supported for this publication.

PhET-iO

This section may be omitted if the sim has not been instrumented for PhET-iO, but is likely good to glance at no matter.

• Does instrumentation follow the conventions described
  in PhET-iO Instrumentation Guide?
  This could be an extensive bullet. At the very least, be sure to know what amount of instrumentation this sim
  supports. Describing this further goes beyond the scope of this document.
• PhET-iO instantiates different objects and wires up listeners that are not present in the PhET-branded simulation.
  It needs to be tested separately for memory leaks. To help isolate the nature of the memory leak, this test should be
  run separately from the PhET brand memory leak test. Test with a colorized Data Stream, and Studio (easily accessed
  from phetmarks). Compare to testing results done by the responsible developer and previous releases.
• Make sure unused PhetioObject instances are disposed, which unregisters their tandems.
• Make sure JOIST dt values are used instead of Date.now() or other Date functions. Perhaps try
  phet.joist.elapsedTime. Though this has already been mentioned, it is necessary for reproducible playback via input
  events and deserves a comment in this PhET-iO section.
• Are random numbers using DOT/dotRandom as an imported module (not a global), and all doing so after modules are
  declared (non-statically)? For example, the following methods (and perhaps others) should not be
  used: Math.random, _.shuffle, _.sample, _.random. This also deserves re-iteration due to its effect on
  record/playback for PhET-iO.
• Like JSON, keys for undefined values are omitted when serializing objects across frames. Consider this when
  determining whether toStateObject should use null or undefined values.
• PhET prefers to use the term "position" to refer to the physical (x,y) position of objects. This applies to both
  brands, but is more important for the PhET-iO API. See location vs position phet-info#126
• Are your IOType state methods violating the API of the core type by accessing private fields? #287
• When defining a boolean Property to indicate whether something is enabled with a tandem name of enabledProperty,
  use AXON/EnabledProperty. This should be done in both the model and the view. If you're using a DerivedProperty,
  skip this item.
• Do not use translated strings in phetioDocumentaton - it changes the PhET-iO API!

[samreid]

Code review is complete. I feel the architecture and overall setup is in good shape. I've opened side issues or TODO comments for recommendations for further improvements.

One difficulty I have in navigating the code is that it is difficult to map from the screen to the code and back. For instance, here is a panel in one of the buoyancy screens:

Since this is an untitled control, it is difficult to know where it lies in the files, and vice versa (looking at the files, difficult to know it is referring to that panel). I'm not sure how to solve that for this case, but it does seem to be less of a problem for the titled panels or titled accordion boxes. Just writing this down as a potential topic to keep in mind for future designs--nothing to do about it at the moment.

Of the remaining TODOs listed for this issue, some are relatively minor and can be addressed here. Others will likely want to move into a dedicated side issue. Some are questions, and some we may decide are not worth the effort. For next steps, can @AgustinVallejo and @zepumph please review the TODOs that reference this issue and take next steps?

[AgustinVallejo]

We moved all the remaining TODOs into #317. Thank you very much @samreid for your thorough review!! Closing

[phet-dev]

Reopening because there is a TODO marked for this issue.

[AgustinVallejo]

Reopening because there is a TODO marked for this issue.

Ironic.

Closing again