JSLint: Appendix C - JavaScript: The Good Parts

JavaScript's biggest problem is its dependence on global variables, particularly implied global variables. If a variable is not explicitly declared (usually with the var statement), then JavaScript assumes that the variable was global. This can mask misspelled names and other problems.

JSLint expects that all variables and functions will be declared before they are used or invoked. This allows it to detect implied global variables. It is also good practice because it makes programs easier to read.

Sometimes a file is dependent on global variables and functions that are defined elsewhere. You can identify these to JSLint by including a comment in your file that lists the global functions and objects that your program depends on, but that are not defined in your program or script file.

A global declaration comment can be used to list all of the names that you are intentionally using as global variables. JSLint can use this information to identify misspellings and forgotten var declarations. A global declaration can look like this:

/*global getElementByAttribute, breakCycles, hanoi */

A global declaration starts with /*global. Notice that there is no space before the g. You can have as many /*global comments as you like. They must appear before the use of the variables they specify.

Some globals can be predefined for you (see the later section "the section called “Options”"). Select the "Assume a browser" (browser) option to predefine the standard global properties that are supplied by web browsers, such as window and document and alert. Select the "Assume Rhino" (rhino) option to predefine the global properties provided by the Rhino environment. Select the "Assume a Yahoo Widget" (widget) option to predefine the global properties provided by the Yahoo! Widgets environment.

Since JavaScript is a loosely typed dynamic-object language, it is not possible to determine at compile time if property names are spelled correctly. JSLint provides some assistance with this.

At the bottom of its report, JSLint displays a /*members*/ comment. It contains all of the names and string literals that were used with dot notation, subscript notation, and object literals to name the members of objects. You can look through the list for misspellings. Member names that were used only once are shown in italics. This is to make misspellings easier to spot.

You can copy the /*members*/ comment into your script file. JSLint will check the spelling of all property names against the list. That way, you can have JSLint look for misspellings for you:

/*members doTell, iDoDeclare, mercySakes,
    myGoodness, ohGoOn, wellShutMyMouth */

The implementation of JSLint accepts an option object that allows you to determine the subset of JavaScript that is acceptable to you. It is also possible to set those options within the source of a script.

An option specification can look like this:

/*jslint nomen: true, evil: false */

An option specification starts with /*jslint. Notice that there is no space before the j. The specification contains a sequence of name/value pairs, where the names are JSLint options and the values are true or false. An option specification takes precedence over the option object. All of the options default to false. Table C.1, “JSLint options” lists the options available in using JSLint.

JavaScript uses a C-like syntax, which requires the use of semicolons to delimit statements. JavaScript attempts to make semicolons optional with a semicolon insertion mechanism. This is dangerous.

Like C, JavaScript has ++ and -- and ( operators, which can be prefixes or suffixes. The disambiguation is done by the semicolon.

In JavaScript, a linefeed can be whitespace, or it can act as a semicolon. This replaces one ambiguity with another.

JSLint expects that every statement be followed by ; except for for, function, if, switch, try, and while. JSLint does not expect to see unnecessary semicolons or the empty statement.

As a further defense against the masking of errors by the semicolon insertion mechanism, JSLint expects long statements to be broken only after one of these punctuation characters or operators:

, . ; : { } ( [ = < > ? ! + - * / % ˜ ^ | &
== != <= >= += -= *= /= %= ^= |= &= << >> || &&
=== !== <<= >>= >>> >>>=

JSLint does not expect to see a long statement broken after an identifier, a string, a number, a closer, or a suffix operator:

) ] ++ −−

JSLint allows you to turn on the "Tolerate sloppy line breaking" (laxbreak) option.

Semicolon insertion can mask copy/paste errors. If you always break lines after operators, then JSLint can do a better job of finding those errors.

The comma operator can lead to excessively tricky expressions. It can also mask some programming errors.

JSLint expects to see the comma used as a separator, but not as an operator (except in the initialization and incrementation parts of the for statement). It does not expect to see elided elements in array literals. Extra commas should not be used. A comma should not appear after the last element of an array literal or object literal because it can be misinterpreted by some browsers.

JSLint expects that if and for statements will be made with blocks—that is, with statements enclosed in braces ({}).

JavaScript allows an if to be written like this:

if (condition)
    statement;

That form is known to contribute to mistakes in projects where many programmers are working on the same code. That is why JSLint expects the use of a block:

if (condition) {
    statements;
}

Experience shows that this form is more resilient.

In many languages, a block introduces a scope. Variables introduced in a block are not visible outside of the block.

In JavaScript, blocks do not introduce a scope. There is only function-scope. A variable introduced anywhere in a function is visible everywhere in the function. JavaScript's blocks confuse experienced programmers and lead to errors because the familiar syntax makes a false promise.

JSLint expects blocks with function, if, switch, while, for, do, and try statements and nowhere else. An exception is made for an unblocked if statement on an else or for in.

An expression statement is expected to be an assignment, a function/method call, or delete. All other expression statements are considered errors.

The for in statement allows for looping through the names of all of the properties of an object. Unfortunately, it also loops through all of the members that were inherited through the prototype chain. This has the bad side effect of serving up method functions when the interest is in the data members.

The body of every for in statement should be wrapped in an if statement that does filtering. if can select for a particular type or range of values, it can exclude functions, or it can exclude properties from the prototype. For example:

for (name in object) {
    if (object.hasOwnProperty(name)) {
        ....
    }
}

A common error in switch statements is to forget to place a break statement after each case, resulting in unintended fall-through. JSLint expects that the statement before the next case or default is one of these: break, return, or throw.

JavaScript allows var definitions to occur anywhere within a function. JSLint is stricter.

JSLint expects that:

JSLint does not expect:

The with statement was intended to provide a shorthand in accessing members in deeply nested objects. Unfortunately, it behaves very badly when setting new members. Never use the with statement. Use a var instead.

JSLint does not expect to see a with statement.

JSLint does not expect to see an assignment statement in the condition part of an if or while statement. This is because it is more likely that:

if (a = b) {
    ...
}

was intended to be:

if (a == b) {
    ...
}

If you really intend an assignment, wrap it in another set of parentheses:

if ((a = b)) {
    ...
}

The == and != operators do type coercion before comparing. This is bad because it causes ' \f\r \n\t ' == 0 to be true. This can mask type errors.

When comparing to any of the following values, always use the === or !== operators, which do not do type coercion:

0 '' undefined null false true

If you want the type coercion, then use the short form. Instead of:

(foo != 0)

just say:

(foo)

And instead of:

(foo == 0)

say:

(!foo)

Use of the === and !== operators is always preferred. There is a "Disallow == and != " (eqeqeq) option, which requires the use of === and !== in all cases.

JavaScript allows any statement to have a label, and labels have a separate namespace. JSLint is stricter.

JSLint expects labels only on statements that interact with break: switch, while, do, and for. JSLint expects that labels will be distinct from variables and parameters.

JSLint expects that a return, break, continue, or throw statement will be followed by a } or case or default.

JSLint expects that + will not be followed by + or ++, and that - will not be followed by - or --. A misplaced space can turn + + into ++, an error that is difficult to see. Use parentheses to avoid confusion.

The ++ (increment) and -- (decrement) operators have been known to contribute to bad code by encouraging excessive trickiness. They are second only to faulty architecture in enabling viruses and other security menaces. The JSLint option plusplus prohibits the use of these operators.

JavaScript does not have an integer type, but it does have bitwise operators. The bitwise operators convert their operands from floating-point to integers and back, so they are not nearly as efficient as they are in C or other languages. They are rarely useful in browser applications. The similarity to the logical operators can mask some programming errors. The bitwise option prohibits the use of these operators.

The eval function and its relatives (Function, setTimeout, and setInterval) provide access to the JavaScript compiler. This is sometimes useful, but in most cases it indicates the presence of extremely bad coding. The eval function is the most misused feature of JavaScript.

In most C-like languages, void is a type. In JavaScript, void is a prefix operator that always returns undefined. JSLint does not expect to see void because it is confusing and not very useful.

Regular expressions are written in a terse and cryptic notation. JSLint looks for problems that may cause portability problems. It also attempts to resolve visual ambiguities by recommending explicit escapement.

JavaScript's syntax for regular expression literals overloads the / character. To avoid ambiguity, JSLint expects that the character preceding a regular expression literal is a ( or = or : or , character.

Constructors are functions that are designed to be used with the new prefix. The new prefix creates a new object based on the function's prototype, and binds that object to the function's implied this parameter. If you neglect to use the new prefix, no new object will be made, and this will be bound to the global object. This is a serious mistake.

JSLint enforces the convention that constructor functions be given names with initial uppercase letters. JSLint does not expect to see a function invocation with an initial uppercase name unless it has the new prefix. JSLint does not expect to see the new prefix used with functions whose names do not start with initial uppercase.

JSLint does not expect to see the wrapper forms new Number, new String, or new Boolean.

JSLint does not expect to see new Object (use {} instead).

JSLint does not expect to see new Array (use [] instead).

JSLint does not do flow analysis to determine that variables are assigned values before they are used. This is because variables are given a value (undefined) that is a reasonable default for many applications.

JSLint does not do any kind of global analysis. It does not attempt to determine that functions used with new are really constructors (except by enforcing capitalization conventions), or that method names are spelled correctly.

JSLint is able to handle HTML text. It can inspect the JavaScript content contained within <script>...</script> tags and event handlers. It also inspects the HTML content, looking for problems that are known to interfere with JavaScript:

JSLint is less anal than the sycophantic conformity demanded by XHTML, but more strict than the popular browsers.

JSLint also checks for the occurrence of </ in string literals. You should always write <\/ instead. The extra backslash is ignored by the JavaScript compiler, but not by the HTML parser. Tricks like this should not be necessary, and yet they are.

There is an option that allows use of uppercase tag names. There is also an option that allows the use of inline HTML event handlers.

JSLint can also check that JSON data structures are well formed. If the first character JSLint sees is { or [, then it strictly enforces the JSON rules. See Appendix E, JSON.

If JSLint is able to complete its scan, it generates a function report. It lists the following for each function:

The report will also include a list of all of the member names that were used.