liza/doc/hacking.texi

@c  This document is part of the Liza Data Collection Framework manual.
@c  Copyright (C) 2018 R-T Specialty, LLC.
@c
@c    Permission is granted to copy, distribute and/or modify this document
@c    under the terms of the GNU Free Documentation License, Version 1.3
@c    or any later version published by the Free Software Foundation;
@c    with no Invariant Sections, no Front-Cover Texts, and no Back-Cover
@c    Texts.  A copy of the license is included in the section entitled ``GNU
@c    Free Documentation License''.

@node Hacking
@chapter Hacking Liza
@helpwanted

This chapter provides general information and guidance for
  [prospective] developers of Liza.

@dnindex Libraries
@dnindex GNU ease.js
@dnindex Mocha
@dnindex Chai
For writing classes; interfaces; and traits,
  developers should familiarize themselves with @easejs{}.
For writing unit tests,
  developers should be familiarize themselves with
  @mocha{} and @chai{}.
For more information on the libraries used by Liza,
  see @ref{Libraries}.

@dnindex Copyright Header
@dnindex Source Files
@dnindex Source File Naming
Most source files have a general structure that must be followed.
For example,
  all such files must have a copyright header and
    must be named after the class they define or system under test.
For more information,
  @pxref{Source Files}.

Generally speaking,
  developers should be familiar with vanilla ECMAScript;
  DOM APIs;
  and have a basic understanding of Node.js for well-rounded Liza
    development.
Writing this manual requires basic understanding of Texinfo.
References for these topics and others are provided in
  @pxref{Developer Resources}.

@menu
* Source Files::         Conventions for project files
* Libraries::            The few libraries used by Liza
* Developer Resources::  Where to look for more information
* TypeScript Migration:: Information on migrating to TypeScript
@end menu


@node Source Files
@section Source Files
@helpwanted

@dnindex Source Files
This section describes conventions for organizing files,
  both in directory structure and in content.

@menu
* Copyright Header::                   Important header at the top of all
                                       source files
* ECMAScript Strict Mode:Strict Mode.  Always indicate strict mode
@end menu

@node Copyright Header
@subsection Copyright Header
@dnindex Copyright Header
@dnindex License
@dnindex GNU General Public License version 3
@dnindex GNU General Public License version 3, Or Later
Every source file should begin with a copyright header including the
  appropriate years and license information.
This ensures that this information is always available even if the
  file becomes separated from the source distribution (e.g. is
  distributed independently).
Further,
  it is necessary to indicate that the source file is distributed under
  the GNU General Public License version@tie{}3 @emph{or later}@mdash{
    }that ``or later'' clause does not exist as part of the license itself,
      and so the mere presence of the license in @file{COPYING} is
      insufficient.

The copyright headers vary slightly between JavaScript and Texinfo
  source files,
    represented in @ref{f:cheader-js} and @ref{f:cheader-texi}
    respectively.

@float Figure, f:cheader-js
@example
/**
 * DESCRIPTION OF FILE
 *
 *  Copyright (C) 2017, 2018 R-T Specialty, LLC.
 *
 *  This file is part of the Liza Data Collection Framework
 *
 *  Liza is free software: you can redistribute it and/or modify
 *  it under the terms of the GNU General Public License as published by
 *  the Free Software Foundation, either version 3 of the License, or
 *  (at your option) any later version.
 *
 *  This program is distributed in the hope that it will be useful,
 *  but WITHOUT ANY WARRANTY; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *  GNU General Public License for more details.
 *
 *  You should have received a copy of the GNU General Public License
 *  along with this program.  If not, see <http://www.gnu.org/licenses/>.
 */
@end example
@caption{Example copyright header for JavaScript files}
@end float

@float Figure, f:cheader-texi
@example
@@c  This document is part of the Liza Data Collection Framework manual.
@@c  Copyright (C) 2018 R-T Specialty, LLC.
@@c
@@c    Permission is granted to copy, distribute and/or modify this document
@@c    under the terms of the GNU Free Documentation License, Version 1.3
@@c    or any later version published by the Free Software Foundation;
@@c    with no Invariant Sections, no Front-Cover Texts, and no Back-Cover
@@c    Texts.  A copy of the license is included in the section entitled ``GNU
@@c    Free Documentation License''.
@end example
@caption{Example copyright header for JavaScript files}
@end float

For more information, see ``How to Apply These Terms to Your New
  Programs'' under the @gplvthree{}.


@node Strict Mode
@subsection ECMAScript Strict Mode
@dnindex Strict Mode, ECMAScript
ECMAScript@tie{}5's
  @url{https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Strict_mode,Strict Mode}
  throws errors in more situations that may lead to buggy code,
    allows for better optimization of ECMAScript code during runtime,
    and prohibits syntax that conflicts with future ECMAScript versions.
It also enables certain features,
  like using @code{let} inside blocks.

It should always be enabled going forward as shown in @ref{f:strict-mode}.
The statement should immediately follow the copyright header
  (@pxref{Copyright Header}),
    before any other code.

@float Figure, f:strict-mode
@example
// Copyright header

'use strict';

// ...
@end example
@caption{Enabling strict mode}
@end float



@node Libraries
@section Libraries Used
@dnindex Libraries
Liza does not use many libraries.
The primary reason for this was that few libraries useful to Liza
  existed during its initial development@mdash{
    }Node.js and its community was still very young.
With that said,
  care should be taken to ensure that libraries are added only after a
    careful analysis of its costs and benefits,
      as they add volatility to the whole system and may also
      introduce security vulnerabilities outside of our control.
They further introduce maintenance obligations for keeping up with
  newer versions of those libraries and addressing
  backwards-compatibility concerns.


@subsection System Libraries
@dnindex GNU ease.js
Liza was originally developed using JavaScript
  (first ECMAScript@tie{}3, and then ECMAScript@tie{}5).
JavaScript does not natively support the classical object-oriented
  model familiar to users of more traditional classical
  object-oriented languages like Java, C++, C#, and@tie{}PHP.
Liza is built using @easejs{},
  which provides those familiar features.
The primary language used by developers in the office that created
  Liza is@tie{}PHP,
    which motivated the creation of ease.js to ease the burden of
    entry.

Consequently,
  Liza is written in a classical object-oriented style rather than
  using prototypes.
The @code{class} keyword introduced in ECMAScript@tie{} is largely
  syntatic sugar around the prototype model and does not address the
  primary concerns of ease.js,
    nor does it provide traits.

@dnindex TypeScript
@emph{The project is now migrating toward TypeScript},
  so new code should not use ease.js unless required
    and an effort should be made to move existing code away from
      ease.js.
For more information on this migration,
  see @xref{TypeScript Migration}.


@subsection Testing Libraries
@dnindex Mocha
@dnindex Chai
@mocha{} is used as the test runner for JavaScript unit tests.
@chai{} is the assertion library.
This differs from PHP development where a single system (PHPUnit)
  encompasses both of these needs.

Chai offers a few different styles of assertions (``should'',
  ``expect'', and ``assert'');
      Liza uses @url{http://www.chaijs.com/guide/styles/#expect,``expect''}.

@devnotice{A library to aid in mocking TypeScript classes needs to be
             researched.}


@subsection UI Libraries
@dnindex jQuery
jQuery was used in the past,
  but has been largely purged from the system (and continues to be
  removed) due to strong performance issues.
Further,
  now that browser APIs have stabalized and Liza no longer needs to
    support as far back as Internet Explorer@tie{}6,
      the standard DOM APIs are more than sufficient.

Liza instead provides its own UI and DOM abstractions
  (@srcrefraw{src/ui}) that have been optimized for Liza's needs.

There are modern frameworks that may overlap with the type of UI
  operations that Liza performs,
    as well as certain DOM optimizations that it performs;
  however,
    it is unlikely that such frameworks (e.g. React, Angular, Meteor)
      will ever be integrated,
        as the cost of doing so exceeds the marginal benefit.


@node Developer Resources
@section Developer Resources
@dnindex Resources, Developer
@dnindex MDN
@url{https://developer.mozilla.org/en-US/docs/Web,MDN}@footnote{
  Formerly the ``Mozilla Developer Network''; see
    @url{https://blog.mozilla.org/opendesign/future-mdn-focus-web-docs/,''The Future of MDN: A Focus on Web Docs''}
    for the history of the rename.}
  is an essential resource for web development in general,
    especially for JavaScript/ECMAScript and the various Web APIs.
It contains resources for all levels,
  including for those @url{https://developer.mozilla.org/en-US/docs/Learn/JavaScript/First_steps,unfamiliar with JavaScript}.
All developers should familiarize themselves with the resources
  available on MDN so that they understand what type of information is
  readily accessible for future reference.

@dnindex TypeScript
An overview of TypeScript can be found in its
  @url{https://www.typescriptlang.org/docs/handbook/basic-types.html,Handbook}.
The language borrows concepts from a number of others,
  so many concepts may be familiar to you.
TypeScript uses structural typing (duck typing).
In Liza,
  we also choose to implement nominal typing using ``branding''
    (@srcrefraw{src/types/misc.d.ts}).
A @url{https://github.com/microsoft/TypeScript/blob/master/doc/spec.md,language specification}
  is also available.

@dnindex Node.js
The Server (@pxref{Server}) uses Node.js.
Although it's largely abstracted away,
  there may be times where you need to touch on it,
    in which case the
    @url{https://nodejs.org/en/docs/,Node.js documentation} will be helpful.
However,
  it is important to note the version of Node.js that Liza is currently using,
    as it may be woefully out of date and require looking at older
    versions of the documentation.

@dnindex Texinfo, GNU
@cindex Documentation
This manual is written using
  @url{https://www.gnu.org/software/texinfo/,Texinfo},
    which is the documentation format of the GNU@tie{}operating system.
The format is structured and well-suited for software documentation
  with output in a variety of formats.
Looking at the source code of this manual will be helpful@mdash{
  }it provides the general structure and numerous macros that are
    specific to Liza.

@dnindex MongoDB
@cindex Database
Data are persisted using @url{https://www.mongodb.com/,MongoDB}.
Database operations in Liza are abstracted away,
  but it's helpful to understand how to query the database directly to
    understand how the system works and composes its data,
  and for the purposes of debugging.

For information on specific libraries used by Liza,
  @pxref{Libraries}.


@node TypeScript Migration
@section TypeScript Migration
@dnindex TypeScript
@helpwanted{}

This section contains notes regarding a migration to TypeScript.
It is intended to serve as a guide@mdash{
  }it is not prescriptive.


@subsection Migrating Away From GNU ease.js
Liza was originally written in @easejs.
TypeScript now provides many features that ease.js was written to address,
  though not all (most notably traits).

Since ease.js was designed with JavaScript interoperability in mind,
  and TypeScript generates prototypes from classes,
  TypeScript classes serve as drop-in replacements under most
    circumstances.
However,
  subtypes must be migrated at the same time as their parents,
    otherwise type checking in TypeScript cannot properly be performed.
If this is a concern,
  @url{https://www.typescriptlang.org/docs/handbook/advanced-types.html#type-guards-and-type-assertions,type assertions}
  can potentially be used to coerce types during a transition period
    in conjunction with ease.js'
    @url{https://www.gnu.org/software/easejs/manual/easejs.html#Type-Checks-and-Polymorphism,@samp{Class.isA}}.

Interfaces do not exist at runtime in TypeScript,
  but they do in easejs.
Consequently,
  you can continue to export an ease.js interface while also exporting
    a TypeScript interface.
To do this,
  continue to export using @samp{module.exports} rather than
    TypeScript's @samp{export =}.

ease.js implements stackable Scala-like traits.
Traits are @emph{not} provided by TypeScript.
Traits will therefore have to be refactored into,
  for example,
  decorators or strategies.


@subsection Structural Typing
@dnindex Typing, Duck
@dnindex Typing, Structural
TypeScript implements
  @url{https://en.wikipedia.org/wiki/Structural_typing,structural typing},
    also called duck typing.
This means that any two types sharing the same ``shape'' are
  compatible with one-another.

For classes,
  this can be mitigated by defining private members,
    which then ensures that compatible types are indeed subtypes.

Interfaces can be used in either the traditional OOP sense,
  or as a means to define the shape of some arbitrary object.
Since interfaces do not define implementation details,
  the distinction isn't important@mdash{
    }it does not matter if we receive an instance of an object
      implementing an interface,
        or some object arbitrary that just happens to adhere to it.

In other instances where we want to distinguish between two values
  with otherwise compatible APIs,
    Nominal Typing below.


@subsection Nominal Typing
@dnindex Typing, Nominal
It is sometimes desirable to distinguish between two otherwise
  compatible types.
Consider, for example, a user@tie{}id and a Unix timestamp.
Both are of type @code{number},
  but it's desirable to ensure that one is not used where another is
    expected.

TypeScript doesn't directly support
  @url{https://en.wikipedia.org/wiki/Nominal_typing,nominal typing},
    where compatibility of data types are determined by name.
Liza uses a convention called ``branding'',
  abstracted behind a @code{NominalType} generic
    (defined in @srcrefraw{src/types/misc.d.ts}).

@float Figure, f:nom-type
@verbatim
type UnixTimestamp = NominalType<number, 'UnixTimestamp'>;
type Milliseconds  = NominalType<number, 'Milliseconds'>;

function timeElapsed( start: UnixTimestamp, end: UnixTimestamp ): Milliseconds
{
    return end - start;
}

const start = <UnixTimestamp>1571325570000;
const end   = <UnixTimestamp>1571514320000;

// this is okay
const elapsed = timeElapsed( start, end );

// this is not, since elapsed is of type Milliseconds
timeElapsed( start, elapsed );

@end verbatim
@caption{Example of nominal typing}
@end float

Consider the example in @ref{f:nom-type}.
Both @code{UnixTimestamp} and @code{Milliseconds} are a @code{number} type,
  but they have been defined in such a way that their names are part
    of the type definition.
Not only does the compiler prevent bugs caused from mixing data,
  but nominal types also help to make the code self-documenting.

If you want to have self-documenting types @emph{without} employing
  nominal typing,
    use type aliases.

There are no prescriptive rules for whether a type should be defined
  nominally.

In some cases,
  it's useful to use nominal types after having validated data,
    so that the compiler can enforce that assumption from that point forward.
This can be done using
  @url{https://www.typescriptlang.org/docs/handbook/advanced-types.html#type-guards-and-type-assertions,type assertions}.

@float Figure, f:nom-assert
@verbatim
type PositiveInteger = NominalType<number, 'PositiveInteger'>;

const isPositiveInteger = ( x: number ): n is PositiveInteger => n > 0;

const lookupIndex<T>( arr: T[], i: PositiveInteger ): T => arr[ i ];

// untrusted input from the user
const user_input = readSomeValue();

if ( isPositiveInteger( user_input ) )
{
    // user_input is now of type PositiveInteger
    return lookupIndex( data, user_input );
}
@end verbatim
@caption{Validating nominal types}
@end float

In @ref{f:nom-assert} above,
  we only assume something to be a @code{PositiveInteger} after having
    checked its value.
After that point,
  we can use TypeScript's type system to ensure at compile time that
    we are only using positive integers in certain contexts.

@devnotice{Never cast values
             (e.g. using @samp{<PositiveInteger>user_input})
             when type predicates are provided,
               since that undermines type safety.}